1 ==============================
2 LLVM Language Reference Manual
3 ==============================
12 This document is a reference manual for the LLVM assembly language. LLVM
13 is a Static Single Assignment (SSA) based representation that provides
14 type safety, low-level operations, flexibility, and the capability of
15 representing 'all' high-level languages cleanly. It is the common code
16 representation used throughout all phases of the LLVM compilation
22 The LLVM code representation is designed to be used in three different
23 forms: as an in-memory compiler IR, as an on-disk bitcode representation
24 (suitable for fast loading by a Just-In-Time compiler), and as a human
25 readable assembly language representation. This allows LLVM to provide a
26 powerful intermediate representation for efficient compiler
27 transformations and analysis, while providing a natural means to debug
28 and visualize the transformations. The three different forms of LLVM are
29 all equivalent. This document describes the human readable
30 representation and notation.
32 The LLVM representation aims to be light-weight and low-level while
33 being expressive, typed, and extensible at the same time. It aims to be
34 a "universal IR" of sorts, by being at a low enough level that
35 high-level ideas may be cleanly mapped to it (similar to how
36 microprocessors are "universal IR's", allowing many source languages to
37 be mapped to them). By providing type information, LLVM can be used as
38 the target of optimizations: for example, through pointer analysis, it
39 can be proven that a C automatic variable is never accessed outside of
40 the current function, allowing it to be promoted to a simple SSA value
41 instead of a memory location.
48 It is important to note that this document describes 'well formed' LLVM
49 assembly language. There is a difference between what the parser accepts
50 and what is considered 'well formed'. For example, the following
51 instruction is syntactically okay, but not well formed:
57 because the definition of ``%x`` does not dominate all of its uses. The
58 LLVM infrastructure provides a verification pass that may be used to
59 verify that an LLVM module is well formed. This pass is automatically
60 run by the parser after parsing input assembly and by the optimizer
61 before it outputs bitcode. The violations pointed out by the verifier
62 pass indicate bugs in transformation passes or input to the parser.
69 LLVM identifiers come in two basic types: global and local. Global
70 identifiers (functions, global variables) begin with the ``'@'``
71 character. Local identifiers (register names, types) begin with the
72 ``'%'`` character. Additionally, there are three different formats for
73 identifiers, for different purposes:
75 #. Named values are represented as a string of characters with their
76 prefix. For example, ``%foo``, ``@DivisionByZero``,
77 ``%a.really.long.identifier``. The actual regular expression used is
78 '``[%@][a-zA-Z$._][a-zA-Z$._0-9]*``'. Identifiers which require other
79 characters in their names can be surrounded with quotes. Special
80 characters may be escaped using ``"\xx"`` where ``xx`` is the ASCII
81 code for the character in hexadecimal. In this way, any character can
82 be used in a name value, even quotes themselves.
83 #. Unnamed values are represented as an unsigned numeric value with
84 their prefix. For example, ``%12``, ``@2``, ``%44``.
85 #. Constants, which are described in the section Constants_ below.
87 LLVM requires that values start with a prefix for two reasons: Compilers
88 don't need to worry about name clashes with reserved words, and the set
89 of reserved words may be expanded in the future without penalty.
90 Additionally, unnamed identifiers allow a compiler to quickly come up
91 with a temporary variable without having to avoid symbol table
94 Reserved words in LLVM are very similar to reserved words in other
95 languages. There are keywords for different opcodes ('``add``',
96 '``bitcast``', '``ret``', etc...), for primitive type names ('``void``',
97 '``i32``', etc...), and others. These reserved words cannot conflict
98 with variable names, because none of them start with a prefix character
101 Here is an example of LLVM code to multiply the integer variable
108 %result = mul i32 %X, 8
110 After strength reduction:
114 %result = shl i32 %X, 3
120 %0 = add i32 %X, %X ; yields {i32}:%0
121 %1 = add i32 %0, %0 ; yields {i32}:%1
122 %result = add i32 %1, %1
124 This last way of multiplying ``%X`` by 8 illustrates several important
125 lexical features of LLVM:
127 #. Comments are delimited with a '``;``' and go until the end of line.
128 #. Unnamed temporaries are created when the result of a computation is
129 not assigned to a named value.
130 #. Unnamed temporaries are numbered sequentially (using a per-function
131 incrementing counter, starting with 0). Note that basic blocks are
132 included in this numbering. For example, if the entry basic block is not
133 given a label name, then it will get number 0.
135 It also shows a convention that we follow in this document. When
136 demonstrating instructions, we will follow an instruction with a comment
137 that defines the type and name of value produced.
145 LLVM programs are composed of ``Module``'s, each of which is a
146 translation unit of the input programs. Each module consists of
147 functions, global variables, and symbol table entries. Modules may be
148 combined together with the LLVM linker, which merges function (and
149 global variable) definitions, resolves forward declarations, and merges
150 symbol table entries. Here is an example of the "hello world" module:
154 ; Declare the string constant as a global constant.
155 @.str = private unnamed_addr constant [13 x i8] c"hello world\0A\00"
157 ; External declaration of the puts function
158 declare i32 @puts(i8* nocapture) nounwind
160 ; Definition of main function
161 define i32 @main() { ; i32()*
162 ; Convert [13 x i8]* to i8 *...
163 %cast210 = getelementptr [13 x i8]* @.str, i64 0, i64 0
165 ; Call puts function to write out the string to stdout.
166 call i32 @puts(i8* %cast210)
171 !1 = metadata !{i32 42}
174 This example is made up of a :ref:`global variable <globalvars>` named
175 "``.str``", an external declaration of the "``puts``" function, a
176 :ref:`function definition <functionstructure>` for "``main``" and
177 :ref:`named metadata <namedmetadatastructure>` "``foo``".
179 In general, a module is made up of a list of global values (where both
180 functions and global variables are global values). Global values are
181 represented by a pointer to a memory location (in this case, a pointer
182 to an array of char, and a pointer to a function), and have one of the
183 following :ref:`linkage types <linkage>`.
190 All Global Variables and Functions have one of the following types of
194 Global values with "``private``" linkage are only directly
195 accessible by objects in the current module. In particular, linking
196 code into a module with an private global value may cause the
197 private to be renamed as necessary to avoid collisions. Because the
198 symbol is private to the module, all references can be updated. This
199 doesn't show up in any symbol table in the object file.
201 Similar to private, but the value shows as a local symbol
202 (``STB_LOCAL`` in the case of ELF) in the object file. This
203 corresponds to the notion of the '``static``' keyword in C.
204 ``available_externally``
205 Globals with "``available_externally``" linkage are never emitted
206 into the object file corresponding to the LLVM module. They exist to
207 allow inlining and other optimizations to take place given knowledge
208 of the definition of the global, which is known to be somewhere
209 outside the module. Globals with ``available_externally`` linkage
210 are allowed to be discarded at will, and are otherwise the same as
211 ``linkonce_odr``. This linkage type is only allowed on definitions,
214 Globals with "``linkonce``" linkage are merged with other globals of
215 the same name when linkage occurs. This can be used to implement
216 some forms of inline functions, templates, or other code which must
217 be generated in each translation unit that uses it, but where the
218 body may be overridden with a more definitive definition later.
219 Unreferenced ``linkonce`` globals are allowed to be discarded. Note
220 that ``linkonce`` linkage does not actually allow the optimizer to
221 inline the body of this function into callers because it doesn't
222 know if this definition of the function is the definitive definition
223 within the program or whether it will be overridden by a stronger
224 definition. To enable inlining and other optimizations, use
225 "``linkonce_odr``" linkage.
227 "``weak``" linkage has the same merging semantics as ``linkonce``
228 linkage, except that unreferenced globals with ``weak`` linkage may
229 not be discarded. This is used for globals that are declared "weak"
232 "``common``" linkage is most similar to "``weak``" linkage, but they
233 are used for tentative definitions in C, such as "``int X;``" at
234 global scope. Symbols with "``common``" linkage are merged in the
235 same way as ``weak symbols``, and they may not be deleted if
236 unreferenced. ``common`` symbols may not have an explicit section,
237 must have a zero initializer, and may not be marked
238 ':ref:`constant <globalvars>`'. Functions and aliases may not have
241 .. _linkage_appending:
244 "``appending``" linkage may only be applied to global variables of
245 pointer to array type. When two global variables with appending
246 linkage are linked together, the two global arrays are appended
247 together. This is the LLVM, typesafe, equivalent of having the
248 system linker append together "sections" with identical names when
251 The semantics of this linkage follow the ELF object file model: the
252 symbol is weak until linked, if not linked, the symbol becomes null
253 instead of being an undefined reference.
254 ``linkonce_odr``, ``weak_odr``
255 Some languages allow differing globals to be merged, such as two
256 functions with different semantics. Other languages, such as
257 ``C++``, ensure that only equivalent globals are ever merged (the
258 "one definition rule" --- "ODR"). Such languages can use the
259 ``linkonce_odr`` and ``weak_odr`` linkage types to indicate that the
260 global will only be merged with equivalent globals. These linkage
261 types are otherwise the same as their non-``odr`` versions.
263 If none of the above identifiers are used, the global is externally
264 visible, meaning that it participates in linkage and can be used to
265 resolve external symbol references.
267 It is illegal for a function *declaration* to have any linkage type
268 other than ``external`` or ``extern_weak``.
275 LLVM :ref:`functions <functionstructure>`, :ref:`calls <i_call>` and
276 :ref:`invokes <i_invoke>` can all have an optional calling convention
277 specified for the call. The calling convention of any pair of dynamic
278 caller/callee must match, or the behavior of the program is undefined.
279 The following calling conventions are supported by LLVM, and more may be
282 "``ccc``" - The C calling convention
283 This calling convention (the default if no other calling convention
284 is specified) matches the target C calling conventions. This calling
285 convention supports varargs function calls and tolerates some
286 mismatch in the declared prototype and implemented declaration of
287 the function (as does normal C).
288 "``fastcc``" - The fast calling convention
289 This calling convention attempts to make calls as fast as possible
290 (e.g. by passing things in registers). This calling convention
291 allows the target to use whatever tricks it wants to produce fast
292 code for the target, without having to conform to an externally
293 specified ABI (Application Binary Interface). `Tail calls can only
294 be optimized when this, the GHC or the HiPE convention is
295 used. <CodeGenerator.html#id80>`_ This calling convention does not
296 support varargs and requires the prototype of all callees to exactly
297 match the prototype of the function definition.
298 "``coldcc``" - The cold calling convention
299 This calling convention attempts to make code in the caller as
300 efficient as possible under the assumption that the call is not
301 commonly executed. As such, these calls often preserve all registers
302 so that the call does not break any live ranges in the caller side.
303 This calling convention does not support varargs and requires the
304 prototype of all callees to exactly match the prototype of the
305 function definition. Furthermore the inliner doesn't consider such function
307 "``cc 10``" - GHC convention
308 This calling convention has been implemented specifically for use by
309 the `Glasgow Haskell Compiler (GHC) <http://www.haskell.org/ghc>`_.
310 It passes everything in registers, going to extremes to achieve this
311 by disabling callee save registers. This calling convention should
312 not be used lightly but only for specific situations such as an
313 alternative to the *register pinning* performance technique often
314 used when implementing functional programming languages. At the
315 moment only X86 supports this convention and it has the following
318 - On *X86-32* only supports up to 4 bit type parameters. No
319 floating point types are supported.
320 - On *X86-64* only supports up to 10 bit type parameters and 6
321 floating point parameters.
323 This calling convention supports `tail call
324 optimization <CodeGenerator.html#id80>`_ but requires both the
325 caller and callee are using it.
326 "``cc 11``" - The HiPE calling convention
327 This calling convention has been implemented specifically for use by
328 the `High-Performance Erlang
329 (HiPE) <http://www.it.uu.se/research/group/hipe/>`_ compiler, *the*
330 native code compiler of the `Ericsson's Open Source Erlang/OTP
331 system <http://www.erlang.org/download.shtml>`_. It uses more
332 registers for argument passing than the ordinary C calling
333 convention and defines no callee-saved registers. The calling
334 convention properly supports `tail call
335 optimization <CodeGenerator.html#id80>`_ but requires that both the
336 caller and the callee use it. It uses a *register pinning*
337 mechanism, similar to GHC's convention, for keeping frequently
338 accessed runtime components pinned to specific hardware registers.
339 At the moment only X86 supports this convention (both 32 and 64
341 "``webkit_jscc``" - WebKit's JavaScript calling convention
342 This calling convention has been implemented for `WebKit FTL JIT
343 <https://trac.webkit.org/wiki/FTLJIT>`_. It passes arguments on the
344 stack right to left (as cdecl does), and returns a value in the
345 platform's customary return register.
346 "``anyregcc``" - Dynamic calling convention for code patching
347 This is a special convention that supports patching an arbitrary code
348 sequence in place of a call site. This convention forces the call
349 arguments into registers but allows them to be dynamcially
350 allocated. This can currently only be used with calls to
351 llvm.experimental.patchpoint because only this intrinsic records
352 the location of its arguments in a side table. See :doc:`StackMaps`.
353 "``preserve_mostcc``" - The `PreserveMost` calling convention
354 This calling convention attempts to make the code in the caller as little
355 intrusive as possible. This calling convention behaves identical to the `C`
356 calling convention on how arguments and return values are passed, but it
357 uses a different set of caller/callee-saved registers. This alleviates the
358 burden of saving and recovering a large register set before and after the
359 call in the caller. If the arguments are passed in callee-saved registers,
360 then they will be preserved by the callee across the call. This doesn't
361 apply for values returned in callee-saved registers.
363 - On X86-64 the callee preserves all general purpose registers, except for
364 R11. R11 can be used as a scratch register. Floating-point registers
365 (XMMs/YMMs) are not preserved and need to be saved by the caller.
367 The idea behind this convention is to support calls to runtime functions
368 that have a hot path and a cold path. The hot path is usually a small piece
369 of code that doesn't many registers. The cold path might need to call out to
370 another function and therefore only needs to preserve the caller-saved
371 registers, which haven't already been saved by the caller. The
372 `PreserveMost` calling convention is very similar to the `cold` calling
373 convention in terms of caller/callee-saved registers, but they are used for
374 different types of function calls. `coldcc` is for function calls that are
375 rarely executed, whereas `preserve_mostcc` function calls are intended to be
376 on the hot path and definitely executed a lot. Furthermore `preserve_mostcc`
377 doesn't prevent the inliner from inlining the function call.
379 This calling convention will be used by a future version of the ObjectiveC
380 runtime and should therefore still be considered experimental at this time.
381 Although this convention was created to optimize certain runtime calls to
382 the ObjectiveC runtime, it is not limited to this runtime and might be used
383 by other runtimes in the future too. The current implementation only
384 supports X86-64, but the intention is to support more architectures in the
386 "``preserve_allcc``" - The `PreserveAll` calling convention
387 This calling convention attempts to make the code in the caller even less
388 intrusive than the `PreserveMost` calling convention. This calling
389 convention also behaves identical to the `C` calling convention on how
390 arguments and return values are passed, but it uses a different set of
391 caller/callee-saved registers. This removes the burden of saving and
392 recovering a large register set before and after the call in the caller. If
393 the arguments are passed in callee-saved registers, then they will be
394 preserved by the callee across the call. This doesn't apply for values
395 returned in callee-saved registers.
397 - On X86-64 the callee preserves all general purpose registers, except for
398 R11. R11 can be used as a scratch register. Furthermore it also preserves
399 all floating-point registers (XMMs/YMMs).
401 The idea behind this convention is to support calls to runtime functions
402 that don't need to call out to any other functions.
404 This calling convention, like the `PreserveMost` calling convention, will be
405 used by a future version of the ObjectiveC runtime and should be considered
406 experimental at this time.
407 "``cc <n>``" - Numbered convention
408 Any calling convention may be specified by number, allowing
409 target-specific calling conventions to be used. Target specific
410 calling conventions start at 64.
412 More calling conventions can be added/defined on an as-needed basis, to
413 support Pascal conventions or any other well-known target-independent
416 .. _visibilitystyles:
421 All Global Variables and Functions have one of the following visibility
424 "``default``" - Default style
425 On targets that use the ELF object file format, default visibility
426 means that the declaration is visible to other modules and, in
427 shared libraries, means that the declared entity may be overridden.
428 On Darwin, default visibility means that the declaration is visible
429 to other modules. Default visibility corresponds to "external
430 linkage" in the language.
431 "``hidden``" - Hidden style
432 Two declarations of an object with hidden visibility refer to the
433 same object if they are in the same shared object. Usually, hidden
434 visibility indicates that the symbol will not be placed into the
435 dynamic symbol table, so no other module (executable or shared
436 library) can reference it directly.
437 "``protected``" - Protected style
438 On ELF, protected visibility indicates that the symbol will be
439 placed in the dynamic symbol table, but that references within the
440 defining module will bind to the local symbol. That is, the symbol
441 cannot be overridden by another module.
443 A symbol with ``internal`` or ``private`` linkage must have ``default``
451 All Global Variables, Functions and Aliases can have one of the following
455 "``dllimport``" causes the compiler to reference a function or variable via
456 a global pointer to a pointer that is set up by the DLL exporting the
457 symbol. On Microsoft Windows targets, the pointer name is formed by
458 combining ``__imp_`` and the function or variable name.
460 "``dllexport``" causes the compiler to provide a global pointer to a pointer
461 in a DLL, so that it can be referenced with the ``dllimport`` attribute. On
462 Microsoft Windows targets, the pointer name is formed by combining
463 ``__imp_`` and the function or variable name. Since this storage class
464 exists for defining a dll interface, the compiler, assembler and linker know
465 it is externally referenced and must refrain from deleting the symbol.
470 LLVM IR allows you to specify both "identified" and "literal" :ref:`structure
471 types <t_struct>`. Literal types are uniqued structurally, but identified types
472 are never uniqued. An :ref:`opaque structural type <t_opaque>` can also be used
473 to forward declare a type which is not yet available.
475 An example of a identified structure specification is:
479 %mytype = type { %mytype*, i32 }
481 Prior to the LLVM 3.0 release, identified types were structurally uniqued. Only
482 literal types are uniqued in recent versions of LLVM.
489 Global variables define regions of memory allocated at compilation time
492 Global variables definitions must be initialized, may have an explicit section
493 to be placed in, and may have an optional explicit alignment specified.
495 Global variables in other translation units can also be declared, in which
496 case they don't have an initializer.
498 A variable may be defined as ``thread_local``, which means that it will
499 not be shared by threads (each thread will have a separated copy of the
500 variable). Not all targets support thread-local variables. Optionally, a
501 TLS model may be specified:
504 For variables that are only used within the current shared library.
506 For variables in modules that will not be loaded dynamically.
508 For variables defined in the executable and only used within it.
510 The models correspond to the ELF TLS models; see `ELF Handling For
511 Thread-Local Storage <http://people.redhat.com/drepper/tls.pdf>`_ for
512 more information on under which circumstances the different models may
513 be used. The target may choose a different TLS model if the specified
514 model is not supported, or if a better choice of model can be made.
516 A variable may be defined as a global ``constant``, which indicates that
517 the contents of the variable will **never** be modified (enabling better
518 optimization, allowing the global data to be placed in the read-only
519 section of an executable, etc). Note that variables that need runtime
520 initialization cannot be marked ``constant`` as there is a store to the
523 LLVM explicitly allows *declarations* of global variables to be marked
524 constant, even if the final definition of the global is not. This
525 capability can be used to enable slightly better optimization of the
526 program, but requires the language definition to guarantee that
527 optimizations based on the 'constantness' are valid for the translation
528 units that do not include the definition.
530 As SSA values, global variables define pointer values that are in scope
531 (i.e. they dominate) all basic blocks in the program. Global variables
532 always define a pointer to their "content" type because they describe a
533 region of memory, and all memory objects in LLVM are accessed through
536 Global variables can be marked with ``unnamed_addr`` which indicates
537 that the address is not significant, only the content. Constants marked
538 like this can be merged with other constants if they have the same
539 initializer. Note that a constant with significant address *can* be
540 merged with a ``unnamed_addr`` constant, the result being a constant
541 whose address is significant.
543 A global variable may be declared to reside in a target-specific
544 numbered address space. For targets that support them, address spaces
545 may affect how optimizations are performed and/or what target
546 instructions are used to access the variable. The default address space
547 is zero. The address space qualifier must precede any other attributes.
549 LLVM allows an explicit section to be specified for globals. If the
550 target supports it, it will emit globals to the section specified.
552 By default, global initializers are optimized by assuming that global
553 variables defined within the module are not modified from their
554 initial values before the start of the global initializer. This is
555 true even for variables potentially accessible from outside the
556 module, including those with external linkage or appearing in
557 ``@llvm.used`` or dllexported variables. This assumption may be suppressed
558 by marking the variable with ``externally_initialized``.
560 An explicit alignment may be specified for a global, which must be a
561 power of 2. If not present, or if the alignment is set to zero, the
562 alignment of the global is set by the target to whatever it feels
563 convenient. If an explicit alignment is specified, the global is forced
564 to have exactly that alignment. Targets and optimizers are not allowed
565 to over-align the global if the global has an assigned section. In this
566 case, the extra alignment could be observable: for example, code could
567 assume that the globals are densely packed in their section and try to
568 iterate over them as an array, alignment padding would break this
571 Globals can also have a :ref:`DLL storage class <dllstorageclass>`.
575 [@<GlobalVarName> =] [Linkage] [Visibility] [DLLStorageClass] [ThreadLocal]
576 [AddrSpace] [unnamed_addr] [ExternallyInitialized]
577 <global | constant> <Type>
578 [, section "name"] [, align <Alignment>]
580 For example, the following defines a global in a numbered address space
581 with an initializer, section, and alignment:
585 @G = addrspace(5) constant float 1.0, section "foo", align 4
587 The following example just declares a global variable
591 @G = external global i32
593 The following example defines a thread-local global with the
594 ``initialexec`` TLS model:
598 @G = thread_local(initialexec) global i32 0, align 4
600 .. _functionstructure:
605 LLVM function definitions consist of the "``define``" keyword, an
606 optional :ref:`linkage type <linkage>`, an optional :ref:`visibility
607 style <visibility>`, an optional :ref:`DLL storage class <dllstorageclass>`,
608 an optional :ref:`calling convention <callingconv>`,
609 an optional ``unnamed_addr`` attribute, a return type, an optional
610 :ref:`parameter attribute <paramattrs>` for the return type, a function
611 name, a (possibly empty) argument list (each with optional :ref:`parameter
612 attributes <paramattrs>`), optional :ref:`function attributes <fnattrs>`,
613 an optional section, an optional alignment, an optional :ref:`garbage
614 collector name <gc>`, an optional :ref:`prefix <prefixdata>`, an opening
615 curly brace, a list of basic blocks, and a closing curly brace.
617 LLVM function declarations consist of the "``declare``" keyword, an
618 optional :ref:`linkage type <linkage>`, an optional :ref:`visibility
619 style <visibility>`, an optional :ref:`DLL storage class <dllstorageclass>`,
620 an optional :ref:`calling convention <callingconv>`,
621 an optional ``unnamed_addr`` attribute, a return type, an optional
622 :ref:`parameter attribute <paramattrs>` for the return type, a function
623 name, a possibly empty list of arguments, an optional alignment, an optional
624 :ref:`garbage collector name <gc>` and an optional :ref:`prefix <prefixdata>`.
626 A function definition contains a list of basic blocks, forming the CFG (Control
627 Flow Graph) for the function. Each basic block may optionally start with a label
628 (giving the basic block a symbol table entry), contains a list of instructions,
629 and ends with a :ref:`terminator <terminators>` instruction (such as a branch or
630 function return). If an explicit label is not provided, a block is assigned an
631 implicit numbered label, using the next value from the same counter as used for
632 unnamed temporaries (:ref:`see above<identifiers>`). For example, if a function
633 entry block does not have an explicit label, it will be assigned label "%0",
634 then the first unnamed temporary in that block will be "%1", etc.
636 The first basic block in a function is special in two ways: it is
637 immediately executed on entrance to the function, and it is not allowed
638 to have predecessor basic blocks (i.e. there can not be any branches to
639 the entry block of a function). Because the block can have no
640 predecessors, it also cannot have any :ref:`PHI nodes <i_phi>`.
642 LLVM allows an explicit section to be specified for functions. If the
643 target supports it, it will emit functions to the section specified.
645 An explicit alignment may be specified for a function. If not present,
646 or if the alignment is set to zero, the alignment of the function is set
647 by the target to whatever it feels convenient. If an explicit alignment
648 is specified, the function is forced to have at least that much
649 alignment. All alignments must be a power of 2.
651 If the ``unnamed_addr`` attribute is given, the address is know to not
652 be significant and two identical functions can be merged.
656 define [linkage] [visibility] [DLLStorageClass]
658 <ResultType> @<FunctionName> ([argument list])
659 [unnamed_addr] [fn Attrs] [section "name"] [align N]
660 [gc] [prefix Constant] { ... }
667 Aliases act as "second name" for the aliasee value (which can be either
668 function, global variable, another alias or bitcast of global value).
669 Aliases may have an optional :ref:`linkage type <linkage>`, an optional
670 :ref:`visibility style <visibility>`, and an optional :ref:`DLL storage class
675 @<Name> = [Visibility] [DLLStorageClass] alias [Linkage] <AliaseeTy> @<Aliasee>
677 The linkage must be one of ``private``, ``internal``, ``linkonce``, ``weak``,
678 ``linkonce_odr``, ``weak_odr``, ``external``. Note that some system linkers
679 might not correctly handle dropping a weak symbol that is aliased by a non-weak
682 Alias that are not ``unnamed_addr`` are guaranteed to have the same address as
685 The aliasee must be a definition.
687 Aliases are not allowed to point to aliases with linkages that can be
688 overridden. Since they are only a second name, the possibility of the
689 intermediate alias being overridden cannot be represented in an object file.
691 .. _namedmetadatastructure:
696 Named metadata is a collection of metadata. :ref:`Metadata
697 nodes <metadata>` (but not metadata strings) are the only valid
698 operands for a named metadata.
702 ; Some unnamed metadata nodes, which are referenced by the named metadata.
703 !0 = metadata !{metadata !"zero"}
704 !1 = metadata !{metadata !"one"}
705 !2 = metadata !{metadata !"two"}
707 !name = !{!0, !1, !2}
714 The return type and each parameter of a function type may have a set of
715 *parameter attributes* associated with them. Parameter attributes are
716 used to communicate additional information about the result or
717 parameters of a function. Parameter attributes are considered to be part
718 of the function, not of the function type, so functions with different
719 parameter attributes can have the same function type.
721 Parameter attributes are simple keywords that follow the type specified.
722 If multiple parameter attributes are needed, they are space separated.
727 declare i32 @printf(i8* noalias nocapture, ...)
728 declare i32 @atoi(i8 zeroext)
729 declare signext i8 @returns_signed_char()
731 Note that any attributes for the function result (``nounwind``,
732 ``readonly``) come immediately after the argument list.
734 Currently, only the following parameter attributes are defined:
737 This indicates to the code generator that the parameter or return
738 value should be zero-extended to the extent required by the target's
739 ABI (which is usually 32-bits, but is 8-bits for a i1 on x86-64) by
740 the caller (for a parameter) or the callee (for a return value).
742 This indicates to the code generator that the parameter or return
743 value should be sign-extended to the extent required by the target's
744 ABI (which is usually 32-bits) by the caller (for a parameter) or
745 the callee (for a return value).
747 This indicates that this parameter or return value should be treated
748 in a special target-dependent fashion during while emitting code for
749 a function call or return (usually, by putting it in a register as
750 opposed to memory, though some targets use it to distinguish between
751 two different kinds of registers). Use of this attribute is
754 This indicates that the pointer parameter should really be passed by
755 value to the function. The attribute implies that a hidden copy of
756 the pointee is made between the caller and the callee, so the callee
757 is unable to modify the value in the caller. This attribute is only
758 valid on LLVM pointer arguments. It is generally used to pass
759 structs and arrays by value, but is also valid on pointers to
760 scalars. The copy is considered to belong to the caller not the
761 callee (for example, ``readonly`` functions should not write to
762 ``byval`` parameters). This is not a valid attribute for return
765 The byval attribute also supports specifying an alignment with the
766 align attribute. It indicates the alignment of the stack slot to
767 form and the known alignment of the pointer specified to the call
768 site. If the alignment is not specified, then the code generator
769 makes a target-specific assumption.
775 The ``inalloca`` argument attribute allows the caller to take the
776 address of outgoing stack arguments. An ``inalloca`` argument must
777 be a pointer to stack memory produced by an ``alloca`` instruction.
778 The alloca, or argument allocation, must also be tagged with the
779 inalloca keyword. Only the past argument may have the ``inalloca``
780 attribute, and that argument is guaranteed to be passed in memory.
782 An argument allocation may be used by a call at most once because
783 the call may deallocate it. The ``inalloca`` attribute cannot be
784 used in conjunction with other attributes that affect argument
785 storage, like ``inreg``, ``nest``, ``sret``, or ``byval``. The
786 ``inalloca`` attribute also disables LLVM's implicit lowering of
787 large aggregate return values, which means that frontend authors
788 must lower them with ``sret`` pointers.
790 When the call site is reached, the argument allocation must have
791 been the most recent stack allocation that is still live, or the
792 results are undefined. It is possible to allocate additional stack
793 space after an argument allocation and before its call site, but it
794 must be cleared off with :ref:`llvm.stackrestore
797 See :doc:`InAlloca` for more information on how to use this
801 This indicates that the pointer parameter specifies the address of a
802 structure that is the return value of the function in the source
803 program. This pointer must be guaranteed by the caller to be valid:
804 loads and stores to the structure may be assumed by the callee
805 not to trap and to be properly aligned. This may only be applied to
806 the first parameter. This is not a valid attribute for return
812 This indicates that pointer values :ref:`based <pointeraliasing>` on
813 the argument or return value do not alias pointer values which are
814 not *based* on it, ignoring certain "irrelevant" dependencies. For a
815 call to the parent function, dependencies between memory references
816 from before or after the call and from those during the call are
817 "irrelevant" to the ``noalias`` keyword for the arguments and return
818 value used in that call. The caller shares the responsibility with
819 the callee for ensuring that these requirements are met. For further
820 details, please see the discussion of the NoAlias response in :ref:`alias
821 analysis <Must, May, or No>`.
823 Note that this definition of ``noalias`` is intentionally similar
824 to the definition of ``restrict`` in C99 for function arguments,
825 though it is slightly weaker.
827 For function return values, C99's ``restrict`` is not meaningful,
828 while LLVM's ``noalias`` is.
830 This indicates that the callee does not make any copies of the
831 pointer that outlive the callee itself. This is not a valid
832 attribute for return values.
837 This indicates that the pointer parameter can be excised using the
838 :ref:`trampoline intrinsics <int_trampoline>`. This is not a valid
839 attribute for return values and can only be applied to one parameter.
842 This indicates that the function always returns the argument as its return
843 value. This is an optimization hint to the code generator when generating
844 the caller, allowing tail call optimization and omission of register saves
845 and restores in some cases; it is not checked or enforced when generating
846 the callee. The parameter and the function return type must be valid
847 operands for the :ref:`bitcast instruction <i_bitcast>`. This is not a
848 valid attribute for return values and can only be applied to one parameter.
852 Garbage Collector Names
853 -----------------------
855 Each function may specify a garbage collector name, which is simply a
860 define void @f() gc "name" { ... }
862 The compiler declares the supported values of *name*. Specifying a
863 collector which will cause the compiler to alter its output in order to
864 support the named garbage collection algorithm.
871 Prefix data is data associated with a function which the code generator
872 will emit immediately before the function body. The purpose of this feature
873 is to allow frontends to associate language-specific runtime metadata with
874 specific functions and make it available through the function pointer while
875 still allowing the function pointer to be called. To access the data for a
876 given function, a program may bitcast the function pointer to a pointer to
877 the constant's type. This implies that the IR symbol points to the start
880 To maintain the semantics of ordinary function calls, the prefix data must
881 have a particular format. Specifically, it must begin with a sequence of
882 bytes which decode to a sequence of machine instructions, valid for the
883 module's target, which transfer control to the point immediately succeeding
884 the prefix data, without performing any other visible action. This allows
885 the inliner and other passes to reason about the semantics of the function
886 definition without needing to reason about the prefix data. Obviously this
887 makes the format of the prefix data highly target dependent.
889 Prefix data is laid out as if it were an initializer for a global variable
890 of the prefix data's type. No padding is automatically placed between the
891 prefix data and the function body. If padding is required, it must be part
894 A trivial example of valid prefix data for the x86 architecture is ``i8 144``,
895 which encodes the ``nop`` instruction:
899 define void @f() prefix i8 144 { ... }
901 Generally prefix data can be formed by encoding a relative branch instruction
902 which skips the metadata, as in this example of valid prefix data for the
903 x86_64 architecture, where the first two bytes encode ``jmp .+10``:
907 %0 = type <{ i8, i8, i8* }>
909 define void @f() prefix %0 <{ i8 235, i8 8, i8* @md}> { ... }
911 A function may have prefix data but no body. This has similar semantics
912 to the ``available_externally`` linkage in that the data may be used by the
913 optimizers but will not be emitted in the object file.
920 Attribute groups are groups of attributes that are referenced by objects within
921 the IR. They are important for keeping ``.ll`` files readable, because a lot of
922 functions will use the same set of attributes. In the degenerative case of a
923 ``.ll`` file that corresponds to a single ``.c`` file, the single attribute
924 group will capture the important command line flags used to build that file.
926 An attribute group is a module-level object. To use an attribute group, an
927 object references the attribute group's ID (e.g. ``#37``). An object may refer
928 to more than one attribute group. In that situation, the attributes from the
929 different groups are merged.
931 Here is an example of attribute groups for a function that should always be
932 inlined, has a stack alignment of 4, and which shouldn't use SSE instructions:
936 ; Target-independent attributes:
937 attributes #0 = { alwaysinline alignstack=4 }
939 ; Target-dependent attributes:
940 attributes #1 = { "no-sse" }
942 ; Function @f has attributes: alwaysinline, alignstack=4, and "no-sse".
943 define void @f() #0 #1 { ... }
950 Function attributes are set to communicate additional information about
951 a function. Function attributes are considered to be part of the
952 function, not of the function type, so functions with different function
953 attributes can have the same function type.
955 Function attributes are simple keywords that follow the type specified.
956 If multiple attributes are needed, they are space separated. For
961 define void @f() noinline { ... }
962 define void @f() alwaysinline { ... }
963 define void @f() alwaysinline optsize { ... }
964 define void @f() optsize { ... }
967 This attribute indicates that, when emitting the prologue and
968 epilogue, the backend should forcibly align the stack pointer.
969 Specify the desired alignment, which must be a power of two, in
972 This attribute indicates that the inliner should attempt to inline
973 this function into callers whenever possible, ignoring any active
974 inlining size threshold for this caller.
