KUNPCKBW/KUNPCKWD/KUNPCKDQ—Unpack for Mask Registers

Opcode/Instruction	Op/En	64/32 bit Mode Support	CPUID Feature Flag	Description
VEX.NDS.L1.66.0F.W0 4B /r KUNPCKBW k1, k2, k3	RVR	V/V	AVX512F	Unpack and interleave 8 bits masks in k2 and k3 and write word result in k1.
VEX.NDS.L1.0F.W0 4B /r KUNPCKWD k1, k2, k3	RVR	V/V	AVX512BW	Unpack and interleave 16 bits in k2 and k3 and write double-word result in k1.
VEX.NDS.L1.0F.W1 4B /r KUNPCKDQ k1, k2, k3	RVR	V/V	AVX512BW	Unpack and interleave 32 bits masks in k2 and k3 and write quadword result in k1.

Instruction Operand Encoding

Op/En	Operand 1	Operand 2	Operand 3
RVR	ModRM:reg (w)	VEX.1vvv (r)	ModRM:r/m (r, ModRM:[7:6] must be 11b)

Description

Unpacks the lower 8/16/32 bits of the second and third operands (source operands) into the low part of the first operand (destination operand), starting from the low bytes. The result is zero-extended in the destination.

Operation

KUNPCKBW

DEST[7:0] (cid:197) SRC2[7:0]
DEST[15:8] (cid:197) SRC1[7:0]
DEST[MAX_KL-1:16] (cid:197) 0

KUNPCKWD

DEST[15:0] (cid:197) SRC2[15:0]
DEST[31:16] (cid:197) SRC1[15:0] DEST[MAX_KL-1:32] (cid:197) 0

KUNPCKDQ

DEST[31:0] (cid:197) SRC2[31:0]
DEST[63:32] (cid:197) SRC1[31:0] DEST[MAX_KL-1:64] (cid:197) 0

Intel C/C++ Compiler Intrinsic Equivalent

KUNPCKBW __mmask16 _mm512_kunpackb(__mmask16 a, __mmask16 b);
KUNPCKDQ __mmask64 _mm512_kunpackd(__mmask64 a, __mmask64 b);
KUNPCKWD __mmask32 _mm512_kunpackw(__mmask32 a, __mmask32 b);

Flags Affected

None

SIMD Floating-Point Exceptions

None

Other Exceptions

See Exceptions Type K20.