UNPCKHPD—Unpack and Interleave High Packed Double-Precision Floating-Point Values

Opcode/Instruction

RM

UNPCKHPD xmm1, xmm2/m128

Op /En

64/32 bit Mode Support

V/V

CPUID Feature Flag

SSE2

Description

Unpacks and Interleaves double-precision floating-point values from high quadwords of xmm1 and xmm2/m128.

VEX.NDS.128.66.0F.WIG 15 /r

VUNPCKHPD xmm1,xmm2, xmm3/m128

RVM V/V AVX Unpacks and Interleaves double-precision floating-point values from high quadwords of xmm2 and xmm3/m128.

VEX.NDS.256.66.0F.WIG 15 /r

VUNPCKHPD ymm1,ymm2, ymm3/m256

RVM V/V AVX Unpacks and Interleaves double-precision floating-point values from high quadwords of ymm2 and ymm3/m256.

EVEX.NDS.128.66.0F.W1 15 /r

VUNPCKHPD xmm1 {k1}{z}, xmm2, xmm3/m128/m64bcst

FV V/V AVX512VL AVX512F Unpacks and Interleaves double precision floating-point values from high quadwords of xmm2 and xmm3/m128/m64bcst subject to writemask k1.

EVEX.NDS.256.66.0F.W1 15 /r

VUNPCKHPD ymm1 {k1}{z}, ymm2, ymm3/m256/m64bcst

FV V/V AVX512VL AVX512F Unpacks and Interleaves double precision floating-point values from high quadwords of ymm2 and ymm3/m256/m64bcst subject to writemask k1.

EVEX.NDS.512.66.0F.W1 15 /r

VUNPCKHPD zmm1 {k1}{z}, zmm2, zmm3/m512/m64bcst

FV V/V AVX512F Unpacks and Interleaves double-precision floating-point values from high quadwords of zmm2 and zmm3/m512/m64bcst subject to writemask k1.

Instruction Operand Encoding

Op/En Operand 1 Operand 2 Operand 3 Operand 4
RM ModRM:reg (r, w) ModRM:r/m (r) NA NA
RVM ModRM:reg (w) VEX.vvvv (r) ModRM:r/m (r) NA
FV ModRM:reg (w) EVEX.vvvv (r) ModRM:r/m (r) NA

Description

Performs an interleaved unpack of the high double-precision floating-point values from the first source operand and the second source operand. See Figure 4-15 in the Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 2B.

128-bit Legacy SSE version: The second source can be an XMM register or an 128-bit memory location. The desti-nation is not distinct from the first source XMM register and the upper bits (MAX_VL-1:128) of the corresponding ZMM register destination are unmodified. When unpacking from a memory operand, an implementation may fetch only the appropriate 64 bits; however, alignment to 16-byte boundary and normal segment checking will still be enforced.

VEX.128 encoded version: The first source operand is a XMM register. The second source operand can be a XMM register or a 128-bit memory location. The destination operand is a XMM register. The upper bits (MAX_VL-1:128) of the corresponding ZMM register destination are zeroed.

VEX.256 encoded version: The first source operand is a YMM register. The second source operand can be a YMM register or a 256-bit memory location. The destination operand is a YMM register.

EVEX.512 encoded version: The first source operand is a ZMM register. The second source operand is a ZMM register, a 512-bit memory location, or a 512-bit vector broadcasted from a 64-bit memory location. The destina-tion operand is a ZMM register, conditionally updated using writemask k1.

EVEX.256 encoded version: The first source operand is a YMM register. The second source operand is a YMM register, a 256-bit memory location, or a 256-bit vector broadcasted from a 64-bit memory location. The destina-tion operand is a YMM register, conditionally updated using writemask k1.

EVEX.128 encoded version: The first source operand is a XMM register. The second source operand is a XMM register, a 128-bit memory location, or a 128-bit vector broadcasted from a 64-bit memory location. The destina-tion operand is a XMM register, conditionally updated using writemask k1.