976 This indicates that the callee function at a call site should be
977 recognized as a built-in function, even though the function's declaration
978 uses the ``nobuiltin`` attribute. This is only valid at call sites for
979 direct calls to functions which are declared with the ``nobuiltin``
982 This attribute indicates that this function is rarely called. When
983 computing edge weights, basic blocks post-dominated by a cold
984 function call are also considered to be cold; and, thus, given low
987 This attribute indicates that the source code contained a hint that
988 inlining this function is desirable (such as the "inline" keyword in
989 C/C++). It is just a hint; it imposes no requirements on the
992 This attribute suggests that optimization passes and code generator
993 passes make choices that keep the code size of this function as small
994 as possible and perform optimizations that may sacrifice runtime
995 performance in order to minimize the size of the generated code.
997 This attribute disables prologue / epilogue emission for the
998 function. This can have very system-specific consequences.
1000 This indicates that the callee function at a call site is not recognized as
1001 a built-in function. LLVM will retain the original call and not replace it
1002 with equivalent code based on the semantics of the built-in function, unless
1003 the call site uses the ``builtin`` attribute. This is valid at call sites
1004 and on function declarations and definitions.
1006 This attribute indicates that calls to the function cannot be
1007 duplicated. A call to a ``noduplicate`` function may be moved
1008 within its parent function, but may not be duplicated within
1009 its parent function.
1011 A function containing a ``noduplicate`` call may still
1012 be an inlining candidate, provided that the call is not
1013 duplicated by inlining. That implies that the function has
1014 internal linkage and only has one call site, so the original
1015 call is dead after inlining.
1017 This attributes disables implicit floating point instructions.
1019 This attribute indicates that the inliner should never inline this
1020 function in any situation. This attribute may not be used together
1021 with the ``alwaysinline`` attribute.
1023 This attribute suppresses lazy symbol binding for the function. This
1024 may make calls to the function faster, at the cost of extra program
1025 startup time if the function is not called during program startup.
1027 This attribute indicates that the code generator should not use a
1028 red zone, even if the target-specific ABI normally permits it.
1030 This function attribute indicates that the function never returns
1031 normally. This produces undefined behavior at runtime if the
1032 function ever does dynamically return.
1034 This function attribute indicates that the function never returns
1035 with an unwind or exceptional control flow. If the function does
1036 unwind, its runtime behavior is undefined.
1038 This function attribute indicates that the function is not optimized
1039 by any optimization or code generator passes with the
1040 exception of interprocedural optimization passes.
1041 This attribute cannot be used together with the ``alwaysinline``
1042 attribute; this attribute is also incompatible
1043 with the ``minsize`` attribute and the ``optsize`` attribute.
1045 This attribute requires the ``noinline`` attribute to be specified on
1046 the function as well, so the function is never inlined into any caller.
1047 Only functions with the ``alwaysinline`` attribute are valid
1048 candidates for inlining into the body of this function.
1050 This attribute suggests that optimization passes and code generator
1051 passes make choices that keep the code size of this function low,
1052 and otherwise do optimizations specifically to reduce code size as
1053 long as they do not significantly impact runtime performance.
1055 On a function, this attribute indicates that the function computes its
1056 result (or decides to unwind an exception) based strictly on its arguments,
1057 without dereferencing any pointer arguments or otherwise accessing
1058 any mutable state (e.g. memory, control registers, etc) visible to
1059 caller functions. It does not write through any pointer arguments
1060 (including ``byval`` arguments) and never changes any state visible
1061 to callers. This means that it cannot unwind exceptions by calling
1062 the ``C++`` exception throwing methods.
1064 On an argument, this attribute indicates that the function does not
1065 dereference that pointer argument, even though it may read or write the
1066 memory that the pointer points to if accessed through other pointers.
1068 On a function, this attribute indicates that the function does not write
1069 through any pointer arguments (including ``byval`` arguments) or otherwise
1070 modify any state (e.g. memory, control registers, etc) visible to
1071 caller functions. It may dereference pointer arguments and read
1072 state that may be set in the caller. A readonly function always
1073 returns the same value (or unwinds an exception identically) when
1074 called with the same set of arguments and global state. It cannot
1075 unwind an exception by calling the ``C++`` exception throwing
1078 On an argument, this attribute indicates that the function does not write
1079 through this pointer argument, even though it may write to the memory that
1080 the pointer points to.
1082 This attribute indicates that this function can return twice. The C
1083 ``setjmp`` is an example of such a function. The compiler disables
1084 some optimizations (like tail calls) in the caller of these
1086 ``sanitize_address``
1087 This attribute indicates that AddressSanitizer checks
1088 (dynamic address safety analysis) are enabled for this function.
1090 This attribute indicates that MemorySanitizer checks (dynamic detection
1091 of accesses to uninitialized memory) are enabled for this function.
1093 This attribute indicates that ThreadSanitizer checks
1094 (dynamic thread safety analysis) are enabled for this function.
1096 This attribute indicates that the function should emit a stack
1097 smashing protector. It is in the form of a "canary" --- a random value
1098 placed on the stack before the local variables that's checked upon
1099 return from the function to see if it has been overwritten. A
1100 heuristic is used to determine if a function needs stack protectors
1101 or not. The heuristic used will enable protectors for functions with:
1103 - Character arrays larger than ``ssp-buffer-size`` (default 8).
1104 - Aggregates containing character arrays larger than ``ssp-buffer-size``.
1105 - Calls to alloca() with variable sizes or constant sizes greater than
1106 ``ssp-buffer-size``.
1108 Variables that are identified as requiring a protector will be arranged
1109 on the stack such that they are adjacent to the stack protector guard.
1111 If a function that has an ``ssp`` attribute is inlined into a
1112 function that doesn't have an ``ssp`` attribute, then the resulting
1113 function will have an ``ssp`` attribute.
1115 This attribute indicates that the function should *always* emit a
1116 stack smashing protector. This overrides the ``ssp`` function
1119 Variables that are identified as requiring a protector will be arranged
1120 on the stack such that they are adjacent to the stack protector guard.
1121 The specific layout rules are:
1123 #. Large arrays and structures containing large arrays
1124 (``>= ssp-buffer-size``) are closest to the stack protector.
1125 #. Small arrays and structures containing small arrays
1126 (``< ssp-buffer-size``) are 2nd closest to the protector.
1127 #. Variables that have had their address taken are 3rd closest to the
1130 If a function that has an ``sspreq`` attribute is inlined into a
1131 function that doesn't have an ``sspreq`` attribute or which has an
1132 ``ssp`` or ``sspstrong`` attribute, then the resulting function will have
1133 an ``sspreq`` attribute.
1135 This attribute indicates that the function should emit a stack smashing
1136 protector. This attribute causes a strong heuristic to be used when
1137 determining if a function needs stack protectors. The strong heuristic
1138 will enable protectors for functions with:
1140 - Arrays of any size and type
1141 - Aggregates containing an array of any size and type.
1142 - Calls to alloca().
1143 - Local variables that have had their address taken.
1145 Variables that are identified as requiring a protector will be arranged
1146 on the stack such that they are adjacent to the stack protector guard.
1147 The specific layout rules are:
1149 #. Large arrays and structures containing large arrays
1150 (``>= ssp-buffer-size``) are closest to the stack protector.
1151 #. Small arrays and structures containing small arrays
1152 (``< ssp-buffer-size``) are 2nd closest to the protector.
1153 #. Variables that have had their address taken are 3rd closest to the
1156 This overrides the ``ssp`` function attribute.
1158 If a function that has an ``sspstrong`` attribute is inlined into a
1159 function that doesn't have an ``sspstrong`` attribute, then the
1160 resulting function will have an ``sspstrong`` attribute.
1162 This attribute indicates that the ABI being targeted requires that
1163 an unwind table entry be produce for this function even if we can
1164 show that no exceptions passes by it. This is normally the case for
1165 the ELF x86-64 abi, but it can be disabled for some compilation
1170 Module-Level Inline Assembly
1171 ----------------------------
1173 Modules may contain "module-level inline asm" blocks, which corresponds
1174 to the GCC "file scope inline asm" blocks. These blocks are internally
1175 concatenated by LLVM and treated as a single unit, but may be separated
1176 in the ``.ll`` file if desired. The syntax is very simple:
1178 .. code-block:: llvm
1180 module asm "inline asm code goes here"
1181 module asm "more can go here"
1183 The strings can contain any character by escaping non-printable
1184 characters. The escape sequence used is simply "\\xx" where "xx" is the
1185 two digit hex code for the number.
1187 The inline asm code is simply printed to the machine code .s file when
1188 assembly code is generated.
1190 .. _langref_datalayout:
1195 A module may specify a target specific data layout string that specifies
1196 how data is to be laid out in memory. The syntax for the data layout is
1199 .. code-block:: llvm
1201 target datalayout = "layout specification"
1203 The *layout specification* consists of a list of specifications
1204 separated by the minus sign character ('-'). Each specification starts
1205 with a letter and may include other information after the letter to
1206 define some aspect of the data layout. The specifications accepted are
1210 Specifies that the target lays out data in big-endian form. That is,
1211 the bits with the most significance have the lowest address
1214 Specifies that the target lays out data in little-endian form. That
1215 is, the bits with the least significance have the lowest address
1218 Specifies the natural alignment of the stack in bits. Alignment
1219 promotion of stack variables is limited to the natural stack
1220 alignment to avoid dynamic stack realignment. The stack alignment
1221 must be a multiple of 8-bits. If omitted, the natural stack
1222 alignment defaults to "unspecified", which does not prevent any
1223 alignment promotions.
1224 ``p[n]:<size>:<abi>:<pref>``
1225 This specifies the *size* of a pointer and its ``<abi>`` and
1226 ``<pref>``\erred alignments for address space ``n``. All sizes are in
1227 bits. The address space, ``n`` is optional, and if not specified,
1228 denotes the default address space 0. The value of ``n`` must be
1229 in the range [1,2^23).
1230 ``i<size>:<abi>:<pref>``
1231 This specifies the alignment for an integer type of a given bit
1232 ``<size>``. The value of ``<size>`` must be in the range [1,2^23).
1233 ``v<size>:<abi>:<pref>``
1234 This specifies the alignment for a vector type of a given bit
1236 ``f<size>:<abi>:<pref>``
1237 This specifies the alignment for a floating point type of a given bit
1238 ``<size>``. Only values of ``<size>`` that are supported by the target
1239 will work. 32 (float) and 64 (double) are supported on all targets; 80
1240 or 128 (different flavors of long double) are also supported on some
1243 This specifies the alignment for an object of aggregate type.
1245 If present, specifies that llvm names are mangled in the output. The
1248 * ``e``: ELF mangling: Private symbols get a ``.L`` prefix.
1249 * ``m``: Mips mangling: Private symbols get a ``$`` prefix.
1250 * ``o``: Mach-O mangling: Private symbols get ``L`` prefix. Other
1251 symbols get a ``_`` prefix.
1252 * ``w``: Windows COFF prefix: Similar to Mach-O, but stdcall and fastcall
1253 functions also get a suffix based on the frame size.
1254 ``n<size1>:<size2>:<size3>...``
1255 This specifies a set of native integer widths for the target CPU in
1256 bits. For example, it might contain ``n32`` for 32-bit PowerPC,
1257 ``n32:64`` for PowerPC 64, or ``n8:16:32:64`` for X86-64. Elements of
1258 this set are considered to support most general arithmetic operations
1261 On every specification that takes a ``<abi>:<pref>``, specifying the
1262 ``<pref>`` alignment is optional. If omitted, the preceding ``:``
1263 should be omitted too and ``<pref>`` will be equal to ``<abi>``.
1265 When constructing the data layout for a given target, LLVM starts with a
1266 default set of specifications which are then (possibly) overridden by
1267 the specifications in the ``datalayout`` keyword. The default
1268 specifications are given in this list:
1270 - ``E`` - big endian
1271 - ``p:64:64:64`` - 64-bit pointers with 64-bit alignment.
1272 - ``p[n]:64:64:64`` - Other address spaces are assumed to be the
1273 same as the default address space.
1274 - ``S0`` - natural stack alignment is unspecified
1275 - ``i1:8:8`` - i1 is 8-bit (byte) aligned
1276 - ``i8:8:8`` - i8 is 8-bit (byte) aligned
1277 - ``i16:16:16`` - i16 is 16-bit aligned
1278 - ``i32:32:32`` - i32 is 32-bit aligned
1279 - ``i64:32:64`` - i64 has ABI alignment of 32-bits but preferred
1280 alignment of 64-bits
1281 - ``f16:16:16`` - half is 16-bit aligned
1282 - ``f32:32:32`` - float is 32-bit aligned
1283 - ``f64:64:64`` - double is 64-bit aligned
1284 - ``f128:128:128`` - quad is 128-bit aligned
1285 - ``v64:64:64`` - 64-bit vector is 64-bit aligned
1286 - ``v128:128:128`` - 128-bit vector is 128-bit aligned
1287 - ``a:0:64`` - aggregates are 64-bit aligned
1289 When LLVM is determining the alignment for a given type, it uses the
1292 #. If the type sought is an exact match for one of the specifications,
1293 that specification is used.
1294 #. If no match is found, and the type sought is an integer type, then
1295 the smallest integer type that is larger than the bitwidth of the
1296 sought type is used. If none of the specifications are larger than
1297 the bitwidth then the largest integer type is used. For example,
1298 given the default specifications above, the i7 type will use the
1299 alignment of i8 (next largest) while both i65 and i256 will use the
1300 alignment of i64 (largest specified).
1301 #. If no match is found, and the type sought is a vector type, then the
1302 largest vector type that is smaller than the sought vector type will
1303 be used as a fall back. This happens because <128 x double> can be
1304 implemented in terms of 64 <2 x double>, for example.
1306 The function of the data layout string may not be what you expect.
1307 Notably, this is not a specification from the frontend of what alignment
1308 the code generator should use.
1310 Instead, if specified, the target data layout is required to match what
1311 the ultimate *code generator* expects. This string is used by the
1312 mid-level optimizers to improve code, and this only works if it matches
1313 what the ultimate code generator uses. If you would like to generate IR
1314 that does not embed this target-specific detail into the IR, then you
1315 don't have to specify the string. This will disable some optimizations
1316 that require precise layout information, but this also prevents those
1317 optimizations from introducing target specificity into the IR.
1324 A module may specify a target triple string that describes the target
1325 host. The syntax for the target triple is simply:
1327 .. code-block:: llvm
1329 target triple = "x86_64-apple-macosx10.7.0"
1331 The *target triple* string consists of a series of identifiers delimited
1332 by the minus sign character ('-'). The canonical forms are:
1336 ARCHITECTURE-VENDOR-OPERATING_SYSTEM
1337 ARCHITECTURE-VENDOR-OPERATING_SYSTEM-ENVIRONMENT
1339 This information is passed along to the backend so that it generates
1340 code for the proper architecture. It's possible to override this on the
1341 command line with the ``-mtriple`` command line option.
1343 .. _pointeraliasing:
1345 Pointer Aliasing Rules
1346 ----------------------
1348 Any memory access must be done through a pointer value associated with
1349 an address range of the memory access, otherwise the behavior is
1350 undefined. Pointer values are associated with address ranges according
1351 to the following rules:
1353 - A pointer value is associated with the addresses associated with any
1354 value it is *based* on.
1355 - An address of a global variable is associated with the address range
1356 of the variable's storage.
1357 - The result value of an allocation instruction is associated with the
1358 address range of the allocated storage.
1359 - A null pointer in the default address-space is associated with no
1361 - An integer constant other than zero or a pointer value returned from
1362 a function not defined within LLVM may be associated with address
1363 ranges allocated through mechanisms other than those provided by
1364 LLVM. Such ranges shall not overlap with any ranges of addresses
1365 allocated by mechanisms provided by LLVM.
1367 A pointer value is *based* on another pointer value according to the
1370 - A pointer value formed from a ``getelementptr`` operation is *based*
1371 on the first operand of the ``getelementptr``.
1372 - The result value of a ``bitcast`` is *based* on the operand of the
1374 - A pointer value formed by an ``inttoptr`` is *based* on all pointer
1375 values that contribute (directly or indirectly) to the computation of
1376 the pointer's value.
1377 - The "*based* on" relationship is transitive.
1379 Note that this definition of *"based"* is intentionally similar to the
1380 definition of *"based"* in C99, though it is slightly weaker.
1382 LLVM IR does not associate types with memory. The result type of a
1383 ``load`` merely indicates the size and alignment of the memory from
1384 which to load, as well as the interpretation of the value. The first
1385 operand type of a ``store`` similarly only indicates the size and
1386 alignment of the store.
1388 Consequently, type-based alias analysis, aka TBAA, aka
1389 ``-fstrict-aliasing``, is not applicable to general unadorned LLVM IR.
1390 :ref:`Metadata <metadata>` may be used to encode additional information
1391 which specialized optimization passes may use to implement type-based
1396 Volatile Memory Accesses
1397 ------------------------
1399 Certain memory accesses, such as :ref:`load <i_load>`'s,
1400 :ref:`store <i_store>`'s, and :ref:`llvm.memcpy <int_memcpy>`'s may be
1401 marked ``volatile``. The optimizers must not change the number of
1402 volatile operations or change their order of execution relative to other
1403 volatile operations. The optimizers *may* change the order of volatile
1404 operations relative to non-volatile operations. This is not Java's
1405 "volatile" and has no cross-thread synchronization behavior.
1407 IR-level volatile loads and stores cannot safely be optimized into
1408 llvm.memcpy or llvm.memmove intrinsics even when those intrinsics are
1409 flagged volatile. Likewise, the backend should never split or merge
1410 target-legal volatile load/store instructions.
1412 .. admonition:: Rationale
1414 Platforms may rely on volatile loads and stores of natively supported
1415 data width to be executed as single instruction. For example, in C
1416 this holds for an l-value of volatile primitive type with native
1417 hardware support, but not necessarily for aggregate types. The
1418 frontend upholds these expectations, which are intentionally
1419 unspecified in the IR. The rules above ensure that IR transformation
1420 do not violate the frontend's contract with the language.
1424 Memory Model for Concurrent Operations
1425 --------------------------------------
1427 The LLVM IR does not define any way to start parallel threads of
1428 execution or to register signal handlers. Nonetheless, there are
1429 platform-specific ways to create them, and we define LLVM IR's behavior
1430 in their presence. This model is inspired by the C++0x memory model.
1432 For a more informal introduction to this model, see the :doc:`Atomics`.
1434 We define a *happens-before* partial order as the least partial order
1437 - Is a superset of single-thread program order, and
1438 - When a *synchronizes-with* ``b``, includes an edge from ``a`` to
1439 ``b``. *Synchronizes-with* pairs are introduced by platform-specific
1440 techniques, like pthread locks, thread creation, thread joining,
1441 etc., and by atomic instructions. (See also :ref:`Atomic Memory Ordering
1442 Constraints <ordering>`).
1444 Note that program order does not introduce *happens-before* edges
1445 between a thread and signals executing inside that thread.
1447 Every (defined) read operation (load instructions, memcpy, atomic
1448 loads/read-modify-writes, etc.) R reads a series of bytes written by
1449 (defined) write operations (store instructions, atomic
1450 stores/read-modify-writes, memcpy, etc.). For the purposes of this
1451 section, initialized globals are considered to have a write of the
1452 initializer which is atomic and happens before any other read or write
1453 of the memory in question. For each byte of a read R, R\ :sub:`byte`
1454 may see any write to the same byte, except:
1456 - If write\ :sub:`1` happens before write\ :sub:`2`, and
1457 write\ :sub:`2` happens before R\ :sub:`byte`, then
1458 R\ :sub:`byte` does not see write\ :sub:`1`.
1459 - If R\ :sub:`byte` happens before write\ :sub:`3`, then
1460 R\ :sub:`byte` does not see write\ :sub:`3`.
1462 Given that definition, R\ :sub:`byte` is defined as follows:
1464 - If R is volatile, the result is target-dependent. (Volatile is
1465 supposed to give guarantees which can support ``sig_atomic_t`` in
1466 C/C++, and may be used for accesses to addresses which do not behave
1467 like normal memory. It does not generally provide cross-thread
1469 - Otherwise, if there is no write to the same byte that happens before
1470 R\ :sub:`byte`, R\ :sub:`byte` returns ``undef`` for that byte.
1471 - Otherwise, if R\ :sub:`byte` may see exactly one write,
1472 R\ :sub:`byte` returns the value written by that write.
1473 - Otherwise, if R is atomic, and all the writes R\ :sub:`byte` may
1474 see are atomic, it chooses one of the values written. See the :ref:`Atomic
1475 Memory Ordering Constraints <ordering>` section for additional
1476 constraints on how the choice is made.
1477 - Otherwise R\ :sub:`byte` returns ``undef``.
1479 R returns the value composed of the series of bytes it read. This
1480 implies that some bytes within the value may be ``undef`` **without**
1481 the entire value being ``undef``. Note that this only defines the
1482 semantics of the operation; it doesn't mean that targets will emit more
1483 than one instruction to read the series of bytes.
1485 Note that in cases where none of the atomic intrinsics are used, this
1486 model places only one restriction on IR transformations on top of what
1487 is required for single-threaded execution: introducing a store to a byte
1488 which might not otherwise be stored is not allowed in general.
1489 (Specifically, in the case where another thread might write to and read
1490 from an address, introducing a store can change a load that may see
1491 exactly one write into a load that may see multiple writes.)
1495 Atomic Memory Ordering Constraints
1496 ----------------------------------
1498 Atomic instructions (:ref:`cmpxchg <i_cmpxchg>`,
1499 :ref:`atomicrmw <i_atomicrmw>`, :ref:`fence <i_fence>`,
1500 :ref:`atomic load <i_load>`, and :ref:`atomic store <i_store>`) take
1501 ordering parameters that determine which other atomic instructions on
1502 the same address they *synchronize with*. These semantics are borrowed
1503 from Java and C++0x, but are somewhat more colloquial. If these
1504 descriptions aren't precise enough, check those specs (see spec
1505 references in the :doc:`atomics guide <Atomics>`).
1506 :ref:`fence <i_fence>` instructions treat these orderings somewhat
1507 differently since they don't take an address. See that instruction's
1508 documentation for details.
1510 For a simpler introduction to the ordering constraints, see the
1514 The set of values that can be read is governed by the happens-before
1515 partial order. A value cannot be read unless some operation wrote
1516 it. This is intended to provide a guarantee strong enough to model
1517 Java's non-volatile shared variables. This ordering cannot be
1518 specified for read-modify-write operations; it is not strong enough
1519 to make them atomic in any interesting way.
1521 In addition to the guarantees of ``unordered``, there is a single
1522 total order for modifications by ``monotonic`` operations on each
1523 address. All modification orders must be compatible with the
1524 happens-before order. There is no guarantee that the modification
1525 orders can be combined to a global total order for the whole program
1526 (and this often will not be possible). The read in an atomic
1527 read-modify-write operation (:ref:`cmpxchg <i_cmpxchg>` and
1528 :ref:`atomicrmw <i_atomicrmw>`) reads the value in the modification
1529 order immediately before the value it writes. If one atomic read
1530 happens before another atomic read of the same address, the later
1531 read must see the same value or a later value in the address's
1532 modification order. This disallows reordering of ``monotonic`` (or
1533 stronger) operations on the same address. If an address is written
1534 ``monotonic``-ally by one thread, and other threads ``monotonic``-ally
1535 read that address repeatedly, the other threads must eventually see
1536 the write. This corresponds to the C++0x/C1x
1537 ``memory_order_relaxed``.
1539 In addition to the guarantees of ``monotonic``, a
1540 *synchronizes-with* edge may be formed with a ``release`` operation.
1541 This is intended to model C++'s ``memory_order_acquire``.
1543 In addition to the guarantees of ``monotonic``, if this operation
1544 writes a value which is subsequently read by an ``acquire``
1545 operation, it *synchronizes-with* that operation. (This isn't a
1546 complete description; see the C++0x definition of a release
1547 sequence.) This corresponds to the C++0x/C1x
1548 ``memory_order_release``.
1549 ``acq_rel`` (acquire+release)
1550 Acts as both an ``acquire`` and ``release`` operation on its
1551 address. This corresponds to the C++0x/C1x ``memory_order_acq_rel``.
1552 ``seq_cst`` (sequentially consistent)
1553 In addition to the guarantees of ``acq_rel`` (``acquire`` for an
1554 operation which only reads, ``release`` for an operation which only
1555 writes), there is a global total order on all
1556 sequentially-consistent operations on all addresses, which is
1557 consistent with the *happens-before* partial order and with the
1558 modification orders of all the affected addresses. Each
1559 sequentially-consistent read sees the last preceding write to the
1560 same address in this global order. This corresponds to the C++0x/C1x
1561 ``memory_order_seq_cst`` and Java volatile.
1565 If an atomic operation is marked ``singlethread``, it only *synchronizes
1566 with* or participates in modification and seq\_cst total orderings with
1567 other operations running in the same thread (for example, in signal
1575 LLVM IR floating-point binary ops (:ref:`fadd <i_fadd>`,
1576 :ref:`fsub <i_fsub>`, :ref:`fmul <i_fmul>`, :ref:`fdiv <i_fdiv>`,
1577 :ref:`frem <i_frem>`) have the following flags that can set to enable
1578 otherwise unsafe floating point operations
1581 No NaNs - Allow optimizations to assume the arguments and result are not
1582 NaN. Such optimizations are required to retain defined behavior over
1583 NaNs, but the value of the result is undefined.
1586 No Infs - Allow optimizations to assume the arguments and result are not
1587 +/-Inf. Such optimizations are required to retain defined behavior over
1588 +/-Inf, but the value of the result is undefined.
1591 No Signed Zeros - Allow optimizations to treat the sign of a zero
1592 argument or result as insignificant.
1595 Allow Reciprocal - Allow optimizations to use the reciprocal of an
1596 argument rather than perform division.
1599 Fast - Allow algebraically equivalent transformations that may
1600 dramatically change results in floating point (e.g. reassociate). This
1601 flag implies all the others.
1608 The LLVM type system is one of the most important features of the
1609 intermediate representation. Being typed enables a number of
1610 optimizations to be performed on the intermediate representation
1611 directly, without having to do extra analyses on the side before the
1612 transformation. A strong type system makes it easier to read the
1613 generated code and enables novel analyses and transformations that are
1614 not feasible to perform on normal three address code representations.
1624 The void type does not represent any value and has no size.
1642 The function type can be thought of as a function signature. It consists of a
1643 return type and a list of formal parameter types. The return type of a function
1644 type is a void type or first class type --- except for :ref:`label <t_label>`
1645 and :ref:`metadata <t_metadata>` types.
1651 <returntype> (<parameter list>)
1653 ...where '``<parameter list>``' is a comma-separated list of type
1654 specifiers. Optionally, the parameter list may include a type ``...``, which
1655 indicates that the function takes a variable number of arguments. Variable
1656 argument functions can access their arguments with the :ref:`variable argument
1657 handling intrinsic <int_varargs>` functions. '``<returntype>``' is any type
1658 except :ref:`label <t_label>` and :ref:`metadata <t_metadata>`.
1662 +---------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1663 | ``i32 (i32)`` | function taking an ``i32``, returning an ``i32`` |
1664 +---------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1665 | ``float (i16, i32 *) *`` | :ref:`Pointer <t_pointer>` to a function that takes an ``i16`` and a :ref:`pointer <t_pointer>` to ``i32``, returning ``float``. |
1666 +---------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1667 | ``i32 (i8*, ...)`` | A vararg function that takes at least one :ref:`pointer <t_pointer>` to ``i8`` (char in C), which returns an integer. This is the signature for ``printf`` in LLVM. |
1668 +---------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1669 | ``{i32, i32} (i32)`` | A function taking an ``i32``, returning a :ref:`structure <t_struct>` containing two ``i32`` values |
1670 +---------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1677 The :ref:`first class <t_firstclass>` types are perhaps the most important.
1678 Values of these types are the only ones which can be produced by
1686 These are the types that are valid in registers from CodeGen's perspective.
1695 The integer type is a very simple type that simply specifies an
1696 arbitrary bit width for the integer type desired. Any bit width from 1
1697 bit to 2\ :sup:`23`\ -1 (about 8 million) can be specified.
1705 The number of bits the integer will occupy is specified by the ``N``
1711 +----------------+------------------------------------------------+
1712 | ``i1`` | a single-bit integer. |
1713 +----------------+------------------------------------------------+
1714 | ``i32`` | a 32-bit integer. |
1715 +----------------+------------------------------------------------+
1716 | ``i1942652`` | a really big integer of over 1 million bits. |
1717 +----------------+------------------------------------------------+
1721 Floating Point Types
1722 """"""""""""""""""""
1731 - 16-bit floating point value
1734 - 32-bit floating point value
1737 - 64-bit floating point value
1740 - 128-bit floating point value (112-bit mantissa)
1743 - 80-bit floating point value (X87)
1746 - 128-bit floating point value (two 64-bits)
1753 The x86_mmx type represents a value held in an MMX register on an x86
1754 machine. The operations allowed on it are quite limited: parameters and
1755 return values, load and store, and bitcast. User-specified MMX
1756 instructions are represented as intrinsic or asm calls with arguments
1757 and/or results of this type. There are no arrays, vectors or constants
1774 The pointer type is used to specify memory locations. Pointers are
1775 commonly used to reference objects in memory.
1777 Pointer types may have an optional address space attribute defining the
1778 numbered address space where the pointed-to object resides. The default
1779 address space is number zero. The semantics of non-zero address spaces
1780 are target-specific.
1782 Note that LLVM does not permit pointers to void (``void*``) nor does it
1783 permit pointers to labels (``label*``). Use ``i8*`` instead.
1793 +-------------------------+--------------------------------------------------------------------------------------------------------------+
1794 | ``[4 x i32]*`` | A :ref:`pointer <t_pointer>` to :ref:`array <t_array>` of four ``i32`` values. |
1795 +-------------------------+--------------------------------------------------------------------------------------------------------------+
1796 | ``i32 (i32*) *`` | A :ref:`pointer <t_pointer>` to a :ref:`function <t_function>` that takes an ``i32*``, returning an ``i32``. |
1797 +-------------------------+--------------------------------------------------------------------------------------------------------------+
1798 | ``i32 addrspace(5)*`` | A :ref:`pointer <t_pointer>` to an ``i32`` value that resides in address space #5. |
1799 +-------------------------+--------------------------------------------------------------------------------------------------------------+
1808 A vector type is a simple derived type that represents a vector of
1809 elements. Vector types are used when multiple primitive data are
1810 operated in parallel using a single instruction (SIMD). A vector type
1811 requires a size (number of elements) and an underlying primitive data
1812 type. Vector types are considered :ref:`first class <t_firstclass>`.
1818 < <# elements> x <elementtype> >
1820 The number of elements is a constant integer value larger than 0;
1821 elementtype may be any integer or floating point type, or a pointer to
1822 these types. Vectors of size zero are not allowed.
1826 +-------------------+--------------------------------------------------+
1827 | ``<4 x i32>`` | Vector of 4 32-bit integer values. |
1828 +-------------------+--------------------------------------------------+
1829 | ``<8 x float>`` | Vector of 8 32-bit floating-point values. |
1830 +-------------------+--------------------------------------------------+
1831 | ``<2 x i64>`` | Vector of 2 64-bit integer values. |
1832 +-------------------+--------------------------------------------------+
1833 | ``<4 x i64*>`` | Vector of 4 pointers to 64-bit integer values. |
1834 +-------------------+--------------------------------------------------+
1843 The label type represents code labels.
1858 The metadata type represents embedded metadata. No derived types may be
1859 created from metadata except for :ref:`function <t_function>` arguments.
1872 Aggregate Types are a subset of derived types that can contain multiple
1873 member types. :ref:`Arrays <t_array>` and :ref:`structs <t_struct>` are
1874 aggregate types. :ref:`Vectors <t_vector>` are not considered to be
1884 The array type is a very simple derived type that arranges elements
1885 sequentially in memory. The array type requires a size (number of
1886 elements) and an underlying data type.
1892 [<# elements> x <elementtype>]
1894 The number of elements is a constant integer value; ``elementtype`` may
1895 be any type with a size.
1899 +------------------+--------------------------------------+
1900 | ``[40 x i32]`` | Array of 40 32-bit integer values. |
1901 +------------------+--------------------------------------+
1902 | ``[41 x i32]`` | Array of 41 32-bit integer values. |
1903 +------------------+--------------------------------------+
1904 | ``[4 x i8]`` | Array of 4 8-bit integer values. |
1905 +------------------+--------------------------------------+
1907 Here are some examples of multidimensional arrays:
1909 +-----------------------------+----------------------------------------------------------+
1910 | ``[3 x [4 x i32]]`` | 3x4 array of 32-bit integer values. |
1911 +-----------------------------+----------------------------------------------------------+
1912 | ``[12 x [10 x float]]`` | 12x10 array of single precision floating point values. |
1913 +-----------------------------+----------------------------------------------------------+
1914 | ``[2 x [3 x [4 x i16]]]`` | 2x3x4 array of 16-bit integer values. |
1915 +-----------------------------+----------------------------------------------------------+
1917 There is no restriction on indexing beyond the end of the array implied
1918 by a static type (though there are restrictions on indexing beyond the
1919 bounds of an allocated object in some cases). This means that
1920 single-dimension 'variable sized array' addressing can be implemented in
1921 LLVM with a zero length array type. An implementation of 'pascal style
1922 arrays' in LLVM could use the type "``{ i32, [0 x float]}``", for
1932 The structure type is used to represent a collection of data members
1933 together in memory. The elements of a structure may be any type that has
1936 Structures in memory are accessed using '``load``' and '``store``' by
1937 getting a pointer to a field with the '``getelementptr``' instruction.
1938 Structures in registers are accessed using the '``extractvalue``' and
1939 '``insertvalue``' instructions.
1941 Structures may optionally be "packed" structures, which indicate that
1942 the alignment of the struct is one byte, and that there is no padding
1943 between the elements. In non-packed structs, padding between field types
1944 is inserted as defined by the DataLayout string in the module, which is
1945 required to match what the underlying code generator expects.
1947 Structures can either be "literal" or "identified". A literal structure
1948 is defined inline with other types (e.g. ``{i32, i32}*``) whereas
1949 identified types are always defined at the top level with a name.
1950 Literal types are uniqued by their contents and can never be recursive
1951 or opaque since there is no way to write one. Identified types can be
1952 recursive, can be opaqued, and are never uniqued.
1958 %T1 = type { <type list> } ; Identified normal struct type
1959 %T2 = type <{ <type list> }> ; Identified packed struct type
1963 +------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1964 | ``{ i32, i32, i32 }`` | A triple of three ``i32`` values |
1965 +------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1966 | ``{ float, i32 (i32) * }`` | A pair, where the first element is a ``float`` and the second element is a :ref:`pointer <t_pointer>` to a :ref:`function <t_function>` that takes an ``i32``, returning an ``i32``. |
1967 +------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1968 | ``<{ i8, i32 }>`` | A packed struct known to be 5 bytes in size. |
1969 +------------------------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1973 Opaque Structure Types
1974 """"""""""""""""""""""
1978 Opaque structure types are used to represent named structure types that
1979 do not have a body specified. This corresponds (for example) to the C
1980 notion of a forward declared structure.
