MAXPD—Maximum of Packed Double-Precision Floating-Point Values

Opcode/Instruction Op /En 64/32 bit Mode Support CPUID Feature Flag Description
66 0F 5F /r MAXPD xmm1, xmm2/m128 RM V/V SSE2 Return the maximum double-precision floating-point values between xmm1 and xmm2/m128.
VEX.NDS.128.66.0F.WIG 5F /r VMAXPD xmm1, xmm2, xmm3/m128 RVM V/V AVX Return the maximum double-precision floating-point values between xmm2 and xmm3/m128.
VEX.NDS.256.66.0F.WIG 5F /r VMAXPD ymm1, ymm2, ymm3/m256 RVM V/V AVX Return the maximum packed double-precision floating-point values between ymm2 and ymm3/m256.
EVEX.NDS.128.66.0F.W1 5F /r VMAXPD xmm1 {k1}{z}, xmm2, xmm3/m128/m64bcst FV V/V AVX512VL AVX512F Return the maximum packed double-precision floating-point values between xmm2 and xmm3/m128/m64bcst and store result in xmm1 subject to writemask k1.
EVEX.NDS.256.66.0F.W1 5F /r VMAXPD ymm1 {k1}{z}, ymm2, ymm3/m256/m64bcst FV V/V AVX512VL AVX512F Return the maximum packed double-precision floating-point values between ymm2 and ymm3/m256/m64bcst and store result in ymm1 subject to writemask k1.
EVEX.NDS.512.66.0F.W1 5F /r VMAXPD zmm1 {k1}{z}, zmm2, zmm3/m512/m64bcst{sae} FV V/V AVX512F Return the maximum packed double-precision floating-point values between zmm2 and zmm3/m512/m64bcst and store result in zmm1 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 ModRM:r/m (r) NA
FV ModRM:reg (w) EVEX.vvvv ModRM:r/m (r) NA

Description

Performs a SIMD compare of the packed double-precision floating-point values in the first source operand and the second source operand and returns the maximum value for each pair of values to the destination operand.

If the values being compared are both 0.0s (of either sign), the value in the second operand (source operand) is returned. If a value in the second operand is an SNaN, then SNaN is forwarded unchanged to the destination (that is, a QNaN version of the SNaN is not returned).

If only one value is a NaN (SNaN or QNaN) for this instruction, the second operand (source operand), either a NaN or a valid floating-point value, is written to the result. If instead of this behavior, it is required that the NaN source operand (from either the first or second operand) be returned, the action of MAXPD can be emulated using a sequence of instructions, such as a comparison followed by AND, ANDN and OR.

EVEX encoded versions: The first source operand (the second operand) is a ZMM/YMM/XMM register. The second source operand can be a ZMM/YMM/XMM register, a 512/256/128-bit memory location or a 512/256/128-bit vector broadcasted from a 64-bit memory location. The destination operand is a ZMM/YMM/XMM register conditionally updated with writemask k1.

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. The upper bits (MAX_VL-1:256) of the corresponding ZMM register destination are zeroed.

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.

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.

Operation

MAX(SRC1, SRC2)

{

IF ((SRC1 = 0.0) and (SRC2 = 0.0)) THEN DEST (cid:197)SRC2;

ELSE IF (SRC1 = SNaN) THEN DEST (cid:197)SRC2; FI;

ELSE IF (SRC2 = SNaN) THEN DEST (cid:197)SRC2; FI;

ELSE IF (SRC1 > SRC2) THEN DEST (cid:197)SRC1;

ELSE DEST (cid:197)SRC2;

FI;

}

VMAXPD (EVEX encoded versions)

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

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

i (cid:197) j * 64

IF k1[j] OR *no writemask*

THEN

IF (EVEX.b = 1) AND (SRC2 *is memory*)

THEN

DEST[i+63:i] (cid:197) MAX(SRC1[i+63:i], SRC2[63:0])

ELSE

DEST[i+63:i] (cid:197) MAX(SRC1[i+63:i], SRC2[i+63:i])

FI;

ELSE

IF *merging-masking*

; merging-masking

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

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

; zeroing-masking

FI

FI;

ENDFOR

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

VMAXPD (VEX.256 encoded version)

DEST[63:0] (cid:197)MAX(SRC1[63:0], SRC2[63:0])

DEST[127:64] (cid:197)MAX(SRC1[127:64], SRC2[127:64])

DEST[191:128] (cid:197)MAX(SRC1[191:128], SRC2[191:128])

DEST[255:192] (cid:197)MAX(SRC1[255:192], SRC2[255:192])

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

VMAXPD (VEX.128 encoded version)

DEST[63:0] (cid:197)MAX(SRC1[63:0], SRC2[63:0])

DEST[127:64] (cid:197)MAX(SRC1[127:64], SRC2[127:64])

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

MAXPD (128-bit Legacy SSE version)

DEST[63:0] (cid:197)MAX(DEST[63:0], SRC[63:0])

DEST[127:64] (cid:197)MAX(DEST[127:64], SRC[127:64])

DEST[MAX_VL-1:128] (Unmodified)

Intel C/C++ Compiler Intrinsic Equivalent

VMAXPD __m512d _mm512_max_pd( __m512d a, __m512d b);

VMAXPD __m512d _mm512_mask_max_pd(__m512d s, __mmask8 k, __m512d a, __m512d b,);

VMAXPD __m512d _mm512_maskz_max_pd( __mmask8 k, __m512d a, __m512d b);

VMAXPD __m512d _mm512_max_round_pd( __m512d a, __m512d b, int);

VMAXPD __m512d _mm512_mask_max_round_pd(__m512d s, __mmask8 k, __m512d a, __m512d b, int);

VMAXPD __m512d _mm512_maskz_max_round_pd( __mmask8 k, __m512d a, __m512d b, int);

VMAXPD __m256d _mm256_mask_max_pd(__m5256d s, __mmask8 k, __m256d a, __m256d b);

VMAXPD __m256d _mm256_maskz_max_pd( __mmask8 k, __m256d a, __m256d b);

VMAXPD __m128d _mm_mask_max_pd(__m128d s, __mmask8 k, __m128d a, __m128d b);

VMAXPD __m128d _mm_maskz_max_pd( __mmask8 k, __m128d a, __m128d b);

VMAXPD __m256d _mm256_max_pd (__m256d a, __m256d b);

(V)MAXPD __m128d _mm_max_pd (__m128d a, __m128d b);

SIMD Floating-Point Exceptions

Invalid (including QNaN Source Operand), Denormal

Other Exceptions

Non-EVEX-encoded instruction, see Exceptions Type 2.

EVEX-encoded instruction, see Exceptions Type E2.