'llvm' Dialect

This dialect maps LLVM IR into MLIR by defining the corresponding operations and types. LLVM IR metadata is usually represented as MLIR attributes, which offer additional structure verification.

We use “LLVM IR” to designate the intermediate representation of LLVM and “LLVM dialect” or “LLVM IR dialect” to refer to this MLIR dialect.

Unless explicitly stated otherwise, the semantics of the LLVM dialect operations must correspond to the semantics of LLVM IR instructions and any divergence is considered a bug. The dialect also contains auxiliary operations that smoothen the differences in the IR structure, e.g., MLIR does not have phi operations and LLVM IR does not have a constant operation. These auxiliary operations are systematically prefixed with mlir, e.g. llvm.mlir.constant where llvm. is the dialect namespace prefix.

Dependency on LLVM IR ¶

LLVM dialect is not expected to depend on any object that requires an LLVMContext, such as an LLVM IR instruction or type. Instead, MLIR provides thread-safe alternatives compatible with the rest of the infrastructure. The dialect is allowed to depend on the LLVM IR objects that don’t require a context, such as data layout and triple description.

Module Structure ¶

IR modules use the built-in MLIR ModuleOp and support all its features. In particular, modules can be named, nested and are subject to symbol visibility. Modules can contain any operations, including LLVM functions and globals.

Data Layout and Triple ¶

An IR module may have an optional data layout and triple information attached using MLIR attributes llvm.data_layout and llvm.triple, respectively. Both are string attributes with the same syntax as in LLVM IR and are verified to be correct. They can be defined as follows.

module attributes {llvm.data_layout = "e",
                   llvm.target_triple = "aarch64-linux-android"} {
  // module contents
}

Functions ¶

LLVM functions are represented by a special operation, llvm.func, that has syntax similar to that of the built-in function operation but supports LLVM-related features such as linkage and variadic argument lists. See detailed description in the operation list below .

PHI Nodes and Block Arguments ¶

MLIR uses block arguments instead of PHI nodes to communicate values between blocks. Therefore, the LLVM dialect has no operation directly equivalent to phi in LLVM IR. Instead, all terminators can pass values as successor operands as these values will be forwarded as block arguments when the control flow is transferred.

For example:

^bb1:
  %0 = llvm.addi %arg0, %cst : i32
  llvm.br ^bb2[%0: i32]

// If the control flow comes from ^bb1, %arg1 == %0.
^bb2(%arg1: i32)
  // ...

is equivalent to LLVM IR

%0:
  %1 = add i32 %arg0, %cst
  br %3

%3:
  %arg1 = phi [%1, %0], //...

Since there is no need to use the block identifier to differentiate the source of different values, the LLVM dialect supports terminators that transfer the control flow to the same block with different arguments. For example:

^bb1:
  llvm.cond_br %cond, ^bb2[%0: i32], ^bb2[%1: i32]

^bb2(%arg0: i32):
  // ...

Context-Level Values ¶

Some value kinds in LLVM IR, such as constants and undefs, are uniqued in context and used directly in relevant operations. MLIR does not support such values for thread-safety and concept parsimony reasons. Instead, regular values are produced by dedicated operations that have the corresponding semantics: llvm.mlir.constant , llvm.mlir.undef , llvm.mlir.null . Note how these operations are prefixed with mlir. to indicate that they don’t belong to LLVM IR but are only necessary to model it in MLIR. The values produced by these operations are usable just like any other value.

Examples:

// Create an undefined value of structure type with a 32-bit integer followed
// by a float.
%0 = llvm.mlir.undef : !llvm.struct<(i32, f32)>

// Null pointer to i8.
%1 = llvm.mlir.null : !llvm.ptr<i8>

// Null pointer to a function with signature void().
%2 = llvm.mlir.null : !llvm.ptr<func<void ()>>

// Constant 42 as i32.
%3 = llvm.mlir.constant(42 : i32) : i32

// Splat dense vector constant.
%3 = llvm.mlir.constant(dense<1.0> : vector<4xf32>) : vector<4xf32>

Note that constants list the type twice. This is an artifact of the LLVM dialect not using built-in types, which are used for typed MLIR attributes. The syntax will be reevaluated after considering composite constants.

Globals ¶

Global variables are also defined using a special operation, llvm.mlir.global , located at the module level. Globals are MLIR symbols and are identified by their name.

Since functions need to be isolated-from-above, i.e. values defined outside the function cannot be directly used inside the function, an additional operation, llvm.mlir.addressof , is provided to locally define a value containing the address of a global. The actual value can then be loaded from that pointer, or a new value can be stored into it if the global is not declared constant. This is similar to LLVM IR where globals are accessed through name and have a pointer type.

Linkage ¶

Module-level named objects in the LLVM dialect, namely functions and globals, have an optional linkage attribute derived from LLVM IR linkage types . Linkage is specified by the same keyword as in LLVM IR and is located between the operation name (llvm.func or llvm.global) and the symbol name. If no linkage keyword is present, external linkage is assumed by default. Linakge is distinct from MLIR symbol visibility.

Attribute Pass-Through ¶

The LLVM dialect provides a mechanism to forward function-level attributes to LLVM IR using the passthrough attribute. This is an array attribute containing either string attributes or array attributes. In the former case, the value of the string is interpreted as the name of LLVM IR function attribute. In the latter case, the array is expected to contain exactly two string attributes, the first corresponding to the name of LLVM IR function attribute, and the second corresponding to its value. Note that even integer LLVM IR function attributes have their value represented in the string form.