1991 +--------------+-------------------+
1992 | ``opaque`` | An opaque type. |
1993 +--------------+-------------------+
2000 LLVM has several different basic types of constants. This section
2001 describes them all and their syntax.
2006 **Boolean constants**
2007 The two strings '``true``' and '``false``' are both valid constants
2009 **Integer constants**
2010 Standard integers (such as '4') are constants of the
2011 :ref:`integer <t_integer>` type. Negative numbers may be used with
2013 **Floating point constants**
2014 Floating point constants use standard decimal notation (e.g.
2015 123.421), exponential notation (e.g. 1.23421e+2), or a more precise
2016 hexadecimal notation (see below). The assembler requires the exact
2017 decimal value of a floating-point constant. For example, the
2018 assembler accepts 1.25 but rejects 1.3 because 1.3 is a repeating
2019 decimal in binary. Floating point constants must have a :ref:`floating
2020 point <t_floating>` type.
2021 **Null pointer constants**
2022 The identifier '``null``' is recognized as a null pointer constant
2023 and must be of :ref:`pointer type <t_pointer>`.
2025 The one non-intuitive notation for constants is the hexadecimal form of
2026 floating point constants. For example, the form
2027 '``double 0x432ff973cafa8000``' is equivalent to (but harder to read
2028 than) '``double 4.5e+15``'. The only time hexadecimal floating point
2029 constants are required (and the only time that they are generated by the
2030 disassembler) is when a floating point constant must be emitted but it
2031 cannot be represented as a decimal floating point number in a reasonable
2032 number of digits. For example, NaN's, infinities, and other special
2033 values are represented in their IEEE hexadecimal format so that assembly
2034 and disassembly do not cause any bits to change in the constants.
2036 When using the hexadecimal form, constants of types half, float, and
2037 double are represented using the 16-digit form shown above (which
2038 matches the IEEE754 representation for double); half and float values
2039 must, however, be exactly representable as IEEE 754 half and single
2040 precision, respectively. Hexadecimal format is always used for long
2041 double, and there are three forms of long double. The 80-bit format used
2042 by x86 is represented as ``0xK`` followed by 20 hexadecimal digits. The
2043 128-bit format used by PowerPC (two adjacent doubles) is represented by
2044 ``0xM`` followed by 32 hexadecimal digits. The IEEE 128-bit format is
2045 represented by ``0xL`` followed by 32 hexadecimal digits. Long doubles
2046 will only work if they match the long double format on your target.
2047 The IEEE 16-bit format (half precision) is represented by ``0xH``
2048 followed by 4 hexadecimal digits. All hexadecimal formats are big-endian
2049 (sign bit at the left).
2051 There are no constants of type x86_mmx.
2053 .. _complexconstants:
2058 Complex constants are a (potentially recursive) combination of simple
2059 constants and smaller complex constants.
2061 **Structure constants**
2062 Structure constants are represented with notation similar to
2063 structure type definitions (a comma separated list of elements,
2064 surrounded by braces (``{}``)). For example:
2065 "``{ i32 4, float 17.0, i32* @G }``", where "``@G``" is declared as
2066 "``@G = external global i32``". Structure constants must have
2067 :ref:`structure type <t_struct>`, and the number and types of elements
2068 must match those specified by the type.
2070 Array constants are represented with notation similar to array type
2071 definitions (a comma separated list of elements, surrounded by
2072 square brackets (``[]``)). For example:
2073 "``[ i32 42, i32 11, i32 74 ]``". Array constants must have
2074 :ref:`array type <t_array>`, and the number and types of elements must
2075 match those specified by the type.
2076 **Vector constants**
2077 Vector constants are represented with notation similar to vector
2078 type definitions (a comma separated list of elements, surrounded by
2079 less-than/greater-than's (``<>``)). For example:
2080 "``< i32 42, i32 11, i32 74, i32 100 >``". Vector constants
2081 must have :ref:`vector type <t_vector>`, and the number and types of
2082 elements must match those specified by the type.
2083 **Zero initialization**
2084 The string '``zeroinitializer``' can be used to zero initialize a
2085 value to zero of *any* type, including scalar and
2086 :ref:`aggregate <t_aggregate>` types. This is often used to avoid
2087 having to print large zero initializers (e.g. for large arrays) and
2088 is always exactly equivalent to using explicit zero initializers.
2090 A metadata node is a structure-like constant with :ref:`metadata
2091 type <t_metadata>`. For example:
2092 "``metadata !{ i32 0, metadata !"test" }``". Unlike other
2093 constants that are meant to be interpreted as part of the
2094 instruction stream, metadata is a place to attach additional
2095 information such as debug info.
2097 Global Variable and Function Addresses
2098 --------------------------------------
2100 The addresses of :ref:`global variables <globalvars>` and
2101 :ref:`functions <functionstructure>` are always implicitly valid
2102 (link-time) constants. These constants are explicitly referenced when
2103 the :ref:`identifier for the global <identifiers>` is used and always have
2104 :ref:`pointer <t_pointer>` type. For example, the following is a legal LLVM
2107 .. code-block:: llvm
2111 @Z = global [2 x i32*] [ i32* @X, i32* @Y ]
2118 The string '``undef``' can be used anywhere a constant is expected, and
2119 indicates that the user of the value may receive an unspecified
2120 bit-pattern. Undefined values may be of any type (other than '``label``'
2121 or '``void``') and be used anywhere a constant is permitted.
2123 Undefined values are useful because they indicate to the compiler that
2124 the program is well defined no matter what value is used. This gives the
2125 compiler more freedom to optimize. Here are some examples of
2126 (potentially surprising) transformations that are valid (in pseudo IR):
2128 .. code-block:: llvm
2138 This is safe because all of the output bits are affected by the undef
2139 bits. Any output bit can have a zero or one depending on the input bits.
2141 .. code-block:: llvm
2152 These logical operations have bits that are not always affected by the
2153 input. For example, if ``%X`` has a zero bit, then the output of the
2154 '``and``' operation will always be a zero for that bit, no matter what
2155 the corresponding bit from the '``undef``' is. As such, it is unsafe to
2156 optimize or assume that the result of the '``and``' is '``undef``'.
2157 However, it is safe to assume that all bits of the '``undef``' could be
2158 0, and optimize the '``and``' to 0. Likewise, it is safe to assume that
2159 all the bits of the '``undef``' operand to the '``or``' could be set,
2160 allowing the '``or``' to be folded to -1.
2162 .. code-block:: llvm
2164 %A = select undef, %X, %Y
2165 %B = select undef, 42, %Y
2166 %C = select %X, %Y, undef
2176 This set of examples shows that undefined '``select``' (and conditional
2177 branch) conditions can go *either way*, but they have to come from one
2178 of the two operands. In the ``%A`` example, if ``%X`` and ``%Y`` were
2179 both known to have a clear low bit, then ``%A`` would have to have a
2180 cleared low bit. However, in the ``%C`` example, the optimizer is
2181 allowed to assume that the '``undef``' operand could be the same as
2182 ``%Y``, allowing the whole '``select``' to be eliminated.
2184 .. code-block:: llvm
2186 %A = xor undef, undef
2203 This example points out that two '``undef``' operands are not
2204 necessarily the same. This can be surprising to people (and also matches
2205 C semantics) where they assume that "``X^X``" is always zero, even if
2206 ``X`` is undefined. This isn't true for a number of reasons, but the
2207 short answer is that an '``undef``' "variable" can arbitrarily change
2208 its value over its "live range". This is true because the variable
2209 doesn't actually *have a live range*. Instead, the value is logically
2210 read from arbitrary registers that happen to be around when needed, so
2211 the value is not necessarily consistent over time. In fact, ``%A`` and
2212 ``%C`` need to have the same semantics or the core LLVM "replace all
2213 uses with" concept would not hold.
2215 .. code-block:: llvm
2223 These examples show the crucial difference between an *undefined value*
2224 and *undefined behavior*. An undefined value (like '``undef``') is
2225 allowed to have an arbitrary bit-pattern. This means that the ``%A``
2226 operation can be constant folded to '``undef``', because the '``undef``'
2227 could be an SNaN, and ``fdiv`` is not (currently) defined on SNaN's.
2228 However, in the second example, we can make a more aggressive
2229 assumption: because the ``undef`` is allowed to be an arbitrary value,
2230 we are allowed to assume that it could be zero. Since a divide by zero
2231 has *undefined behavior*, we are allowed to assume that the operation
2232 does not execute at all. This allows us to delete the divide and all
2233 code after it. Because the undefined operation "can't happen", the
2234 optimizer can assume that it occurs in dead code.
2236 .. code-block:: llvm
2238 a: store undef -> %X
2239 b: store %X -> undef
2244 These examples reiterate the ``fdiv`` example: a store *of* an undefined
2245 value can be assumed to not have any effect; we can assume that the
2246 value is overwritten with bits that happen to match what was already
2247 there. However, a store *to* an undefined location could clobber
2248 arbitrary memory, therefore, it has undefined behavior.
2255 Poison values are similar to :ref:`undef values <undefvalues>`, however
2256 they also represent the fact that an instruction or constant expression
2257 which cannot evoke side effects has nevertheless detected a condition
2258 which results in undefined behavior.
2260 There is currently no way of representing a poison value in the IR; they
2261 only exist when produced by operations such as :ref:`add <i_add>` with
2264 Poison value behavior is defined in terms of value *dependence*:
2266 - Values other than :ref:`phi <i_phi>` nodes depend on their operands.
2267 - :ref:`Phi <i_phi>` nodes depend on the operand corresponding to
2268 their dynamic predecessor basic block.
2269 - Function arguments depend on the corresponding actual argument values
2270 in the dynamic callers of their functions.
2271 - :ref:`Call <i_call>` instructions depend on the :ref:`ret <i_ret>`
2272 instructions that dynamically transfer control back to them.
2273 - :ref:`Invoke <i_invoke>` instructions depend on the
2274 :ref:`ret <i_ret>`, :ref:`resume <i_resume>`, or exception-throwing
2275 call instructions that dynamically transfer control back to them.
2276 - Non-volatile loads and stores depend on the most recent stores to all
2277 of the referenced memory addresses, following the order in the IR
2278 (including loads and stores implied by intrinsics such as
2279 :ref:`@llvm.memcpy <int_memcpy>`.)
2280 - An instruction with externally visible side effects depends on the
2281 most recent preceding instruction with externally visible side
2282 effects, following the order in the IR. (This includes :ref:`volatile
2283 operations <volatile>`.)
2284 - An instruction *control-depends* on a :ref:`terminator
2285 instruction <terminators>` if the terminator instruction has
2286 multiple successors and the instruction is always executed when
2287 control transfers to one of the successors, and may not be executed
2288 when control is transferred to another.
2289 - Additionally, an instruction also *control-depends* on a terminator
2290 instruction if the set of instructions it otherwise depends on would
2291 be different if the terminator had transferred control to a different
2293 - Dependence is transitive.
2295 Poison Values have the same behavior as :ref:`undef values <undefvalues>`,
2296 with the additional affect that any instruction which has a *dependence*
2297 on a poison value has undefined behavior.
2299 Here are some examples:
2301 .. code-block:: llvm
2304 %poison = sub nuw i32 0, 1 ; Results in a poison value.
2305 %still_poison = and i32 %poison, 0 ; 0, but also poison.
2306 %poison_yet_again = getelementptr i32* @h, i32 %still_poison
2307 store i32 0, i32* %poison_yet_again ; memory at @h[0] is poisoned
2309 store i32 %poison, i32* @g ; Poison value stored to memory.
2310 %poison2 = load i32* @g ; Poison value loaded back from memory.
2312 store volatile i32 %poison, i32* @g ; External observation; undefined behavior.
2314 %narrowaddr = bitcast i32* @g to i16*
2315 %wideaddr = bitcast i32* @g to i64*
2316 %poison3 = load i16* %narrowaddr ; Returns a poison value.
2317 %poison4 = load i64* %wideaddr ; Returns a poison value.
2319 %cmp = icmp slt i32 %poison, 0 ; Returns a poison value.
2320 br i1 %cmp, label %true, label %end ; Branch to either destination.
2323 store volatile i32 0, i32* @g ; This is control-dependent on %cmp, so
2324 ; it has undefined behavior.
2328 %p = phi i32 [ 0, %entry ], [ 1, %true ]
2329 ; Both edges into this PHI are
2330 ; control-dependent on %cmp, so this
2331 ; always results in a poison value.
2333 store volatile i32 0, i32* @g ; This would depend on the store in %true
2334 ; if %cmp is true, or the store in %entry
2335 ; otherwise, so this is undefined behavior.
2337 br i1 %cmp, label %second_true, label %second_end
2338 ; The same branch again, but this time the
2339 ; true block doesn't have side effects.
2346 store volatile i32 0, i32* @g ; This time, the instruction always depends
2347 ; on the store in %end. Also, it is
2348 ; control-equivalent to %end, so this is
2349 ; well-defined (ignoring earlier undefined
2350 ; behavior in this example).
2354 Addresses of Basic Blocks
2355 -------------------------
2357 ``blockaddress(@function, %block)``
2359 The '``blockaddress``' constant computes the address of the specified
2360 basic block in the specified function, and always has an ``i8*`` type.
2361 Taking the address of the entry block is illegal.
2363 This value only has defined behavior when used as an operand to the
2364 ':ref:`indirectbr <i_indirectbr>`' instruction, or for comparisons
2365 against null. Pointer equality tests between labels addresses results in
2366 undefined behavior --- though, again, comparison against null is ok, and
2367 no label is equal to the null pointer. This may be passed around as an
2368 opaque pointer sized value as long as the bits are not inspected. This
2369 allows ``ptrtoint`` and arithmetic to be performed on these values so
2370 long as the original value is reconstituted before the ``indirectbr``
2373 Finally, some targets may provide defined semantics when using the value
2374 as the operand to an inline assembly, but that is target specific.
2378 Constant Expressions
2379 --------------------
2381 Constant expressions are used to allow expressions involving other
2382 constants to be used as constants. Constant expressions may be of any
2383 :ref:`first class <t_firstclass>` type and may involve any LLVM operation
2384 that does not have side effects (e.g. load and call are not supported).
2385 The following is the syntax for constant expressions:
2387 ``trunc (CST to TYPE)``
2388 Truncate a constant to another type. The bit size of CST must be
2389 larger than the bit size of TYPE. Both types must be integers.
2390 ``zext (CST to TYPE)``
2391 Zero extend a constant to another type. The bit size of CST must be
2392 smaller than the bit size of TYPE. Both types must be integers.
2393 ``sext (CST to TYPE)``
2394 Sign extend a constant to another type. The bit size of CST must be
2395 smaller than the bit size of TYPE. Both types must be integers.
2396 ``fptrunc (CST to TYPE)``
2397 Truncate a floating point constant to another floating point type.
2398 The size of CST must be larger than the size of TYPE. Both types
2399 must be floating point.
2400 ``fpext (CST to TYPE)``
2401 Floating point extend a constant to another type. The size of CST
2402 must be smaller or equal to the size of TYPE. Both types must be
2404 ``fptoui (CST to TYPE)``
2405 Convert a floating point constant to the corresponding unsigned
2406 integer constant. TYPE must be a scalar or vector integer type. CST
2407 must be of scalar or vector floating point type. Both CST and TYPE
2408 must be scalars, or vectors of the same number of elements. If the
2409 value won't fit in the integer type, the results are undefined.
2410 ``fptosi (CST to TYPE)``
2411 Convert a floating point constant to the corresponding signed
2412 integer constant. TYPE must be a scalar or vector integer type. CST
2413 must be of scalar or vector floating point type. Both CST and TYPE
2414 must be scalars, or vectors of the same number of elements. If the
2415 value won't fit in the integer type, the results are undefined.
2416 ``uitofp (CST to TYPE)``
2417 Convert an unsigned integer constant to the corresponding floating
2418 point constant. TYPE must be a scalar or vector floating point type.
2419 CST must be of scalar or vector integer type. Both CST and TYPE must
2420 be scalars, or vectors of the same number of elements. If the value
2421 won't fit in the floating point type, the results are undefined.
2422 ``sitofp (CST to TYPE)``
2423 Convert a signed integer constant to the corresponding floating
2424 point constant. TYPE must be a scalar or vector floating point type.
2425 CST must be of scalar or vector integer type. Both CST and TYPE must
2426 be scalars, or vectors of the same number of elements. If the value
2427 won't fit in the floating point type, the results are undefined.
2428 ``ptrtoint (CST to TYPE)``
2429 Convert a pointer typed constant to the corresponding integer
2430 constant. ``TYPE`` must be an integer type. ``CST`` must be of
2431 pointer type. The ``CST`` value is zero extended, truncated, or
2432 unchanged to make it fit in ``TYPE``.
2433 ``inttoptr (CST to TYPE)``
2434 Convert an integer constant to a pointer constant. TYPE must be a
2435 pointer type. CST must be of integer type. The CST value is zero
2436 extended, truncated, or unchanged to make it fit in a pointer size.
2437 This one is *really* dangerous!
2438 ``bitcast (CST to TYPE)``
2439 Convert a constant, CST, to another TYPE. The constraints of the
2440 operands are the same as those for the :ref:`bitcast
2441 instruction <i_bitcast>`.
2442 ``addrspacecast (CST to TYPE)``
2443 Convert a constant pointer or constant vector of pointer, CST, to another
2444 TYPE in a different address space. The constraints of the operands are the
2445 same as those for the :ref:`addrspacecast instruction <i_addrspacecast>`.
2446 ``getelementptr (CSTPTR, IDX0, IDX1, ...)``, ``getelementptr inbounds (CSTPTR, IDX0, IDX1, ...)``
2447 Perform the :ref:`getelementptr operation <i_getelementptr>` on
2448 constants. As with the :ref:`getelementptr <i_getelementptr>`
2449 instruction, the index list may have zero or more indexes, which are
2450 required to make sense for the type of "CSTPTR".
2451 ``select (COND, VAL1, VAL2)``
2452 Perform the :ref:`select operation <i_select>` on constants.
2453 ``icmp COND (VAL1, VAL2)``
2454 Performs the :ref:`icmp operation <i_icmp>` on constants.
2455 ``fcmp COND (VAL1, VAL2)``
2456 Performs the :ref:`fcmp operation <i_fcmp>` on constants.
2457 ``extractelement (VAL, IDX)``
2458 Perform the :ref:`extractelement operation <i_extractelement>` on
2460 ``insertelement (VAL, ELT, IDX)``
2461 Perform the :ref:`insertelement operation <i_insertelement>` on
2463 ``shufflevector (VEC1, VEC2, IDXMASK)``
2464 Perform the :ref:`shufflevector operation <i_shufflevector>` on
2466 ``extractvalue (VAL, IDX0, IDX1, ...)``
2467 Perform the :ref:`extractvalue operation <i_extractvalue>` on
2468 constants. The index list is interpreted in a similar manner as
2469 indices in a ':ref:`getelementptr <i_getelementptr>`' operation. At
2470 least one index value must be specified.
2471 ``insertvalue (VAL, ELT, IDX0, IDX1, ...)``
2472 Perform the :ref:`insertvalue operation <i_insertvalue>` on constants.
2473 The index list is interpreted in a similar manner as indices in a
2474 ':ref:`getelementptr <i_getelementptr>`' operation. At least one index
2475 value must be specified.
2476 ``OPCODE (LHS, RHS)``
2477 Perform the specified operation of the LHS and RHS constants. OPCODE
2478 may be any of the :ref:`binary <binaryops>` or :ref:`bitwise
2479 binary <bitwiseops>` operations. The constraints on operands are
2480 the same as those for the corresponding instruction (e.g. no bitwise
2481 operations on floating point values are allowed).
2488 Inline Assembler Expressions
2489 ----------------------------
2491 LLVM supports inline assembler expressions (as opposed to :ref:`Module-Level
2492 Inline Assembly <moduleasm>`) through the use of a special value. This
2493 value represents the inline assembler as a string (containing the
2494 instructions to emit), a list of operand constraints (stored as a
2495 string), a flag that indicates whether or not the inline asm expression
2496 has side effects, and a flag indicating whether the function containing
2497 the asm needs to align its stack conservatively. An example inline
2498 assembler expression is:
2500 .. code-block:: llvm
2502 i32 (i32) asm "bswap $0", "=r,r"
2504 Inline assembler expressions may **only** be used as the callee operand
2505 of a :ref:`call <i_call>` or an :ref:`invoke <i_invoke>` instruction.
2506 Thus, typically we have:
2508 .. code-block:: llvm
2510 %X = call i32 asm "bswap $0", "=r,r"(i32 %Y)
2512 Inline asms with side effects not visible in the constraint list must be
2513 marked as having side effects. This is done through the use of the
2514 '``sideeffect``' keyword, like so:
2516 .. code-block:: llvm
2518 call void asm sideeffect "eieio", ""()
2520 In some cases inline asms will contain code that will not work unless
2521 the stack is aligned in some way, such as calls or SSE instructions on
2522 x86, yet will not contain code that does that alignment within the asm.
2523 The compiler should make conservative assumptions about what the asm
2524 might contain and should generate its usual stack alignment code in the
2525 prologue if the '``alignstack``' keyword is present:
2527 .. code-block:: llvm
2529 call void asm alignstack "eieio", ""()
2531 Inline asms also support using non-standard assembly dialects. The
2532 assumed dialect is ATT. When the '``inteldialect``' keyword is present,
2533 the inline asm is using the Intel dialect. Currently, ATT and Intel are
2534 the only supported dialects. An example is:
2536 .. code-block:: llvm
2538 call void asm inteldialect "eieio", ""()
2540 If multiple keywords appear the '``sideeffect``' keyword must come
2541 first, the '``alignstack``' keyword second and the '``inteldialect``'
2547 The call instructions that wrap inline asm nodes may have a
2548 "``!srcloc``" MDNode attached to it that contains a list of constant
2549 integers. If present, the code generator will use the integer as the
2550 location cookie value when report errors through the ``LLVMContext``
2551 error reporting mechanisms. This allows a front-end to correlate backend
2552 errors that occur with inline asm back to the source code that produced
2555 .. code-block:: llvm
2557 call void asm sideeffect "something bad", ""(), !srcloc !42
2559 !42 = !{ i32 1234567 }
2561 It is up to the front-end to make sense of the magic numbers it places
2562 in the IR. If the MDNode contains multiple constants, the code generator
2563 will use the one that corresponds to the line of the asm that the error
2568 Metadata Nodes and Metadata Strings
2569 -----------------------------------
2571 LLVM IR allows metadata to be attached to instructions in the program
2572 that can convey extra information about the code to the optimizers and
2573 code generator. One example application of metadata is source-level
2574 debug information. There are two metadata primitives: strings and nodes.
2575 All metadata has the ``metadata`` type and is identified in syntax by a
2576 preceding exclamation point ('``!``').
2578 A metadata string is a string surrounded by double quotes. It can
2579 contain any character by escaping non-printable characters with
2580 "``\xx``" where "``xx``" is the two digit hex code. For example:
2583 Metadata nodes are represented with notation similar to structure
2584 constants (a comma separated list of elements, surrounded by braces and
2585 preceded by an exclamation point). Metadata nodes can have any values as
2586 their operand. For example:
2588 .. code-block:: llvm
2590 !{ metadata !"test\00", i32 10}
2592 A :ref:`named metadata <namedmetadatastructure>` is a collection of
2593 metadata nodes, which can be looked up in the module symbol table. For
2596 .. code-block:: llvm
2598 !foo = metadata !{!4, !3}
2600 Metadata can be used as function arguments. Here ``llvm.dbg.value``
2601 function is using two metadata arguments:
2603 .. code-block:: llvm
2605 call void @llvm.dbg.value(metadata !24, i64 0, metadata !25)
2607 Metadata can be attached with an instruction. Here metadata ``!21`` is
2608 attached to the ``add`` instruction using the ``!dbg`` identifier:
2610 .. code-block:: llvm
2612 %indvar.next = add i64 %indvar, 1, !dbg !21
2614 More information about specific metadata nodes recognized by the
2615 optimizers and code generator is found below.
2620 In LLVM IR, memory does not have types, so LLVM's own type system is not
2621 suitable for doing TBAA. Instead, metadata is added to the IR to
2622 describe a type system of a higher level language. This can be used to
2623 implement typical C/C++ TBAA, but it can also be used to implement
2624 custom alias analysis behavior for other languages.
2626 The current metadata format is very simple. TBAA metadata nodes have up
2627 to three fields, e.g.:
2629 .. code-block:: llvm
2631 !0 = metadata !{ metadata !"an example type tree" }
2632 !1 = metadata !{ metadata !"int", metadata !0 }
2633 !2 = metadata !{ metadata !"float", metadata !0 }
2634 !3 = metadata !{ metadata !"const float", metadata !2, i64 1 }
2636 The first field is an identity field. It can be any value, usually a
2637 metadata string, which uniquely identifies the type. The most important
2638 name in the tree is the name of the root node. Two trees with different
2639 root node names are entirely disjoint, even if they have leaves with
2642 The second field identifies the type's parent node in the tree, or is
2643 null or omitted for a root node. A type is considered to alias all of
2644 its descendants and all of its ancestors in the tree. Also, a type is
2645 considered to alias all types in other trees, so that bitcode produced
2646 from multiple front-ends is handled conservatively.
2648 If the third field is present, it's an integer which if equal to 1
2649 indicates that the type is "constant" (meaning
2650 ``pointsToConstantMemory`` should return true; see `other useful
2651 AliasAnalysis methods <AliasAnalysis.html#OtherItfs>`_).
2653 '``tbaa.struct``' Metadata
2654 ^^^^^^^^^^^^^^^^^^^^^^^^^^
2656 The :ref:`llvm.memcpy <int_memcpy>` is often used to implement
2657 aggregate assignment operations in C and similar languages, however it
2658 is defined to copy a contiguous region of memory, which is more than
2659 strictly necessary for aggregate types which contain holes due to
2660 padding. Also, it doesn't contain any TBAA information about the fields
2663 ``!tbaa.struct`` metadata can describe which memory subregions in a
2664 memcpy are padding and what the TBAA tags of the struct are.
2666 The current metadata format is very simple. ``!tbaa.struct`` metadata
2667 nodes are a list of operands which are in conceptual groups of three.
2668 For each group of three, the first operand gives the byte offset of a
2669 field in bytes, the second gives its size in bytes, and the third gives
2672 .. code-block:: llvm
2674 !4 = metadata !{ i64 0, i64 4, metadata !1, i64 8, i64 4, metadata !2 }
2676 This describes a struct with two fields. The first is at offset 0 bytes
2677 with size 4 bytes, and has tbaa tag !1. The second is at offset 8 bytes
2678 and has size 4 bytes and has tbaa tag !2.
2680 Note that the fields need not be contiguous. In this example, there is a
2681 4 byte gap between the two fields. This gap represents padding which
2682 does not carry useful data and need not be preserved.
2684 '``fpmath``' Metadata
2685 ^^^^^^^^^^^^^^^^^^^^^
2687 ``fpmath`` metadata may be attached to any instruction of floating point
2688 type. It can be used to express the maximum acceptable error in the
2689 result of that instruction, in ULPs, thus potentially allowing the
2690 compiler to use a more efficient but less accurate method of computing
2691 it. ULP is defined as follows:
2693 If ``x`` is a real number that lies between two finite consecutive
2694 floating-point numbers ``a`` and ``b``, without being equal to one
2695 of them, then ``ulp(x) = |b - a|``, otherwise ``ulp(x)`` is the
2696 distance between the two non-equal finite floating-point numbers
2697 nearest ``x``. Moreover, ``ulp(NaN)`` is ``NaN``.
2699 The metadata node shall consist of a single positive floating point
2700 number representing the maximum relative error, for example:
2702 .. code-block:: llvm
2704 !0 = metadata !{ float 2.5 } ; maximum acceptable inaccuracy is 2.5 ULPs
2706 '``range``' Metadata
2707 ^^^^^^^^^^^^^^^^^^^^
2709 ``range`` metadata may be attached only to loads of integer types. It
2710 expresses the possible ranges the loaded value is in. The ranges are
2711 represented with a flattened list of integers. The loaded value is known
2712 to be in the union of the ranges defined by each consecutive pair. Each
2713 pair has the following properties:
2715 - The type must match the type loaded by the instruction.
2716 - The pair ``a,b`` represents the range ``[a,b)``.
2717 - Both ``a`` and ``b`` are constants.
2718 - The range is allowed to wrap.
2719 - The range should not represent the full or empty set. That is,
2722 In addition, the pairs must be in signed order of the lower bound and
2723 they must be non-contiguous.
2727 .. code-block:: llvm
2729 %a = load i8* %x, align 1, !range !0 ; Can only be 0 or 1
2730 %b = load i8* %y, align 1, !range !1 ; Can only be 255 (-1), 0 or 1
2731 %c = load i8* %z, align 1, !range !2 ; Can only be 0, 1, 3, 4 or 5
2732 %d = load i8* %z, align 1, !range !3 ; Can only be -2, -1, 3, 4 or 5
2734 !0 = metadata !{ i8 0, i8 2 }
2735 !1 = metadata !{ i8 255, i8 2 }
2736 !2 = metadata !{ i8 0, i8 2, i8 3, i8 6 }
2737 !3 = metadata !{ i8 -2, i8 0, i8 3, i8 6 }
2742 It is sometimes useful to attach information to loop constructs. Currently,
2743 loop metadata is implemented as metadata attached to the branch instruction
2744 in the loop latch block. This type of metadata refer to a metadata node that is
2745 guaranteed to be separate for each loop. The loop identifier metadata is
2746 specified with the name ``llvm.loop``.
2748 The loop identifier metadata is implemented using a metadata that refers to
2749 itself to avoid merging it with any other identifier metadata, e.g.,
2750 during module linkage or function inlining. That is, each loop should refer
2751 to their own identification metadata even if they reside in separate functions.
2752 The following example contains loop identifier metadata for two separate loop
2755 .. code-block:: llvm
2757 !0 = metadata !{ metadata !0 }
2758 !1 = metadata !{ metadata !1 }
2760 The loop identifier metadata can be used to specify additional per-loop
2761 metadata. Any operands after the first operand can be treated as user-defined
2762 metadata. For example the ``llvm.vectorizer.unroll`` metadata is understood
2763 by the loop vectorizer to indicate how many times to unroll the loop:
2765 .. code-block:: llvm
2767 br i1 %exitcond, label %._crit_edge, label %.lr.ph, !llvm.loop !0
2769 !0 = metadata !{ metadata !0, metadata !1 }
2770 !1 = metadata !{ metadata !"llvm.vectorizer.unroll", i32 2 }
2775 Metadata types used to annotate memory accesses with information helpful
2776 for optimizations are prefixed with ``llvm.mem``.
2778 '``llvm.mem.parallel_loop_access``' Metadata
2779 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2781 For a loop to be parallel, in addition to using
2782 the ``llvm.loop`` metadata to mark the loop latch branch instruction,
2783 also all of the memory accessing instructions in the loop body need to be
2784 marked with the ``llvm.mem.parallel_loop_access`` metadata. If there
2785 is at least one memory accessing instruction not marked with the metadata,
2786 the loop must be considered a sequential loop. This causes parallel loops to be
2787 converted to sequential loops due to optimization passes that are unaware of
2788 the parallel semantics and that insert new memory instructions to the loop
2791 Example of a loop that is considered parallel due to its correct use of
2792 both ``llvm.loop`` and ``llvm.mem.parallel_loop_access``
2793 metadata types that refer to the same loop identifier metadata.
2795 .. code-block:: llvm
2799 %val0 = load i32* %arrayidx, !llvm.mem.parallel_loop_access !0
2801 store i32 %val0, i32* %arrayidx1, !llvm.mem.parallel_loop_access !0
2803 br i1 %exitcond, label %for.end, label %for.body, !llvm.loop !0
2807 !0 = metadata !{ metadata !0 }
2809 It is also possible to have nested parallel loops. In that case the
2810 memory accesses refer to a list of loop identifier metadata nodes instead of
2811 the loop identifier metadata node directly:
2813 .. code-block:: llvm
2817 %val1 = load i32* %arrayidx3, !llvm.mem.parallel_loop_access !2
2819 br label %inner.for.body
2823 %val0 = load i32* %arrayidx1, !llvm.mem.parallel_loop_access !0
2825 store i32 %val0, i32* %arrayidx2, !llvm.mem.parallel_loop_access !0
2827 br i1 %exitcond, label %inner.for.end, label %inner.for.body, !llvm.loop !1
2831 store i32 %val1, i32* %arrayidx4, !llvm.mem.parallel_loop_access !2
2833 br i1 %exitcond, label %outer.for.end, label %outer.for.body, !llvm.loop !2
2835 outer.for.end: ; preds = %for.body
2837 !0 = metadata !{ metadata !1, metadata !2 } ; a list of loop identifiers
2838 !1 = metadata !{ metadata !1 } ; an identifier for the inner loop
2839 !2 = metadata !{ metadata !2 } ; an identifier for the outer loop
2841 '``llvm.vectorizer``'
2842 ^^^^^^^^^^^^^^^^^^^^^
2844 Metadata prefixed with ``llvm.vectorizer`` is used to control per-loop
2845 vectorization parameters such as vectorization factor and unroll factor.
2847 ``llvm.vectorizer`` metadata should be used in conjunction with ``llvm.loop``
2848 loop identification metadata.
2850 '``llvm.vectorizer.unroll``' Metadata
2851 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2853 This metadata instructs the loop vectorizer to unroll the specified
2854 loop exactly ``N`` times.
2856 The first operand is the string ``llvm.vectorizer.unroll`` and the second
2857 operand is an integer specifying the unroll factor. For example:
2859 .. code-block:: llvm
2861 !0 = metadata !{ metadata !"llvm.vectorizer.unroll", i32 4 }
2863 Note that setting ``llvm.vectorizer.unroll`` to 1 disables unrolling of the
2866 If ``llvm.vectorizer.unroll`` is set to 0 then the amount of unrolling will be
2867 determined automatically.
2869 '``llvm.vectorizer.width``' Metadata
2870 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2872 This metadata sets the target width of the vectorizer to ``N``. Without
2873 this metadata, the vectorizer will choose a width automatically.
2874 Regardless of this metadata, the vectorizer will only vectorize loops if
2875 it believes it is valid to do so.
2877 The first operand is the string ``llvm.vectorizer.width`` and the second
2878 operand is an integer specifying the width. For example:
2880 .. code-block:: llvm
2882 !0 = metadata !{ metadata !"llvm.vectorizer.width", i32 4 }
2884 Note that setting ``llvm.vectorizer.width`` to 1 disables vectorization of the
2887 If ``llvm.vectorizer.width`` is set to 0 then the width will be determined
2890 Module Flags Metadata
2891 =====================
2893 Information about the module as a whole is difficult to convey to LLVM's
2894 subsystems. The LLVM IR isn't sufficient to transmit this information.
2895 The ``llvm.module.flags`` named metadata exists in order to facilitate
2896 this. These flags are in the form of key / value pairs --- much like a
2897 dictionary --- making it easy for any subsystem who cares about a flag to
2900 The ``llvm.module.flags`` metadata contains a list of metadata triplets.
2901 Each triplet has the following form:
2903 - The first element is a *behavior* flag, which specifies the behavior
2904 when two (or more) modules are merged together, and it encounters two
2905 (or more) metadata with the same ID. The supported behaviors are
2907 - The second element is a metadata string that is a unique ID for the
2908 metadata. Each module may only have one flag entry for each unique ID (not
2909 including entries with the **Require** behavior).
