reprospect.test.sass.matchers.convert_fp_to_int module

class reprospect.test.sass.matchers.convert_fp_to_int.ConvertFpToInt(*, dst_dtype: TypeInfo | int | DTypeLike | None = None, src_dtype: TypeInfo | int | DTypeLike | None = None, dst: str = 'R[0-9]+', src: str = 'R[0-9]+')View on GitHub

Bases: InstructionMatcher

Convert floating-point value to integer value.

Note

The .NTZ opcode modifier means:

round to nearest integer in the direction of zero

References:

• [AASB22]

• https://docs.nvidia.com/cuda/parallel-thread-execution/#rounding-modifiers

Note

It is not decorated with dataclasses.dataclass() because of https://github.com/mypyc/mypyc/issues/1061.

__init__(*, dst_dtype: TypeInfo | int | DTypeLike | None = None, src_dtype: TypeInfo | int | DTypeLike | None = None, dst: str = 'R[0-9]+', src: str = 'R[0-9]+') → NoneView on GitHub

Parameters:

• src_dtype – Type of the source floating-point value.

• dst_dtype – Type of the destination integer value.

match(inst: Instruction | str) → InstructionMatch | NoneView on GitHub

matcher: Final[OpcodeModsWithOperandsMatcher]