Example:

llvm.func @func() attributes {
  passthrough = ["noinline",           // value-less attribute
                 ["alignstack", "4"],  // integer attribute with value
                 ["other", "attr"]]    // attribute unknown to LLVM
} {
  llvm.return
}

If the attribute is not known to LLVM IR, it will be attached as a string attribute.

Types ¶

LLVM dialect uses built-in types whenever possible and defines a set of complementary types, which correspond to the LLVM IR types that cannot be directly represented with built-in types. Similarly to other MLIR context-owned objects, the creation and manipulation of LLVM dialect types is thread-safe.

MLIR does not support module-scoped named type declarations, e.g. %s = type {i32, i32} in LLVM IR. Instead, types must be fully specified at each use, except for recursive types where only the first reference to a named type needs to be fully specified. MLIR type aliases can be used to achieve more compact syntax.

The general syntax of LLVM dialect types is !llvm., followed by a type kind identifier (e.g., ptr for pointer or struct for structure) and by an optional list of type parameters in angle brackets. The dialect follows MLIR style for types with nested angle brackets and keyword specifiers rather than using different bracket styles to differentiate types. Types inside the angle brackets may omit the !llvm. prefix for brevity: the parser first attempts to find a type (starting with ! or a built-in type) and falls back to accepting a keyword. For example, !llvm.ptr<!llvm.ptr<i32>> and !llvm.ptr<ptr<i32>> are equivalent, with the latter being the canonical form, and denote a pointer to a pointer to a 32-bit integer.

Built-in Type Compatibility ¶

LLVM dialect accepts a subset of built-in types that are referred to as LLVM dialect-compatible types. The following types are compatible:

• Signless integers - iN (IntegerType).
• Floating point types - bfloat, half, float, double , f80, f128 (FloatType).
• 1D vectors of signless integers or floating point types - vector<NxT> (VectorType).

Note that only a subset of types that can be represented by a given class is compatible. For example, signed and unsigned integers are not compatible. LLVM provides a function, bool LLVM::isCompatibleType(Type), that can be used as a compatibility check.

Each LLVM IR type corresponds to exactly one MLIR type, either built-in or LLVM dialect type. For example, because i32 is LLVM-compatible, there is no !llvm.i32 type. However, !llvm.ptr<T> is defined in the LLVM dialect as there is no corresponding built-in type.

Additional Simple Types ¶

The following non-parametric types derived from the LLVM IR are available in the LLVM dialect:

• !llvm.x86_mmx (LLVMX86MMXType) - value held in an MMX register on x86 machine.
• !llvm.ppc_fp128 (LLVMPPCFP128Type) - 128-bit floating-point value (two 64 bits).
• !llvm.token (LLVMTokenType) - a non-inspectable value associated with an operation.
• !llvm.metadata (LLVMMetadataType) - LLVM IR metadata, to be used only if the metadata cannot be represented as structured MLIR attributes.
• !llvm.void (LLVMVoidType) - does not represent any value; can only appear in function results.

These types represent a single value (or an absence thereof in case of void) and correspond to their LLVM IR counterparts.

Additional Parametric Types ¶

These types are parameterized by the types they contain, e.g., the pointee or the element type, which can be either compatible built-in or LLVM dialect types.

Pointer Types ¶

Pointer types specify an address in memory.

Pointer types are parametric types parameterized by the element type and the address space. The address space is an integer, but this choice may be reconsidered if MLIR implements named address spaces. Their syntax is as follows:

  llvm-ptr-type ::= `!llvm.ptr<` type (`,` integer-literal)? `>`

where the optional integer literal corresponds to the memory space. Both cases are represented by LLVMPointerType internally.

Array Types ¶

Array types represent sequences of elements in memory. Array elements can be addressed with a value unknown at compile time, and can be nested. Only 1D arrays are allowed though.

Array types are parameterized by the fixed size and the element type. Syntactically, their representation is the following:

  llvm-array-type ::= `!llvm.array<` integer-literal `x` type `>`

and they are internally represented as LLVMArrayType.

Function Types ¶

Function types represent the type of a function, i.e. its signature.

Function types are parameterized by the result type, the list of argument types and by an optional “variadic” flag. Unlike built-in FunctionType, LLVM dialect functions (LLVMFunctionType) always have single result, which may be !llvm.void if the function does not return anything. The syntax is as follows:

  llvm-func-type ::= `!llvm.func<` type `(` type-list (`,` `...`)? `)` `>`

For example,

!llvm.func<void ()>           // a function with no arguments;
!llvm.func<i32 (f32, i32)>    // a function with two arguments and a result;
!llvm.func<void (i32, ...)>   // a variadic function with at least one argument.

In the LLVM dialect, functions are not first-class objects and one cannot have a value of function type. Instead, one can take the address of a function and operate on pointers to functions.

Vector Types ¶

Vector types represent sequences of elements, typically when multiple data elements are processed by a single instruction (SIMD). Vectors are thought of as stored in registers and therefore vector elements can only be addressed through constant indices.

Vector types are parameterized by the size, which may be either fixed or a multiple of some fixed size in case of scalable vectors, and the element type. Vectors cannot be nested and only 1D vectors are supported. Scalable vectors are still considered 1D.