2910 - The third element is the value of the flag.
2912 When two (or more) modules are merged together, the resulting
2913 ``llvm.module.flags`` metadata is the union of the modules' flags. That is, for
2914 each unique metadata ID string, there will be exactly one entry in the merged
2915 modules ``llvm.module.flags`` metadata table, and the value for that entry will
2916 be determined by the merge behavior flag, as described below. The only exception
2917 is that entries with the *Require* behavior are always preserved.
2919 The following behaviors are supported:
2930 Emits an error if two values disagree, otherwise the resulting value
2931 is that of the operands.
2935 Emits a warning if two values disagree. The result value will be the
2936 operand for the flag from the first module being linked.
2940 Adds a requirement that another module flag be present and have a
2941 specified value after linking is performed. The value must be a
2942 metadata pair, where the first element of the pair is the ID of the
2943 module flag to be restricted, and the second element of the pair is
2944 the value the module flag should be restricted to. This behavior can
2945 be used to restrict the allowable results (via triggering of an
2946 error) of linking IDs with the **Override** behavior.
2950 Uses the specified value, regardless of the behavior or value of the
2951 other module. If both modules specify **Override**, but the values
2952 differ, an error will be emitted.
2956 Appends the two values, which are required to be metadata nodes.
2960 Appends the two values, which are required to be metadata
2961 nodes. However, duplicate entries in the second list are dropped
2962 during the append operation.
2964 It is an error for a particular unique flag ID to have multiple behaviors,
2965 except in the case of **Require** (which adds restrictions on another metadata
2966 value) or **Override**.
2968 An example of module flags:
2970 .. code-block:: llvm
2972 !0 = metadata !{ i32 1, metadata !"foo", i32 1 }
2973 !1 = metadata !{ i32 4, metadata !"bar", i32 37 }
2974 !2 = metadata !{ i32 2, metadata !"qux", i32 42 }
2975 !3 = metadata !{ i32 3, metadata !"qux",
2977 metadata !"foo", i32 1
2980 !llvm.module.flags = !{ !0, !1, !2, !3 }
2982 - Metadata ``!0`` has the ID ``!"foo"`` and the value '1'. The behavior
2983 if two or more ``!"foo"`` flags are seen is to emit an error if their
2984 values are not equal.
2986 - Metadata ``!1`` has the ID ``!"bar"`` and the value '37'. The
2987 behavior if two or more ``!"bar"`` flags are seen is to use the value
2990 - Metadata ``!2`` has the ID ``!"qux"`` and the value '42'. The
2991 behavior if two or more ``!"qux"`` flags are seen is to emit a
2992 warning if their values are not equal.
2994 - Metadata ``!3`` has the ID ``!"qux"`` and the value:
2998 metadata !{ metadata !"foo", i32 1 }
3000 The behavior is to emit an error if the ``llvm.module.flags`` does not
3001 contain a flag with the ID ``!"foo"`` that has the value '1' after linking is
3004 Objective-C Garbage Collection Module Flags Metadata
3005 ----------------------------------------------------
3007 On the Mach-O platform, Objective-C stores metadata about garbage
3008 collection in a special section called "image info". The metadata
3009 consists of a version number and a bitmask specifying what types of
3010 garbage collection are supported (if any) by the file. If two or more
3011 modules are linked together their garbage collection metadata needs to
3012 be merged rather than appended together.
3014 The Objective-C garbage collection module flags metadata consists of the
3015 following key-value pairs:
3024 * - ``Objective-C Version``
3025 - **[Required]** --- The Objective-C ABI version. Valid values are 1 and 2.
3027 * - ``Objective-C Image Info Version``
3028 - **[Required]** --- The version of the image info section. Currently
3031 * - ``Objective-C Image Info Section``
3032 - **[Required]** --- The section to place the metadata. Valid values are
3033 ``"__OBJC, __image_info, regular"`` for Objective-C ABI version 1, and
3034 ``"__DATA,__objc_imageinfo, regular, no_dead_strip"`` for
3035 Objective-C ABI version 2.
3037 * - ``Objective-C Garbage Collection``
3038 - **[Required]** --- Specifies whether garbage collection is supported or
3039 not. Valid values are 0, for no garbage collection, and 2, for garbage
3040 collection supported.
3042 * - ``Objective-C GC Only``
3043 - **[Optional]** --- Specifies that only garbage collection is supported.
3044 If present, its value must be 6. This flag requires that the
3045 ``Objective-C Garbage Collection`` flag have the value 2.
3047 Some important flag interactions:
3049 - If a module with ``Objective-C Garbage Collection`` set to 0 is
3050 merged with a module with ``Objective-C Garbage Collection`` set to
3051 2, then the resulting module has the
3052 ``Objective-C Garbage Collection`` flag set to 0.
3053 - A module with ``Objective-C Garbage Collection`` set to 0 cannot be
3054 merged with a module with ``Objective-C GC Only`` set to 6.
3056 Automatic Linker Flags Module Flags Metadata
3057 --------------------------------------------
3059 Some targets support embedding flags to the linker inside individual object
3060 files. Typically this is used in conjunction with language extensions which
3061 allow source files to explicitly declare the libraries they depend on, and have
3062 these automatically be transmitted to the linker via object files.
3064 These flags are encoded in the IR using metadata in the module flags section,
3065 using the ``Linker Options`` key. The merge behavior for this flag is required
3066 to be ``AppendUnique``, and the value for the key is expected to be a metadata
3067 node which should be a list of other metadata nodes, each of which should be a
3068 list of metadata strings defining linker options.
3070 For example, the following metadata section specifies two separate sets of
3071 linker options, presumably to link against ``libz`` and the ``Cocoa``
3074 !0 = metadata !{ i32 6, metadata !"Linker Options",
3076 metadata !{ metadata !"-lz" },
3077 metadata !{ metadata !"-framework", metadata !"Cocoa" } } }
3078 !llvm.module.flags = !{ !0 }
3080 The metadata encoding as lists of lists of options, as opposed to a collapsed
3081 list of options, is chosen so that the IR encoding can use multiple option
3082 strings to specify e.g., a single library, while still having that specifier be
3083 preserved as an atomic element that can be recognized by a target specific
3084 assembly writer or object file emitter.
3086 Each individual option is required to be either a valid option for the target's
3087 linker, or an option that is reserved by the target specific assembly writer or
3088 object file emitter. No other aspect of these options is defined by the IR.
3090 .. _intrinsicglobalvariables:
3092 Intrinsic Global Variables
3093 ==========================
3095 LLVM has a number of "magic" global variables that contain data that
3096 affect code generation or other IR semantics. These are documented here.
3097 All globals of this sort should have a section specified as
3098 "``llvm.metadata``". This section and all globals that start with
3099 "``llvm.``" are reserved for use by LLVM.
3103 The '``llvm.used``' Global Variable
3104 -----------------------------------
3106 The ``@llvm.used`` global is an array which has
3107 :ref:`appending linkage <linkage_appending>`. This array contains a list of
3108 pointers to named global variables, functions and aliases which may optionally
3109 have a pointer cast formed of bitcast or getelementptr. For example, a legal
3112 .. code-block:: llvm
3117 @llvm.used = appending global [2 x i8*] [
3119 i8* bitcast (i32* @Y to i8*)
3120 ], section "llvm.metadata"
3122 If a symbol appears in the ``@llvm.used`` list, then the compiler, assembler,
3123 and linker are required to treat the symbol as if there is a reference to the
3124 symbol that it cannot see (which is why they have to be named). For example, if
3125 a variable has internal linkage and no references other than that from the
3126 ``@llvm.used`` list, it cannot be deleted. This is commonly used to represent
3127 references from inline asms and other things the compiler cannot "see", and
3128 corresponds to "``attribute((used))``" in GNU C.
3130 On some targets, the code generator must emit a directive to the
3131 assembler or object file to prevent the assembler and linker from
3132 molesting the symbol.
3134 .. _gv_llvmcompilerused:
3136 The '``llvm.compiler.used``' Global Variable
3137 --------------------------------------------
3139 The ``@llvm.compiler.used`` directive is the same as the ``@llvm.used``
3140 directive, except that it only prevents the compiler from touching the
3141 symbol. On targets that support it, this allows an intelligent linker to
3142 optimize references to the symbol without being impeded as it would be
3145 This is a rare construct that should only be used in rare circumstances,
3146 and should not be exposed to source languages.
3148 .. _gv_llvmglobalctors:
3150 The '``llvm.global_ctors``' Global Variable
3151 -------------------------------------------
3153 .. code-block:: llvm
3155 %0 = type { i32, void ()*, i8* }
3156 @llvm.global_ctors = appending global [1 x %0] [%0 { i32 65535, void ()* @ctor, i8* @data }]
3158 The ``@llvm.global_ctors`` array contains a list of constructor
3159 functions, priorities, and an optional associated global or function.
3160 The functions referenced by this array will be called in ascending order
3161 of priority (i.e. lowest first) when the module is loaded. The order of
3162 functions with the same priority is not defined.
3164 If the third field is present, non-null, and points to a global variable
3165 or function, the initializer function will only run if the associated
3166 data from the current module is not discarded.
3168 .. _llvmglobaldtors:
3170 The '``llvm.global_dtors``' Global Variable
3171 -------------------------------------------
3173 .. code-block:: llvm
3175 %0 = type { i32, void ()*, i8* }
3176 @llvm.global_dtors = appending global [1 x %0] [%0 { i32 65535, void ()* @dtor, i8* @data }]
3178 The ``@llvm.global_dtors`` array contains a list of destructor
3179 functions, priorities, and an optional associated global or function.
3180 The functions referenced by this array will be called in descending
3181 order of priority (i.e. highest first) when the module is loaded. The
3182 order of functions with the same priority is not defined.
3184 If the third field is present, non-null, and points to a global variable
3185 or function, the destructor function will only run if the associated
3186 data from the current module is not discarded.
3188 Instruction Reference
3189 =====================
3191 The LLVM instruction set consists of several different classifications
3192 of instructions: :ref:`terminator instructions <terminators>`, :ref:`binary
3193 instructions <binaryops>`, :ref:`bitwise binary
3194 instructions <bitwiseops>`, :ref:`memory instructions <memoryops>`, and
3195 :ref:`other instructions <otherops>`.
3199 Terminator Instructions
3200 -----------------------
3202 As mentioned :ref:`previously <functionstructure>`, every basic block in a
3203 program ends with a "Terminator" instruction, which indicates which
3204 block should be executed after the current block is finished. These
3205 terminator instructions typically yield a '``void``' value: they produce
3206 control flow, not values (the one exception being the
3207 ':ref:`invoke <i_invoke>`' instruction).
3209 The terminator instructions are: ':ref:`ret <i_ret>`',
3210 ':ref:`br <i_br>`', ':ref:`switch <i_switch>`',
3211 ':ref:`indirectbr <i_indirectbr>`', ':ref:`invoke <i_invoke>`',
3212 ':ref:`resume <i_resume>`', and ':ref:`unreachable <i_unreachable>`'.
3216 '``ret``' Instruction
3217 ^^^^^^^^^^^^^^^^^^^^^
3224 ret <type> <value> ; Return a value from a non-void function
3225 ret void ; Return from void function
3230 The '``ret``' instruction is used to return control flow (and optionally
3231 a value) from a function back to the caller.
3233 There are two forms of the '``ret``' instruction: one that returns a
3234 value and then causes control flow, and one that just causes control
3240 The '``ret``' instruction optionally accepts a single argument, the
3241 return value. The type of the return value must be a ':ref:`first
3242 class <t_firstclass>`' type.
3244 A function is not :ref:`well formed <wellformed>` if it it has a non-void
3245 return type and contains a '``ret``' instruction with no return value or
3246 a return value with a type that does not match its type, or if it has a
3247 void return type and contains a '``ret``' instruction with a return
3253 When the '``ret``' instruction is executed, control flow returns back to
3254 the calling function's context. If the caller is a
3255 ":ref:`call <i_call>`" instruction, execution continues at the
3256 instruction after the call. If the caller was an
3257 ":ref:`invoke <i_invoke>`" instruction, execution continues at the
3258 beginning of the "normal" destination block. If the instruction returns
3259 a value, that value shall set the call or invoke instruction's return
3265 .. code-block:: llvm
3267 ret i32 5 ; Return an integer value of 5
3268 ret void ; Return from a void function
3269 ret { i32, i8 } { i32 4, i8 2 } ; Return a struct of values 4 and 2
3273 '``br``' Instruction
3274 ^^^^^^^^^^^^^^^^^^^^
3281 br i1 <cond>, label <iftrue>, label <iffalse>
3282 br label <dest> ; Unconditional branch
3287 The '``br``' instruction is used to cause control flow to transfer to a
3288 different basic block in the current function. There are two forms of
3289 this instruction, corresponding to a conditional branch and an
3290 unconditional branch.
3295 The conditional branch form of the '``br``' instruction takes a single
3296 '``i1``' value and two '``label``' values. The unconditional form of the
3297 '``br``' instruction takes a single '``label``' value as a target.
3302 Upon execution of a conditional '``br``' instruction, the '``i1``'
3303 argument is evaluated. If the value is ``true``, control flows to the
3304 '``iftrue``' ``label`` argument. If "cond" is ``false``, control flows
3305 to the '``iffalse``' ``label`` argument.
3310 .. code-block:: llvm
3313 %cond = icmp eq i32 %a, %b
3314 br i1 %cond, label %IfEqual, label %IfUnequal
3322 '``switch``' Instruction
3323 ^^^^^^^^^^^^^^^^^^^^^^^^
3330 switch <intty> <value>, label <defaultdest> [ <intty> <val>, label <dest> ... ]
3335 The '``switch``' instruction is used to transfer control flow to one of
3336 several different places. It is a generalization of the '``br``'
3337 instruction, allowing a branch to occur to one of many possible
3343 The '``switch``' instruction uses three parameters: an integer
3344 comparison value '``value``', a default '``label``' destination, and an
3345 array of pairs of comparison value constants and '``label``'s. The table
3346 is not allowed to contain duplicate constant entries.
3351 The ``switch`` instruction specifies a table of values and destinations.
3352 When the '``switch``' instruction is executed, this table is searched
3353 for the given value. If the value is found, control flow is transferred
3354 to the corresponding destination; otherwise, control flow is transferred
3355 to the default destination.
3360 Depending on properties of the target machine and the particular
3361 ``switch`` instruction, this instruction may be code generated in
3362 different ways. For example, it could be generated as a series of
3363 chained conditional branches or with a lookup table.
3368 .. code-block:: llvm
3370 ; Emulate a conditional br instruction
3371 %Val = zext i1 %value to i32
3372 switch i32 %Val, label %truedest [ i32 0, label %falsedest ]
3374 ; Emulate an unconditional br instruction
3375 switch i32 0, label %dest [ ]
3377 ; Implement a jump table:
3378 switch i32 %val, label %otherwise [ i32 0, label %onzero
3380 i32 2, label %ontwo ]
3384 '``indirectbr``' Instruction
3385 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
3392 indirectbr <somety>* <address>, [ label <dest1>, label <dest2>, ... ]
3397 The '``indirectbr``' instruction implements an indirect branch to a
3398 label within the current function, whose address is specified by
3399 "``address``". Address must be derived from a
3400 :ref:`blockaddress <blockaddress>` constant.
3405 The '``address``' argument is the address of the label to jump to. The
3406 rest of the arguments indicate the full set of possible destinations
3407 that the address may point to. Blocks are allowed to occur multiple
3408 times in the destination list, though this isn't particularly useful.
3410 This destination list is required so that dataflow analysis has an
3411 accurate understanding of the CFG.
3416 Control transfers to the block specified in the address argument. All
3417 possible destination blocks must be listed in the label list, otherwise
3418 this instruction has undefined behavior. This implies that jumps to
3419 labels defined in other functions have undefined behavior as well.
3424 This is typically implemented with a jump through a register.
3429 .. code-block:: llvm
3431 indirectbr i8* %Addr, [ label %bb1, label %bb2, label %bb3 ]
3435 '``invoke``' Instruction
3436 ^^^^^^^^^^^^^^^^^^^^^^^^
3443 <result> = invoke [cconv] [ret attrs] <ptr to function ty> <function ptr val>(<function args>) [fn attrs]
3444 to label <normal label> unwind label <exception label>
3449 The '``invoke``' instruction causes control to transfer to a specified
3450 function, with the possibility of control flow transfer to either the
3451 '``normal``' label or the '``exception``' label. If the callee function
3452 returns with the "``ret``" instruction, control flow will return to the
3453 "normal" label. If the callee (or any indirect callees) returns via the
3454 ":ref:`resume <i_resume>`" instruction or other exception handling
3455 mechanism, control is interrupted and continued at the dynamically
3456 nearest "exception" label.
3458 The '``exception``' label is a `landing
3459 pad <ExceptionHandling.html#overview>`_ for the exception. As such,
3460 '``exception``' label is required to have the
3461 ":ref:`landingpad <i_landingpad>`" instruction, which contains the
3462 information about the behavior of the program after unwinding happens,
3463 as its first non-PHI instruction. The restrictions on the
3464 "``landingpad``" instruction's tightly couples it to the "``invoke``"
3465 instruction, so that the important information contained within the
3466 "``landingpad``" instruction can't be lost through normal code motion.
3471 This instruction requires several arguments:
3473 #. The optional "cconv" marker indicates which :ref:`calling
3474 convention <callingconv>` the call should use. If none is
3475 specified, the call defaults to using C calling conventions.
3476 #. The optional :ref:`Parameter Attributes <paramattrs>` list for return
3477 values. Only '``zeroext``', '``signext``', and '``inreg``' attributes
3479 #. '``ptr to function ty``': shall be the signature of the pointer to
3480 function value being invoked. In most cases, this is a direct
3481 function invocation, but indirect ``invoke``'s are just as possible,
3482 branching off an arbitrary pointer to function value.
3483 #. '``function ptr val``': An LLVM value containing a pointer to a
3484 function to be invoked.
3485 #. '``function args``': argument list whose types match the function
3486 signature argument types and parameter attributes. All arguments must
3487 be of :ref:`first class <t_firstclass>` type. If the function signature
3488 indicates the function accepts a variable number of arguments, the
3489 extra arguments can be specified.
3490 #. '``normal label``': the label reached when the called function
3491 executes a '``ret``' instruction.
3492 #. '``exception label``': the label reached when a callee returns via
3493 the :ref:`resume <i_resume>` instruction or other exception handling
3495 #. The optional :ref:`function attributes <fnattrs>` list. Only
3496 '``noreturn``', '``nounwind``', '``readonly``' and '``readnone``'
3497 attributes are valid here.
3502 This instruction is designed to operate as a standard '``call``'
3503 instruction in most regards. The primary difference is that it
3504 establishes an association with a label, which is used by the runtime
3505 library to unwind the stack.
3507 This instruction is used in languages with destructors to ensure that
3508 proper cleanup is performed in the case of either a ``longjmp`` or a
3509 thrown exception. Additionally, this is important for implementation of
3510 '``catch``' clauses in high-level languages that support them.
3512 For the purposes of the SSA form, the definition of the value returned
3513 by the '``invoke``' instruction is deemed to occur on the edge from the
3514 current block to the "normal" label. If the callee unwinds then no
3515 return value is available.
3520 .. code-block:: llvm
3522 %retval = invoke i32 @Test(i32 15) to label %Continue
3523 unwind label %TestCleanup ; {i32}:retval set
3524 %retval = invoke coldcc i32 %Testfnptr(i32 15) to label %Continue
3525 unwind label %TestCleanup ; {i32}:retval set
3529 '``resume``' Instruction
3530 ^^^^^^^^^^^^^^^^^^^^^^^^
3537 resume <type> <value>
3542 The '``resume``' instruction is a terminator instruction that has no
3548 The '``resume``' instruction requires one argument, which must have the
3549 same type as the result of any '``landingpad``' instruction in the same
3555 The '``resume``' instruction resumes propagation of an existing
3556 (in-flight) exception whose unwinding was interrupted with a
3557 :ref:`landingpad <i_landingpad>` instruction.
3562 .. code-block:: llvm
3564 resume { i8*, i32 } %exn
3568 '``unreachable``' Instruction
3569 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
3581 The '``unreachable``' instruction has no defined semantics. This
3582 instruction is used to inform the optimizer that a particular portion of
3583 the code is not reachable. This can be used to indicate that the code
3584 after a no-return function cannot be reached, and other facts.
3589 The '``unreachable``' instruction has no defined semantics.
3596 Binary operators are used to do most of the computation in a program.
3597 They require two operands of the same type, execute an operation on
3598 them, and produce a single value. The operands might represent multiple
3599 data, as is the case with the :ref:`vector <t_vector>` data type. The
3600 result value has the same type as its operands.
3602 There are several different binary operators:
3606 '``add``' Instruction
3607 ^^^^^^^^^^^^^^^^^^^^^
3614 <result> = add <ty> <op1>, <op2> ; yields {ty}:result
3615 <result> = add nuw <ty> <op1>, <op2> ; yields {ty}:result
3616 <result> = add nsw <ty> <op1>, <op2> ; yields {ty}:result
3617 <result> = add nuw nsw <ty> <op1>, <op2> ; yields {ty}:result
3622 The '``add``' instruction returns the sum of its two operands.
3627 The two arguments to the '``add``' instruction must be
3628 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
3629 arguments must have identical types.
3634 The value produced is the integer sum of the two operands.
3636 If the sum has unsigned overflow, the result returned is the
3637 mathematical result modulo 2\ :sup:`n`\ , where n is the bit width of
3640 Because LLVM integers use a two's complement representation, this
3641 instruction is appropriate for both signed and unsigned integers.
3643 ``nuw`` and ``nsw`` stand for "No Unsigned Wrap" and "No Signed Wrap",
3644 respectively. If the ``nuw`` and/or ``nsw`` keywords are present, the
3645 result value of the ``add`` is a :ref:`poison value <poisonvalues>` if
3646 unsigned and/or signed overflow, respectively, occurs.
3651 .. code-block:: llvm
3653 <result> = add i32 4, %var ; yields {i32}:result = 4 + %var
3657 '``fadd``' Instruction
3658 ^^^^^^^^^^^^^^^^^^^^^^
3665 <result> = fadd [fast-math flags]* <ty> <op1>, <op2> ; yields {ty}:result
3670 The '``fadd``' instruction returns the sum of its two operands.
3675 The two arguments to the '``fadd``' instruction must be :ref:`floating
3676 point <t_floating>` or :ref:`vector <t_vector>` of floating point values.
3677 Both arguments must have identical types.
3682 The value produced is the floating point sum of the two operands. This
3683 instruction can also take any number of :ref:`fast-math flags <fastmath>`,
3684 which are optimization hints to enable otherwise unsafe floating point
3690 .. code-block:: llvm
3692 <result> = fadd float 4.0, %var ; yields {float}:result = 4.0 + %var
3694 '``sub``' Instruction
3695 ^^^^^^^^^^^^^^^^^^^^^
3702 <result> = sub <ty> <op1>, <op2> ; yields {ty}:result
3703 <result> = sub nuw <ty> <op1>, <op2> ; yields {ty}:result
3704 <result> = sub nsw <ty> <op1>, <op2> ; yields {ty}:result
3705 <result> = sub nuw nsw <ty> <op1>, <op2> ; yields {ty}:result
3710 The '``sub``' instruction returns the difference of its two operands.
3712 Note that the '``sub``' instruction is used to represent the '``neg``'
3713 instruction present in most other intermediate representations.
3718 The two arguments to the '``sub``' instruction must be
3719 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
3720 arguments must have identical types.
3725 The value produced is the integer difference of the two operands.
3727 If the difference has unsigned overflow, the result returned is the
3728 mathematical result modulo 2\ :sup:`n`\ , where n is the bit width of
3731 Because LLVM integers use a two's complement representation, this
3732 instruction is appropriate for both signed and unsigned integers.
3734 ``nuw`` and ``nsw`` stand for "No Unsigned Wrap" and "No Signed Wrap",
3735 respectively. If the ``nuw`` and/or ``nsw`` keywords are present, the
3736 result value of the ``sub`` is a :ref:`poison value <poisonvalues>` if
3737 unsigned and/or signed overflow, respectively, occurs.
3742 .. code-block:: llvm
3744 <result> = sub i32 4, %var ; yields {i32}:result = 4 - %var
3745 <result> = sub i32 0, %val ; yields {i32}:result = -%var
3749 '``fsub``' Instruction
3750 ^^^^^^^^^^^^^^^^^^^^^^
3757 <result> = fsub [fast-math flags]* <ty> <op1>, <op2> ; yields {ty}:result
3762 The '``fsub``' instruction returns the difference of its two operands.
3764 Note that the '``fsub``' instruction is used to represent the '``fneg``'
3765 instruction present in most other intermediate representations.
3770 The two arguments to the '``fsub``' instruction must be :ref:`floating
3771 point <t_floating>` or :ref:`vector <t_vector>` of floating point values.
3772 Both arguments must have identical types.
3777 The value produced is the floating point difference of the two operands.
3778 This instruction can also take any number of :ref:`fast-math
3779 flags <fastmath>`, which are optimization hints to enable otherwise
3780 unsafe floating point optimizations:
3785 .. code-block:: llvm
3787 <result> = fsub float 4.0, %var ; yields {float}:result = 4.0 - %var
3788 <result> = fsub float -0.0, %val ; yields {float}:result = -%var
3790 '``mul``' Instruction
3791 ^^^^^^^^^^^^^^^^^^^^^
3798 <result> = mul <ty> <op1>, <op2> ; yields {ty}:result
3799 <result> = mul nuw <ty> <op1>, <op2> ; yields {ty}:result
3800 <result> = mul nsw <ty> <op1>, <op2> ; yields {ty}:result
3801 <result> = mul nuw nsw <ty> <op1>, <op2> ; yields {ty}:result
3806 The '``mul``' instruction returns the product of its two operands.
3811 The two arguments to the '``mul``' instruction must be
3812 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
3813 arguments must have identical types.
3818 The value produced is the integer product of the two operands.
3820 If the result of the multiplication has unsigned overflow, the result
3821 returned is the mathematical result modulo 2\ :sup:`n`\ , where n is the
3822 bit width of the result.
3824 Because LLVM integers use a two's complement representation, and the
3825 result is the same width as the operands, this instruction returns the
3826 correct result for both signed and unsigned integers. If a full product
3827 (e.g. ``i32`` * ``i32`` -> ``i64``) is needed, the operands should be
3828 sign-extended or zero-extended as appropriate to the width of the full
3831 ``nuw`` and ``nsw`` stand for "No Unsigned Wrap" and "No Signed Wrap",
3832 respectively. If the ``nuw`` and/or ``nsw`` keywords are present, the
3833 result value of the ``mul`` is a :ref:`poison value <poisonvalues>` if
3834 unsigned and/or signed overflow, respectively, occurs.
3839 .. code-block:: llvm
3841 <result> = mul i32 4, %var ; yields {i32}:result = 4 * %var
3845 '``fmul``' Instruction
3846 ^^^^^^^^^^^^^^^^^^^^^^
3853 <result> = fmul [fast-math flags]* <ty> <op1>, <op2> ; yields {ty}:result
3858 The '``fmul``' instruction returns the product of its two operands.
3863 The two arguments to the '``fmul``' instruction must be :ref:`floating
3864 point <t_floating>` or :ref:`vector <t_vector>` of floating point values.
3865 Both arguments must have identical types.
3870 The value produced is the floating point product of the two operands.
3871 This instruction can also take any number of :ref:`fast-math
3872 flags <fastmath>`, which are optimization hints to enable otherwise
3873 unsafe floating point optimizations:
3878 .. code-block:: llvm
3880 <result> = fmul float 4.0, %var ; yields {float}:result = 4.0 * %var
3882 '``udiv``' Instruction
3883 ^^^^^^^^^^^^^^^^^^^^^^
3890 <result> = udiv <ty> <op1>, <op2> ; yields {ty}:result
3891 <result> = udiv exact <ty> <op1>, <op2> ; yields {ty}:result
3896 The '``udiv``' instruction returns the quotient of its two operands.
3901 The two arguments to the '``udiv``' instruction must be
3902 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
3903 arguments must have identical types.
3908 The value produced is the unsigned integer quotient of the two operands.
3910 Note that unsigned integer division and signed integer division are
3911 distinct operations; for signed integer division, use '``sdiv``'.
3913 Division by zero leads to undefined behavior.
3915 If the ``exact`` keyword is present, the result value of the ``udiv`` is
3916 a :ref:`poison value <poisonvalues>` if %op1 is not a multiple of %op2 (as
3917 such, "((a udiv exact b) mul b) == a").
3922 .. code-block:: llvm
3924 <result> = udiv i32 4, %var ; yields {i32}:result = 4 / %var
3926 '``sdiv``' Instruction
3927 ^^^^^^^^^^^^^^^^^^^^^^
3934 <result> = sdiv <ty> <op1>, <op2> ; yields {ty}:result
3935 <result> = sdiv exact <ty> <op1>, <op2> ; yields {ty}:result
3940 The '``sdiv``' instruction returns the quotient of its two operands.
3945 The two arguments to the '``sdiv``' instruction must be
3946 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
3947 arguments must have identical types.
3952 The value produced is the signed integer quotient of the two operands
3953 rounded towards zero.
3955 Note that signed integer division and unsigned integer division are
3956 distinct operations; for unsigned integer division, use '``udiv``'.
3958 Division by zero leads to undefined behavior. Overflow also leads to
3959 undefined behavior; this is a rare case, but can occur, for example, by
3960 doing a 32-bit division of -2147483648 by -1.
3962 If the ``exact`` keyword is present, the result value of the ``sdiv`` is
3963 a :ref:`poison value <poisonvalues>` if the result would be rounded.
3968 .. code-block:: llvm
3970 <result> = sdiv i32 4, %var ; yields {i32}:result = 4 / %var
3974 '``fdiv``' Instruction
3975 ^^^^^^^^^^^^^^^^^^^^^^
3982 <result> = fdiv [fast-math flags]* <ty> <op1>, <op2> ; yields {ty}:result
3987 The '``fdiv``' instruction returns the quotient of its two operands.
3992 The two arguments to the '``fdiv``' instruction must be :ref:`floating
3993 point <t_floating>` or :ref:`vector <t_vector>` of floating point values.
3994 Both arguments must have identical types.
3999 The value produced is the floating point quotient of the two operands.
4000 This instruction can also take any number of :ref:`fast-math
4001 flags <fastmath>`, which are optimization hints to enable otherwise
4002 unsafe floating point optimizations:
4007 .. code-block:: llvm
4009 <result> = fdiv float 4.0, %var ; yields {float}:result = 4.0 / %var
4011 '``urem``' Instruction
4012 ^^^^^^^^^^^^^^^^^^^^^^
4019 <result> = urem <ty> <op1>, <op2> ; yields {ty}:result
4024 The '``urem``' instruction returns the remainder from the unsigned
4025 division of its two arguments.
4030 The two arguments to the '``urem``' instruction must be
4031 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
4032 arguments must have identical types.
4037 This instruction returns the unsigned integer *remainder* of a division.
4038 This instruction always performs an unsigned division to get the
4041 Note that unsigned integer remainder and signed integer remainder are
4042 distinct operations; for signed integer remainder, use '``srem``'.
4044 Taking the remainder of a division by zero leads to undefined behavior.
4049 .. code-block:: llvm
4051 <result> = urem i32 4, %var ; yields {i32}:result = 4 % %var
4053 '``srem``' Instruction
4054 ^^^^^^^^^^^^^^^^^^^^^^
4061 <result> = srem <ty> <op1>, <op2> ; yields {ty}:result
4066 The '``srem``' instruction returns the remainder from the signed
4067 division of its two operands. This instruction can also take
4068 :ref:`vector <t_vector>` versions of the values in which case the elements
4074 The two arguments to the '``srem``' instruction must be
4075 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
4076 arguments must have identical types.
4081 This instruction returns the *remainder* of a division (where the result
4082 is either zero or has the same sign as the dividend, ``op1``), not the
4083 *modulo* operator (where the result is either zero or has the same sign
4084 as the divisor, ``op2``) of a value. For more information about the
4085 difference, see `The Math
4086 Forum <http://mathforum.org/dr.math/problems/anne.4.28.99.html>`_. For a
4087 table of how this is implemented in various languages, please see
4089 operation <http://en.wikipedia.org/wiki/Modulo_operation>`_.
4091 Note that signed integer remainder and unsigned integer remainder are
4092 distinct operations; for unsigned integer remainder, use '``urem``'.
4094 Taking the remainder of a division by zero leads to undefined behavior.
4095 Overflow also leads to undefined behavior; this is a rare case, but can
4096 occur, for example, by taking the remainder of a 32-bit division of
4097 -2147483648 by -1. (The remainder doesn't actually overflow, but this
4098 rule lets srem be implemented using instructions that return both the
4099 result of the division and the remainder.)
4104 .. code-block:: llvm
4106 <result> = srem i32 4, %var ; yields {i32}:result = 4 % %var
4110 '``frem``' Instruction
4111 ^^^^^^^^^^^^^^^^^^^^^^
4118 <result> = frem [fast-math flags]* <ty> <op1>, <op2> ; yields {ty}:result
4123 The '``frem``' instruction returns the remainder from the division of
4129 The two arguments to the '``frem``' instruction must be :ref:`floating
4130 point <t_floating>` or :ref:`vector <t_vector>` of floating point values.
4131 Both arguments must have identical types.
4136 This instruction returns the *remainder* of a division. The remainder
4137 has the same sign as the dividend. This instruction can also take any
4138 number of :ref:`fast-math flags <fastmath>`, which are optimization hints
4139 to enable otherwise unsafe floating point optimizations:
4144 .. code-block:: llvm
4146 <result> = frem float 4.0, %var ; yields {float}:result = 4.0 % %var
4150 Bitwise Binary Operations
4151 -------------------------
4153 Bitwise binary operators are used to do various forms of bit-twiddling
4154 in a program. They are generally very efficient instructions and can
4155 commonly be strength reduced from other instructions. They require two
4156 operands of the same type, execute an operation on them, and produce a
4157 single value. The resulting value is the same type as its operands.
4159 '``shl``' Instruction
4160 ^^^^^^^^^^^^^^^^^^^^^
4167 <result> = shl <ty> <op1>, <op2> ; yields {ty}:result
4168 <result> = shl nuw <ty> <op1>, <op2> ; yields {ty}:result
4169 <result> = shl nsw <ty> <op1>, <op2> ; yields {ty}:result
4170 <result> = shl nuw nsw <ty> <op1>, <op2> ; yields {ty}:result
4175 The '``shl``' instruction returns the first operand shifted to the left
4176 a specified number of bits.
4181 Both arguments to the '``shl``' instruction must be the same
4182 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer type.
4183 '``op2``' is treated as an unsigned value.
