vpmadd52luq

Packed Multiply-Add Unsigned 52-bit Integers (Low)

VPMADD52LUQ zmm1 {k1}, zmm2, zmm3/m512

Fused multiply-add for 52-bit integers (Low 52 bits).

Details

Performs a fused multiply-add operation on 52-bit unsigned integers, extracting the lower 52 bits of the 104-bit product and accumulating into the destination. For each 64-bit lane: compute (zmm2[63:0] * zmm3[63:0]) + (zmm1[63:0]), extract bits [51:0], and write back to zmm1. Supports masking via k1; no arithmetic flags are modified.

Pseudocode Operation

for (i = 0; i < 8; i++) {
  idx = 64*i;
  product = zmm2[idx+63 : idx] * zmm3[idx+63 : idx];
  sum = (product + zmm1[idx+63 : idx]);
  result = sum & 0xFFFFFFFFFFFFF;  // 52 bits
  zmm1[idx+63 : idx] = result;
  if (k1[i]) zmm1[idx+63 : idx] = result;
}

Example

VPMADD52LUQ zmm1, zmm2, zmm3/m512

Encoding

Binary Layout

EVEX

Format EVEX

Opcode EVEX.512.66.0F38.W1 B4 /r

Extension AVX-512-IFMA

Operands

dest
ZMM
src1
ZMM
src2
ZMM/Mem

Reference (Intel® SDM)

Instruction Forms

Opcode	Instruction	Op/En	64/32-bit Mode	CPUID	Description
VEX.128.66.0F38.W1 B4 /r	VPMADD52LUQ xmm1, xmm2, xmm3/m128	A	V/V	AVX_IFMA	Multiply unsigned 52-bit integers in xmm2 and xmm3/m128 and add the low 52 bits of the 104-bit product to the qword unsigned integers in xmm1.
VEX.256.66.0F38.W1 B4 /r	VPMADD52LUQ ymm1, ymm2, ymm3/m256	A	V/V	AVX_IFMA	Multiply unsigned 52-bit integers in ymm2 and ymm3/m256 and add the low 52 bits of the 104-bit product to the qword unsigned integers in ymm1.
EVEX.128.66.0F38.W1 B4 /r	VPMADD52LUQ xmm1 {k1}{z}, xmm2,xmm3/m128/m64bcst	B	V/V	(AVX512_IFMA AND AVX512VL) OR AVX10.1	Multiply unsigned 52-bit integers in xmm2 and xmm3/m128 and add the low 52 bits of the 104-bit product to the qword unsigned integers in xmm1 using writemask k1.
EVEX.256.66.0F38.W1 B4 /r	VPMADD52LUQ ymm1 {k1}{z}, ymm2, ymm3/m256/m64bcst	B	V/V	(AVX512_IFMA AND AVX512VL) OR AVX10.1	Multiply unsigned 52-bit integers in ymm2 and ymm3/m256 and add the low 52 bits of the 104-bit product to the qword unsigned integers in ymm1 using writemask k1.
EVEX.512.66.0F38.W1 B4 /r	VPMADD52LUQ zmm1 {k1}{z}, zmm2,zmm3/m512/m64bcst	B	V/V	AVX512_IFMA OR AVX10.1	Multiply unsigned 52-bit integers in zmm2 and zmm3/m512 and add the low 52 bits of the 104-bit product to the qword unsigned integers in zmm1 using writemask k1.

Instruction Operand Encoding

Op/En	Tuple Type	Operand 1	Operand 2	Operand 3	Operand 4
A	N/A	ModRM:reg (r, w)	VEX.vvvv (r)	ModRM:r/m (r)	N/A
B	Full	ModRM:reg (r, w)	EVEX.vvvv (r)	ModRM:r/m(r)	N/A

Description

Multiplies packed unsigned 52-bit integers in each qword element of the first source operand (the second operand) with the packed unsigned 52-bit integers in the corresponding elements of the second source operand (the third operand) to form packed 104-bit intermediate results. The low 52-bit, unsigned integer of each 104-bit product is added to the corresponding qword unsigned integer of the destination operand (the first operand) under the writemask k1. The first source 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 at 64-bit granularity. VPMADD52LUQ—Packed Multiply of Unsigned 52-Bit Integers and Add the Low 52-Bit Products to Qword Accumulators Vol. 2C 5-568

