Mojo function

mulhi

mulhi(a: SIMD[uint16, 1], b: SIMD[uint16, 1]) -> SIMD[uint32, 1]

Calculates the most significant 32 bits of the product of two 16-bit unsigned integers.

This function multiplies two 16-bit unsigned integers and returns the high 32 bits of their product. This is useful for fixed-point arithmetic and overflow detection.

Note: On NVIDIA GPUs, this maps directly to the MULHI.U16 PTX instruction. On other architectures, it performs the multiplication using 32-bit arithmetic.

Args:

a (SIMD[uint16, 1]): First 16-bit unsigned integer operand.
b (SIMD[uint16, 1]): Second 16-bit unsigned integer operand.

Returns:

The high 32 bits of the product a * b

mulhi(a: SIMD[int16, 1], b: SIMD[int16, 1]) -> SIMD[int32, 1]

Calculates the most significant 32 bits of the product of two 16-bit signed integers.

This function multiplies two 16-bit signed integers and returns the high 32 bits of their product. This is useful for fixed-point arithmetic and overflow detection.

Note: On NVIDIA GPUs, this maps directly to the MULHI.S16 PTX instruction. On other architectures, it performs the multiplication using 32-bit arithmetic.

Args:

a (SIMD[int16, 1]): First 16-bit signed integer operand.
b (SIMD[int16, 1]): Second 16-bit signed integer operand.

Returns:

The high 32 bits of the product a * b

mulhi(a: SIMD[uint32, 1], b: SIMD[uint32, 1]) -> SIMD[uint32, 1]

Calculates the most significant 32 bits of the product of two 32-bit unsigned integers.

This function multiplies two 32-bit unsigned integers and returns the high 32 bits of their product. This is useful for fixed-point arithmetic and overflow detection.

Note: On NVIDIA GPUs, this maps directly to the MULHI.U32 PTX instruction. On other architectures, it performs the multiplication using 64-bit arithmetic.

Args:

a (SIMD[uint32, 1]): First 32-bit unsigned integer operand.
b (SIMD[uint32, 1]): Second 32-bit unsigned integer operand.

Returns:

The high 32 bits of the product a * b

mulhi(a: SIMD[int32, 1], b: SIMD[int32, 1]) -> SIMD[int32, 1]

Calculates the most significant 32 bits of the product of two 32-bit signed integers.

This function multiplies two 32-bit signed integers and returns the high 32 bits of their product. This is useful for fixed-point arithmetic and overflow detection.

Note: On NVIDIA GPUs, this maps directly to the MULHI.S32 PTX instruction. On other architectures, it performs the multiplication using 64-bit arithmetic.

Args:

a (SIMD[int32, 1]): First 32-bit signed integer operand.
b (SIMD[int32, 1]): Second 32-bit signed integer operand.

Returns:

The high 32 bits of the product a * b