4188 The value produced is ``op1`` \* 2\ :sup:`op2` mod 2\ :sup:`n`,
4189 where ``n`` is the width of the result. If ``op2`` is (statically or
4190 dynamically) negative or equal to or larger than the number of bits in
4191 ``op1``, the result is undefined. If the arguments are vectors, each
4192 vector element of ``op1`` is shifted by the corresponding shift amount
4195 If the ``nuw`` keyword is present, then the shift produces a :ref:`poison
4196 value <poisonvalues>` if it shifts out any non-zero bits. If the
4197 ``nsw`` keyword is present, then the shift produces a :ref:`poison
4198 value <poisonvalues>` if it shifts out any bits that disagree with the
4199 resultant sign bit. As such, NUW/NSW have the same semantics as they
4200 would if the shift were expressed as a mul instruction with the same
4201 nsw/nuw bits in (mul %op1, (shl 1, %op2)).
4206 .. code-block:: llvm
4208 <result> = shl i32 4, %var ; yields {i32}: 4 << %var
4209 <result> = shl i32 4, 2 ; yields {i32}: 16
4210 <result> = shl i32 1, 10 ; yields {i32}: 1024
4211 <result> = shl i32 1, 32 ; undefined
4212 <result> = shl <2 x i32> < i32 1, i32 1>, < i32 1, i32 2> ; yields: result=<2 x i32> < i32 2, i32 4>
4214 '``lshr``' Instruction
4215 ^^^^^^^^^^^^^^^^^^^^^^
4222 <result> = lshr <ty> <op1>, <op2> ; yields {ty}:result
4223 <result> = lshr exact <ty> <op1>, <op2> ; yields {ty}:result
4228 The '``lshr``' instruction (logical shift right) returns the first
4229 operand shifted to the right a specified number of bits with zero fill.
4234 Both arguments to the '``lshr``' instruction must be the same
4235 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer type.
4236 '``op2``' is treated as an unsigned value.
4241 This instruction always performs a logical shift right operation. The
4242 most significant bits of the result will be filled with zero bits after
4243 the shift. If ``op2`` is (statically or dynamically) equal to or larger
4244 than the number of bits in ``op1``, the result is undefined. If the
4245 arguments are vectors, each vector element of ``op1`` is shifted by the
4246 corresponding shift amount in ``op2``.
4248 If the ``exact`` keyword is present, the result value of the ``lshr`` is
4249 a :ref:`poison value <poisonvalues>` if any of the bits shifted out are
4255 .. code-block:: llvm
4257 <result> = lshr i32 4, 1 ; yields {i32}:result = 2
4258 <result> = lshr i32 4, 2 ; yields {i32}:result = 1
4259 <result> = lshr i8 4, 3 ; yields {i8}:result = 0
4260 <result> = lshr i8 -2, 1 ; yields {i8}:result = 0x7F
4261 <result> = lshr i32 1, 32 ; undefined
4262 <result> = lshr <2 x i32> < i32 -2, i32 4>, < i32 1, i32 2> ; yields: result=<2 x i32> < i32 0x7FFFFFFF, i32 1>
4264 '``ashr``' Instruction
4265 ^^^^^^^^^^^^^^^^^^^^^^
4272 <result> = ashr <ty> <op1>, <op2> ; yields {ty}:result
4273 <result> = ashr exact <ty> <op1>, <op2> ; yields {ty}:result
4278 The '``ashr``' instruction (arithmetic shift right) returns the first
4279 operand shifted to the right a specified number of bits with sign
4285 Both arguments to the '``ashr``' instruction must be the same
4286 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer type.
4287 '``op2``' is treated as an unsigned value.
4292 This instruction always performs an arithmetic shift right operation,
4293 The most significant bits of the result will be filled with the sign bit
4294 of ``op1``. If ``op2`` is (statically or dynamically) equal to or larger
4295 than the number of bits in ``op1``, the result is undefined. If the
4296 arguments are vectors, each vector element of ``op1`` is shifted by the
4297 corresponding shift amount in ``op2``.
4299 If the ``exact`` keyword is present, the result value of the ``ashr`` is
4300 a :ref:`poison value <poisonvalues>` if any of the bits shifted out are
4306 .. code-block:: llvm
4308 <result> = ashr i32 4, 1 ; yields {i32}:result = 2
4309 <result> = ashr i32 4, 2 ; yields {i32}:result = 1
4310 <result> = ashr i8 4, 3 ; yields {i8}:result = 0
4311 <result> = ashr i8 -2, 1 ; yields {i8}:result = -1
4312 <result> = ashr i32 1, 32 ; undefined
4313 <result> = ashr <2 x i32> < i32 -2, i32 4>, < i32 1, i32 3> ; yields: result=<2 x i32> < i32 -1, i32 0>
4315 '``and``' Instruction
4316 ^^^^^^^^^^^^^^^^^^^^^
4323 <result> = and <ty> <op1>, <op2> ; yields {ty}:result
4328 The '``and``' instruction returns the bitwise logical and of its two
4334 The two arguments to the '``and``' instruction must be
4335 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
4336 arguments must have identical types.
4341 The truth table used for the '``and``' instruction is:
4358 .. code-block:: llvm
4360 <result> = and i32 4, %var ; yields {i32}:result = 4 & %var
4361 <result> = and i32 15, 40 ; yields {i32}:result = 8
4362 <result> = and i32 4, 8 ; yields {i32}:result = 0
4364 '``or``' Instruction
4365 ^^^^^^^^^^^^^^^^^^^^
4372 <result> = or <ty> <op1>, <op2> ; yields {ty}:result
4377 The '``or``' instruction returns the bitwise logical inclusive or of its
4383 The two arguments to the '``or``' instruction must be
4384 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
4385 arguments must have identical types.
4390 The truth table used for the '``or``' instruction is:
4409 <result> = or i32 4, %var ; yields {i32}:result = 4 | %var
4410 <result> = or i32 15, 40 ; yields {i32}:result = 47
4411 <result> = or i32 4, 8 ; yields {i32}:result = 12
4413 '``xor``' Instruction
4414 ^^^^^^^^^^^^^^^^^^^^^
4421 <result> = xor <ty> <op1>, <op2> ; yields {ty}:result
4426 The '``xor``' instruction returns the bitwise logical exclusive or of
4427 its two operands. The ``xor`` is used to implement the "one's
4428 complement" operation, which is the "~" operator in C.
4433 The two arguments to the '``xor``' instruction must be
4434 :ref:`integer <t_integer>` or :ref:`vector <t_vector>` of integer values. Both
4435 arguments must have identical types.
4440 The truth table used for the '``xor``' instruction is:
4457 .. code-block:: llvm
4459 <result> = xor i32 4, %var ; yields {i32}:result = 4 ^ %var
4460 <result> = xor i32 15, 40 ; yields {i32}:result = 39
4461 <result> = xor i32 4, 8 ; yields {i32}:result = 12
4462 <result> = xor i32 %V, -1 ; yields {i32}:result = ~%V
4467 LLVM supports several instructions to represent vector operations in a
4468 target-independent manner. These instructions cover the element-access
4469 and vector-specific operations needed to process vectors effectively.
4470 While LLVM does directly support these vector operations, many
4471 sophisticated algorithms will want to use target-specific intrinsics to
4472 take full advantage of a specific target.
4474 .. _i_extractelement:
4476 '``extractelement``' Instruction
4477 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
4484 <result> = extractelement <n x <ty>> <val>, <ty2> <idx> ; yields <ty>
4489 The '``extractelement``' instruction extracts a single scalar element
4490 from a vector at a specified index.
4495 The first operand of an '``extractelement``' instruction is a value of
4496 :ref:`vector <t_vector>` type. The second operand is an index indicating
4497 the position from which to extract the element. The index may be a
4498 variable of any integer type.
4503 The result is a scalar of the same type as the element type of ``val``.
4504 Its value is the value at position ``idx`` of ``val``. If ``idx``
4505 exceeds the length of ``val``, the results are undefined.
4510 .. code-block:: llvm
4512 <result> = extractelement <4 x i32> %vec, i32 0 ; yields i32
4514 .. _i_insertelement:
4516 '``insertelement``' Instruction
4517 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
4524 <result> = insertelement <n x <ty>> <val>, <ty> <elt>, <ty2> <idx> ; yields <n x <ty>>
4529 The '``insertelement``' instruction inserts a scalar element into a
4530 vector at a specified index.
4535 The first operand of an '``insertelement``' instruction is a value of
4536 :ref:`vector <t_vector>` type. The second operand is a scalar value whose
4537 type must equal the element type of the first operand. The third operand
4538 is an index indicating the position at which to insert the value. The
4539 index may be a variable of any integer type.
4544 The result is a vector of the same type as ``val``. Its element values
4545 are those of ``val`` except at position ``idx``, where it gets the value
4546 ``elt``. If ``idx`` exceeds the length of ``val``, the results are
4552 .. code-block:: llvm
4554 <result> = insertelement <4 x i32> %vec, i32 1, i32 0 ; yields <4 x i32>
4556 .. _i_shufflevector:
4558 '``shufflevector``' Instruction
4559 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
4566 <result> = shufflevector <n x <ty>> <v1>, <n x <ty>> <v2>, <m x i32> <mask> ; yields <m x <ty>>
4571 The '``shufflevector``' instruction constructs a permutation of elements
4572 from two input vectors, returning a vector with the same element type as
4573 the input and length that is the same as the shuffle mask.
4578 The first two operands of a '``shufflevector``' instruction are vectors
4579 with the same type. The third argument is a shuffle mask whose element
4580 type is always 'i32'. The result of the instruction is a vector whose
4581 length is the same as the shuffle mask and whose element type is the
4582 same as the element type of the first two operands.
4584 The shuffle mask operand is required to be a constant vector with either
4585 constant integer or undef values.
4590 The elements of the two input vectors are numbered from left to right
4591 across both of the vectors. The shuffle mask operand specifies, for each
4592 element of the result vector, which element of the two input vectors the
4593 result element gets. The element selector may be undef (meaning "don't
4594 care") and the second operand may be undef if performing a shuffle from
4600 .. code-block:: llvm
4602 <result> = shufflevector <4 x i32> %v1, <4 x i32> %v2,
4603 <4 x i32> <i32 0, i32 4, i32 1, i32 5> ; yields <4 x i32>
4604 <result> = shufflevector <4 x i32> %v1, <4 x i32> undef,
4605 <4 x i32> <i32 0, i32 1, i32 2, i32 3> ; yields <4 x i32> - Identity shuffle.
4606 <result> = shufflevector <8 x i32> %v1, <8 x i32> undef,
4607 <4 x i32> <i32 0, i32 1, i32 2, i32 3> ; yields <4 x i32>
4608 <result> = shufflevector <4 x i32> %v1, <4 x i32> %v2,
4609 <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7 > ; yields <8 x i32>
4611 Aggregate Operations
4612 --------------------
4614 LLVM supports several instructions for working with
4615 :ref:`aggregate <t_aggregate>` values.
4619 '``extractvalue``' Instruction
4620 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
4627 <result> = extractvalue <aggregate type> <val>, <idx>{, <idx>}*
4632 The '``extractvalue``' instruction extracts the value of a member field
4633 from an :ref:`aggregate <t_aggregate>` value.
4638 The first operand of an '``extractvalue``' instruction is a value of
4639 :ref:`struct <t_struct>` or :ref:`array <t_array>` type. The operands are
4640 constant indices to specify which value to extract in a similar manner
4641 as indices in a '``getelementptr``' instruction.
4643 The major differences to ``getelementptr`` indexing are:
4645 - Since the value being indexed is not a pointer, the first index is
4646 omitted and assumed to be zero.
4647 - At least one index must be specified.
4648 - Not only struct indices but also array indices must be in bounds.
4653 The result is the value at the position in the aggregate specified by
4659 .. code-block:: llvm
4661 <result> = extractvalue {i32, float} %agg, 0 ; yields i32
4665 '``insertvalue``' Instruction
4666 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
4673 <result> = insertvalue <aggregate type> <val>, <ty> <elt>, <idx>{, <idx>}* ; yields <aggregate type>
4678 The '``insertvalue``' instruction inserts a value into a member field in
4679 an :ref:`aggregate <t_aggregate>` value.
4684 The first operand of an '``insertvalue``' instruction is a value of
4685 :ref:`struct <t_struct>` or :ref:`array <t_array>` type. The second operand is
4686 a first-class value to insert. The following operands are constant
4687 indices indicating the position at which to insert the value in a
4688 similar manner as indices in a '``extractvalue``' instruction. The value
4689 to insert must have the same type as the value identified by the
4695 The result is an aggregate of the same type as ``val``. Its value is
4696 that of ``val`` except that the value at the position specified by the
4697 indices is that of ``elt``.
4702 .. code-block:: llvm
4704 %agg1 = insertvalue {i32, float} undef, i32 1, 0 ; yields {i32 1, float undef}
4705 %agg2 = insertvalue {i32, float} %agg1, float %val, 1 ; yields {i32 1, float %val}
4706 %agg3 = insertvalue {i32, {float}} %agg1, float %val, 1, 0 ; yields {i32 1, float %val}
4710 Memory Access and Addressing Operations
4711 ---------------------------------------
4713 A key design point of an SSA-based representation is how it represents
4714 memory. In LLVM, no memory locations are in SSA form, which makes things
4715 very simple. This section describes how to read, write, and allocate
4720 '``alloca``' Instruction
4721 ^^^^^^^^^^^^^^^^^^^^^^^^
4728 <result> = alloca [inalloca] <type> [, <ty> <NumElements>] [, align <alignment>] ; yields {type*}:result
4733 The '``alloca``' instruction allocates memory on the stack frame of the
4734 currently executing function, to be automatically released when this
4735 function returns to its caller. The object is always allocated in the
4736 generic address space (address space zero).
4741 The '``alloca``' instruction allocates ``sizeof(<type>)*NumElements``
4742 bytes of memory on the runtime stack, returning a pointer of the
4743 appropriate type to the program. If "NumElements" is specified, it is
4744 the number of elements allocated, otherwise "NumElements" is defaulted
4745 to be one. If a constant alignment is specified, the value result of the
4746 allocation is guaranteed to be aligned to at least that boundary. If not
4747 specified, or if zero, the target can choose to align the allocation on
4748 any convenient boundary compatible with the type.
4750 '``type``' may be any sized type.
4755 Memory is allocated; a pointer is returned. The operation is undefined
4756 if there is insufficient stack space for the allocation. '``alloca``'d
4757 memory is automatically released when the function returns. The
4758 '``alloca``' instruction is commonly used to represent automatic
4759 variables that must have an address available. When the function returns
4760 (either with the ``ret`` or ``resume`` instructions), the memory is
4761 reclaimed. Allocating zero bytes is legal, but the result is undefined.
4762 The order in which memory is allocated (ie., which way the stack grows)
4768 .. code-block:: llvm
4770 %ptr = alloca i32 ; yields {i32*}:ptr
4771 %ptr = alloca i32, i32 4 ; yields {i32*}:ptr
4772 %ptr = alloca i32, i32 4, align 1024 ; yields {i32*}:ptr
4773 %ptr = alloca i32, align 1024 ; yields {i32*}:ptr
4777 '``load``' Instruction
4778 ^^^^^^^^^^^^^^^^^^^^^^
4785 <result> = load [volatile] <ty>* <pointer>[, align <alignment>][, !nontemporal !<index>][, !invariant.load !<index>]
4786 <result> = load atomic [volatile] <ty>* <pointer> [singlethread] <ordering>, align <alignment>
4787 !<index> = !{ i32 1 }
4792 The '``load``' instruction is used to read from memory.
4797 The argument to the ``load`` instruction specifies the memory address
4798 from which to load. The pointer must point to a :ref:`first
4799 class <t_firstclass>` type. If the ``load`` is marked as ``volatile``,
4800 then the optimizer is not allowed to modify the number or order of
4801 execution of this ``load`` with other :ref:`volatile
4802 operations <volatile>`.
4804 If the ``load`` is marked as ``atomic``, it takes an extra
4805 :ref:`ordering <ordering>` and optional ``singlethread`` argument. The
4806 ``release`` and ``acq_rel`` orderings are not valid on ``load``
4807 instructions. Atomic loads produce :ref:`defined <memmodel>` results
4808 when they may see multiple atomic stores. The type of the pointee must
4809 be an integer type whose bit width is a power of two greater than or
4810 equal to eight and less than or equal to a target-specific size limit.
4811 ``align`` must be explicitly specified on atomic loads, and the load has
4812 undefined behavior if the alignment is not set to a value which is at
4813 least the size in bytes of the pointee. ``!nontemporal`` does not have
4814 any defined semantics for atomic loads.
4816 The optional constant ``align`` argument specifies the alignment of the
4817 operation (that is, the alignment of the memory address). A value of 0
4818 or an omitted ``align`` argument means that the operation has the ABI
4819 alignment for the target. It is the responsibility of the code emitter
4820 to ensure that the alignment information is correct. Overestimating the
4821 alignment results in undefined behavior. Underestimating the alignment
4822 may produce less efficient code. An alignment of 1 is always safe.
4824 The optional ``!nontemporal`` metadata must reference a single
4825 metadata name ``<index>`` corresponding to a metadata node with one
4826 ``i32`` entry of value 1. The existence of the ``!nontemporal``
4827 metadata on the instruction tells the optimizer and code generator
4828 that this load is not expected to be reused in the cache. The code
4829 generator may select special instructions to save cache bandwidth, such
4830 as the ``MOVNT`` instruction on x86.
4832 The optional ``!invariant.load`` metadata must reference a single
4833 metadata name ``<index>`` corresponding to a metadata node with no
4834 entries. The existence of the ``!invariant.load`` metadata on the
4835 instruction tells the optimizer and code generator that this load
4836 address points to memory which does not change value during program
4837 execution. The optimizer may then move this load around, for example, by
4838 hoisting it out of loops using loop invariant code motion.
4843 The location of memory pointed to is loaded. If the value being loaded
4844 is of scalar type then the number of bytes read does not exceed the
4845 minimum number of bytes needed to hold all bits of the type. For
4846 example, loading an ``i24`` reads at most three bytes. When loading a
4847 value of a type like ``i20`` with a size that is not an integral number
4848 of bytes, the result is undefined if the value was not originally
4849 written using a store of the same type.
4854 .. code-block:: llvm
4856 %ptr = alloca i32 ; yields {i32*}:ptr
4857 store i32 3, i32* %ptr ; yields {void}
4858 %val = load i32* %ptr ; yields {i32}:val = i32 3
4862 '``store``' Instruction
4863 ^^^^^^^^^^^^^^^^^^^^^^^
4870 store [volatile] <ty> <value>, <ty>* <pointer>[, align <alignment>][, !nontemporal !<index>] ; yields {void}
4871 store atomic [volatile] <ty> <value>, <ty>* <pointer> [singlethread] <ordering>, align <alignment> ; yields {void}
4876 The '``store``' instruction is used to write to memory.
4881 There are two arguments to the ``store`` instruction: a value to store
4882 and an address at which to store it. The type of the ``<pointer>``
4883 operand must be a pointer to the :ref:`first class <t_firstclass>` type of
4884 the ``<value>`` operand. If the ``store`` is marked as ``volatile``,
4885 then the optimizer is not allowed to modify the number or order of
4886 execution of this ``store`` with other :ref:`volatile
4887 operations <volatile>`.
4889 If the ``store`` is marked as ``atomic``, it takes an extra
4890 :ref:`ordering <ordering>` and optional ``singlethread`` argument. The
4891 ``acquire`` and ``acq_rel`` orderings aren't valid on ``store``
4892 instructions. Atomic loads produce :ref:`defined <memmodel>` results
4893 when they may see multiple atomic stores. The type of the pointee must
4894 be an integer type whose bit width is a power of two greater than or
4895 equal to eight and less than or equal to a target-specific size limit.
4896 ``align`` must be explicitly specified on atomic stores, and the store
4897 has undefined behavior if the alignment is not set to a value which is
4898 at least the size in bytes of the pointee. ``!nontemporal`` does not
4899 have any defined semantics for atomic stores.
4901 The optional constant ``align`` argument specifies the alignment of the
4902 operation (that is, the alignment of the memory address). A value of 0
4903 or an omitted ``align`` argument means that the operation has the ABI
4904 alignment for the target. It is the responsibility of the code emitter
4905 to ensure that the alignment information is correct. Overestimating the
4906 alignment results in undefined behavior. Underestimating the
4907 alignment may produce less efficient code. An alignment of 1 is always
4910 The optional ``!nontemporal`` metadata must reference a single metadata
4911 name ``<index>`` corresponding to a metadata node with one ``i32`` entry of
4912 value 1. The existence of the ``!nontemporal`` metadata on the instruction
4913 tells the optimizer and code generator that this load is not expected to
4914 be reused in the cache. The code generator may select special
4915 instructions to save cache bandwidth, such as the MOVNT instruction on
4921 The contents of memory are updated to contain ``<value>`` at the
4922 location specified by the ``<pointer>`` operand. If ``<value>`` is
4923 of scalar type then the number of bytes written does not exceed the
4924 minimum number of bytes needed to hold all bits of the type. For
4925 example, storing an ``i24`` writes at most three bytes. When writing a
4926 value of a type like ``i20`` with a size that is not an integral number
4927 of bytes, it is unspecified what happens to the extra bits that do not
4928 belong to the type, but they will typically be overwritten.
4933 .. code-block:: llvm
4935 %ptr = alloca i32 ; yields {i32*}:ptr
4936 store i32 3, i32* %ptr ; yields {void}
4937 %val = load i32* %ptr ; yields {i32}:val = i32 3
4941 '``fence``' Instruction
4942 ^^^^^^^^^^^^^^^^^^^^^^^
4949 fence [singlethread] <ordering> ; yields {void}
4954 The '``fence``' instruction is used to introduce happens-before edges
4960 '``fence``' instructions take an :ref:`ordering <ordering>` argument which
4961 defines what *synchronizes-with* edges they add. They can only be given
4962 ``acquire``, ``release``, ``acq_rel``, and ``seq_cst`` orderings.
4967 A fence A which has (at least) ``release`` ordering semantics
4968 *synchronizes with* a fence B with (at least) ``acquire`` ordering
4969 semantics if and only if there exist atomic operations X and Y, both
4970 operating on some atomic object M, such that A is sequenced before X, X
4971 modifies M (either directly or through some side effect of a sequence
4972 headed by X), Y is sequenced before B, and Y observes M. This provides a
4973 *happens-before* dependency between A and B. Rather than an explicit
4974 ``fence``, one (but not both) of the atomic operations X or Y might
4975 provide a ``release`` or ``acquire`` (resp.) ordering constraint and
4976 still *synchronize-with* the explicit ``fence`` and establish the
4977 *happens-before* edge.
4979 A ``fence`` which has ``seq_cst`` ordering, in addition to having both
4980 ``acquire`` and ``release`` semantics specified above, participates in
4981 the global program order of other ``seq_cst`` operations and/or fences.
4983 The optional ":ref:`singlethread <singlethread>`" argument specifies
4984 that the fence only synchronizes with other fences in the same thread.
4985 (This is useful for interacting with signal handlers.)
4990 .. code-block:: llvm
4992 fence acquire ; yields {void}
4993 fence singlethread seq_cst ; yields {void}
4997 '``cmpxchg``' Instruction
4998 ^^^^^^^^^^^^^^^^^^^^^^^^^
5005 cmpxchg [volatile] <ty>* <pointer>, <ty> <cmp>, <ty> <new> [singlethread] <success ordering> <failure ordering> ; yields {ty}
5010 The '``cmpxchg``' instruction is used to atomically modify memory. It
5011 loads a value in memory and compares it to a given value. If they are
5012 equal, it stores a new value into the memory.
5017 There are three arguments to the '``cmpxchg``' instruction: an address
5018 to operate on, a value to compare to the value currently be at that
5019 address, and a new value to place at that address if the compared values
5020 are equal. The type of '<cmp>' must be an integer type whose bit width
5021 is a power of two greater than or equal to eight and less than or equal
5022 to a target-specific size limit. '<cmp>' and '<new>' must have the same
5023 type, and the type of '<pointer>' must be a pointer to that type. If the
5024 ``cmpxchg`` is marked as ``volatile``, then the optimizer is not allowed
5025 to modify the number or order of execution of this ``cmpxchg`` with
5026 other :ref:`volatile operations <volatile>`.
5028 The success and failure :ref:`ordering <ordering>` arguments specify how this
5029 ``cmpxchg`` synchronizes with other atomic operations. The both ordering
5030 parameters must be at least ``monotonic``, the ordering constraint on failure
5031 must be no stronger than that on success, and the failure ordering cannot be
5032 either ``release`` or ``acq_rel``.
5034 The optional "``singlethread``" argument declares that the ``cmpxchg``
5035 is only atomic with respect to code (usually signal handlers) running in
5036 the same thread as the ``cmpxchg``. Otherwise the cmpxchg is atomic with
5037 respect to all other code in the system.
5039 The pointer passed into cmpxchg must have alignment greater than or
5040 equal to the size in memory of the operand.
5045 The contents of memory at the location specified by the '``<pointer>``'
5046 operand is read and compared to '``<cmp>``'; if the read value is the
5047 equal, '``<new>``' is written. The original value at the location is
5050 A successful ``cmpxchg`` is a read-modify-write instruction for the purpose of
5051 identifying release sequences. A failed ``cmpxchg`` is equivalent to an atomic
5052 load with an ordering parameter determined the second ordering parameter.
5057 .. code-block:: llvm
5060 %orig = atomic load i32* %ptr unordered ; yields {i32}
5064 %cmp = phi i32 [ %orig, %entry ], [%old, %loop]
5065 %squared = mul i32 %cmp, %cmp
5066 %old = cmpxchg i32* %ptr, i32 %cmp, i32 %squared acq_rel monotonic ; yields {i32}
5067 %success = icmp eq i32 %cmp, %old
5068 br i1 %success, label %done, label %loop
5075 '``atomicrmw``' Instruction
5076 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
5083 atomicrmw [volatile] <operation> <ty>* <pointer>, <ty> <value> [singlethread] <ordering> ; yields {ty}
5088 The '``atomicrmw``' instruction is used to atomically modify memory.
5093 There are three arguments to the '``atomicrmw``' instruction: an
5094 operation to apply, an address whose value to modify, an argument to the
5095 operation. The operation must be one of the following keywords:
5109 The type of '<value>' must be an integer type whose bit width is a power
5110 of two greater than or equal to eight and less than or equal to a
5111 target-specific size limit. The type of the '``<pointer>``' operand must
5112 be a pointer to that type. If the ``atomicrmw`` is marked as
5113 ``volatile``, then the optimizer is not allowed to modify the number or
5114 order of execution of this ``atomicrmw`` with other :ref:`volatile
5115 operations <volatile>`.
5120 The contents of memory at the location specified by the '``<pointer>``'
5121 operand are atomically read, modified, and written back. The original
5122 value at the location is returned. The modification is specified by the
5125 - xchg: ``*ptr = val``
5126 - add: ``*ptr = *ptr + val``
5127 - sub: ``*ptr = *ptr - val``
5128 - and: ``*ptr = *ptr & val``
5129 - nand: ``*ptr = ~(*ptr & val)``
5130 - or: ``*ptr = *ptr | val``
5131 - xor: ``*ptr = *ptr ^ val``
5132 - max: ``*ptr = *ptr > val ? *ptr : val`` (using a signed comparison)
5133 - min: ``*ptr = *ptr < val ? *ptr : val`` (using a signed comparison)
5134 - umax: ``*ptr = *ptr > val ? *ptr : val`` (using an unsigned
5136 - umin: ``*ptr = *ptr < val ? *ptr : val`` (using an unsigned
5142 .. code-block:: llvm
5144 %old = atomicrmw add i32* %ptr, i32 1 acquire ; yields {i32}
5146 .. _i_getelementptr:
5148 '``getelementptr``' Instruction
5149 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5156 <result> = getelementptr <pty>* <ptrval>{, <ty> <idx>}*
5157 <result> = getelementptr inbounds <pty>* <ptrval>{, <ty> <idx>}*
5158 <result> = getelementptr <ptr vector> ptrval, <vector index type> idx
5163 The '``getelementptr``' instruction is used to get the address of a
5164 subelement of an :ref:`aggregate <t_aggregate>` data structure. It performs
5165 address calculation only and does not access memory.
5170 The first argument is always a pointer or a vector of pointers, and
5171 forms the basis of the calculation. The remaining arguments are indices
5172 that indicate which of the elements of the aggregate object are indexed.
5173 The interpretation of each index is dependent on the type being indexed
5174 into. The first index always indexes the pointer value given as the
5175 first argument, the second index indexes a value of the type pointed to
5176 (not necessarily the value directly pointed to, since the first index
5177 can be non-zero), etc. The first type indexed into must be a pointer
5178 value, subsequent types can be arrays, vectors, and structs. Note that
5179 subsequent types being indexed into can never be pointers, since that
5180 would require loading the pointer before continuing calculation.
5182 The type of each index argument depends on the type it is indexing into.
5183 When indexing into a (optionally packed) structure, only ``i32`` integer
5184 **constants** are allowed (when using a vector of indices they must all
5185 be the **same** ``i32`` integer constant). When indexing into an array,
5186 pointer or vector, integers of any width are allowed, and they are not
5187 required to be constant. These integers are treated as signed values
5190 For example, let's consider a C code fragment and how it gets compiled
5206 int *foo(struct ST *s) {
5207 return &s[1].Z.B[5][13];
5210 The LLVM code generated by Clang is:
5212 .. code-block:: llvm
5214 %struct.RT = type { i8, [10 x [20 x i32]], i8 }
5215 %struct.ST = type { i32, double, %struct.RT }
5217 define i32* @foo(%struct.ST* %s) nounwind uwtable readnone optsize ssp {
5219 %arrayidx = getelementptr inbounds %struct.ST* %s, i64 1, i32 2, i32 1, i64 5, i64 13
5226 In the example above, the first index is indexing into the
5227 '``%struct.ST*``' type, which is a pointer, yielding a '``%struct.ST``'
5228 = '``{ i32, double, %struct.RT }``' type, a structure. The second index
5229 indexes into the third element of the structure, yielding a
5230 '``%struct.RT``' = '``{ i8 , [10 x [20 x i32]], i8 }``' type, another
5231 structure. The third index indexes into the second element of the
5232 structure, yielding a '``[10 x [20 x i32]]``' type, an array. The two
5233 dimensions of the array are subscripted into, yielding an '``i32``'
5234 type. The '``getelementptr``' instruction returns a pointer to this
5235 element, thus computing a value of '``i32*``' type.
5237 Note that it is perfectly legal to index partially through a structure,
5238 returning a pointer to an inner element. Because of this, the LLVM code
5239 for the given testcase is equivalent to:
5241 .. code-block:: llvm
5243 define i32* @foo(%struct.ST* %s) {
5244 %t1 = getelementptr %struct.ST* %s, i32 1 ; yields %struct.ST*:%t1
5245 %t2 = getelementptr %struct.ST* %t1, i32 0, i32 2 ; yields %struct.RT*:%t2
5246 %t3 = getelementptr %struct.RT* %t2, i32 0, i32 1 ; yields [10 x [20 x i32]]*:%t3
5247 %t4 = getelementptr [10 x [20 x i32]]* %t3, i32 0, i32 5 ; yields [20 x i32]*:%t4
5248 %t5 = getelementptr [20 x i32]* %t4, i32 0, i32 13 ; yields i32*:%t5
5252 If the ``inbounds`` keyword is present, the result value of the
5253 ``getelementptr`` is a :ref:`poison value <poisonvalues>` if the base
5254 pointer is not an *in bounds* address of an allocated object, or if any
5255 of the addresses that would be formed by successive addition of the
5256 offsets implied by the indices to the base address with infinitely
5257 precise signed arithmetic are not an *in bounds* address of that
5258 allocated object. The *in bounds* addresses for an allocated object are
5259 all the addresses that point into the object, plus the address one byte
5260 past the end. In cases where the base is a vector of pointers the
5261 ``inbounds`` keyword applies to each of the computations element-wise.
5263 If the ``inbounds`` keyword is not present, the offsets are added to the
5264 base address with silently-wrapping two's complement arithmetic. If the
5265 offsets have a different width from the pointer, they are sign-extended
5266 or truncated to the width of the pointer. The result value of the
5267 ``getelementptr`` may be outside the object pointed to by the base
5268 pointer. The result value may not necessarily be used to access memory
5269 though, even if it happens to point into allocated storage. See the
5270 :ref:`Pointer Aliasing Rules <pointeraliasing>` section for more
5273 The getelementptr instruction is often confusing. For some more insight
5274 into how it works, see :doc:`the getelementptr FAQ <GetElementPtr>`.
5279 .. code-block:: llvm
5281 ; yields [12 x i8]*:aptr
5282 %aptr = getelementptr {i32, [12 x i8]}* %saptr, i64 0, i32 1
5284 %vptr = getelementptr {i32, <2 x i8>}* %svptr, i64 0, i32 1, i32 1
5286 %eptr = getelementptr [12 x i8]* %aptr, i64 0, i32 1
5288 %iptr = getelementptr [10 x i32]* @arr, i16 0, i16 0
5290 In cases where the pointer argument is a vector of pointers, each index
5291 must be a vector with the same number of elements. For example:
5293 .. code-block:: llvm
5295 %A = getelementptr <4 x i8*> %ptrs, <4 x i64> %offsets,
5297 Conversion Operations
5298 ---------------------
5300 The instructions in this category are the conversion instructions
5301 (casting) which all take a single operand and a type. They perform
5302 various bit conversions on the operand.
5304 '``trunc .. to``' Instruction
5305 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5312 <result> = trunc <ty> <value> to <ty2> ; yields ty2
5317 The '``trunc``' instruction truncates its operand to the type ``ty2``.
5322 The '``trunc``' instruction takes a value to trunc, and a type to trunc
5323 it to. Both types must be of :ref:`integer <t_integer>` types, or vectors
5324 of the same number of integers. The bit size of the ``value`` must be
5325 larger than the bit size of the destination type, ``ty2``. Equal sized
5326 types are not allowed.
5331 The '``trunc``' instruction truncates the high order bits in ``value``
5332 and converts the remaining bits to ``ty2``. Since the source size must
5333 be larger than the destination size, ``trunc`` cannot be a *no-op cast*.
5334 It will always truncate bits.
5339 .. code-block:: llvm
5341 %X = trunc i32 257 to i8 ; yields i8:1
5342 %Y = trunc i32 123 to i1 ; yields i1:true
5343 %Z = trunc i32 122 to i1 ; yields i1:false
5344 %W = trunc <2 x i16> <i16 8, i16 7> to <2 x i8> ; yields <i8 8, i8 7>
5346 '``zext .. to``' Instruction
5347 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5354 <result> = zext <ty> <value> to <ty2> ; yields ty2
5359 The '``zext``' instruction zero extends its operand to type ``ty2``.
5364 The '``zext``' instruction takes a value to cast, and a type to cast it
5365 to. Both types must be of :ref:`integer <t_integer>` types, or vectors of
5366 the same number of integers. The bit size of the ``value`` must be
5367 smaller than the bit size of the destination type, ``ty2``.
