VBROADCASTF32X2

VBROADCASTF32X2 takes two singles, the low 64 bits of an XMM register or of memory, and repeats the pair in every 64-bit slot of a YMM or ZMM register, under a mask of singles. It belongs to AVX512DQ (plus AVX512VL for YMM) and has no XMM destination. VBROADCASTI32X2 is the integer form, which does have one. Unmasked, the result is bit for bit what VBROADCASTSD gives from the same 64 bits; the difference is the mask, which can split a pair. VBROADCASTF32X4 repeats four singles.

Recorded on Zen 5#

• A split pair. vbroadcastf32x2 zmm14{k3}, qword ptr [mem], with AAEBh in the low 16 bits of K3, wrote the second single of memory in every odd element and the first only in elements 0 and 6, the even ones K3 selects; elements 2, 4, 8 and others kept their old values.
• From a register, zero-masked. vbroadcastf32x2 ymm14{k7}{z}, xmm8 with E3h in K7 wrote elements 0, 1, 5, 6 and 7 from the low pair of XMM8, odd elements from the second single, zeroed elements 2 to 4 and cleared ZMM14 above bit 255.
• Unmasked YMM. The EVEX form wrote four copies of the pair and cleared bits 511:256.

In compiled code#

At -O2 with -march=x86-64-v4, GCC 14 compiles _mm512_broadcast_f32x2 to VBROADCASTF32X2, from memory or from a register, while Clang 23 emits VBROADCASTSD for both, the same bits. With a write mask, _mm256_mask_broadcast_f32x2, both use VBROADCASTF32X2. A loop over a repeating pair, h[i] = f[i] * p2[i & 1], stays scalar in both.

AMD’s Zen 5 latency spreadsheet gives one macro-op on pipe FP1 or FP2, two per cycle, for the YMM and ZMM forms.

Forms

Latency and throughput

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 // 32):
    if k1[i]:
        zmm1.d[i] = src2.d[i % 2]
    elif zeroing:
        zmm1.d[i] = 0
    # else: zmm1.d[i] keeps its value
# 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-E6

Attributes

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

All forms

write masking

Memory forms

memory fault suppression

Intrinsics

The C intrinsics of Clang 23.1.2 that compile to VBROADCASTF32X2, 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 AVX10.1 instructions

• KXORQ
• KXORW
• VADDPD
• VADDPH
• VADDPS
• VADDSD
• VADDSH
• VADDSS
• VAESDEC
• VAESDECLAST
• VAESENC
• VAESENCLAST
• VALIGND
• VALIGNQ
• VANDNPD
• VANDNPS
• VANDPD
• VANDPS
• VBLENDMPD
• VBLENDMPS
• VBROADCASTF32X4
• VBROADCASTF32X8
• VBROADCASTF64X2
• VBROADCASTF64X4
• VBROADCASTI32X2
• VBROADCASTI32X4
• VBROADCASTI32X8
• VBROADCASTI64X2
• VBROADCASTI64X4
• VBROADCASTSD
• VBROADCASTSS
• VCMPPD
• VCMPPH
• VCMPPS
• VCMPSD
• VCMPSH
• VCMPSS
• VCOMISD
• VCOMISH
• VCOMISS

All 686 AVX10.1 instructions