LLVM dialect uses built-in vector types for fixed-size vectors of built-in types, and provides additional types for fixed-sized vectors of LLVM dialect types (LLVMFixedVectorType) and scalable vectors of any types (LLVMScalableVectorType). These two additional types share the following syntax:

  llvm-vec-type ::= `!llvm.vec<` (`?` `x`)? integer-literal `x` type `>`

Note that the sets of element types supported by built-in and LLVM dialect vector types are mutually exclusive, e.g., the built-in vector type does not accept !llvm.ptr<i32> and the LLVM dialect fixed-width vector type does not accept i32.

The following functions are provided to operate on any kind of the vector types compatible with the LLVM dialect:

• bool LLVM::isCompatibleVectorType(Type) - checks whether a type is a vector type compatible with the LLVM dialect;
• Type LLVM::getVectorElementType(Type) - returns the element type of any vector type compatible with the LLVM dialect;
• llvm::ElementCount LLVM::getVectorNumElements(Type) - returns the number of elements in any vector type compatible with the LLVM dialect;
• Type LLVM::getFixedVectorType(Type, unsigned) - gets a fixed vector type with the given element type and size; the resulting type is either a built-in or an LLVM dialect vector type depending on which one supports the given element type.

Examples of Compatible Vector Types ¶

vector<42 x i32>                   // Vector of 42 32-bit integers.
!llvm.vec<42 x ptr<i32>>           // Vector of 42 pointers to 32-bit integers.
!llvm.vec<? x 4 x i32>             // Scalable vector of 32-bit integers with
                                   // size divisible by 4.
!llvm.array<2 x vector<2 x i32>>   // Array of 2 vectors of 2 32-bit integers.
!llvm.array<2 x vec<2 x ptr<i32>>> // Array of 2 vectors of 2 pointers to 32-bit
                                   // integers.

Structure Types ¶

The structure type is used to represent a collection of data members together in memory. The elements of a structure may be any type that has a size.

Structure types are represented in a single dedicated class mlir::LLVM::LLVMStructType. Internally, the struct type stores a (potentially empty) name, a (potentially empty) list of contained types and a bitmask indicating whether the struct is named, opaque, packed or uninitialized. Structure types that don’t have a name are referred to as literal structs. Such structures are uniquely identified by their contents. Identified structs on the other hand are uniquely identified by the name.

Identified Structure Types ¶

Identified structure types are uniqued using their name in a given context. Attempting to construct an identified structure with the same name a structure that already exists in the context will result in the existing structure being returned. MLIR does not auto-rename identified structs in case of name conflicts because there is no naming scope equivalent to a module in LLVM IR since MLIR modules can be arbitrarily nested.

Programmatically, identified structures can be constructed in an uninitialized state. In this case, they are given a name but the body must be set up by a later call, using MLIR’s type mutation mechanism. Such uninitialized types can be used in type construction, but must be eventually initialized for IR to be valid. This mechanism allows for constructing recursive or mutually referring structure types: an uninitialized type can be used in its own initialization.

Once the type is initialized, its body cannot be changed anymore. Any further attempts to modify the body will fail and return failure to the caller unless the type is initialized with the exact same body. Type initialization is thread-safe; however, if a concurrent thread initializes the type before the current thread, the initialization may return failure.

The syntax for identified structure types is as follows.

llvm-ident-struct-type ::= `!llvm.struct<` string-literal, `opaque` `>`
                         | `!llvm.struct<` string-literal, `packed`?
                           `(` type-or-ref-list  `)` `>`
type-or-ref-list ::= <maybe empty comma-separated list of type-or-ref>
type-or-ref ::= <any compatible type with optional !llvm.>
              | `!llvm.`? `struct<` string-literal `>`

The body of the identified struct is printed in full unless the it is transitively contained in the same struct. In the latter case, only the identifier is printed. For example, the structure containing the pointer to itself is represented as !llvm.struct<"A", (ptr<"A">)>, and the structure A containing two pointers to the structure B containing a pointer to the structure A is represented as !llvm.struct<"A", (ptr<"B", (ptr<"A">)>, ptr<"B", (ptr<"A">))>. Note that the structure B is “unrolled” for both elements. A structure with the same name but different body is a syntax error. The user must ensure structure name uniqueness across all modules processed in a given MLIR context. Structure names are arbitrary string literals and may include, e.g., spaces and keywords.

Identified structs may be opaque. In this case, the body is unknown but the structure type is considered initialized and is valid in the IR.

Literal Structure Types ¶

Literal structures are uniqued according to the list of elements they contain, and can optionally be packed. The syntax for such structs is as follows.

llvm-literal-struct-type ::= `!llvm.struct<` `packed`? `(` type-list `)` `>`
type-list ::= <maybe empty comma-separated list of types with optional !llvm.>

Literal structs cannot be recursive, but can contain other structs. Therefore, they must be constructed in a single step with the entire list of contained elements provided.

Examples of Structure Types ¶

!llvm.struct<>                  // NOT allowed
!llvm.struct<()>                // empty, literal
!llvm.struct<(i32)>             // literal
!llvm.struct<(struct<(i32)>)>   // struct containing a struct
!llvm.struct<packed (i8, i32)>  // packed struct
!llvm.struct<"a">               // recursive reference, only allowed within
                                // another struct, NOT allowed at top level
!llvm.struct<"a", ptr<struct<"a">>>  // supported example of recursive reference
!llvm.struct<"a", ()>           // empty, named (necessary to differentiate from
                                // recursive reference)
!llvm.struct<"a", opaque>       // opaque, named
!llvm.struct<"a", (i32)>        // named
!llvm.struct<"a", packed (i8, i32)>  // named, packed

Unsupported Types ¶

LLVM IR label type does not have a counterpart in the LLVM dialect since, in MLIR, blocks are not values and don’t need a type.