5372 The ``zext`` fills the high order bits of the ``value`` with zero bits
5373 until it reaches the size of the destination type, ``ty2``.
5375 When zero extending from i1, the result will always be either 0 or 1.
5380 .. code-block:: llvm
5382 %X = zext i32 257 to i64 ; yields i64:257
5383 %Y = zext i1 true to i32 ; yields i32:1
5384 %Z = zext <2 x i16> <i16 8, i16 7> to <2 x i32> ; yields <i32 8, i32 7>
5386 '``sext .. to``' Instruction
5387 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5394 <result> = sext <ty> <value> to <ty2> ; yields ty2
5399 The '``sext``' sign extends ``value`` to the type ``ty2``.
5404 The '``sext``' instruction takes a value to cast, and a type to cast it
5405 to. Both types must be of :ref:`integer <t_integer>` types, or vectors of
5406 the same number of integers. The bit size of the ``value`` must be
5407 smaller than the bit size of the destination type, ``ty2``.
5412 The '``sext``' instruction performs a sign extension by copying the sign
5413 bit (highest order bit) of the ``value`` until it reaches the bit size
5414 of the type ``ty2``.
5416 When sign extending from i1, the extension always results in -1 or 0.
5421 .. code-block:: llvm
5423 %X = sext i8 -1 to i16 ; yields i16 :65535
5424 %Y = sext i1 true to i32 ; yields i32:-1
5425 %Z = sext <2 x i16> <i16 8, i16 7> to <2 x i32> ; yields <i32 8, i32 7>
5427 '``fptrunc .. to``' Instruction
5428 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5435 <result> = fptrunc <ty> <value> to <ty2> ; yields ty2
5440 The '``fptrunc``' instruction truncates ``value`` to type ``ty2``.
5445 The '``fptrunc``' instruction takes a :ref:`floating point <t_floating>`
5446 value to cast and a :ref:`floating point <t_floating>` type to cast it to.
5447 The size of ``value`` must be larger than the size of ``ty2``. This
5448 implies that ``fptrunc`` cannot be used to make a *no-op cast*.
5453 The '``fptrunc``' instruction truncates a ``value`` from a larger
5454 :ref:`floating point <t_floating>` type to a smaller :ref:`floating
5455 point <t_floating>` type. If the value cannot fit within the
5456 destination type, ``ty2``, then the results are undefined.
5461 .. code-block:: llvm
5463 %X = fptrunc double 123.0 to float ; yields float:123.0
5464 %Y = fptrunc double 1.0E+300 to float ; yields undefined
5466 '``fpext .. to``' Instruction
5467 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5474 <result> = fpext <ty> <value> to <ty2> ; yields ty2
5479 The '``fpext``' extends a floating point ``value`` to a larger floating
5485 The '``fpext``' instruction takes a :ref:`floating point <t_floating>`
5486 ``value`` to cast, and a :ref:`floating point <t_floating>` type to cast it
5487 to. The source type must be smaller than the destination type.
5492 The '``fpext``' instruction extends the ``value`` from a smaller
5493 :ref:`floating point <t_floating>` type to a larger :ref:`floating
5494 point <t_floating>` type. The ``fpext`` cannot be used to make a
5495 *no-op cast* because it always changes bits. Use ``bitcast`` to make a
5496 *no-op cast* for a floating point cast.
5501 .. code-block:: llvm
5503 %X = fpext float 3.125 to double ; yields double:3.125000e+00
5504 %Y = fpext double %X to fp128 ; yields fp128:0xL00000000000000004000900000000000
5506 '``fptoui .. to``' Instruction
5507 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5514 <result> = fptoui <ty> <value> to <ty2> ; yields ty2
5519 The '``fptoui``' converts a floating point ``value`` to its unsigned
5520 integer equivalent of type ``ty2``.
5525 The '``fptoui``' instruction takes a value to cast, which must be a
5526 scalar or vector :ref:`floating point <t_floating>` value, and a type to
5527 cast it to ``ty2``, which must be an :ref:`integer <t_integer>` type. If
5528 ``ty`` is a vector floating point type, ``ty2`` must be a vector integer
5529 type with the same number of elements as ``ty``
5534 The '``fptoui``' instruction converts its :ref:`floating
5535 point <t_floating>` operand into the nearest (rounding towards zero)
5536 unsigned integer value. If the value cannot fit in ``ty2``, the results
5542 .. code-block:: llvm
5544 %X = fptoui double 123.0 to i32 ; yields i32:123
5545 %Y = fptoui float 1.0E+300 to i1 ; yields undefined:1
5546 %Z = fptoui float 1.04E+17 to i8 ; yields undefined:1
5548 '``fptosi .. to``' Instruction
5549 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5556 <result> = fptosi <ty> <value> to <ty2> ; yields ty2
5561 The '``fptosi``' instruction converts :ref:`floating point <t_floating>`
5562 ``value`` to type ``ty2``.
5567 The '``fptosi``' instruction takes a value to cast, which must be a
5568 scalar or vector :ref:`floating point <t_floating>` value, and a type to
5569 cast it to ``ty2``, which must be an :ref:`integer <t_integer>` type. If
5570 ``ty`` is a vector floating point type, ``ty2`` must be a vector integer
5571 type with the same number of elements as ``ty``
5576 The '``fptosi``' instruction converts its :ref:`floating
5577 point <t_floating>` operand into the nearest (rounding towards zero)
5578 signed integer value. If the value cannot fit in ``ty2``, the results
5584 .. code-block:: llvm
5586 %X = fptosi double -123.0 to i32 ; yields i32:-123
5587 %Y = fptosi float 1.0E-247 to i1 ; yields undefined:1
5588 %Z = fptosi float 1.04E+17 to i8 ; yields undefined:1
5590 '``uitofp .. to``' Instruction
5591 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5598 <result> = uitofp <ty> <value> to <ty2> ; yields ty2
5603 The '``uitofp``' instruction regards ``value`` as an unsigned integer
5604 and converts that value to the ``ty2`` type.
5609 The '``uitofp``' instruction takes a value to cast, which must be a
5610 scalar or vector :ref:`integer <t_integer>` value, and a type to cast it to
5611 ``ty2``, which must be an :ref:`floating point <t_floating>` type. If
5612 ``ty`` is a vector integer type, ``ty2`` must be a vector floating point
5613 type with the same number of elements as ``ty``
5618 The '``uitofp``' instruction interprets its operand as an unsigned
5619 integer quantity and converts it to the corresponding floating point
5620 value. If the value cannot fit in the floating point value, the results
5626 .. code-block:: llvm
5628 %X = uitofp i32 257 to float ; yields float:257.0
5629 %Y = uitofp i8 -1 to double ; yields double:255.0
5631 '``sitofp .. to``' Instruction
5632 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5639 <result> = sitofp <ty> <value> to <ty2> ; yields ty2
5644 The '``sitofp``' instruction regards ``value`` as a signed integer and
5645 converts that value to the ``ty2`` type.
5650 The '``sitofp``' instruction takes a value to cast, which must be a
5651 scalar or vector :ref:`integer <t_integer>` value, and a type to cast it to
5652 ``ty2``, which must be an :ref:`floating point <t_floating>` type. If
5653 ``ty`` is a vector integer type, ``ty2`` must be a vector floating point
5654 type with the same number of elements as ``ty``
5659 The '``sitofp``' instruction interprets its operand as a signed integer
5660 quantity and converts it to the corresponding floating point value. If
5661 the value cannot fit in the floating point value, the results are
5667 .. code-block:: llvm
5669 %X = sitofp i32 257 to float ; yields float:257.0
5670 %Y = sitofp i8 -1 to double ; yields double:-1.0
5674 '``ptrtoint .. to``' Instruction
5675 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5682 <result> = ptrtoint <ty> <value> to <ty2> ; yields ty2
5687 The '``ptrtoint``' instruction converts the pointer or a vector of
5688 pointers ``value`` to the integer (or vector of integers) type ``ty2``.
5693 The '``ptrtoint``' instruction takes a ``value`` to cast, which must be
5694 a a value of type :ref:`pointer <t_pointer>` or a vector of pointers, and a
5695 type to cast it to ``ty2``, which must be an :ref:`integer <t_integer>` or
5696 a vector of integers type.
5701 The '``ptrtoint``' instruction converts ``value`` to integer type
5702 ``ty2`` by interpreting the pointer value as an integer and either
5703 truncating or zero extending that value to the size of the integer type.
5704 If ``value`` is smaller than ``ty2`` then a zero extension is done. If
5705 ``value`` is larger than ``ty2`` then a truncation is done. If they are
5706 the same size, then nothing is done (*no-op cast*) other than a type
5712 .. code-block:: llvm
5714 %X = ptrtoint i32* %P to i8 ; yields truncation on 32-bit architecture
5715 %Y = ptrtoint i32* %P to i64 ; yields zero extension on 32-bit architecture
5716 %Z = ptrtoint <4 x i32*> %P to <4 x i64>; yields vector zero extension for a vector of addresses on 32-bit architecture
5720 '``inttoptr .. to``' Instruction
5721 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5728 <result> = inttoptr <ty> <value> to <ty2> ; yields ty2
5733 The '``inttoptr``' instruction converts an integer ``value`` to a
5734 pointer type, ``ty2``.
5739 The '``inttoptr``' instruction takes an :ref:`integer <t_integer>` value to
5740 cast, and a type to cast it to, which must be a :ref:`pointer <t_pointer>`
5746 The '``inttoptr``' instruction converts ``value`` to type ``ty2`` by
5747 applying either a zero extension or a truncation depending on the size
5748 of the integer ``value``. If ``value`` is larger than the size of a
5749 pointer then a truncation is done. If ``value`` is smaller than the size
5750 of a pointer then a zero extension is done. If they are the same size,
5751 nothing is done (*no-op cast*).
5756 .. code-block:: llvm
5758 %X = inttoptr i32 255 to i32* ; yields zero extension on 64-bit architecture
5759 %Y = inttoptr i32 255 to i32* ; yields no-op on 32-bit architecture
5760 %Z = inttoptr i64 0 to i32* ; yields truncation on 32-bit architecture
5761 %Z = inttoptr <4 x i32> %G to <4 x i8*>; yields truncation of vector G to four pointers
5765 '``bitcast .. to``' Instruction
5766 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5773 <result> = bitcast <ty> <value> to <ty2> ; yields ty2
5778 The '``bitcast``' instruction converts ``value`` to type ``ty2`` without
5784 The '``bitcast``' instruction takes a value to cast, which must be a
5785 non-aggregate first class value, and a type to cast it to, which must
5786 also be a non-aggregate :ref:`first class <t_firstclass>` type. The
5787 bit sizes of ``value`` and the destination type, ``ty2``, must be
5788 identical. If the source type is a pointer, the destination type must
5789 also be a pointer of the same size. This instruction supports bitwise
5790 conversion of vectors to integers and to vectors of other types (as
5791 long as they have the same size).
5796 The '``bitcast``' instruction converts ``value`` to type ``ty2``. It
5797 is always a *no-op cast* because no bits change with this
5798 conversion. The conversion is done as if the ``value`` had been stored
5799 to memory and read back as type ``ty2``. Pointer (or vector of
5800 pointers) types may only be converted to other pointer (or vector of
5801 pointers) types with the same address space through this instruction.
5802 To convert pointers to other types, use the :ref:`inttoptr <i_inttoptr>`
5803 or :ref:`ptrtoint <i_ptrtoint>` instructions first.
5808 .. code-block:: llvm
5810 %X = bitcast i8 255 to i8 ; yields i8 :-1
5811 %Y = bitcast i32* %x to sint* ; yields sint*:%x
5812 %Z = bitcast <2 x int> %V to i64; ; yields i64: %V
5813 %Z = bitcast <2 x i32*> %V to <2 x i64*> ; yields <2 x i64*>
5815 .. _i_addrspacecast:
5817 '``addrspacecast .. to``' Instruction
5818 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
5825 <result> = addrspacecast <pty> <ptrval> to <pty2> ; yields pty2
5830 The '``addrspacecast``' instruction converts ``ptrval`` from ``pty`` in
5831 address space ``n`` to type ``pty2`` in address space ``m``.
5836 The '``addrspacecast``' instruction takes a pointer or vector of pointer value
5837 to cast and a pointer type to cast it to, which must have a different
5843 The '``addrspacecast``' instruction converts the pointer value
5844 ``ptrval`` to type ``pty2``. It can be a *no-op cast* or a complex
5845 value modification, depending on the target and the address space
5846 pair. Pointer conversions within the same address space must be
5847 performed with the ``bitcast`` instruction. Note that if the address space
5848 conversion is legal then both result and operand refer to the same memory
5854 .. code-block:: llvm
5856 %X = addrspacecast i32* %x to i32 addrspace(1)* ; yields i32 addrspace(1)*:%x
5857 %Y = addrspacecast i32 addrspace(1)* %y to i64 addrspace(2)* ; yields i64 addrspace(2)*:%y
5858 %Z = addrspacecast <4 x i32*> %z to <4 x float addrspace(3)*> ; yields <4 x float addrspace(3)*>:%z
5865 The instructions in this category are the "miscellaneous" instructions,
5866 which defy better classification.
5870 '``icmp``' Instruction
5871 ^^^^^^^^^^^^^^^^^^^^^^
5878 <result> = icmp <cond> <ty> <op1>, <op2> ; yields {i1} or {<N x i1>}:result
5883 The '``icmp``' instruction returns a boolean value or a vector of
5884 boolean values based on comparison of its two integer, integer vector,
5885 pointer, or pointer vector operands.
5890 The '``icmp``' instruction takes three operands. The first operand is
5891 the condition code indicating the kind of comparison to perform. It is
5892 not a value, just a keyword. The possible condition code are:
5895 #. ``ne``: not equal
5896 #. ``ugt``: unsigned greater than
5897 #. ``uge``: unsigned greater or equal
5898 #. ``ult``: unsigned less than
5899 #. ``ule``: unsigned less or equal
5900 #. ``sgt``: signed greater than
5901 #. ``sge``: signed greater or equal
5902 #. ``slt``: signed less than
5903 #. ``sle``: signed less or equal
5905 The remaining two arguments must be :ref:`integer <t_integer>` or
5906 :ref:`pointer <t_pointer>` or integer :ref:`vector <t_vector>` typed. They
5907 must also be identical types.
5912 The '``icmp``' compares ``op1`` and ``op2`` according to the condition
5913 code given as ``cond``. The comparison performed always yields either an
5914 :ref:`i1 <t_integer>` or vector of ``i1`` result, as follows:
5916 #. ``eq``: yields ``true`` if the operands are equal, ``false``
5917 otherwise. No sign interpretation is necessary or performed.
5918 #. ``ne``: yields ``true`` if the operands are unequal, ``false``
5919 otherwise. No sign interpretation is necessary or performed.
5920 #. ``ugt``: interprets the operands as unsigned values and yields
5921 ``true`` if ``op1`` is greater than ``op2``.
5922 #. ``uge``: interprets the operands as unsigned values and yields
5923 ``true`` if ``op1`` is greater than or equal to ``op2``.
5924 #. ``ult``: interprets the operands as unsigned values and yields
5925 ``true`` if ``op1`` is less than ``op2``.
5926 #. ``ule``: interprets the operands as unsigned values and yields
5927 ``true`` if ``op1`` is less than or equal to ``op2``.
5928 #. ``sgt``: interprets the operands as signed values and yields ``true``
5929 if ``op1`` is greater than ``op2``.
5930 #. ``sge``: interprets the operands as signed values and yields ``true``
5931 if ``op1`` is greater than or equal to ``op2``.
5932 #. ``slt``: interprets the operands as signed values and yields ``true``
5933 if ``op1`` is less than ``op2``.
5934 #. ``sle``: interprets the operands as signed values and yields ``true``
5935 if ``op1`` is less than or equal to ``op2``.
5937 If the operands are :ref:`pointer <t_pointer>` typed, the pointer values
5938 are compared as if they were integers.
5940 If the operands are integer vectors, then they are compared element by
5941 element. The result is an ``i1`` vector with the same number of elements
5942 as the values being compared. Otherwise, the result is an ``i1``.
5947 .. code-block:: llvm
5949 <result> = icmp eq i32 4, 5 ; yields: result=false
5950 <result> = icmp ne float* %X, %X ; yields: result=false
5951 <result> = icmp ult i16 4, 5 ; yields: result=true
5952 <result> = icmp sgt i16 4, 5 ; yields: result=false
5953 <result> = icmp ule i16 -4, 5 ; yields: result=false
5954 <result> = icmp sge i16 4, 5 ; yields: result=false
5956 Note that the code generator does not yet support vector types with the
5957 ``icmp`` instruction.
5961 '``fcmp``' Instruction
5962 ^^^^^^^^^^^^^^^^^^^^^^
5969 <result> = fcmp <cond> <ty> <op1>, <op2> ; yields {i1} or {<N x i1>}:result
5974 The '``fcmp``' instruction returns a boolean value or vector of boolean
5975 values based on comparison of its operands.
5977 If the operands are floating point scalars, then the result type is a
5978 boolean (:ref:`i1 <t_integer>`).
5980 If the operands are floating point vectors, then the result type is a
5981 vector of boolean with the same number of elements as the operands being
5987 The '``fcmp``' instruction takes three operands. The first operand is
5988 the condition code indicating the kind of comparison to perform. It is
5989 not a value, just a keyword. The possible condition code are:
5991 #. ``false``: no comparison, always returns false
5992 #. ``oeq``: ordered and equal
5993 #. ``ogt``: ordered and greater than
5994 #. ``oge``: ordered and greater than or equal
5995 #. ``olt``: ordered and less than
5996 #. ``ole``: ordered and less than or equal
5997 #. ``one``: ordered and not equal
5998 #. ``ord``: ordered (no nans)
5999 #. ``ueq``: unordered or equal
6000 #. ``ugt``: unordered or greater than
6001 #. ``uge``: unordered or greater than or equal
6002 #. ``ult``: unordered or less than
6003 #. ``ule``: unordered or less than or equal
6004 #. ``une``: unordered or not equal
6005 #. ``uno``: unordered (either nans)
6006 #. ``true``: no comparison, always returns true
6008 *Ordered* means that neither operand is a QNAN while *unordered* means
6009 that either operand may be a QNAN.
6011 Each of ``val1`` and ``val2`` arguments must be either a :ref:`floating
6012 point <t_floating>` type or a :ref:`vector <t_vector>` of floating point
6013 type. They must have identical types.
6018 The '``fcmp``' instruction compares ``op1`` and ``op2`` according to the
6019 condition code given as ``cond``. If the operands are vectors, then the
6020 vectors are compared element by element. Each comparison performed
6021 always yields an :ref:`i1 <t_integer>` result, as follows:
6023 #. ``false``: always yields ``false``, regardless of operands.
6024 #. ``oeq``: yields ``true`` if both operands are not a QNAN and ``op1``
6025 is equal to ``op2``.
6026 #. ``ogt``: yields ``true`` if both operands are not a QNAN and ``op1``
6027 is greater than ``op2``.
6028 #. ``oge``: yields ``true`` if both operands are not a QNAN and ``op1``
6029 is greater than or equal to ``op2``.
6030 #. ``olt``: yields ``true`` if both operands are not a QNAN and ``op1``
6031 is less than ``op2``.
6032 #. ``ole``: yields ``true`` if both operands are not a QNAN and ``op1``
6033 is less than or equal to ``op2``.
6034 #. ``one``: yields ``true`` if both operands are not a QNAN and ``op1``
6035 is not equal to ``op2``.
6036 #. ``ord``: yields ``true`` if both operands are not a QNAN.
6037 #. ``ueq``: yields ``true`` if either operand is a QNAN or ``op1`` is
6039 #. ``ugt``: yields ``true`` if either operand is a QNAN or ``op1`` is
6040 greater than ``op2``.
6041 #. ``uge``: yields ``true`` if either operand is a QNAN or ``op1`` is
6042 greater than or equal to ``op2``.
6043 #. ``ult``: yields ``true`` if either operand is a QNAN or ``op1`` is
6045 #. ``ule``: yields ``true`` if either operand is a QNAN or ``op1`` is
6046 less than or equal to ``op2``.
6047 #. ``une``: yields ``true`` if either operand is a QNAN or ``op1`` is
6048 not equal to ``op2``.
6049 #. ``uno``: yields ``true`` if either operand is a QNAN.
6050 #. ``true``: always yields ``true``, regardless of operands.
6055 .. code-block:: llvm
6057 <result> = fcmp oeq float 4.0, 5.0 ; yields: result=false
6058 <result> = fcmp one float 4.0, 5.0 ; yields: result=true
6059 <result> = fcmp olt float 4.0, 5.0 ; yields: result=true
6060 <result> = fcmp ueq double 1.0, 2.0 ; yields: result=false
6062 Note that the code generator does not yet support vector types with the
6063 ``fcmp`` instruction.
6067 '``phi``' Instruction
6068 ^^^^^^^^^^^^^^^^^^^^^
6075 <result> = phi <ty> [ <val0>, <label0>], ...
6080 The '``phi``' instruction is used to implement the φ node in the SSA
6081 graph representing the function.
6086 The type of the incoming values is specified with the first type field.
6087 After this, the '``phi``' instruction takes a list of pairs as
6088 arguments, with one pair for each predecessor basic block of the current
6089 block. Only values of :ref:`first class <t_firstclass>` type may be used as
6090 the value arguments to the PHI node. Only labels may be used as the
6093 There must be no non-phi instructions between the start of a basic block
6094 and the PHI instructions: i.e. PHI instructions must be first in a basic
6097 For the purposes of the SSA form, the use of each incoming value is
6098 deemed to occur on the edge from the corresponding predecessor block to
6099 the current block (but after any definition of an '``invoke``'
6100 instruction's return value on the same edge).
6105 At runtime, the '``phi``' instruction logically takes on the value
6106 specified by the pair corresponding to the predecessor basic block that
6107 executed just prior to the current block.
6112 .. code-block:: llvm
6114 Loop: ; Infinite loop that counts from 0 on up...
6115 %indvar = phi i32 [ 0, %LoopHeader ], [ %nextindvar, %Loop ]
6116 %nextindvar = add i32 %indvar, 1
6121 '``select``' Instruction
6122 ^^^^^^^^^^^^^^^^^^^^^^^^
6129 <result> = select selty <cond>, <ty> <val1>, <ty> <val2> ; yields ty
6131 selty is either i1 or {<N x i1>}
6136 The '``select``' instruction is used to choose one value based on a
6137 condition, without IR-level branching.
6142 The '``select``' instruction requires an 'i1' value or a vector of 'i1'
6143 values indicating the condition, and two values of the same :ref:`first
6144 class <t_firstclass>` type. If the val1/val2 are vectors and the
6145 condition is a scalar, then entire vectors are selected, not individual
6151 If the condition is an i1 and it evaluates to 1, the instruction returns
6152 the first value argument; otherwise, it returns the second value
6155 If the condition is a vector of i1, then the value arguments must be
6156 vectors of the same size, and the selection is done element by element.
6161 .. code-block:: llvm
6163 %X = select i1 true, i8 17, i8 42 ; yields i8:17
6167 '``call``' Instruction
6168 ^^^^^^^^^^^^^^^^^^^^^^
6175 <result> = [tail | musttail] call [cconv] [ret attrs] <ty> [<fnty>*] <fnptrval>(<function args>) [fn attrs]
6180 The '``call``' instruction represents a simple function call.
6185 This instruction requires several arguments:
6187 #. The optional ``tail`` and ``musttail`` markers indicate that the optimizers
6188 should perform tail call optimization. The ``tail`` marker is a hint that
6189 `can be ignored <CodeGenerator.html#sibcallopt>`_. The ``musttail`` marker
6190 means that the call must be tail call optimized in order for the program to
6191 be correct. The ``musttail`` marker provides these guarantees:
6193 #. The call will not cause unbounded stack growth if it is part of a
6194 recursive cycle in the call graph.
6195 #. Arguments with the :ref:`inalloca <attr_inalloca>` attribute are
6198 Both markers imply that the callee does not access allocas or varargs from
6199 the caller. Calls marked ``musttail`` must obey the following additional
6202 - The call must immediately precede a :ref:`ret <i_ret>` instruction,
6203 or a pointer bitcast followed by a ret instruction.
6204 - The ret instruction must return the (possibly bitcasted) value
6205 produced by the call or void.
6206 - The caller and callee prototypes must match. Pointer types of
6207 parameters or return types may differ in pointee type, but not
6209 - The calling conventions of the caller and callee must match.
6210 - All ABI-impacting function attributes, such as sret, byval, inreg,
6211 returned, and inalloca, must match.
6213 Tail call optimization for calls marked ``tail`` is guaranteed to occur if
6214 the following conditions are met:
6216 - Caller and callee both have the calling convention ``fastcc``.
6217 - The call is in tail position (ret immediately follows call and ret
6218 uses value of call or is void).
6219 - Option ``-tailcallopt`` is enabled, or
6220 ``llvm::GuaranteedTailCallOpt`` is ``true``.
6221 - `Platform specific constraints are
6222 met. <CodeGenerator.html#tailcallopt>`_
6224 #. The optional "cconv" marker indicates which :ref:`calling
6225 convention <callingconv>` the call should use. If none is
6226 specified, the call defaults to using C calling conventions. The
6227 calling convention of the call must match the calling convention of
6228 the target function, or else the behavior is undefined.
6229 #. The optional :ref:`Parameter Attributes <paramattrs>` list for return
6230 values. Only '``zeroext``', '``signext``', and '``inreg``' attributes
6232 #. '``ty``': the type of the call instruction itself which is also the
6233 type of the return value. Functions that return no value are marked
6235 #. '``fnty``': shall be the signature of the pointer to function value
6236 being invoked. The argument types must match the types implied by
6237 this signature. This type can be omitted if the function is not
6238 varargs and if the function type does not return a pointer to a
6240 #. '``fnptrval``': An LLVM value containing a pointer to a function to
6241 be invoked. In most cases, this is a direct function invocation, but
6242 indirect ``call``'s are just as possible, calling an arbitrary pointer
6244 #. '``function args``': argument list whose types match the function
6245 signature argument types and parameter attributes. All arguments must
6246 be of :ref:`first class <t_firstclass>` type. If the function signature
6247 indicates the function accepts a variable number of arguments, the
6248 extra arguments can be specified.
6249 #. The optional :ref:`function attributes <fnattrs>` list. Only
6250 '``noreturn``', '``nounwind``', '``readonly``' and '``readnone``'
6251 attributes are valid here.
6256 The '``call``' instruction is used to cause control flow to transfer to
6257 a specified function, with its incoming arguments bound to the specified
6258 values. Upon a '``ret``' instruction in the called function, control
6259 flow continues with the instruction after the function call, and the
6260 return value of the function is bound to the result argument.
6265 .. code-block:: llvm
6267 %retval = call i32 @test(i32 %argc)
6268 call i32 (i8*, ...)* @printf(i8* %msg, i32 12, i8 42) ; yields i32
6269 %X = tail call i32 @foo() ; yields i32
6270 %Y = tail call fastcc i32 @foo() ; yields i32
6271 call void %foo(i8 97 signext)
6273 %struct.A = type { i32, i8 }
6274 %r = call %struct.A @foo() ; yields { 32, i8 }
6275 %gr = extractvalue %struct.A %r, 0 ; yields i32
6276 %gr1 = extractvalue %struct.A %r, 1 ; yields i8
6277 %Z = call void @foo() noreturn ; indicates that %foo never returns normally
6278 %ZZ = call zeroext i32 @bar() ; Return value is %zero extended
6280 llvm treats calls to some functions with names and arguments that match
6281 the standard C99 library as being the C99 library functions, and may
6282 perform optimizations or generate code for them under that assumption.
6283 This is something we'd like to change in the future to provide better
6284 support for freestanding environments and non-C-based languages.
6288 '``va_arg``' Instruction
6289 ^^^^^^^^^^^^^^^^^^^^^^^^
6296 <resultval> = va_arg <va_list*> <arglist>, <argty>
6301 The '``va_arg``' instruction is used to access arguments passed through
6302 the "variable argument" area of a function call. It is used to implement
6303 the ``va_arg`` macro in C.
6308 This instruction takes a ``va_list*`` value and the type of the
6309 argument. It returns a value of the specified argument type and
6310 increments the ``va_list`` to point to the next argument. The actual
6311 type of ``va_list`` is target specific.
6316 The '``va_arg``' instruction loads an argument of the specified type
6317 from the specified ``va_list`` and causes the ``va_list`` to point to
6318 the next argument. For more information, see the variable argument
6319 handling :ref:`Intrinsic Functions <int_varargs>`.
6321 It is legal for this instruction to be called in a function which does
6322 not take a variable number of arguments, for example, the ``vfprintf``
6325 ``va_arg`` is an LLVM instruction instead of an :ref:`intrinsic
6326 function <intrinsics>` because it takes a type as an argument.
6331 See the :ref:`variable argument processing <int_varargs>` section.
6333 Note that the code generator does not yet fully support va\_arg on many
6334 targets. Also, it does not currently support va\_arg with aggregate
6335 types on any target.
6339 '``landingpad``' Instruction
6340 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6347 <resultval> = landingpad <resultty> personality <type> <pers_fn> <clause>+
6348 <resultval> = landingpad <resultty> personality <type> <pers_fn> cleanup <clause>*
6350 <clause> := catch <type> <value>
6351 <clause> := filter <array constant type> <array constant>
6356 The '``landingpad``' instruction is used by `LLVM's exception handling
6357 system <ExceptionHandling.html#overview>`_ to specify that a basic block
6358 is a landing pad --- one where the exception lands, and corresponds to the
6359 code found in the ``catch`` portion of a ``try``/``catch`` sequence. It
6360 defines values supplied by the personality function (``pers_fn``) upon
6361 re-entry to the function. The ``resultval`` has the type ``resultty``.
6366 This instruction takes a ``pers_fn`` value. This is the personality
6367 function associated with the unwinding mechanism. The optional
6368 ``cleanup`` flag indicates that the landing pad block is a cleanup.
6370 A ``clause`` begins with the clause type --- ``catch`` or ``filter`` --- and
6371 contains the global variable representing the "type" that may be caught
6372 or filtered respectively. Unlike the ``catch`` clause, the ``filter``
6373 clause takes an array constant as its argument. Use
6374 "``[0 x i8**] undef``" for a filter which cannot throw. The
6375 '``landingpad``' instruction must contain *at least* one ``clause`` or
6376 the ``cleanup`` flag.
6381 The '``landingpad``' instruction defines the values which are set by the
6382 personality function (``pers_fn``) upon re-entry to the function, and
6383 therefore the "result type" of the ``landingpad`` instruction. As with
6384 calling conventions, how the personality function results are
6385 represented in LLVM IR is target specific.
6387 The clauses are applied in order from top to bottom. If two
6388 ``landingpad`` instructions are merged together through inlining, the
6389 clauses from the calling function are appended to the list of clauses.
6390 When the call stack is being unwound due to an exception being thrown,
6391 the exception is compared against each ``clause`` in turn. If it doesn't
6392 match any of the clauses, and the ``cleanup`` flag is not set, then
6393 unwinding continues further up the call stack.
6395 The ``landingpad`` instruction has several restrictions:
6397 - A landing pad block is a basic block which is the unwind destination
6398 of an '``invoke``' instruction.
6399 - A landing pad block must have a '``landingpad``' instruction as its
6400 first non-PHI instruction.
6401 - There can be only one '``landingpad``' instruction within the landing
6403 - A basic block that is not a landing pad block may not include a
6404 '``landingpad``' instruction.
6405 - All '``landingpad``' instructions in a function must have the same
6406 personality function.
6411 .. code-block:: llvm
6413 ;; A landing pad which can catch an integer.
6414 %res = landingpad { i8*, i32 } personality i32 (...)* @__gxx_personality_v0
6416 ;; A landing pad that is a cleanup.
6417 %res = landingpad { i8*, i32 } personality i32 (...)* @__gxx_personality_v0
6419 ;; A landing pad which can catch an integer and can only throw a double.
6420 %res = landingpad { i8*, i32 } personality i32 (...)* @__gxx_personality_v0
6422 filter [1 x i8**] [@_ZTId]
6429 LLVM supports the notion of an "intrinsic function". These functions
6430 have well known names and semantics and are required to follow certain
6431 restrictions. Overall, these intrinsics represent an extension mechanism
6432 for the LLVM language that does not require changing all of the
6433 transformations in LLVM when adding to the language (or the bitcode
6434 reader/writer, the parser, etc...).
6436 Intrinsic function names must all start with an "``llvm.``" prefix. This
6437 prefix is reserved in LLVM for intrinsic names; thus, function names may
6438 not begin with this prefix. Intrinsic functions must always be external
6439 functions: you cannot define the body of intrinsic functions. Intrinsic
6440 functions may only be used in call or invoke instructions: it is illegal
6441 to take the address of an intrinsic function. Additionally, because
6442 intrinsic functions are part of the LLVM language, it is required if any
6443 are added that they be documented here.
6445 Some intrinsic functions can be overloaded, i.e., the intrinsic
6446 represents a family of functions that perform the same operation but on
6447 different data types. Because LLVM can represent over 8 million
6448 different integer types, overloading is used commonly to allow an
6449 intrinsic function to operate on any integer type. One or more of the
6450 argument types or the result type can be overloaded to accept any
6451 integer type. Argument types may also be defined as exactly matching a
6452 previous argument's type or the result type. This allows an intrinsic
6453 function which accepts multiple arguments, but needs all of them to be
6454 of the same type, to only be overloaded with respect to a single
6455 argument or the result.
6457 Overloaded intrinsics will have the names of its overloaded argument
6458 types encoded into its function name, each preceded by a period. Only
6459 those types which are overloaded result in a name suffix. Arguments
6460 whose type is matched against another type do not. For example, the
6461 ``llvm.ctpop`` function can take an integer of any width and returns an
6462 integer of exactly the same integer width. This leads to a family of
6463 functions such as ``i8 @llvm.ctpop.i8(i8 %val)`` and
6464 ``i29 @llvm.ctpop.i29(i29 %val)``. Only one type, the return type, is
6465 overloaded, and only one type suffix is required. Because the argument's
6466 type is matched against the return type, it does not require its own
6469 To learn how to add an intrinsic function, please see the `Extending
6470 LLVM Guide <ExtendingLLVM.html>`_.
6474 Variable Argument Handling Intrinsics
6475 -------------------------------------
6477 Variable argument support is defined in LLVM with the
6478 :ref:`va_arg <i_va_arg>` instruction and these three intrinsic
6479 functions. These functions are related to the similarly named macros
6480 defined in the ``<stdarg.h>`` header file.
6482 All of these functions operate on arguments that use a target-specific
6483 value type "``va_list``". The LLVM assembly language reference manual
6484 does not define what this type is, so all transformations should be
6485 prepared to handle these functions regardless of the type used.
6487 This example shows how the :ref:`va_arg <i_va_arg>` instruction and the
6488 variable argument handling intrinsic functions are used.