Operation

VUNPCKHPD (EVEX encoded versions when SRC2 is a register)

(KL, VL) = (2, 128), (4, 256), (8, 512)

IF VL >= 128

TMP_DEST[63:0] (cid:197) SRC1[127:64]

TMP_DEST[127:64] (cid:197) SRC2[127:64]

FI;

IF VL >= 256

TMP_DEST[191:128] (cid:197) SRC1[255:192]

TMP_DEST[255:192] (cid:197) SRC2[255:192]

FI;

IF VL >= 512

TMP_DEST[319:256] (cid:197) SRC1[383:320]

TMP_DEST[383:320] (cid:197) SRC2[383:320]

TMP_DEST[447:384] (cid:197) SRC1[511:448]

TMP_DEST[511:448] (cid:197) SRC2[511:448]

FI;

FOR j (cid:197) 0 TO KL-1

i (cid:197) j * 64

IF k1[j] OR *no writemask*

THEN DEST[i+63:i] (cid:197) TMP_DEST[i+63:i]

ELSE

IF *merging-masking*

; merging-masking

THEN *DEST[i+63:i] remains unchanged*

ELSE *zeroing-masking*

; zeroing-masking

DEST[i+63:i] (cid:197) 0

FI

FI;

ENDFOR

DEST[MAX_VL-1:VL] (cid:197) 0

VUNPCKHPD (EVEX encoded version when SRC2 is memory)

(KL, VL) = (2, 128), (4, 256), (8, 512)

FOR j (cid:197) 0 TO KL-1

i (cid:197) j * 64

IF (EVEX.b = 1)

THEN TMP_SRC2[i+63:i] (cid:197) SRC2[63:0]

ELSE TMP_SRC2[i+63:i] (cid:197) SRC2[i+63:i]

FI;

ENDFOR;

IF VL >= 128

TMP_DEST[63:0] (cid:197) SRC1[127:64]

TMP_DEST[127:64] (cid:197) TMP_SRC2[127:64]

FI;

IF VL >= 256

TMP_DEST[191:128] (cid:197) SRC1[255:192]

TMP_DEST[255:192] (cid:197) TMP_SRC2[255:192]

FI;

IF VL >= 512

TMP_DEST[319:256] (cid:197) SRC1[383:320]

TMP_DEST[383:320] (cid:197) TMP_SRC2[383:320]

TMP_DEST[447:384] (cid:197) SRC1[511:448]

TMP_DEST[511:448] (cid:197) TMP_SRC2[511:448]

FI;

FOR j (cid:197) 0 TO KL-1

i (cid:197) j * 64

IF k1[j] OR *no writemask*

THEN DEST[i+63:i] (cid:197) TMP_DEST[i+63:i]

ELSE

IF *merging-masking*

; merging-masking

THEN *DEST[i+63:i] remains unchanged*

ELSE *zeroing-masking*

; zeroing-masking

DEST[i+63:i] (cid:197) 0

FI

FI;

ENDFOR

DEST[MAX_VL-1:VL] (cid:197) 0

VUNPCKHPD (VEX.256 encoded version)

DEST[63:0] (cid:197)SRC1[127:64]

DEST[127:64] (cid:197)SRC2[127:64]

DEST[191:128](cid:197)SRC1[255:192]

DEST[255:192](cid:197)SRC2[255:192]

DEST[MAX_VL-1:256] (cid:197)0

VUNPCKHPD (VEX.128 encoded version)

DEST[63:0] (cid:197)SRC1[127:64]

DEST[127:64] (cid:197)SRC2[127:64]

DEST[MAX_VL-1:128] (cid:197)0

UNPCKHPD (128-bit Legacy SSE version)

DEST[63:0] (cid:197)SRC1[127:64]

DEST[127:64] (cid:197)SRC2[127:64]

DEST[MAX_VL-1:128] (Unmodified)

Intel C/C++ Compiler Intrinsic Equivalent

VUNPCKHPD __m512d _mm512_unpackhi_pd( __m512d a, __m512d b);

VUNPCKHPD __m512d _mm512_mask_unpackhi_pd(__m512d s, __mmask8 k, __m512d a, __m512d b);

VUNPCKHPD __m512d _mm512_maskz_unpackhi_pd(__mmask8 k, __m512d a, __m512d b);

VUNPCKHPD __m256d _mm256_unpackhi_pd(__m256d a, __m256d b)

VUNPCKHPD __m256d _mm256_mask_unpackhi_pd(__m256d s, __mmask8 k, __m256d a, __m256d b);

VUNPCKHPD __m256d _mm256_maskz_unpackhi_pd(__mmask8 k, __m256d a, __m256d b);

UNPCKHPD __m128d _mm_unpackhi_pd(__m128d a, __m128d b)

VUNPCKHPD __m128d _mm_mask_unpackhi_pd(__m128d s, __mmask8 k, __m128d a, __m128d b);

VUNPCKHPD __m128d _mm_maskz_unpackhi_pd(__mmask8 k, __m128d a, __m128d b);

SIMD Floating-Point Exceptions

None

Other Exceptions

Non-EVEX-encoded instructions, see Exceptions Type 4.

EVEX-encoded instructions, see Exceptions Type E4NF.