Operations ¶

All operations in the LLVM IR dialect have a custom form in MLIR. The mnemonic of an operation is that used in LLVM IR prefixed with “llvm.”.

llvm.ashr (::mlir::LLVM::AShrOp) ¶

Syntax:

operation ::= `llvm.ashr` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.access_group (::mlir::LLVM::AccessGroupMetadataOp) ¶

LLVM dialect access group metadata.

Syntax:

operation ::= `llvm.access_group` $sym_name attr-dict

Defines an access group metadata that can be attached to any instruction that potentially accesses memory. The access group may be attached to a memory accessing instruction via the llvm.access.group metadata and a branch instruction in the loop latch block via the llvm.loop.parallel_accesses metadata.

See the following link for more details: https://llvm.org/docs/LangRef.html#llvm-access-group-metadata

Attributes: ¶

llvm.add (::mlir::LLVM::AddOp) ¶

Syntax:

operation ::= `llvm.add` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.addrspacecast (::mlir::LLVM::AddrSpaceCastOp) ¶

Operands: ¶

Results: ¶

llvm.mlir.addressof (::mlir::LLVM::AddressOfOp) ¶

Creates a pointer pointing to a global or a function

Syntax:

operation ::= `llvm.mlir.addressof` $global_name attr-dict `:` type($res)

Creates an SSA value containing a pointer to a global variable or constant defined by llvm.mlir.global. The global value can be defined after its first referenced. If the global value is a constant, storing into it is not allowed.

Examples:

func @foo() {
  // Get the address of a global variable.
  %0 = llvm.mlir.addressof @const : !llvm.ptr<i32>

  // Use it as a regular pointer.
  %1 = llvm.load %0 : !llvm.ptr<i32>

  // Get the address of a function.
  %2 = llvm.mlir.addressof @foo : !llvm.ptr<func<void ()>>

  // The function address can be used for indirect calls.
  llvm.call %2() : () -> ()
}

// Define the global.
llvm.mlir.global @const(42 : i32) : i32

Attributes: ¶

Results: ¶

llvm.alloca (::mlir::LLVM::AllocaOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.and (::mlir::LLVM::AndOp) ¶

Syntax:

operation ::= `llvm.and` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.assume (::mlir::LLVM::AssumeOp) ¶

Operands: ¶

llvm.cmpxchg (::mlir::LLVM::AtomicCmpXchgOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.atomicrmw (::mlir::LLVM::AtomicRMWOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.bitreverse (::mlir::LLVM::BitReverseOp) ¶

Operands: ¶

Results: ¶

llvm.bitcast (::mlir::LLVM::BitcastOp) ¶

Operands: ¶

Results: ¶

llvm.br (::mlir::LLVM::BrOp) ¶

Syntax:

operation ::= `llvm.br` $dest (`(` $destOperands^ `:` type($destOperands) `)`)? attr-dict

Operands: ¶

Successors: ¶

llvm.call (::mlir::LLVM::CallOp) ¶

Call to an LLVM function.

In LLVM IR, functions may return either 0 or 1 value. LLVM IR dialect implements this behavior by providing a variadic call operation for 0- and 1-result functions. Even though MLIR supports multi-result functions, LLVM IR dialect disallows them.

The call instruction supports both direct and indirect calls. Direct calls start with a function name (@-prefixed) and indirect calls start with an SSA value (%-prefixed). The direct callee, if present, is stored as a function attribute callee. The trailing type of the instruction is always the MLIR function type, which may be different from the indirect callee that has the wrapped LLVM IR function type.

Examples:

// Direct call without arguments and with one result.
%0 = llvm.call @foo() : () -> (f32)

// Direct call with arguments and without a result.
llvm.call @bar(%0) : (f32) -> ()

// Indirect call with an argument and without a result.
llvm.call %1(%0) : (f32) -> ()

Attributes: ¶

Operands: ¶

Results: ¶

llvm.cond_br (::mlir::LLVM::CondBrOp) ¶

Syntax:

operation ::= `llvm.cond_br` $condition ( `weights` `(` $branch_weights^ `)` )? `,`
              $trueDest (`(` $trueDestOperands^ `:` type($trueDestOperands) `)`)? `,`
              $falseDest (`(` $falseDestOperands^ `:` type($falseDestOperands) `)`)?
              attr-dict

Attributes: ¶

Operands: ¶

Successors: ¶

llvm.mlir.constant (::mlir::LLVM::ConstantOp) ¶

Defines a constant of LLVM type.

Syntax:

operation ::= `llvm.mlir.constant` `(` $value `)` attr-dict `:` type($res)

Unlike LLVM IR, MLIR does not have first-class constant values. Therefore, all constants must be created as SSA values before being used in other operations. llvm.mlir.constant creates such values for scalars and vectors. It has a mandatory value attribute, which may be an integer, floating point attribute; dense or sparse attribute containing integers or floats. The type of the attribute is one of the corresponding MLIR builtin types. It may be omitted for i64 and f64 types that are implied. The operation produces a new SSA value of the specified LLVM IR dialect type. The type of that value must correspond to the attribute type converted to LLVM IR.

