VFMADD213BF16

VFMADD213BF16 computes op2 × op1 + op3 for each bfloat16 element, 8, 16 or 32 per register, rounds once to nearest even and writes the result over op1. It is the 213 order of AVX10.2’s multiply-add, one of the twelve forms that VFMADD132BF16’s page describes with their shared rules: denormal inputs and results taken as zeros, no exceptions and no MXCSR. Its opcode is A8h in EVEX map 6 with no prefix and W0; VFMADD213PH has the same byte with 66. Only op3 may come from memory or be broadcast from 16 bits.

VFMADD231PS explains the three orders for singles. Under merge masking an element whose mask bit is clear keeps op1, here the other factor.

In our builds#

GCC 14 has no AVX10.2 option. Clang 23 at -O2 with -mavx10.2 compiles the unmasked _mm512_fmadd_pbh(a, b, c) to vfmadd213bf16 zmm0, zmm1, zmm2 and the zero-masked form to the same with {k1} {z}. It also takes this form for the loop Z[i] = X[i] * Y[i] + Z[i] over global __bf16 arrays, with Z read from memory as op3, the multiply and the add contracted by default. A single __bf16 a * b + c instead takes a single-precision FMA between widening and narrowing, and rounds twice. No AVX10.2 processor appears in this site’s recorded cases.

Forms

Data flow

Where the elements of the sources go, element 0 on the right; results that depend on a control value are drawn for an example. How to read.

Operation

What the forms do, in pseudocode that our tests run against results recorded on real processors. How to read.

for i in range(VL // 16):
    if k1[i]:
        a = zmm1.bf16[i]
        b = zmm2.bf16[i]
        c = src3.bf16[i]
        zmm1.bf16[i] = fmadd(b, a, c)
    elif zeroing:
        zmm1.bf16[i] = 0
    # else: zmm1.bf16[i] keeps its value
# m16bcst: src3.bf16[i] is the m16 for every i
# BF16: to nearest even, denormals flushed, no flags; MXCSR unused
# zmm1 is zeroed from bit VL up

Operands

Flags

None: XED records no flag that the forms test or change.

Exceptions

The exception class is XED's name for the set of exception conditions that a form follows. How to read.

All forms

AVX512-E4

Attributes

What else XED records about the forms. How to read.

All forms

write masking; always rounds to nearest even; denormal inputs read as zero; denormal results flushed to zero

Memory forms

embedded broadcast (tuple Full); memory fault suppression

Intrinsics

The C intrinsics of Clang 23.1.2 that compile to VFMADD213BF16, each with the form it compiles to and the Clang options it needs. The forms come from compiling every intrinsic and decoding the result with Intel XED; all intrinsics A–Z explains the method.

CPUID and processors

Each form belongs to an XED ISA set. A form can be used when the processor reports every CPUID bit of one of its ISA set's alternatives.

More AVX10.2 instructions

• VCVTTBF162IUBS
• VCVTTPD2DQS
• VCVTTPD2QQS
• VCVTTPD2UDQS
• VCVTTPD2UQQS
• VCVTTPH2IBS
• VCVTTPH2IUBS
• VCVTTPS2DQS
• VCVTTPS2IBS
• VCVTTPS2IUBS
• VCVTTPS2QQS
• VCVTTPS2UDQS
• VCVTTPS2UQQS
• VCVTTSD2SIS
• VCVTTSD2USIS
• VCVTTSS2SIS
• VCVTTSS2USIS
• VDIVBF16
• VDPPHPS
• VFMADD132BF16
• VFMADD231BF16
• VFMSUB132BF16
• VFMSUB213BF16
• VFMSUB231BF16
• VFNMADD132BF16
• VFNMADD213BF16
• VFNMADD231BF16
• VFNMSUB132BF16
• VFNMSUB213BF16
• VFNMSUB231BF16
• VFPCLASSBF16
• VGETEXPBF16
• VGETMANTBF16
• VMAXBF16
• VMINBF16
• VMINMAXBF16
• VMINMAXPD
• VMINMAXPH
• VMINMAXPS
• VMINMAXSD

All 96 AVX10.2 instructions