llvm2py.ir.type module¶

This module contains all types that exist in the LLVM IR representation. For detailed documentation, please refer to the source https://llvm.org/docs/LangRef.html#type-system

class ArrayType(elem_count: int, elem_ty: Type)¶

Bases: NamedTuple

Parameters:

class FPType(kind: str)¶

Bases: NamedTuple

Parameters:: kind (str) – One of {‘half’, ‘bfloat’, ‘float’, ‘double’, ‘fp128’, ‘x86_fp80’, ‘ppc_fp128’}

class FunctionType(param_tys: list[Type], ret_ty: Type)¶

Bases: NamedTuple

Parameters:

param_tys (list[Type]) – List of function argument types in the appropriate order.
ret_ty (Type) – Type of value returned by function.

class IntegerType(num_bits: int)¶

Bases: NamedTuple

class PtrType(addr_space: int, ty: Type | None = None)¶

Bases: NamedTuple

Almost always, pointers are opaque (untyped).

Parameters:

addr_space (int) – Numbered address space where the pointed-to object resides.
ty (Type | None) – Pointer type (may be used by some gpu targets).

class StructureType(name: str | None, elem_tys: list[Type], is_packed: bool = False)¶

Bases: NamedTuple

Parameters:: name (str | None) – Structure name, if it is a named structure.

class TargetExtensionType(name: str, params: list[int | Type])¶

Bases: NamedTuple

Parameters:

class VectorType(elem_count: int, elem_ty: Type, is_scalable: bool = False)¶

Bases: NamedTuple

Parameters:

elem_count (int) – Vector element count. If the vector is scalable, it is the minimum number of elements in this vector. The actual number of elements in the vector is an integer multiple of this value.
elem_ty (Type) – Vector element type.
is_scalable (bool) – If True, the vector is scalable (vscale).