Examples:

// Integer constant, internal i32 is mandatory
%0 = llvm.mlir.constant(42 : i32) : i32

// It's okay to omit i64.
%1 = llvm.mlir.constant(42) : i64

// Floating point constant.
%2 = llvm.mlir.constant(42.0 : f32) : f32

// Splat dense vector constant.
%3 = llvm.mlir.constant(dense<1.0> : vector<4xf32>) : vector<4xf32>

Attributes: ¶

Results: ¶

llvm.intr.copysign (::mlir::LLVM::CopySignOp) ¶

Operands: ¶

Results: ¶

llvm.intr.coro.begin (::mlir::LLVM::CoroBeginOp) ¶

Syntax:

operation ::= `llvm.intr.coro.begin` $token `,` $mem attr-dict `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.coro.end (::mlir::LLVM::CoroEndOp) ¶

Syntax:

operation ::= `llvm.intr.coro.end` $handle `,` $unwind attr-dict `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.coro.free (::mlir::LLVM::CoroFreeOp) ¶

Syntax:

operation ::= `llvm.intr.coro.free` $id `,` $handle attr-dict `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.coro.id (::mlir::LLVM::CoroIdOp) ¶

Syntax:

operation ::= `llvm.intr.coro.id` $align `,` $promise `,` $coroaddr `,` $fnaddrs attr-dict `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.coro.resume (::mlir::LLVM::CoroResumeOp) ¶

Syntax:

operation ::= `llvm.intr.coro.resume` $handle attr-dict

Operands: ¶

llvm.intr.coro.save (::mlir::LLVM::CoroSaveOp) ¶

Syntax:

operation ::= `llvm.intr.coro.save` $handle attr-dict `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.coro.size (::mlir::LLVM::CoroSizeOp) ¶

Syntax:

operation ::= `llvm.intr.coro.size` attr-dict `:` type($res)

Results: ¶

llvm.intr.coro.suspend (::mlir::LLVM::CoroSuspendOp) ¶

Syntax:

operation ::= `llvm.intr.coro.suspend` $save `,` $final attr-dict `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.cos (::mlir::LLVM::CosOp) ¶

Operands: ¶

Results: ¶

llvm.intr.ctpop (::mlir::LLVM::CtPopOp) ¶

Operands: ¶

Results: ¶

llvm.intr.exp2 (::mlir::LLVM::Exp2Op) ¶

Operands: ¶

Results: ¶

llvm.intr.exp (::mlir::LLVM::ExpOp) ¶

Operands: ¶

Results: ¶

llvm.extractelement (::mlir::LLVM::ExtractElementOp) ¶

Operands: ¶

Results: ¶

llvm.extractvalue (::mlir::LLVM::ExtractValueOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.fabs (::mlir::LLVM::FAbsOp) ¶

Operands: ¶

Results: ¶

llvm.fadd (::mlir::LLVM::FAddOp) ¶

Syntax:

operation ::= `llvm.fadd` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.ceil (::mlir::LLVM::FCeilOp) ¶

Operands: ¶

Results: ¶

llvm.fcmp (::mlir::LLVM::FCmpOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.fdiv (::mlir::LLVM::FDivOp) ¶

Syntax:

operation ::= `llvm.fdiv` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.floor (::mlir::LLVM::FFloorOp) ¶

Operands: ¶

Results: ¶

llvm.intr.fma (::mlir::LLVM::FMAOp) ¶

Operands: ¶

Results: ¶

llvm.intr.fmuladd (::mlir::LLVM::FMulAddOp) ¶

Operands: ¶

Results: ¶

llvm.fmul (::mlir::LLVM::FMulOp) ¶

Syntax:

operation ::= `llvm.fmul` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.fneg (::mlir::LLVM::FNegOp) ¶

Syntax:

operation ::= `llvm.fneg` $operand custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.fpext (::mlir::LLVM::FPExtOp) ¶

Operands: ¶

Results: ¶

llvm.fptosi (::mlir::LLVM::FPToSIOp) ¶

Operands: ¶

Results: ¶

llvm.fptoui (::mlir::LLVM::FPToUIOp) ¶

Operands: ¶

Results: ¶

llvm.fptrunc (::mlir::LLVM::FPTruncOp) ¶

Operands: ¶

Results: ¶

llvm.frem (::mlir::LLVM::FRemOp) ¶

Syntax:

operation ::= `llvm.frem` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.fsub (::mlir::LLVM::FSubOp) ¶

Syntax:

operation ::= `llvm.fsub` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.fence (::mlir::LLVM::FenceOp) ¶

Attributes: ¶

llvm.freeze (::mlir::LLVM::FreezeOp) ¶

Syntax:

operation ::= `llvm.freeze` $val attr-dict `:` type($val)

Operands: ¶

Results: ¶

llvm.getelementptr (::mlir::LLVM::GEPOp) ¶

Syntax:

operation ::= `llvm.getelementptr` $base `[` $indices `]` attr-dict `:` functional-type(operands, results)

Operands: ¶

Results: ¶

llvm.intr.get.active.lane.mask (::mlir::LLVM::GetActiveLaneMaskOp) ¶

Syntax:

operation ::= `llvm.intr.get.active.lane.mask` $base `,` $n attr-dict `:` type($base) `,` type($n) `to` type($res)

Operands: ¶

Results: ¶

llvm.mlir.global (::mlir::LLVM::GlobalOp) ¶

LLVM dialect global.

Since MLIR allows for arbitrary operations to be present at the top level, global variables are defined using the llvm.mlir.global operation. Both global constants and variables can be defined, and the value may also be initialized in both cases.