6490 .. code-block:: llvm
6492 define i32 @test(i32 %X, ...) {
6493 ; Initialize variable argument processing
6495 %ap2 = bitcast i8** %ap to i8*
6496 call void @llvm.va_start(i8* %ap2)
6498 ; Read a single integer argument
6499 %tmp = va_arg i8** %ap, i32
6501 ; Demonstrate usage of llvm.va_copy and llvm.va_end
6503 %aq2 = bitcast i8** %aq to i8*
6504 call void @llvm.va_copy(i8* %aq2, i8* %ap2)
6505 call void @llvm.va_end(i8* %aq2)
6507 ; Stop processing of arguments.
6508 call void @llvm.va_end(i8* %ap2)
6512 declare void @llvm.va_start(i8*)
6513 declare void @llvm.va_copy(i8*, i8*)
6514 declare void @llvm.va_end(i8*)
6518 '``llvm.va_start``' Intrinsic
6519 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6526 declare void @llvm.va_start(i8* <arglist>)
6531 The '``llvm.va_start``' intrinsic initializes ``*<arglist>`` for
6532 subsequent use by ``va_arg``.
6537 The argument is a pointer to a ``va_list`` element to initialize.
6542 The '``llvm.va_start``' intrinsic works just like the ``va_start`` macro
6543 available in C. In a target-dependent way, it initializes the
6544 ``va_list`` element to which the argument points, so that the next call
6545 to ``va_arg`` will produce the first variable argument passed to the
6546 function. Unlike the C ``va_start`` macro, this intrinsic does not need
6547 to know the last argument of the function as the compiler can figure
6550 '``llvm.va_end``' Intrinsic
6551 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
6558 declare void @llvm.va_end(i8* <arglist>)
6563 The '``llvm.va_end``' intrinsic destroys ``*<arglist>``, which has been
6564 initialized previously with ``llvm.va_start`` or ``llvm.va_copy``.
6569 The argument is a pointer to a ``va_list`` to destroy.
6574 The '``llvm.va_end``' intrinsic works just like the ``va_end`` macro
6575 available in C. In a target-dependent way, it destroys the ``va_list``
6576 element to which the argument points. Calls to
6577 :ref:`llvm.va_start <int_va_start>` and
6578 :ref:`llvm.va_copy <int_va_copy>` must be matched exactly with calls to
6583 '``llvm.va_copy``' Intrinsic
6584 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6591 declare void @llvm.va_copy(i8* <destarglist>, i8* <srcarglist>)
6596 The '``llvm.va_copy``' intrinsic copies the current argument position
6597 from the source argument list to the destination argument list.
6602 The first argument is a pointer to a ``va_list`` element to initialize.
6603 The second argument is a pointer to a ``va_list`` element to copy from.
6608 The '``llvm.va_copy``' intrinsic works just like the ``va_copy`` macro
6609 available in C. In a target-dependent way, it copies the source
6610 ``va_list`` element into the destination ``va_list`` element. This
6611 intrinsic is necessary because the `` llvm.va_start`` intrinsic may be
6612 arbitrarily complex and require, for example, memory allocation.
6614 Accurate Garbage Collection Intrinsics
6615 --------------------------------------
6617 LLVM support for `Accurate Garbage Collection <GarbageCollection.html>`_
6618 (GC) requires the implementation and generation of these intrinsics.
6619 These intrinsics allow identification of :ref:`GC roots on the
6620 stack <int_gcroot>`, as well as garbage collector implementations that
6621 require :ref:`read <int_gcread>` and :ref:`write <int_gcwrite>` barriers.
6622 Front-ends for type-safe garbage collected languages should generate
6623 these intrinsics to make use of the LLVM garbage collectors. For more
6624 details, see `Accurate Garbage Collection with
6625 LLVM <GarbageCollection.html>`_.
6627 The garbage collection intrinsics only operate on objects in the generic
6628 address space (address space zero).
6632 '``llvm.gcroot``' Intrinsic
6633 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
6640 declare void @llvm.gcroot(i8** %ptrloc, i8* %metadata)
6645 The '``llvm.gcroot``' intrinsic declares the existence of a GC root to
6646 the code generator, and allows some metadata to be associated with it.
6651 The first argument specifies the address of a stack object that contains
6652 the root pointer. The second pointer (which must be either a constant or
6653 a global value address) contains the meta-data to be associated with the
6659 At runtime, a call to this intrinsic stores a null pointer into the
6660 "ptrloc" location. At compile-time, the code generator generates
6661 information to allow the runtime to find the pointer at GC safe points.
6662 The '``llvm.gcroot``' intrinsic may only be used in a function which
6663 :ref:`specifies a GC algorithm <gc>`.
6667 '``llvm.gcread``' Intrinsic
6668 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
6675 declare i8* @llvm.gcread(i8* %ObjPtr, i8** %Ptr)
6680 The '``llvm.gcread``' intrinsic identifies reads of references from heap
6681 locations, allowing garbage collector implementations that require read
6687 The second argument is the address to read from, which should be an
6688 address allocated from the garbage collector. The first object is a
6689 pointer to the start of the referenced object, if needed by the language
6690 runtime (otherwise null).
6695 The '``llvm.gcread``' intrinsic has the same semantics as a load
6696 instruction, but may be replaced with substantially more complex code by
6697 the garbage collector runtime, as needed. The '``llvm.gcread``'
6698 intrinsic may only be used in a function which :ref:`specifies a GC
6703 '``llvm.gcwrite``' Intrinsic
6704 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6711 declare void @llvm.gcwrite(i8* %P1, i8* %Obj, i8** %P2)
6716 The '``llvm.gcwrite``' intrinsic identifies writes of references to heap
6717 locations, allowing garbage collector implementations that require write
6718 barriers (such as generational or reference counting collectors).
6723 The first argument is the reference to store, the second is the start of
6724 the object to store it to, and the third is the address of the field of
6725 Obj to store to. If the runtime does not require a pointer to the
6726 object, Obj may be null.
6731 The '``llvm.gcwrite``' intrinsic has the same semantics as a store
6732 instruction, but may be replaced with substantially more complex code by
6733 the garbage collector runtime, as needed. The '``llvm.gcwrite``'
6734 intrinsic may only be used in a function which :ref:`specifies a GC
6737 Code Generator Intrinsics
6738 -------------------------
6740 These intrinsics are provided by LLVM to expose special features that
6741 may only be implemented with code generator support.
6743 '``llvm.returnaddress``' Intrinsic
6744 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6751 declare i8 *@llvm.returnaddress(i32 <level>)
6756 The '``llvm.returnaddress``' intrinsic attempts to compute a
6757 target-specific value indicating the return address of the current
6758 function or one of its callers.
6763 The argument to this intrinsic indicates which function to return the
6764 address for. Zero indicates the calling function, one indicates its
6765 caller, etc. The argument is **required** to be a constant integer
6771 The '``llvm.returnaddress``' intrinsic either returns a pointer
6772 indicating the return address of the specified call frame, or zero if it
6773 cannot be identified. The value returned by this intrinsic is likely to
6774 be incorrect or 0 for arguments other than zero, so it should only be
6775 used for debugging purposes.
6777 Note that calling this intrinsic does not prevent function inlining or
6778 other aggressive transformations, so the value returned may not be that
6779 of the obvious source-language caller.
6781 '``llvm.frameaddress``' Intrinsic
6782 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6789 declare i8* @llvm.frameaddress(i32 <level>)
6794 The '``llvm.frameaddress``' intrinsic attempts to return the
6795 target-specific frame pointer value for the specified stack frame.
6800 The argument to this intrinsic indicates which function to return the
6801 frame pointer for. Zero indicates the calling function, one indicates
6802 its caller, etc. The argument is **required** to be a constant integer
6808 The '``llvm.frameaddress``' intrinsic either returns a pointer
6809 indicating the frame address of the specified call frame, or zero if it
6810 cannot be identified. The value returned by this intrinsic is likely to
6811 be incorrect or 0 for arguments other than zero, so it should only be
6812 used for debugging purposes.
6814 Note that calling this intrinsic does not prevent function inlining or
6815 other aggressive transformations, so the value returned may not be that
6816 of the obvious source-language caller.
6818 .. _int_read_register:
6819 .. _int_write_register:
6821 '``llvm.read_register``' and '``llvm.write_register``' Intrinsics
6822 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6829 declare i32 @llvm.read_register.i32(metadata)
6830 declare i64 @llvm.read_register.i64(metadata)
6831 declare void @llvm.write_register.i32(metadata, i32 @value)
6832 declare void @llvm.write_register.i64(metadata, i64 @value)
6833 !0 = metadata !{metadata !"sp\00"}
6838 The '``llvm.read_register``' and '``llvm.write_register``' intrinsics
6839 provides access to the named register. The register must be valid on
6840 the architecture being compiled to. The type needs to be compatible
6841 with the register being read.
6846 The '``llvm.read_register``' intrinsic returns the current value of the
6847 register, where possible. The '``llvm.write_register``' intrinsic sets
6848 the current value of the register, where possible.
6850 This is useful to implement named register global variables that need
6851 to always be mapped to a specific register, as is common practice on
6852 bare-metal programs including OS kernels.
6854 The compiler doesn't check for register availability or use of the used
6855 register in surrounding code, including inline assembly. Because of that,
6856 allocatable registers are not supported.
6858 Warning: So far it only works with the stack pointer on selected
6859 architectures (ARM, ARM64, AArch64, PowerPC and x86_64). Significant amount of
6860 work is needed to support other registers and even more so, allocatable
6865 '``llvm.stacksave``' Intrinsic
6866 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6873 declare i8* @llvm.stacksave()
6878 The '``llvm.stacksave``' intrinsic is used to remember the current state
6879 of the function stack, for use with
6880 :ref:`llvm.stackrestore <int_stackrestore>`. This is useful for
6881 implementing language features like scoped automatic variable sized
6887 This intrinsic returns a opaque pointer value that can be passed to
6888 :ref:`llvm.stackrestore <int_stackrestore>`. When an
6889 ``llvm.stackrestore`` intrinsic is executed with a value saved from
6890 ``llvm.stacksave``, it effectively restores the state of the stack to
6891 the state it was in when the ``llvm.stacksave`` intrinsic executed. In
6892 practice, this pops any :ref:`alloca <i_alloca>` blocks from the stack that
6893 were allocated after the ``llvm.stacksave`` was executed.
6895 .. _int_stackrestore:
6897 '``llvm.stackrestore``' Intrinsic
6898 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6905 declare void @llvm.stackrestore(i8* %ptr)
6910 The '``llvm.stackrestore``' intrinsic is used to restore the state of
6911 the function stack to the state it was in when the corresponding
6912 :ref:`llvm.stacksave <int_stacksave>` intrinsic executed. This is
6913 useful for implementing language features like scoped automatic variable
6914 sized arrays in C99.
6919 See the description for :ref:`llvm.stacksave <int_stacksave>`.
6921 '``llvm.prefetch``' Intrinsic
6922 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6929 declare void @llvm.prefetch(i8* <address>, i32 <rw>, i32 <locality>, i32 <cache type>)
6934 The '``llvm.prefetch``' intrinsic is a hint to the code generator to
6935 insert a prefetch instruction if supported; otherwise, it is a noop.
6936 Prefetches have no effect on the behavior of the program but can change
6937 its performance characteristics.
6942 ``address`` is the address to be prefetched, ``rw`` is the specifier
6943 determining if the fetch should be for a read (0) or write (1), and
6944 ``locality`` is a temporal locality specifier ranging from (0) - no
6945 locality, to (3) - extremely local keep in cache. The ``cache type``
6946 specifies whether the prefetch is performed on the data (1) or
6947 instruction (0) cache. The ``rw``, ``locality`` and ``cache type``
6948 arguments must be constant integers.
6953 This intrinsic does not modify the behavior of the program. In
6954 particular, prefetches cannot trap and do not produce a value. On
6955 targets that support this intrinsic, the prefetch can provide hints to
6956 the processor cache for better performance.
6958 '``llvm.pcmarker``' Intrinsic
6959 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6966 declare void @llvm.pcmarker(i32 <id>)
6971 The '``llvm.pcmarker``' intrinsic is a method to export a Program
6972 Counter (PC) in a region of code to simulators and other tools. The
6973 method is target specific, but it is expected that the marker will use
6974 exported symbols to transmit the PC of the marker. The marker makes no
6975 guarantees that it will remain with any specific instruction after
6976 optimizations. It is possible that the presence of a marker will inhibit
6977 optimizations. The intended use is to be inserted after optimizations to
6978 allow correlations of simulation runs.
6983 ``id`` is a numerical id identifying the marker.
6988 This intrinsic does not modify the behavior of the program. Backends
6989 that do not support this intrinsic may ignore it.
6991 '``llvm.readcyclecounter``' Intrinsic
6992 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
6999 declare i64 @llvm.readcyclecounter()
7004 The '``llvm.readcyclecounter``' intrinsic provides access to the cycle
7005 counter register (or similar low latency, high accuracy clocks) on those
7006 targets that support it. On X86, it should map to RDTSC. On Alpha, it
7007 should map to RPCC. As the backing counters overflow quickly (on the
7008 order of 9 seconds on alpha), this should only be used for small
7014 When directly supported, reading the cycle counter should not modify any
7015 memory. Implementations are allowed to either return a application
7016 specific value or a system wide value. On backends without support, this
7017 is lowered to a constant 0.
7019 Note that runtime support may be conditional on the privilege-level code is
7020 running at and the host platform.
7022 '``llvm.clear_cache``' Intrinsic
7023 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7030 declare void @llvm.clear_cache(i8*, i8*)
7035 The '``llvm.clear_cache``' intrinsic ensures visibility of modifications
7036 in the specified range to the execution unit of the processor. On
7037 targets with non-unified instruction and data cache, the implementation
7038 flushes the instruction cache.
7043 On platforms with coherent instruction and data caches (e.g. x86), this
7044 intrinsic is a nop. On platforms with non-coherent instruction and data
7045 cache (e.g. ARM, MIPS), the intrinsic is lowered either to appropriate
7046 instructions or a system call, if cache flushing requires special
7049 The default behavior is to emit a call to ``__clear_cache`` from the run
7052 This instrinsic does *not* empty the instruction pipeline. Modifications
7053 of the current function are outside the scope of the intrinsic.
7055 Standard C Library Intrinsics
7056 -----------------------------
7058 LLVM provides intrinsics for a few important standard C library
7059 functions. These intrinsics allow source-language front-ends to pass
7060 information about the alignment of the pointer arguments to the code
7061 generator, providing opportunity for more efficient code generation.
7065 '``llvm.memcpy``' Intrinsic
7066 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
7071 This is an overloaded intrinsic. You can use ``llvm.memcpy`` on any
7072 integer bit width and for different address spaces. Not all targets
7073 support all bit widths however.
7077 declare void @llvm.memcpy.p0i8.p0i8.i32(i8* <dest>, i8* <src>,
7078 i32 <len>, i32 <align>, i1 <isvolatile>)
7079 declare void @llvm.memcpy.p0i8.p0i8.i64(i8* <dest>, i8* <src>,
7080 i64 <len>, i32 <align>, i1 <isvolatile>)
7085 The '``llvm.memcpy.*``' intrinsics copy a block of memory from the
7086 source location to the destination location.
7088 Note that, unlike the standard libc function, the ``llvm.memcpy.*``
7089 intrinsics do not return a value, takes extra alignment/isvolatile
7090 arguments and the pointers can be in specified address spaces.
7095 The first argument is a pointer to the destination, the second is a
7096 pointer to the source. The third argument is an integer argument
7097 specifying the number of bytes to copy, the fourth argument is the
7098 alignment of the source and destination locations, and the fifth is a
7099 boolean indicating a volatile access.
7101 If the call to this intrinsic has an alignment value that is not 0 or 1,
7102 then the caller guarantees that both the source and destination pointers
7103 are aligned to that boundary.
7105 If the ``isvolatile`` parameter is ``true``, the ``llvm.memcpy`` call is
7106 a :ref:`volatile operation <volatile>`. The detailed access behavior is not
7107 very cleanly specified and it is unwise to depend on it.
7112 The '``llvm.memcpy.*``' intrinsics copy a block of memory from the
7113 source location to the destination location, which are not allowed to
7114 overlap. It copies "len" bytes of memory over. If the argument is known
7115 to be aligned to some boundary, this can be specified as the fourth
7116 argument, otherwise it should be set to 0 or 1 (both meaning no alignment).
7118 '``llvm.memmove``' Intrinsic
7119 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7124 This is an overloaded intrinsic. You can use llvm.memmove on any integer
7125 bit width and for different address space. Not all targets support all
7130 declare void @llvm.memmove.p0i8.p0i8.i32(i8* <dest>, i8* <src>,
7131 i32 <len>, i32 <align>, i1 <isvolatile>)
7132 declare void @llvm.memmove.p0i8.p0i8.i64(i8* <dest>, i8* <src>,
7133 i64 <len>, i32 <align>, i1 <isvolatile>)
7138 The '``llvm.memmove.*``' intrinsics move a block of memory from the
7139 source location to the destination location. It is similar to the
7140 '``llvm.memcpy``' intrinsic but allows the two memory locations to
7143 Note that, unlike the standard libc function, the ``llvm.memmove.*``
7144 intrinsics do not return a value, takes extra alignment/isvolatile
7145 arguments and the pointers can be in specified address spaces.
7150 The first argument is a pointer to the destination, the second is a
7151 pointer to the source. The third argument is an integer argument
7152 specifying the number of bytes to copy, the fourth argument is the
7153 alignment of the source and destination locations, and the fifth is a
7154 boolean indicating a volatile access.
7156 If the call to this intrinsic has an alignment value that is not 0 or 1,
7157 then the caller guarantees that the source and destination pointers are
7158 aligned to that boundary.
7160 If the ``isvolatile`` parameter is ``true``, the ``llvm.memmove`` call
7161 is a :ref:`volatile operation <volatile>`. The detailed access behavior is
7162 not very cleanly specified and it is unwise to depend on it.
7167 The '``llvm.memmove.*``' intrinsics copy a block of memory from the
7168 source location to the destination location, which may overlap. It
7169 copies "len" bytes of memory over. If the argument is known to be
7170 aligned to some boundary, this can be specified as the fourth argument,
7171 otherwise it should be set to 0 or 1 (both meaning no alignment).
7173 '``llvm.memset.*``' Intrinsics
7174 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7179 This is an overloaded intrinsic. You can use llvm.memset on any integer
7180 bit width and for different address spaces. However, not all targets
7181 support all bit widths.
7185 declare void @llvm.memset.p0i8.i32(i8* <dest>, i8 <val>,
7186 i32 <len>, i32 <align>, i1 <isvolatile>)
7187 declare void @llvm.memset.p0i8.i64(i8* <dest>, i8 <val>,
7188 i64 <len>, i32 <align>, i1 <isvolatile>)
7193 The '``llvm.memset.*``' intrinsics fill a block of memory with a
7194 particular byte value.
7196 Note that, unlike the standard libc function, the ``llvm.memset``
7197 intrinsic does not return a value and takes extra alignment/volatile
7198 arguments. Also, the destination can be in an arbitrary address space.
7203 The first argument is a pointer to the destination to fill, the second
7204 is the byte value with which to fill it, the third argument is an
7205 integer argument specifying the number of bytes to fill, and the fourth
7206 argument is the known alignment of the destination location.
7208 If the call to this intrinsic has an alignment value that is not 0 or 1,
7209 then the caller guarantees that the destination pointer is aligned to
7212 If the ``isvolatile`` parameter is ``true``, the ``llvm.memset`` call is
7213 a :ref:`volatile operation <volatile>`. The detailed access behavior is not
7214 very cleanly specified and it is unwise to depend on it.
7219 The '``llvm.memset.*``' intrinsics fill "len" bytes of memory starting
7220 at the destination location. If the argument is known to be aligned to
7221 some boundary, this can be specified as the fourth argument, otherwise
7222 it should be set to 0 or 1 (both meaning no alignment).
7224 '``llvm.sqrt.*``' Intrinsic
7225 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
7230 This is an overloaded intrinsic. You can use ``llvm.sqrt`` on any
7231 floating point or vector of floating point type. Not all targets support
7236 declare float @llvm.sqrt.f32(float %Val)
7237 declare double @llvm.sqrt.f64(double %Val)
7238 declare x86_fp80 @llvm.sqrt.f80(x86_fp80 %Val)
7239 declare fp128 @llvm.sqrt.f128(fp128 %Val)
7240 declare ppc_fp128 @llvm.sqrt.ppcf128(ppc_fp128 %Val)
7245 The '``llvm.sqrt``' intrinsics return the sqrt of the specified operand,
7246 returning the same value as the libm '``sqrt``' functions would. Unlike
7247 ``sqrt`` in libm, however, ``llvm.sqrt`` has undefined behavior for
7248 negative numbers other than -0.0 (which allows for better optimization,
7249 because there is no need to worry about errno being set).
7250 ``llvm.sqrt(-0.0)`` is defined to return -0.0 like IEEE sqrt.
7255 The argument and return value are floating point numbers of the same
7261 This function returns the sqrt of the specified operand if it is a
7262 nonnegative floating point number.
7264 '``llvm.powi.*``' Intrinsic
7265 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
7270 This is an overloaded intrinsic. You can use ``llvm.powi`` on any
7271 floating point or vector of floating point type. Not all targets support
7276 declare float @llvm.powi.f32(float %Val, i32 %power)
7277 declare double @llvm.powi.f64(double %Val, i32 %power)
7278 declare x86_fp80 @llvm.powi.f80(x86_fp80 %Val, i32 %power)
7279 declare fp128 @llvm.powi.f128(fp128 %Val, i32 %power)
7280 declare ppc_fp128 @llvm.powi.ppcf128(ppc_fp128 %Val, i32 %power)
7285 The '``llvm.powi.*``' intrinsics return the first operand raised to the
7286 specified (positive or negative) power. The order of evaluation of
7287 multiplications is not defined. When a vector of floating point type is
7288 used, the second argument remains a scalar integer value.
7293 The second argument is an integer power, and the first is a value to
7294 raise to that power.
7299 This function returns the first value raised to the second power with an
7300 unspecified sequence of rounding operations.
7302 '``llvm.sin.*``' Intrinsic
7303 ^^^^^^^^^^^^^^^^^^^^^^^^^^
7308 This is an overloaded intrinsic. You can use ``llvm.sin`` on any
7309 floating point or vector of floating point type. Not all targets support
7314 declare float @llvm.sin.f32(float %Val)
7315 declare double @llvm.sin.f64(double %Val)
7316 declare x86_fp80 @llvm.sin.f80(x86_fp80 %Val)
7317 declare fp128 @llvm.sin.f128(fp128 %Val)
7318 declare ppc_fp128 @llvm.sin.ppcf128(ppc_fp128 %Val)
7323 The '``llvm.sin.*``' intrinsics return the sine of the operand.
7328 The argument and return value are floating point numbers of the same
7334 This function returns the sine of the specified operand, returning the
7335 same values as the libm ``sin`` functions would, and handles error
7336 conditions in the same way.
7338 '``llvm.cos.*``' Intrinsic
7339 ^^^^^^^^^^^^^^^^^^^^^^^^^^
7344 This is an overloaded intrinsic. You can use ``llvm.cos`` on any
7345 floating point or vector of floating point type. Not all targets support
7350 declare float @llvm.cos.f32(float %Val)
7351 declare double @llvm.cos.f64(double %Val)
7352 declare x86_fp80 @llvm.cos.f80(x86_fp80 %Val)
7353 declare fp128 @llvm.cos.f128(fp128 %Val)
7354 declare ppc_fp128 @llvm.cos.ppcf128(ppc_fp128 %Val)
7359 The '``llvm.cos.*``' intrinsics return the cosine of the operand.
7364 The argument and return value are floating point numbers of the same
7370 This function returns the cosine of the specified operand, returning the
7371 same values as the libm ``cos`` functions would, and handles error
7372 conditions in the same way.
7374 '``llvm.pow.*``' Intrinsic
7375 ^^^^^^^^^^^^^^^^^^^^^^^^^^
7380 This is an overloaded intrinsic. You can use ``llvm.pow`` on any
7381 floating point or vector of floating point type. Not all targets support
7386 declare float @llvm.pow.f32(float %Val, float %Power)
7387 declare double @llvm.pow.f64(double %Val, double %Power)
7388 declare x86_fp80 @llvm.pow.f80(x86_fp80 %Val, x86_fp80 %Power)
7389 declare fp128 @llvm.pow.f128(fp128 %Val, fp128 %Power)
7390 declare ppc_fp128 @llvm.pow.ppcf128(ppc_fp128 %Val, ppc_fp128 Power)
7395 The '``llvm.pow.*``' intrinsics return the first operand raised to the
7396 specified (positive or negative) power.
7401 The second argument is a floating point power, and the first is a value
7402 to raise to that power.
7407 This function returns the first value raised to the second power,
7408 returning the same values as the libm ``pow`` functions would, and
7409 handles error conditions in the same way.
7411 '``llvm.exp.*``' Intrinsic
7412 ^^^^^^^^^^^^^^^^^^^^^^^^^^
7417 This is an overloaded intrinsic. You can use ``llvm.exp`` on any
7418 floating point or vector of floating point type. Not all targets support
7423 declare float @llvm.exp.f32(float %Val)
7424 declare double @llvm.exp.f64(double %Val)
7425 declare x86_fp80 @llvm.exp.f80(x86_fp80 %Val)
7426 declare fp128 @llvm.exp.f128(fp128 %Val)
7427 declare ppc_fp128 @llvm.exp.ppcf128(ppc_fp128 %Val)
7432 The '``llvm.exp.*``' intrinsics perform the exp function.
7437 The argument and return value are floating point numbers of the same
7443 This function returns the same values as the libm ``exp`` functions
7444 would, and handles error conditions in the same way.
7446 '``llvm.exp2.*``' Intrinsic
7447 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
7452 This is an overloaded intrinsic. You can use ``llvm.exp2`` on any
7453 floating point or vector of floating point type. Not all targets support
7458 declare float @llvm.exp2.f32(float %Val)
7459 declare double @llvm.exp2.f64(double %Val)
7460 declare x86_fp80 @llvm.exp2.f80(x86_fp80 %Val)
7461 declare fp128 @llvm.exp2.f128(fp128 %Val)
7462 declare ppc_fp128 @llvm.exp2.ppcf128(ppc_fp128 %Val)
7467 The '``llvm.exp2.*``' intrinsics perform the exp2 function.
7472 The argument and return value are floating point numbers of the same
7478 This function returns the same values as the libm ``exp2`` functions
7479 would, and handles error conditions in the same way.
7481 '``llvm.log.*``' Intrinsic
7482 ^^^^^^^^^^^^^^^^^^^^^^^^^^
7487 This is an overloaded intrinsic. You can use ``llvm.log`` on any
7488 floating point or vector of floating point type. Not all targets support
7493 declare float @llvm.log.f32(float %Val)
7494 declare double @llvm.log.f64(double %Val)
7495 declare x86_fp80 @llvm.log.f80(x86_fp80 %Val)
7496 declare fp128 @llvm.log.f128(fp128 %Val)
7497 declare ppc_fp128 @llvm.log.ppcf128(ppc_fp128 %Val)
7502 The '``llvm.log.*``' intrinsics perform the log function.
7507 The argument and return value are floating point numbers of the same
7513 This function returns the same values as the libm ``log`` functions
7514 would, and handles error conditions in the same way.
7516 '``llvm.log10.*``' Intrinsic
7517 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7522 This is an overloaded intrinsic. You can use ``llvm.log10`` on any
7523 floating point or vector of floating point type. Not all targets support
7528 declare float @llvm.log10.f32(float %Val)
7529 declare double @llvm.log10.f64(double %Val)
7530 declare x86_fp80 @llvm.log10.f80(x86_fp80 %Val)
7531 declare fp128 @llvm.log10.f128(fp128 %Val)
7532 declare ppc_fp128 @llvm.log10.ppcf128(ppc_fp128 %Val)
7537 The '``llvm.log10.*``' intrinsics perform the log10 function.
7542 The argument and return value are floating point numbers of the same
7548 This function returns the same values as the libm ``log10`` functions
7549 would, and handles error conditions in the same way.
7551 '``llvm.log2.*``' Intrinsic
7552 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
7557 This is an overloaded intrinsic. You can use ``llvm.log2`` on any
7558 floating point or vector of floating point type. Not all targets support
7563 declare float @llvm.log2.f32(float %Val)
7564 declare double @llvm.log2.f64(double %Val)
7565 declare x86_fp80 @llvm.log2.f80(x86_fp80 %Val)
7566 declare fp128 @llvm.log2.f128(fp128 %Val)
7567 declare ppc_fp128 @llvm.log2.ppcf128(ppc_fp128 %Val)
7572 The '``llvm.log2.*``' intrinsics perform the log2 function.
7577 The argument and return value are floating point numbers of the same
7583 This function returns the same values as the libm ``log2`` functions
7584 would, and handles error conditions in the same way.
7586 '``llvm.fma.*``' Intrinsic
7587 ^^^^^^^^^^^^^^^^^^^^^^^^^^
7592 This is an overloaded intrinsic. You can use ``llvm.fma`` on any
7593 floating point or vector of floating point type. Not all targets support
7598 declare float @llvm.fma.f32(float %a, float %b, float %c)
7599 declare double @llvm.fma.f64(double %a, double %b, double %c)
7600 declare x86_fp80 @llvm.fma.f80(x86_fp80 %a, x86_fp80 %b, x86_fp80 %c)
7601 declare fp128 @llvm.fma.f128(fp128 %a, fp128 %b, fp128 %c)
7602 declare ppc_fp128 @llvm.fma.ppcf128(ppc_fp128 %a, ppc_fp128 %b, ppc_fp128 %c)
7607 The '``llvm.fma.*``' intrinsics perform the fused multiply-add
7613 The argument and return value are floating point numbers of the same
7619 This function returns the same values as the libm ``fma`` functions
7620 would, and does not set errno.
7622 '``llvm.fabs.*``' Intrinsic
7623 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
7628 This is an overloaded intrinsic. You can use ``llvm.fabs`` on any
7629 floating point or vector of floating point type. Not all targets support
7634 declare float @llvm.fabs.f32(float %Val)
7635 declare double @llvm.fabs.f64(double %Val)
7636 declare x86_fp80 @llvm.fabs.f80(x86_fp80 %Val)
7637 declare fp128 @llvm.fabs.f128(fp128 %Val)
7638 declare ppc_fp128 @llvm.fabs.ppcf128(ppc_fp128 %Val)
7643 The '``llvm.fabs.*``' intrinsics return the absolute value of the
7649 The argument and return value are floating point numbers of the same
7655 This function returns the same values as the libm ``fabs`` functions
7656 would, and handles error conditions in the same way.
7658 '``llvm.copysign.*``' Intrinsic
7659 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7664 This is an overloaded intrinsic. You can use ``llvm.copysign`` on any
7665 floating point or vector of floating point type. Not all targets support
7670 declare float @llvm.copysign.f32(float %Mag, float %Sgn)
7671 declare double @llvm.copysign.f64(double %Mag, double %Sgn)
7672 declare x86_fp80 @llvm.copysign.f80(x86_fp80 %Mag, x86_fp80 %Sgn)
7673 declare fp128 @llvm.copysign.f128(fp128 %Mag, fp128 %Sgn)
7674 declare ppc_fp128 @llvm.copysign.ppcf128(ppc_fp128 %Mag, ppc_fp128 %Sgn)
7679 The '``llvm.copysign.*``' intrinsics return a value with the magnitude of the
7680 first operand and the sign of the second operand.
7685 The arguments and return value are floating point numbers of the same
7691 This function returns the same values as the libm ``copysign``
7692 functions would, and handles error conditions in the same way.
7694 '``llvm.floor.*``' Intrinsic
7695 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7700 This is an overloaded intrinsic. You can use ``llvm.floor`` on any
7701 floating point or vector of floating point type. Not all targets support
7706 declare float @llvm.floor.f32(float %Val)
7707 declare double @llvm.floor.f64(double %Val)
7708 declare x86_fp80 @llvm.floor.f80(x86_fp80 %Val)
7709 declare fp128 @llvm.floor.f128(fp128 %Val)
7710 declare ppc_fp128 @llvm.floor.ppcf128(ppc_fp128 %Val)
7715 The '``llvm.floor.*``' intrinsics return the floor of the operand.
7720 The argument and return value are floating point numbers of the same
7726 This function returns the same values as the libm ``floor`` functions
7727 would, and handles error conditions in the same way.
7729 '``llvm.ceil.*``' Intrinsic
7730 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
7735 This is an overloaded intrinsic. You can use ``llvm.ceil`` on any
7736 floating point or vector of floating point type. Not all targets support
7741 declare float @llvm.ceil.f32(float %Val)
7742 declare double @llvm.ceil.f64(double %Val)
7743 declare x86_fp80 @llvm.ceil.f80(x86_fp80 %Val)
7744 declare fp128 @llvm.ceil.f128(fp128 %Val)
7745 declare ppc_fp128 @llvm.ceil.ppcf128(ppc_fp128 %Val)
7750 The '``llvm.ceil.*``' intrinsics return the ceiling of the operand.
7755 The argument and return value are floating point numbers of the same
7761 This function returns the same values as the libm ``ceil`` functions
7762 would, and handles error conditions in the same way.
7764 '``llvm.trunc.*``' Intrinsic
7765 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7770 This is an overloaded intrinsic. You can use ``llvm.trunc`` on any
7771 floating point or vector of floating point type. Not all targets support
7776 declare float @llvm.trunc.f32(float %Val)
7777 declare double @llvm.trunc.f64(double %Val)
7778 declare x86_fp80 @llvm.trunc.f80(x86_fp80 %Val)
7779 declare fp128 @llvm.trunc.f128(fp128 %Val)
7780 declare ppc_fp128 @llvm.trunc.ppcf128(ppc_fp128 %Val)
7785 The '``llvm.trunc.*``' intrinsics returns the operand rounded to the
7786 nearest integer not larger in magnitude than the operand.
7791 The argument and return value are floating point numbers of the same
7797 This function returns the same values as the libm ``trunc`` functions
7798 would, and handles error conditions in the same way.
7800 '``llvm.rint.*``' Intrinsic
7801 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
7806 This is an overloaded intrinsic. You can use ``llvm.rint`` on any
7807 floating point or vector of floating point type. Not all targets support
7812 declare float @llvm.rint.f32(float %Val)
7813 declare double @llvm.rint.f64(double %Val)
7814 declare x86_fp80 @llvm.rint.f80(x86_fp80 %Val)
7815 declare fp128 @llvm.rint.f128(fp128 %Val)
7816 declare ppc_fp128 @llvm.rint.ppcf128(ppc_fp128 %Val)
7821 The '``llvm.rint.*``' intrinsics returns the operand rounded to the
7822 nearest integer. It may raise an inexact floating-point exception if the
7823 operand isn't an integer.
7828 The argument and return value are floating point numbers of the same
7834 This function returns the same values as the libm ``rint`` functions
7835 would, and handles error conditions in the same way.
