VFMSUBADD132PS/VFMSUBADD213PS/VFMSUBADD231PS — Fused Multiply-Alternating Subtract/Add of Packed Single-Precision Floating-Point Values

Opcode/Instruction Op/En 64/32 -bit Mode CPUID Feature Flag Description

VEX.DDS.128.66.0F38.W0 97 /r

VFMSUBADD132PS xmm0, xmm1, xmm2/m128

A V/V FMA Multiply packed single-precision floating-point values from xmm0 and xmm2/mem, subtract/add elements in xmm1 and put result in xmm0.

VEX.DDS.128.66.0F38.W0 A7 /r

VFMSUBADD213PS xmm0, xmm1, xmm2/m128

A V/V FMA Multiply packed single-precision floating-point values from xmm0 and xmm1, subtract/add ele-ments in xmm2/mem and put result in xmm0.

VEX.DDS.128.66.0F38.W0 B7 /r

VFMSUBADD231PS xmm0, xmm1, xmm2/m128

A V/V FMA Multiply packed single-precision floating-point values from xmm1 and xmm2/mem, sub-tract/add elements in xmm0 and put result in xmm0.

VEX.DDS.256.66.0F38.W0 97 /r

VFMSUBADD132PS ymm0, ymm1, ymm2/m256

A V/V FMA Multiply packed single-precision floating-point values from ymm0 and ymm2/mem, subtract/add elements in ymm1 and put result in ymm0.

VEX.DDS.256.66.0F38.W0 A7 /r

VFMSUBADD213PS ymm0, ymm1, ymm2/m256

A V/V FMA Multiply packed single-precision floating-point values from ymm0 and ymm1, subtract/add ele-ments in ymm2/mem and put result in ymm0.

VEX.DDS.256.66.0F38.W0 B7 /r

VFMSUBADD231PS ymm0, ymm1, ymm2/m256

A V/V FMA Multiply packed single-precision floating-point values from ymm1 and ymm2/mem, subtract/add elements in ymm0 and put result in ymm0.

Instruction Operand Encoding

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

Description

VFMSUBADD132PS: Multiplies the four or eight packed single-precision floating-point values from the first source operand to the four or eight packed single-precision floating-point values in the third source operand. From the infi-nite precision intermediate result, subtracts the odd single-precision floating-point elements and adds the even single-precision floating-point values in the second source operand, performs rounding and stores the resulting four or eight packed single-precision floating-point values to the destination operand (first source operand).

VFMSUBADD213PS: Multiplies the four or eight packed single-precision floating-point values from the second source operand to the four or eight packed single-precision floating-point values in the first source operand. From the infinite precision intermediate result, subtracts the odd single-precision floating-point elements and adds the even single-precision floating-point values in the third source operand, performs rounding and stores the resulting four or eight packed single-precision floating-point values to the destination operand (first source operand).

VFMSUBADD231PS: Multiplies the four or eight packed single-precision floating-point values from the second source operand to the four or eight packed single-precision floating-point values in the third source operand. From the infinite precision intermediate result, subtracts the odd single-precision floating-point elements and adds the even single-precision floating-point values in the first source operand, performs rounding and stores the resulting four or eight packed single-precision floating-point values to the destination operand (first source operand).

VEX.128 encoded version: The destination operand (also first source operand) is a XMM register and encoded in reg_field. The second source operand is a XMM register and encoded in VEX.vvvv. The third source operand is a XMM register or a 128-bit memory location and encoded in rm_field. The upper 128 bits of the YMM destination register are zeroed.

VEX.256 encoded version: The destination operand (also first source operand) is a YMM register and encoded in reg_field. The second source operand is a YMM register and encoded in VEX.vvvv. The third source operand is a YMM register or a 256-bit memory location and encoded in rm_field.

Compiler tools may optionally support a complementary mnemonic for each instruction mnemonic listed in the opcode/instruction column of the summary table. The behavior of the complementary mnemonic in situations involving NANs are governed by the definition of the instruction mnemonic defined in the opcode/instruction column. See also Section 14.5.1, “FMA Instruction Operand Order and Arithmetic Behavior” in the Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 1.

Operation

In the operations below, "+", "-", and "*" symbols represent addition, subtraction, and multiplication operations with infinite precision inputs and outputs (no rounding).

VFMSUBADD132PS DEST, SRC2, SRC3

IF (VEX.128) THEN
    MAXVL =2
ELSEIF (VEX.256)
    MAXVL = 4
FI
For i = 0 to MAXVL -1{
    n = 64*i;
    DEST[n+31:n] ← RoundFPControl_MXCSR(DEST[n+31:n]*SRC3[n+31:n] + SRC2[n+31:n])
    DEST[n+63:n+32] ← RoundFPControl_MXCSR(DEST[n+63:n+32]*SRC3[n+63:n+32] -SRC2[n+63:n+32])
}
IF (VEX.128) THEN
    DEST[VLMAX-1:128] ← 0
FI

VFMSUBADD213PS DEST, SRC2, SRC3

IF (VEX.128) THEN
    MAXVL =2
ELSEIF (VEX.256)
    MAXVL = 4
FI
For i = 0 to MAXVL -1{
    n = 64*i;
    DEST[n+31:n] ← RoundFPControl_MXCSR(SRC2[n+31:n]*DEST[n+31:n] +SRC3[n+31:n])
    DEST[n+63:n+32] ← RoundFPControl_MXCSR(SRC2[n+63:n+32]*DEST[n+63:n+32] -SRC3[n+63:n+32])
}
IF (VEX.128) THEN
    DEST[VLMAX-1:128] ← 0
FI

VFMSUBADD231PS DEST, SRC2, SRC3

IF (VEX.128) THEN
    MAXVL =2
ELSEIF (VEX.256)
    MAXVL = 4
FI
For i = 0 to MAXVL -1{
    n = 64*i;
    DEST[n+31:n] ← RoundFPControl_MXCSR(SRC2[n+31:n]*SRC3[n+31:n] + DEST[n+31:n])
    DEST[n+63:n+32] ← RoundFPControl_MXCSR(SRC2[n+63:n+32]*SRC3[n+63:n+32] -DEST[n+63:n+32])
}
IF (VEX.128) THEN
    DEST[VLMAX-1:128] ← 0
FI

Intel C/C++ Compiler Intrinsic Equivalent

VFMSUBADD132PS: __m128 _mm_fmsubadd_ps (__m128 a, __m128 b, __m128 c);

VFMSUBADD213PS: __m128 _mm_fmsubadd_ps (__m128 a, __m128 b, __m128 c);

VFMSUBADD231PS: __m128 _mm_fmsubadd_ps (__m128 a, __m128 b, __m128 c);

VFMSUBADD132PS: __m256 _mm256_fmsubadd_ps (__m256 a, __m256 b, __m256 c);

VFMSUBADD213PS: __m256 _mm256_fmsubadd_ps (__m256 a, __m256 b, __m256 c);

VFMSUBADD231PS: __m256 _mm256_fmsubadd_ps (__m256 a, __m256 b, __m256 c);

SIMD Floating-Point Exceptions

Overflow, Underflow, Invalid, Precision, Denormal

Other Exceptions

See Exceptions Type 2