Operation

VPMADDLUQ srcdest, src1, src2 (VEX version)
VL = (128,256)
KL = VL/64

FOR i in 0 .. KL-1:
temp128 := zeroextend64(src1.qword[i][51:0]) *zeroextend64(src2.qword[i][51:0])
srcdest.qword[i] := srcdest.qword[i] +zeroextend64(temp128[51:0])
srcdest[MAXVL:VL] := 0


VPMADD52LUQ (EVEX encoded)
(KL, VL) = (2, 128), (4, 256), (8, 512)
FOR j := 0 TO KL-1
i := j * 64;
IF k1[j] OR *no writemask* THEN
IF src2 is Memory AND EVEX.b=1 THEN
tsrc2[63:0] := ZeroExtend64(src2[51:0]);
ELSE
tsrc2[63:0] := ZeroExtend64(src2[i+51:i];
FI;
Temp128[127:0] := ZeroExtend64(src1[i+51:i]) * tsrc2[63:0];
Temp2[63:0] := DEST[i+63:i] + ZeroExtend64(temp128[51:0]) ;
DEST[i+63:i] := Temp2[63:0];
ELSE
IF *zeroing-masking* THEN
DEST[i+63:i] := 0;
ELSE *merge-masking*
DEST[i+63:i] is unchanged;
FI;
FI;
ENDFOR
DEST[MAX_VL-1:VL] := 0;

Intel C/C++ Compiler Intrinsic Equivalent

VPMADD52LUQ __m128i _mm_madd52lo_avx_epu64 (__m128i __X, __m128i __Y, __m128i __Z);
VPMADD52LUQ __m128i _mm_madd52lo_epu64( __m128i a, __m128i b, __m128i c);
VPMADD52LUQ __m128i _mm_madd52lo_epu64 (__m128i __X, __m128i __Y, __m128i __Z);
VPMADD52LUQ __m128i _mm_mask_madd52lo_epu64(__m128i s, __mmask8 k, __m128i a, __m128i b, __m128i c);
VPMADD52LUQ __m128i _mm_maskz_madd52lo_epu64( __mmask8 k, __m128i a, __m128i b, __m128i c);
VPMADD52LUQ __m256i _mm256_madd52lo_avx_epu64 (__m256i __X, __m256i __Y, __m256i __Z);
VPMADD52LUQ __m256i _mm256_madd52lo_epu64( __m256i a, __m256i b, __m256i c);
VPMADD52LUQ __m256i _mm256_madd52lo_epu64 (__m256i __X, __m256i __Y, __m256i __Z);
VPMADD52LUQ __m256i _mm256_mask_madd52lo_epu64(__m256i s, __mmask8 k, __m256i a, __m256i b, __m256i c);
VPMADD52LUQ __m256i _mm256_maskz_madd52lo_epu64( __mmask8 k, __m256i a, __m256i b, __m256i c);
VPMADD52LUQ __m512i _mm512_madd52lo_epu64( __m512i a, __m512i b, __m512i c);
VPMADD52LUQ __m512i _mm512_mask_madd52lo_epu64(__m512i s, __mmask8 k, __m512i a, __m512i b, __m512i c);
VPMADD52LUQ __m512i _mm512_maskz_madd52lo_epu64( __mmask8 k, __m512i a, __m512i b, __m512i c);

Flags Affected

None. VPMADD52LUQ—Packed Multiply of Unsigned 52-Bit Integers and Add the Low 52-Bit Products to Qword Accumulators Vol. 2C 5-569

Exceptions

SIMD Floating-Point Exceptions

None.

Other Exceptions

VEX-encoded instructions, see Table 2-21, “Type 4 Class Exception Conditions.” EVEX-encoded instructions, see Table 2-51, “Type E4 Class Exception Conditions.” VPMADD52LUQ—Packed Multiply of Unsigned 52-Bit Integers and Add the Low 52-Bit Products to Qword Accumulators Vol. 2C 5-570