7837 '``llvm.nearbyint.*``' Intrinsic
7838 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7843 This is an overloaded intrinsic. You can use ``llvm.nearbyint`` on any
7844 floating point or vector of floating point type. Not all targets support
7849 declare float @llvm.nearbyint.f32(float %Val)
7850 declare double @llvm.nearbyint.f64(double %Val)
7851 declare x86_fp80 @llvm.nearbyint.f80(x86_fp80 %Val)
7852 declare fp128 @llvm.nearbyint.f128(fp128 %Val)
7853 declare ppc_fp128 @llvm.nearbyint.ppcf128(ppc_fp128 %Val)
7858 The '``llvm.nearbyint.*``' intrinsics returns the operand rounded to the
7864 The argument and return value are floating point numbers of the same
7870 This function returns the same values as the libm ``nearbyint``
7871 functions would, and handles error conditions in the same way.
7873 '``llvm.round.*``' Intrinsic
7874 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7879 This is an overloaded intrinsic. You can use ``llvm.round`` on any
7880 floating point or vector of floating point type. Not all targets support
7885 declare float @llvm.round.f32(float %Val)
7886 declare double @llvm.round.f64(double %Val)
7887 declare x86_fp80 @llvm.round.f80(x86_fp80 %Val)
7888 declare fp128 @llvm.round.f128(fp128 %Val)
7889 declare ppc_fp128 @llvm.round.ppcf128(ppc_fp128 %Val)
7894 The '``llvm.round.*``' intrinsics returns the operand rounded to the
7900 The argument and return value are floating point numbers of the same
7906 This function returns the same values as the libm ``round``
7907 functions would, and handles error conditions in the same way.
7909 Bit Manipulation Intrinsics
7910 ---------------------------
7912 LLVM provides intrinsics for a few important bit manipulation
7913 operations. These allow efficient code generation for some algorithms.
7915 '``llvm.bswap.*``' Intrinsics
7916 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7921 This is an overloaded intrinsic function. You can use bswap on any
7922 integer type that is an even number of bytes (i.e. BitWidth % 16 == 0).
7926 declare i16 @llvm.bswap.i16(i16 <id>)
7927 declare i32 @llvm.bswap.i32(i32 <id>)
7928 declare i64 @llvm.bswap.i64(i64 <id>)
7933 The '``llvm.bswap``' family of intrinsics is used to byte swap integer
7934 values with an even number of bytes (positive multiple of 16 bits).
7935 These are useful for performing operations on data that is not in the
7936 target's native byte order.
7941 The ``llvm.bswap.i16`` intrinsic returns an i16 value that has the high
7942 and low byte of the input i16 swapped. Similarly, the ``llvm.bswap.i32``
7943 intrinsic returns an i32 value that has the four bytes of the input i32
7944 swapped, so that if the input bytes are numbered 0, 1, 2, 3 then the
7945 returned i32 will have its bytes in 3, 2, 1, 0 order. The
7946 ``llvm.bswap.i48``, ``llvm.bswap.i64`` and other intrinsics extend this
7947 concept to additional even-byte lengths (6 bytes, 8 bytes and more,
7950 '``llvm.ctpop.*``' Intrinsic
7951 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
7956 This is an overloaded intrinsic. You can use llvm.ctpop on any integer
7957 bit width, or on any vector with integer elements. Not all targets
7958 support all bit widths or vector types, however.
7962 declare i8 @llvm.ctpop.i8(i8 <src>)
7963 declare i16 @llvm.ctpop.i16(i16 <src>)
7964 declare i32 @llvm.ctpop.i32(i32 <src>)
7965 declare i64 @llvm.ctpop.i64(i64 <src>)
7966 declare i256 @llvm.ctpop.i256(i256 <src>)
7967 declare <2 x i32> @llvm.ctpop.v2i32(<2 x i32> <src>)
7972 The '``llvm.ctpop``' family of intrinsics counts the number of bits set
7978 The only argument is the value to be counted. The argument may be of any
7979 integer type, or a vector with integer elements. The return type must
7980 match the argument type.
7985 The '``llvm.ctpop``' intrinsic counts the 1's in a variable, or within
7986 each element of a vector.
7988 '``llvm.ctlz.*``' Intrinsic
7989 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
7994 This is an overloaded intrinsic. You can use ``llvm.ctlz`` on any
7995 integer bit width, or any vector whose elements are integers. Not all
7996 targets support all bit widths or vector types, however.
8000 declare i8 @llvm.ctlz.i8 (i8 <src>, i1 <is_zero_undef>)
8001 declare i16 @llvm.ctlz.i16 (i16 <src>, i1 <is_zero_undef>)
8002 declare i32 @llvm.ctlz.i32 (i32 <src>, i1 <is_zero_undef>)
8003 declare i64 @llvm.ctlz.i64 (i64 <src>, i1 <is_zero_undef>)
8004 declare i256 @llvm.ctlz.i256(i256 <src>, i1 <is_zero_undef>)
8005 declase <2 x i32> @llvm.ctlz.v2i32(<2 x i32> <src>, i1 <is_zero_undef>)
8010 The '``llvm.ctlz``' family of intrinsic functions counts the number of
8011 leading zeros in a variable.
8016 The first argument is the value to be counted. This argument may be of
8017 any integer type, or a vectory with integer element type. The return
8018 type must match the first argument type.
8020 The second argument must be a constant and is a flag to indicate whether
8021 the intrinsic should ensure that a zero as the first argument produces a
8022 defined result. Historically some architectures did not provide a
8023 defined result for zero values as efficiently, and many algorithms are
8024 now predicated on avoiding zero-value inputs.
8029 The '``llvm.ctlz``' intrinsic counts the leading (most significant)
8030 zeros in a variable, or within each element of the vector. If
8031 ``src == 0`` then the result is the size in bits of the type of ``src``
8032 if ``is_zero_undef == 0`` and ``undef`` otherwise. For example,
8033 ``llvm.ctlz(i32 2) = 30``.
8035 '``llvm.cttz.*``' Intrinsic
8036 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
8041 This is an overloaded intrinsic. You can use ``llvm.cttz`` on any
8042 integer bit width, or any vector of integer elements. Not all targets
8043 support all bit widths or vector types, however.
8047 declare i8 @llvm.cttz.i8 (i8 <src>, i1 <is_zero_undef>)
8048 declare i16 @llvm.cttz.i16 (i16 <src>, i1 <is_zero_undef>)
8049 declare i32 @llvm.cttz.i32 (i32 <src>, i1 <is_zero_undef>)
8050 declare i64 @llvm.cttz.i64 (i64 <src>, i1 <is_zero_undef>)
8051 declare i256 @llvm.cttz.i256(i256 <src>, i1 <is_zero_undef>)
8052 declase <2 x i32> @llvm.cttz.v2i32(<2 x i32> <src>, i1 <is_zero_undef>)
8057 The '``llvm.cttz``' family of intrinsic functions counts the number of
8063 The first argument is the value to be counted. This argument may be of
8064 any integer type, or a vectory with integer element type. The return
8065 type must match the first argument type.
8067 The second argument must be a constant and is a flag to indicate whether
8068 the intrinsic should ensure that a zero as the first argument produces a
8069 defined result. Historically some architectures did not provide a
8070 defined result for zero values as efficiently, and many algorithms are
8071 now predicated on avoiding zero-value inputs.
8076 The '``llvm.cttz``' intrinsic counts the trailing (least significant)
8077 zeros in a variable, or within each element of a vector. If ``src == 0``
8078 then the result is the size in bits of the type of ``src`` if
8079 ``is_zero_undef == 0`` and ``undef`` otherwise. For example,
8080 ``llvm.cttz(2) = 1``.
8082 Arithmetic with Overflow Intrinsics
8083 -----------------------------------
8085 LLVM provides intrinsics for some arithmetic with overflow operations.
8087 '``llvm.sadd.with.overflow.*``' Intrinsics
8088 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8093 This is an overloaded intrinsic. You can use ``llvm.sadd.with.overflow``
8094 on any integer bit width.
8098 declare {i16, i1} @llvm.sadd.with.overflow.i16(i16 %a, i16 %b)
8099 declare {i32, i1} @llvm.sadd.with.overflow.i32(i32 %a, i32 %b)
8100 declare {i64, i1} @llvm.sadd.with.overflow.i64(i64 %a, i64 %b)
8105 The '``llvm.sadd.with.overflow``' family of intrinsic functions perform
8106 a signed addition of the two arguments, and indicate whether an overflow
8107 occurred during the signed summation.
8112 The arguments (%a and %b) and the first element of the result structure
8113 may be of integer types of any bit width, but they must have the same
8114 bit width. The second element of the result structure must be of type
8115 ``i1``. ``%a`` and ``%b`` are the two values that will undergo signed
8121 The '``llvm.sadd.with.overflow``' family of intrinsic functions perform
8122 a signed addition of the two variables. They return a structure --- the
8123 first element of which is the signed summation, and the second element
8124 of which is a bit specifying if the signed summation resulted in an
8130 .. code-block:: llvm
8132 %res = call {i32, i1} @llvm.sadd.with.overflow.i32(i32 %a, i32 %b)
8133 %sum = extractvalue {i32, i1} %res, 0
8134 %obit = extractvalue {i32, i1} %res, 1
8135 br i1 %obit, label %overflow, label %normal
8137 '``llvm.uadd.with.overflow.*``' Intrinsics
8138 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8143 This is an overloaded intrinsic. You can use ``llvm.uadd.with.overflow``
8144 on any integer bit width.
8148 declare {i16, i1} @llvm.uadd.with.overflow.i16(i16 %a, i16 %b)
8149 declare {i32, i1} @llvm.uadd.with.overflow.i32(i32 %a, i32 %b)
8150 declare {i64, i1} @llvm.uadd.with.overflow.i64(i64 %a, i64 %b)
8155 The '``llvm.uadd.with.overflow``' family of intrinsic functions perform
8156 an unsigned addition of the two arguments, and indicate whether a carry
8157 occurred during the unsigned summation.
8162 The arguments (%a and %b) and the first element of the result structure
8163 may be of integer types of any bit width, but they must have the same
8164 bit width. The second element of the result structure must be of type
8165 ``i1``. ``%a`` and ``%b`` are the two values that will undergo unsigned
8171 The '``llvm.uadd.with.overflow``' family of intrinsic functions perform
8172 an unsigned addition of the two arguments. They return a structure --- the
8173 first element of which is the sum, and the second element of which is a
8174 bit specifying if the unsigned summation resulted in a carry.
8179 .. code-block:: llvm
8181 %res = call {i32, i1} @llvm.uadd.with.overflow.i32(i32 %a, i32 %b)
8182 %sum = extractvalue {i32, i1} %res, 0
8183 %obit = extractvalue {i32, i1} %res, 1
8184 br i1 %obit, label %carry, label %normal
8186 '``llvm.ssub.with.overflow.*``' Intrinsics
8187 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8192 This is an overloaded intrinsic. You can use ``llvm.ssub.with.overflow``
8193 on any integer bit width.
8197 declare {i16, i1} @llvm.ssub.with.overflow.i16(i16 %a, i16 %b)
8198 declare {i32, i1} @llvm.ssub.with.overflow.i32(i32 %a, i32 %b)
8199 declare {i64, i1} @llvm.ssub.with.overflow.i64(i64 %a, i64 %b)
8204 The '``llvm.ssub.with.overflow``' family of intrinsic functions perform
8205 a signed subtraction of the two arguments, and indicate whether an
8206 overflow occurred during the signed subtraction.
8211 The arguments (%a and %b) and the first element of the result structure
8212 may be of integer types of any bit width, but they must have the same
8213 bit width. The second element of the result structure must be of type
8214 ``i1``. ``%a`` and ``%b`` are the two values that will undergo signed
8220 The '``llvm.ssub.with.overflow``' family of intrinsic functions perform
8221 a signed subtraction of the two arguments. They return a structure --- the
8222 first element of which is the subtraction, and the second element of
8223 which is a bit specifying if the signed subtraction resulted in an
8229 .. code-block:: llvm
8231 %res = call {i32, i1} @llvm.ssub.with.overflow.i32(i32 %a, i32 %b)
8232 %sum = extractvalue {i32, i1} %res, 0
8233 %obit = extractvalue {i32, i1} %res, 1
8234 br i1 %obit, label %overflow, label %normal
8236 '``llvm.usub.with.overflow.*``' Intrinsics
8237 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8242 This is an overloaded intrinsic. You can use ``llvm.usub.with.overflow``
8243 on any integer bit width.
8247 declare {i16, i1} @llvm.usub.with.overflow.i16(i16 %a, i16 %b)
8248 declare {i32, i1} @llvm.usub.with.overflow.i32(i32 %a, i32 %b)
8249 declare {i64, i1} @llvm.usub.with.overflow.i64(i64 %a, i64 %b)
8254 The '``llvm.usub.with.overflow``' family of intrinsic functions perform
8255 an unsigned subtraction of the two arguments, and indicate whether an
8256 overflow occurred during the unsigned subtraction.
8261 The arguments (%a and %b) and the first element of the result structure
8262 may be of integer types of any bit width, but they must have the same
8263 bit width. The second element of the result structure must be of type
8264 ``i1``. ``%a`` and ``%b`` are the two values that will undergo unsigned
8270 The '``llvm.usub.with.overflow``' family of intrinsic functions perform
8271 an unsigned subtraction of the two arguments. They return a structure ---
8272 the first element of which is the subtraction, and the second element of
8273 which is a bit specifying if the unsigned subtraction resulted in an
8279 .. code-block:: llvm
8281 %res = call {i32, i1} @llvm.usub.with.overflow.i32(i32 %a, i32 %b)
8282 %sum = extractvalue {i32, i1} %res, 0
8283 %obit = extractvalue {i32, i1} %res, 1
8284 br i1 %obit, label %overflow, label %normal
8286 '``llvm.smul.with.overflow.*``' Intrinsics
8287 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8292 This is an overloaded intrinsic. You can use ``llvm.smul.with.overflow``
8293 on any integer bit width.
8297 declare {i16, i1} @llvm.smul.with.overflow.i16(i16 %a, i16 %b)
8298 declare {i32, i1} @llvm.smul.with.overflow.i32(i32 %a, i32 %b)
8299 declare {i64, i1} @llvm.smul.with.overflow.i64(i64 %a, i64 %b)
8304 The '``llvm.smul.with.overflow``' family of intrinsic functions perform
8305 a signed multiplication of the two arguments, and indicate whether an
8306 overflow occurred during the signed multiplication.
8311 The arguments (%a and %b) and the first element of the result structure
8312 may be of integer types of any bit width, but they must have the same
8313 bit width. The second element of the result structure must be of type
8314 ``i1``. ``%a`` and ``%b`` are the two values that will undergo signed
8320 The '``llvm.smul.with.overflow``' family of intrinsic functions perform
8321 a signed multiplication of the two arguments. They return a structure ---
8322 the first element of which is the multiplication, and the second element
8323 of which is a bit specifying if the signed multiplication resulted in an
8329 .. code-block:: llvm
8331 %res = call {i32, i1} @llvm.smul.with.overflow.i32(i32 %a, i32 %b)
8332 %sum = extractvalue {i32, i1} %res, 0
8333 %obit = extractvalue {i32, i1} %res, 1
8334 br i1 %obit, label %overflow, label %normal
8336 '``llvm.umul.with.overflow.*``' Intrinsics
8337 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8342 This is an overloaded intrinsic. You can use ``llvm.umul.with.overflow``
8343 on any integer bit width.
8347 declare {i16, i1} @llvm.umul.with.overflow.i16(i16 %a, i16 %b)
8348 declare {i32, i1} @llvm.umul.with.overflow.i32(i32 %a, i32 %b)
8349 declare {i64, i1} @llvm.umul.with.overflow.i64(i64 %a, i64 %b)
8354 The '``llvm.umul.with.overflow``' family of intrinsic functions perform
8355 a unsigned multiplication of the two arguments, and indicate whether an
8356 overflow occurred during the unsigned multiplication.
8361 The arguments (%a and %b) and the first element of the result structure
8362 may be of integer types of any bit width, but they must have the same
8363 bit width. The second element of the result structure must be of type
8364 ``i1``. ``%a`` and ``%b`` are the two values that will undergo unsigned
8370 The '``llvm.umul.with.overflow``' family of intrinsic functions perform
8371 an unsigned multiplication of the two arguments. They return a structure ---
8372 the first element of which is the multiplication, and the second
8373 element of which is a bit specifying if the unsigned multiplication
8374 resulted in an overflow.
8379 .. code-block:: llvm
8381 %res = call {i32, i1} @llvm.umul.with.overflow.i32(i32 %a, i32 %b)
8382 %sum = extractvalue {i32, i1} %res, 0
8383 %obit = extractvalue {i32, i1} %res, 1
8384 br i1 %obit, label %overflow, label %normal
8386 Specialised Arithmetic Intrinsics
8387 ---------------------------------
8389 '``llvm.fmuladd.*``' Intrinsic
8390 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8397 declare float @llvm.fmuladd.f32(float %a, float %b, float %c)
8398 declare double @llvm.fmuladd.f64(double %a, double %b, double %c)
8403 The '``llvm.fmuladd.*``' intrinsic functions represent multiply-add
8404 expressions that can be fused if the code generator determines that (a) the
8405 target instruction set has support for a fused operation, and (b) that the
8406 fused operation is more efficient than the equivalent, separate pair of mul
8407 and add instructions.
8412 The '``llvm.fmuladd.*``' intrinsics each take three arguments: two
8413 multiplicands, a and b, and an addend c.
8422 %0 = call float @llvm.fmuladd.f32(%a, %b, %c)
8424 is equivalent to the expression a \* b + c, except that rounding will
8425 not be performed between the multiplication and addition steps if the
8426 code generator fuses the operations. Fusion is not guaranteed, even if
8427 the target platform supports it. If a fused multiply-add is required the
8428 corresponding llvm.fma.\* intrinsic function should be used
8429 instead. This never sets errno, just as '``llvm.fma.*``'.
8434 .. code-block:: llvm
8436 %r2 = call float @llvm.fmuladd.f32(float %a, float %b, float %c) ; yields {float}:r2 = (a * b) + c
8438 Half Precision Floating Point Intrinsics
8439 ----------------------------------------
8441 For most target platforms, half precision floating point is a
8442 storage-only format. This means that it is a dense encoding (in memory)
8443 but does not support computation in the format.
8445 This means that code must first load the half-precision floating point
8446 value as an i16, then convert it to float with
8447 :ref:`llvm.convert.from.fp16 <int_convert_from_fp16>`. Computation can
8448 then be performed on the float value (including extending to double
8449 etc). To store the value back to memory, it is first converted to float
8450 if needed, then converted to i16 with
8451 :ref:`llvm.convert.to.fp16 <int_convert_to_fp16>`, then storing as an
8454 .. _int_convert_to_fp16:
8456 '``llvm.convert.to.fp16``' Intrinsic
8457 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8464 declare i16 @llvm.convert.to.fp16(f32 %a)
8469 The '``llvm.convert.to.fp16``' intrinsic function performs a conversion
8470 from single precision floating point format to half precision floating
8476 The intrinsic function contains single argument - the value to be
8482 The '``llvm.convert.to.fp16``' intrinsic function performs a conversion
8483 from single precision floating point format to half precision floating
8484 point format. The return value is an ``i16`` which contains the
8490 .. code-block:: llvm
8492 %res = call i16 @llvm.convert.to.fp16(f32 %a)
8493 store i16 %res, i16* @x, align 2
8495 .. _int_convert_from_fp16:
8497 '``llvm.convert.from.fp16``' Intrinsic
8498 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8505 declare f32 @llvm.convert.from.fp16(i16 %a)
8510 The '``llvm.convert.from.fp16``' intrinsic function performs a
8511 conversion from half precision floating point format to single precision
8512 floating point format.
8517 The intrinsic function contains single argument - the value to be
8523 The '``llvm.convert.from.fp16``' intrinsic function performs a
8524 conversion from half single precision floating point format to single
8525 precision floating point format. The input half-float value is
8526 represented by an ``i16`` value.
8531 .. code-block:: llvm
8533 %a = load i16* @x, align 2
8534 %res = call f32 @llvm.convert.from.fp16(i16 %a)
8539 The LLVM debugger intrinsics (which all start with ``llvm.dbg.``
8540 prefix), are described in the `LLVM Source Level
8541 Debugging <SourceLevelDebugging.html#format_common_intrinsics>`_
8544 Exception Handling Intrinsics
8545 -----------------------------
8547 The LLVM exception handling intrinsics (which all start with
8548 ``llvm.eh.`` prefix), are described in the `LLVM Exception
8549 Handling <ExceptionHandling.html#format_common_intrinsics>`_ document.
8553 Trampoline Intrinsics
8554 ---------------------
8556 These intrinsics make it possible to excise one parameter, marked with
8557 the :ref:`nest <nest>` attribute, from a function. The result is a
8558 callable function pointer lacking the nest parameter - the caller does
8559 not need to provide a value for it. Instead, the value to use is stored
8560 in advance in a "trampoline", a block of memory usually allocated on the
8561 stack, which also contains code to splice the nest value into the
8562 argument list. This is used to implement the GCC nested function address
8565 For example, if the function is ``i32 f(i8* nest %c, i32 %x, i32 %y)``
8566 then the resulting function pointer has signature ``i32 (i32, i32)*``.
8567 It can be created as follows:
8569 .. code-block:: llvm
8571 %tramp = alloca [10 x i8], align 4 ; size and alignment only correct for X86
8572 %tramp1 = getelementptr [10 x i8]* %tramp, i32 0, i32 0
8573 call i8* @llvm.init.trampoline(i8* %tramp1, i8* bitcast (i32 (i8*, i32, i32)* @f to i8*), i8* %nval)
8574 %p = call i8* @llvm.adjust.trampoline(i8* %tramp1)
8575 %fp = bitcast i8* %p to i32 (i32, i32)*
8577 The call ``%val = call i32 %fp(i32 %x, i32 %y)`` is then equivalent to
8578 ``%val = call i32 %f(i8* %nval, i32 %x, i32 %y)``.
8582 '``llvm.init.trampoline``' Intrinsic
8583 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8590 declare void @llvm.init.trampoline(i8* <tramp>, i8* <func>, i8* <nval>)
8595 This fills the memory pointed to by ``tramp`` with executable code,
8596 turning it into a trampoline.
8601 The ``llvm.init.trampoline`` intrinsic takes three arguments, all
8602 pointers. The ``tramp`` argument must point to a sufficiently large and
8603 sufficiently aligned block of memory; this memory is written to by the
8604 intrinsic. Note that the size and the alignment are target-specific -
8605 LLVM currently provides no portable way of determining them, so a
8606 front-end that generates this intrinsic needs to have some
8607 target-specific knowledge. The ``func`` argument must hold a function
8608 bitcast to an ``i8*``.
8613 The block of memory pointed to by ``tramp`` is filled with target
8614 dependent code, turning it into a function. Then ``tramp`` needs to be
8615 passed to :ref:`llvm.adjust.trampoline <int_at>` to get a pointer which can
8616 be :ref:`bitcast (to a new function) and called <int_trampoline>`. The new
8617 function's signature is the same as that of ``func`` with any arguments
8618 marked with the ``nest`` attribute removed. At most one such ``nest``
8619 argument is allowed, and it must be of pointer type. Calling the new
8620 function is equivalent to calling ``func`` with the same argument list,
8621 but with ``nval`` used for the missing ``nest`` argument. If, after
8622 calling ``llvm.init.trampoline``, the memory pointed to by ``tramp`` is
8623 modified, then the effect of any later call to the returned function
8624 pointer is undefined.
8628 '``llvm.adjust.trampoline``' Intrinsic
8629 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8636 declare i8* @llvm.adjust.trampoline(i8* <tramp>)
8641 This performs any required machine-specific adjustment to the address of
8642 a trampoline (passed as ``tramp``).
8647 ``tramp`` must point to a block of memory which already has trampoline
8648 code filled in by a previous call to
8649 :ref:`llvm.init.trampoline <int_it>`.
8654 On some architectures the address of the code to be executed needs to be
8655 different to the address where the trampoline is actually stored. This
8656 intrinsic returns the executable address corresponding to ``tramp``
8657 after performing the required machine specific adjustments. The pointer
8658 returned can then be :ref:`bitcast and executed <int_trampoline>`.
8663 This class of intrinsics exists to information about the lifetime of
8664 memory objects and ranges where variables are immutable.
8668 '``llvm.lifetime.start``' Intrinsic
8669 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8676 declare void @llvm.lifetime.start(i64 <size>, i8* nocapture <ptr>)
8681 The '``llvm.lifetime.start``' intrinsic specifies the start of a memory
8687 The first argument is a constant integer representing the size of the
8688 object, or -1 if it is variable sized. The second argument is a pointer
8694 This intrinsic indicates that before this point in the code, the value
8695 of the memory pointed to by ``ptr`` is dead. This means that it is known
8696 to never be used and has an undefined value. A load from the pointer
8697 that precedes this intrinsic can be replaced with ``'undef'``.
8701 '``llvm.lifetime.end``' Intrinsic
8702 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8709 declare void @llvm.lifetime.end(i64 <size>, i8* nocapture <ptr>)
8714 The '``llvm.lifetime.end``' intrinsic specifies the end of a memory
8720 The first argument is a constant integer representing the size of the
8721 object, or -1 if it is variable sized. The second argument is a pointer
8727 This intrinsic indicates that after this point in the code, the value of
8728 the memory pointed to by ``ptr`` is dead. This means that it is known to
8729 never be used and has an undefined value. Any stores into the memory
8730 object following this intrinsic may be removed as dead.
8732 '``llvm.invariant.start``' Intrinsic
8733 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8740 declare {}* @llvm.invariant.start(i64 <size>, i8* nocapture <ptr>)
8745 The '``llvm.invariant.start``' intrinsic specifies that the contents of
8746 a memory object will not change.
8751 The first argument is a constant integer representing the size of the
8752 object, or -1 if it is variable sized. The second argument is a pointer
8758 This intrinsic indicates that until an ``llvm.invariant.end`` that uses
8759 the return value, the referenced memory location is constant and
8762 '``llvm.invariant.end``' Intrinsic
8763 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8770 declare void @llvm.invariant.end({}* <start>, i64 <size>, i8* nocapture <ptr>)
8775 The '``llvm.invariant.end``' intrinsic specifies that the contents of a
8776 memory object are mutable.
8781 The first argument is the matching ``llvm.invariant.start`` intrinsic.
8782 The second argument is a constant integer representing the size of the
8783 object, or -1 if it is variable sized and the third argument is a
8784 pointer to the object.
8789 This intrinsic indicates that the memory is mutable again.
8794 This class of intrinsics is designed to be generic and has no specific
8797 '``llvm.var.annotation``' Intrinsic
8798 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8805 declare void @llvm.var.annotation(i8* <val>, i8* <str>, i8* <str>, i32 <int>)
8810 The '``llvm.var.annotation``' intrinsic.
8815 The first argument is a pointer to a value, the second is a pointer to a
8816 global string, the third is a pointer to a global string which is the
8817 source file name, and the last argument is the line number.
8822 This intrinsic allows annotation of local variables with arbitrary
8823 strings. This can be useful for special purpose optimizations that want
8824 to look for these annotations. These have no other defined use; they are
8825 ignored by code generation and optimization.
8827 '``llvm.ptr.annotation.*``' Intrinsic
8828 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8833 This is an overloaded intrinsic. You can use '``llvm.ptr.annotation``' on a
8834 pointer to an integer of any width. *NOTE* you must specify an address space for
8835 the pointer. The identifier for the default address space is the integer
8840 declare i8* @llvm.ptr.annotation.p<address space>i8(i8* <val>, i8* <str>, i8* <str>, i32 <int>)
8841 declare i16* @llvm.ptr.annotation.p<address space>i16(i16* <val>, i8* <str>, i8* <str>, i32 <int>)
8842 declare i32* @llvm.ptr.annotation.p<address space>i32(i32* <val>, i8* <str>, i8* <str>, i32 <int>)
8843 declare i64* @llvm.ptr.annotation.p<address space>i64(i64* <val>, i8* <str>, i8* <str>, i32 <int>)
8844 declare i256* @llvm.ptr.annotation.p<address space>i256(i256* <val>, i8* <str>, i8* <str>, i32 <int>)
8849 The '``llvm.ptr.annotation``' intrinsic.
8854 The first argument is a pointer to an integer value of arbitrary bitwidth
8855 (result of some expression), the second is a pointer to a global string, the
8856 third is a pointer to a global string which is the source file name, and the
8857 last argument is the line number. It returns the value of the first argument.
8862 This intrinsic allows annotation of a pointer to an integer with arbitrary
8863 strings. This can be useful for special purpose optimizations that want to look
8864 for these annotations. These have no other defined use; they are ignored by code
8865 generation and optimization.
8867 '``llvm.annotation.*``' Intrinsic
8868 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8873 This is an overloaded intrinsic. You can use '``llvm.annotation``' on
8874 any integer bit width.
8878 declare i8 @llvm.annotation.i8(i8 <val>, i8* <str>, i8* <str>, i32 <int>)
8879 declare i16 @llvm.annotation.i16(i16 <val>, i8* <str>, i8* <str>, i32 <int>)
8880 declare i32 @llvm.annotation.i32(i32 <val>, i8* <str>, i8* <str>, i32 <int>)
8881 declare i64 @llvm.annotation.i64(i64 <val>, i8* <str>, i8* <str>, i32 <int>)
8882 declare i256 @llvm.annotation.i256(i256 <val>, i8* <str>, i8* <str>, i32 <int>)
8887 The '``llvm.annotation``' intrinsic.
8892 The first argument is an integer value (result of some expression), the
8893 second is a pointer to a global string, the third is a pointer to a
8894 global string which is the source file name, and the last argument is
8895 the line number. It returns the value of the first argument.
8900 This intrinsic allows annotations to be put on arbitrary expressions
8901 with arbitrary strings. This can be useful for special purpose
8902 optimizations that want to look for these annotations. These have no
8903 other defined use; they are ignored by code generation and optimization.
8905 '``llvm.trap``' Intrinsic
8906 ^^^^^^^^^^^^^^^^^^^^^^^^^
8913 declare void @llvm.trap() noreturn nounwind
8918 The '``llvm.trap``' intrinsic.
8928 This intrinsic is lowered to the target dependent trap instruction. If
8929 the target does not have a trap instruction, this intrinsic will be
8930 lowered to a call of the ``abort()`` function.
8932 '``llvm.debugtrap``' Intrinsic
8933 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8940 declare void @llvm.debugtrap() nounwind
8945 The '``llvm.debugtrap``' intrinsic.
8955 This intrinsic is lowered to code which is intended to cause an
8956 execution trap with the intention of requesting the attention of a
8959 '``llvm.stackprotector``' Intrinsic
8960 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8967 declare void @llvm.stackprotector(i8* <guard>, i8** <slot>)
8972 The ``llvm.stackprotector`` intrinsic takes the ``guard`` and stores it
8973 onto the stack at ``slot``. The stack slot is adjusted to ensure that it
8974 is placed on the stack before local variables.
8979 The ``llvm.stackprotector`` intrinsic requires two pointer arguments.
8980 The first argument is the value loaded from the stack guard
8981 ``@__stack_chk_guard``. The second variable is an ``alloca`` that has
8982 enough space to hold the value of the guard.
8987 This intrinsic causes the prologue/epilogue inserter to force the position of
8988 the ``AllocaInst`` stack slot to be before local variables on the stack. This is
8989 to ensure that if a local variable on the stack is overwritten, it will destroy
8990 the value of the guard. When the function exits, the guard on the stack is
8991 checked against the original guard by ``llvm.stackprotectorcheck``. If they are
8992 different, then ``llvm.stackprotectorcheck`` causes the program to abort by
8993 calling the ``__stack_chk_fail()`` function.
8995 '``llvm.stackprotectorcheck``' Intrinsic
8996 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
9003 declare void @llvm.stackprotectorcheck(i8** <guard>)
9008 The ``llvm.stackprotectorcheck`` intrinsic compares ``guard`` against an already
9009 created stack protector and if they are not equal calls the
9010 ``__stack_chk_fail()`` function.
9015 The ``llvm.stackprotectorcheck`` intrinsic requires one pointer argument, the
9016 the variable ``@__stack_chk_guard``.
9021 This intrinsic is provided to perform the stack protector check by comparing
9022 ``guard`` with the stack slot created by ``llvm.stackprotector`` and if the
9023 values do not match call the ``__stack_chk_fail()`` function.
9025 The reason to provide this as an IR level intrinsic instead of implementing it
9026 via other IR operations is that in order to perform this operation at the IR
9027 level without an intrinsic, one would need to create additional basic blocks to
9028 handle the success/failure cases. This makes it difficult to stop the stack
9029 protector check from disrupting sibling tail calls in Codegen. With this
9030 intrinsic, we are able to generate the stack protector basic blocks late in
9031 codegen after the tail call decision has occurred.
9033 '``llvm.objectsize``' Intrinsic
9034 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
9041 declare i32 @llvm.objectsize.i32(i8* <object>, i1 <min>)
9042 declare i64 @llvm.objectsize.i64(i8* <object>, i1 <min>)
9047 The ``llvm.objectsize`` intrinsic is designed to provide information to
9048 the optimizers to determine at compile time whether a) an operation
9049 (like memcpy) will overflow a buffer that corresponds to an object, or
9050 b) that a runtime check for overflow isn't necessary. An object in this
9051 context means an allocation of a specific class, structure, array, or
9057 The ``llvm.objectsize`` intrinsic takes two arguments. The first
9058 argument is a pointer to or into the ``object``. The second argument is
9059 a boolean and determines whether ``llvm.objectsize`` returns 0 (if true)
9060 or -1 (if false) when the object size is unknown. The second argument
9061 only accepts constants.
9066 The ``llvm.objectsize`` intrinsic is lowered to a constant representing
9067 the size of the object concerned. If the size cannot be determined at
9068 compile time, ``llvm.objectsize`` returns ``i32/i64 -1 or 0`` (depending
9069 on the ``min`` argument).
9071 '``llvm.expect``' Intrinsic
9072 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
9077 This is an overloaded intrinsic. You can use ``llvm.expect`` on any
9082 declare i1 @llvm.expect.i1(i1 <val>, i1 <expected_val>)
9083 declare i32 @llvm.expect.i32(i32 <val>, i32 <expected_val>)
9084 declare i64 @llvm.expect.i64(i64 <val>, i64 <expected_val>)
9089 The ``llvm.expect`` intrinsic provides information about expected (the
9090 most probable) value of ``val``, which can be used by optimizers.
9095 The ``llvm.expect`` intrinsic takes two arguments. The first argument is
9096 a value. The second argument is an expected value, this needs to be a
9097 constant value, variables are not allowed.
9102 This intrinsic is lowered to the ``val``.
9104 '``llvm.donothing``' Intrinsic
9105 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
9112 declare void @llvm.donothing() nounwind readnone
9117 The ``llvm.donothing`` intrinsic doesn't perform any operation. It's the
9118 only intrinsic that can be called with an invoke instruction.
9128 This intrinsic does nothing, and it's removed by optimizers and ignored
9131 Stack Map Intrinsics
9132 --------------------
9134 LLVM provides experimental intrinsics to support runtime patching
9135 mechanisms commonly desired in dynamic language JITs. These intrinsics
9136 are described in :doc:`StackMaps`.