There are two forms of initialization syntax. Simple constants that can be represented as MLIR attributes can be given in-line:

llvm.mlir.global @variable(32.0 : f32) : f32

This initialization and type syntax is similar to llvm.mlir.constant and may use two types: one for MLIR attribute and another for the LLVM value. These types must be compatible.

More complex constants that cannot be represented as MLIR attributes can be given in an initializer region:

// This global is initialized with the equivalent of:
//   i32* getelementptr (i32* @g2, i32 2)
llvm.mlir.global constant @int_gep() : !llvm.ptr<i32> {
  %0 = llvm.mlir.addressof @g2 : !llvm.ptr<i32>
  %1 = llvm.mlir.constant(2 : i32) : i32
  %2 = llvm.getelementptr %0[%1]
     : (!llvm.ptr<i32>, i32) -> !llvm.ptr<i32>
  // The initializer region must end with `llvm.return`.
  llvm.return %2 : !llvm.ptr<i32>
}

Only one of the initializer attribute or initializer region may be provided.

llvm.mlir.global must appear at top-level of the enclosing module. It uses an @-identifier for its value, which will be uniqued by the module with respect to other @-identifiers in it.

Examples:

// Global values use @-identifiers.
llvm.mlir.global constant @cst(42 : i32) : i32

// Non-constant values must also be initialized.
llvm.mlir.global @variable(32.0 : f32) : f32

// Strings are expected to be of wrapped LLVM i8 array type and do not
// automatically include the trailing zero.
llvm.mlir.global @string("abc") : !llvm.array<3 x i8>

// For strings globals, the trailing type may be omitted.
llvm.mlir.global constant @no_trailing_type("foo bar")

// A complex initializer is constructed with an initializer region.
llvm.mlir.global constant @int_gep() : !llvm.ptr<i32> {
  %0 = llvm.mlir.addressof @g2 : !llvm.ptr<i32>
  %1 = llvm.mlir.constant(2 : i32) : i32
  %2 = llvm.getelementptr %0[%1]
     : (!llvm.ptr<i32>, i32) -> !llvm.ptr<i32>
  llvm.return %2 : !llvm.ptr<i32>
}

Similarly to functions, globals have a linkage attribute. In the custom syntax, this attribute is placed between llvm.mlir.global and the optional constant keyword. If the attribute is omitted, external linkage is assumed by default.

Examples:

// A constant with internal linkage will not participate in linking.
llvm.mlir.global internal constant @cst(42 : i32) : i32

// By default, "external" linkage is assumed and the global participates in
// symbol resolution at link-time.
llvm.mlir.global @glob(0 : f32) : f32

// Alignment is optional
llvm.mlir.global private constant @y(dense<1.0> : tensor<8xf32>) : !llvm.array<8 x f32>

Like global variables in LLVM IR, globals can have an (optional) alignment attribute using keyword alignment. The integer value of the alignment must be a positive integer that is a power of 2.

Examples:

// Alignment is optional
llvm.mlir.global private constant @y(dense<1.0> : tensor<8xf32>) { alignment = 32 : i64 } : !llvm.array<8 x f32>

Attributes: ¶

llvm.icmp (::mlir::LLVM::ICmpOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.inline_asm (::mlir::LLVM::InlineAsmOp) ¶

Syntax:

operation ::= `llvm.inline_asm` (`has_side_effects` $has_side_effects^)?
              (`is_align_stack` $is_align_stack^)?
              (`asm_dialect` `=` $asm_dialect^)?
              attr-dict
              $asm_string `,` $constraints
              operands `:` functional-type(operands, results)

The InlineAsmOp mirrors the underlying LLVM semantics with a notable exception: the embedded asm_string is not allowed to define or reference any symbol or any global variable: only the operands of the op may be read, written, or referenced. Attempting to define or reference any symbol or any global behavior is considered undefined behavior at this time.

Attributes: ¶

Operands: ¶

Results: ¶

llvm.insertelement (::mlir::LLVM::InsertElementOp) ¶

Operands: ¶

Results: ¶

llvm.insertvalue (::mlir::LLVM::InsertValueOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.inttoptr (::mlir::LLVM::IntToPtrOp) ¶

Operands: ¶

Results: ¶

llvm.invoke (::mlir::LLVM::InvokeOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

Successors: ¶

llvm.func (::mlir::LLVM::LLVMFuncOp) ¶

LLVM dialect function.

MLIR functions are defined by an operation that is not built into the IR itself. The LLVM dialect provides an llvm.func operation to define functions compatible with LLVM IR. These functions have LLVM dialect function type but use MLIR syntax to express it. They are required to have exactly one result type. LLVM function operation is intended to capture additional properties of LLVM functions, such as linkage and calling convention, that may be modeled differently by the built-in MLIR function.

// The type of @bar is !llvm<"i64 (i64)">
llvm.func @bar(%arg0: i64) -> i64 {
  llvm.return %arg0 : i64
}

// Type type of @foo is !llvm<"void (i64)">
// !llvm.void type is omitted
llvm.func @foo(%arg0: i64) {
  llvm.return
}

// A function with `internal` linkage.
llvm.func internal @internal_func() {
  llvm.return
}

Attributes: ¶

llvm.lshr (::mlir::LLVM::LShrOp) ¶

Syntax:

operation ::= `llvm.lshr` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.landingpad (::mlir::LLVM::LandingpadOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.load (::mlir::LLVM::LoadOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.log10 (::mlir::LLVM::Log10Op) ¶

Operands: ¶

Results: ¶

llvm.intr.log2 (::mlir::LLVM::Log2Op) ¶

Operands: ¶

Results: ¶

llvm.intr.log (::mlir::LLVM::LogOp) ¶

Operands: ¶

Results: ¶

llvm.intr.masked.load (::mlir::LLVM::MaskedLoadOp) ¶

Syntax:

operation ::= `llvm.intr.masked.load` operands attr-dict `:` functional-type(operands, results)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.masked.store (::mlir::LLVM::MaskedStoreOp) ¶

Syntax:

operation ::= `llvm.intr.masked.store` $value `,` $data `,` $mask attr-dict `:` type($value) `,` type($mask) `into` type($data)

Attributes: ¶

Operands: ¶

llvm.intr.matrix.column.major.load (::mlir::LLVM::MatrixColumnMajorLoadOp) ¶

Syntax:

operation ::= `llvm.intr.matrix.column.major.load` $data `,` `<` `stride` `=` $stride `>` attr-dict`:` type($res) `from` type($data) `stride` type($stride)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.matrix.column.major.store (::mlir::LLVM::MatrixColumnMajorStoreOp) ¶

Syntax:

operation ::= `llvm.intr.matrix.column.major.store` $matrix `,` $data `,` `<` `stride` `=` $stride `>` attr-dict`:` type($matrix) `to` type($data) `stride` type($stride)

Attributes: ¶

Operands: ¶

llvm.intr.matrix.multiply (::mlir::LLVM::MatrixMultiplyOp) ¶

Syntax:

operation ::= `llvm.intr.matrix.multiply` $lhs `,` $rhs attr-dict `:` `(` type($lhs) `,` type($rhs) `)` `->` type($res)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.matrix.transpose (::mlir::LLVM::MatrixTransposeOp) ¶

Syntax:

operation ::= `llvm.intr.matrix.transpose` $matrix attr-dict `:` type($matrix) `into` type($res)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.maxnum (::mlir::LLVM::MaxNumOp) ¶

Operands: ¶

Results: ¶

llvm.intr.maximum (::mlir::LLVM::MaximumOp) ¶

Operands: ¶

Results: ¶

llvm.intr.memcpy.inline (::mlir::LLVM::MemcpyInlineOp) ¶

Operands: ¶

llvm.intr.memcpy (::mlir::LLVM::MemcpyOp) ¶

Operands: ¶

llvm.metadata (::mlir::LLVM::MetadataOp) ¶

LLVM dialect metadata.

Syntax:

operation ::= `llvm.metadata` $sym_name attr-dict-with-keyword $body

llvm.metadata op defines one or more metadata nodes. Currently the llvm.access_group metadata op is supported.

Example: llvm.metadata @metadata { llvm.access_group @group1 llvm.access_group @group2 llvm.return }

Attributes: ¶

llvm.intr.minnum (::mlir::LLVM::MinNumOp) ¶

Operands: ¶

Results: ¶

llvm.intr.minimum (::mlir::LLVM::MinimumOp) ¶

Operands: ¶

Results: ¶

llvm.mul (::mlir::LLVM::MulOp) ¶

Syntax:

operation ::= `llvm.mul` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.mlir.null (::mlir::LLVM::NullOp) ¶

Defines a value containing a null pointer to LLVM type.

Syntax:

operation ::= `llvm.mlir.null` attr-dict `:` type($res)

Unlike LLVM IR, MLIR does not have first-class null pointers. They must be explicitly created as SSA values using llvm.mlir.null. This operation has no operands or attributes, and returns a null value of a wrapped LLVM IR pointer type.

Examples:

// Null pointer to i8.
%0 = llvm.mlir.null : !llvm.ptr<i8>

// Null pointer to a function with signature void().
%1 = llvm.mlir.null : !llvm.ptr<func<void ()>>

Results: ¶

llvm.or (::mlir::LLVM::OrOp) ¶

Syntax:

operation ::= `llvm.or` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.pow (::mlir::LLVM::PowOp) ¶

Operands: ¶

Results: ¶

llvm.intr.prefetch (::mlir::LLVM::Prefetch) ¶

Operands: ¶

llvm.ptrtoint (::mlir::LLVM::PtrToIntOp) ¶

Operands: ¶

Results: ¶

llvm.resume (::mlir::LLVM::ResumeOp) ¶

Syntax:

operation ::= `llvm.resume` $value attr-dict `:` type($value)

Operands: ¶

llvm.return (::mlir::LLVM::ReturnOp) ¶

Operands: ¶

llvm.intr.sadd.with.overflow (::mlir::LLVM::SAddWithOverflowOp) ¶

Operands: ¶

Results: ¶

llvm.sdiv (::mlir::LLVM::SDivOp) ¶

Syntax:

operation ::= `llvm.sdiv` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.sext (::mlir::LLVM::SExtOp) ¶

Operands: ¶

Results: ¶

llvm.sitofp (::mlir::LLVM::SIToFPOp) ¶

Operands: ¶

Results: ¶

llvm.intr.smax (::mlir::LLVM::SMaxOp) ¶

Operands: ¶

Results: ¶

llvm.intr.smin (::mlir::LLVM::SMinOp) ¶

Operands: ¶

Results: ¶

llvm.intr.smul.with.overflow (::mlir::LLVM::SMulWithOverflowOp) ¶

Operands: ¶

Results: ¶

llvm.srem (::mlir::LLVM::SRemOp) ¶

Syntax:

operation ::= `llvm.srem` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.ssub.with.overflow (::mlir::LLVM::SSubWithOverflowOp) ¶

Operands: ¶

Results: ¶

llvm.select (::mlir::LLVM::SelectOp) ¶

Syntax:

operation ::= `llvm.select` operands attr-dict `:` type($condition) `,` type($res)

Operands: ¶

Results: ¶

llvm.shl (::mlir::LLVM::ShlOp) ¶

Syntax:

operation ::= `llvm.shl` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.shufflevector (::mlir::LLVM::ShuffleVectorOp) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.sin (::mlir::LLVM::SinOp) ¶

Operands: ¶

Results: ¶

llvm.intr.sqrt (::mlir::LLVM::SqrtOp) ¶

Operands: ¶

Results: ¶

llvm.intr.stackrestore (::mlir::LLVM::StackRestoreOp) ¶

Syntax:

operation ::= `llvm.intr.stackrestore` $ptr attr-dict

Operands: ¶

llvm.intr.stacksave (::mlir::LLVM::StackSaveOp) ¶

Syntax:

operation ::= `llvm.intr.stacksave` attr-dict `:` type($res)

Results: ¶

llvm.store (::mlir::LLVM::StoreOp) ¶

Attributes: ¶

Operands: ¶

llvm.sub (::mlir::LLVM::SubOp) ¶

Syntax:

operation ::= `llvm.sub` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.switch (::mlir::LLVM::SwitchOp) ¶

Syntax:

operation ::= `llvm.switch` $value `,`
              $defaultDestination (`(` $defaultOperands^ `:` type($defaultOperands) `)`)?
              `[` `\n` custom<SwitchOpCases>($case_values, $caseDestinations,
              $caseOperands, type($caseOperands),
              $case_operand_offsets) `]`
              attr-dict

Attributes: ¶

Operands: ¶

Successors: ¶

llvm.trunc (::mlir::LLVM::TruncOp) ¶

Operands: ¶

Results: ¶

llvm.intr.uadd.with.overflow (::mlir::LLVM::UAddWithOverflowOp) ¶

Operands: ¶

Results: ¶

llvm.udiv (::mlir::LLVM::UDivOp) ¶

Syntax:

operation ::= `llvm.udiv` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.uitofp (::mlir::LLVM::UIToFPOp) ¶

Operands: ¶

Results: ¶

llvm.intr.umul.with.overflow (::mlir::LLVM::UMulWithOverflowOp) ¶

Operands: ¶

Results: ¶

llvm.urem (::mlir::LLVM::URemOp) ¶

Syntax:

operation ::= `llvm.urem` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.intr.usub.with.overflow (::mlir::LLVM::USubWithOverflowOp) ¶

Operands: ¶

Results: ¶

llvm.mlir.undef (::mlir::LLVM::UndefOp) ¶

Creates an undefined value of LLVM dialect type.

Syntax:

operation ::= `llvm.mlir.undef` attr-dict `:` type($res)

Unlike LLVM IR, MLIR does not have first-class undefined values. Such values must be created as SSA values using llvm.mlir.undef. This operation has no operands or attributes. It creates an undefined value of the specified LLVM IR dialect type wrapping an LLVM IR structure type.

Example:

// Create a structure with a 32-bit integer followed by a float.
%0 = llvm.mlir.undef : !llvm.struct<(i32, f32)>

Results: ¶

llvm.unreachable (::mlir::LLVM::UnreachableOp) ¶

llvm.xor (::mlir::LLVM::XOrOp) ¶

Syntax:

operation ::= `llvm.xor` $lhs `,` $rhs custom<LLVMOpAttrs>(attr-dict) `:` type($res)

Operands: ¶

Results: ¶

llvm.zext (::mlir::LLVM::ZExtOp) ¶

Operands: ¶

Results: ¶

llvm.intr.masked.compressstore (::mlir::LLVM::masked_compressstore) ¶

Operands: ¶

llvm.intr.masked.expandload (::mlir::LLVM::masked_expandload) ¶

Operands: ¶

Results: ¶

llvm.intr.masked.gather (::mlir::LLVM::masked_gather) ¶

Syntax:

operation ::= `llvm.intr.masked.gather` operands attr-dict `:` functional-type(operands, results)

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.masked.scatter (::mlir::LLVM::masked_scatter) ¶

Syntax:

operation ::= `llvm.intr.masked.scatter` $value `,` $ptrs `,` $mask attr-dict `:` type($value) `,` type($mask) `into` type($ptrs)

Attributes: ¶

Operands: ¶

llvm.intr.vector.reduce.add (::mlir::LLVM::vector_reduce_add) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.and (::mlir::LLVM::vector_reduce_and) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.fadd (::mlir::LLVM::vector_reduce_fadd) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.fmax (::mlir::LLVM::vector_reduce_fmax) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.fmin (::mlir::LLVM::vector_reduce_fmin) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.fmul (::mlir::LLVM::vector_reduce_fmul) ¶

Attributes: ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.mul (::mlir::LLVM::vector_reduce_mul) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.or (::mlir::LLVM::vector_reduce_or) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.smax (::mlir::LLVM::vector_reduce_smax) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.smin (::mlir::LLVM::vector_reduce_smin) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.umax (::mlir::LLVM::vector_reduce_umax) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.umin (::mlir::LLVM::vector_reduce_umin) ¶

Operands: ¶

Results: ¶

llvm.intr.vector.reduce.xor (::mlir::LLVM::vector_reduce_xor) ¶

Operands: ¶

Results: ¶