x86-64.net

Instruction

VFMSUB213PH

Fused multiply-subtract of packed half-precision values: second operand times first minus third, into the first.

Extensions
Forms
6 (6 EVEX)
Data
Intel XED v2026.08.23

Forms

Every encoding of VFMSUB213PH that Intel XED knows, one row per XED instruction form (iform). How to read the notation.
InstructionEncodingRequires
VFMSUB213PH xmm {k}{z}, xmm, xmm VFMSUB213PH_XMMf16_MASKmskw_XMMf16_XMMf16_AVX512EVEX.128.66.MAP6.W0 AA /rAVX10.1 or AVX512_FP16 + AVX512VL
VFMSUB213PH xmm {k}{z}, xmm, m128/m16bcst VFMSUB213PH_XMMf16_MASKmskw_XMMf16_MEMf16_AVX512EVEX.128.66.MAP6.W0 AA /rAVX10.1 or AVX512_FP16 + AVX512VL
VFMSUB213PH ymm {k}{z}, ymm, ymm VFMSUB213PH_YMMf16_MASKmskw_YMMf16_YMMf16_AVX512EVEX.256.66.MAP6.W0 AA /rAVX10.1 or AVX512_FP16 + AVX512VL
VFMSUB213PH ymm {k}{z}, ymm, m256/m16bcst VFMSUB213PH_YMMf16_MASKmskw_YMMf16_MEMf16_AVX512EVEX.256.66.MAP6.W0 AA /rAVX10.1 or AVX512_FP16 + AVX512VL
VFMSUB213PH zmm {k}{z}, zmm, zmm, {er} VFMSUB213PH_ZMMf16_MASKmskw_ZMMf16_ZMMf16_AVX512EVEX.512.66.MAP6.W0 AA /rAVX10.1 or AVX512_FP16
VFMSUB213PH zmm {k}{z}, zmm, m512/m16bcst VFMSUB213PH_ZMMf16_MASKmskw_ZMMf16_MEMf16_AVX512EVEX.512.66.MAP6.W0 AA /rAVX10.1 or AVX512_FP16

Intrinsics

The C intrinsics of Clang 23.1.2 that compile to VFMSUB213PH, 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.

IntrinsicCompiles to
_mm_fmsub_ph __m128h _mm_fmsub_ph(__m128h __A, __m128h __B, __m128h __C)

VFMSUB213PH xmm, xmm, xmm EVEX.128 -mavx512fp16 -mavx512vl

_mm_mask_fmsub_ph __m128h _mm_mask_fmsub_ph(__m128h __A, __mmask8 __U, __m128h __B, __m128h __C)

VFMSUB213PH xmm {k}, xmm, m128/m16bcst EVEX.128 -mavx512fp16 -mavx512vlwhen __C comes from memory

_mm_maskz_fmsub_ph __m128h _mm_maskz_fmsub_ph(__mmask8 __U, __m128h __A, __m128h __B, __m128h __C)

VFMSUB213PH xmm {k}{z}, xmm, xmm EVEX.128 -mavx512fp16 -mavx512vl

_mm256_fmsub_ph __m256h _mm256_fmsub_ph(__m256h __A, __m256h __B, __m256h __C)

VFMSUB213PH ymm, ymm, ymm EVEX.256 -mavx512fp16 -mavx512vl

_mm256_mask_fmsub_ph __m256h _mm256_mask_fmsub_ph(__m256h __A, __mmask16 __U, __m256h __B, __m256h __C)

VFMSUB213PH ymm {k}, ymm, m256/m16bcst EVEX.256 -mavx512fp16 -mavx512vlwhen __C comes from memory

_mm256_maskz_fmsub_ph __m256h _mm256_maskz_fmsub_ph(__mmask16 __U, __m256h __A, __m256h __B, __m256h __C)

VFMSUB213PH ymm {k}{z}, ymm, ymm EVEX.256 -mavx512fp16 -mavx512vl

_mm512_fmsub_ph __m512h _mm512_fmsub_ph(__m512h __A, __m512h __B, __m512h __C)

VFMSUB213PH zmm, zmm, zmm EVEX.512 -mavx512fp16

_mm512_fmsub_round_ph __m512h _mm512_fmsub_round_ph(__m512h A, __m512h B, __m512h C, const int R)

VFMSUB213PH zmm, zmm, zmm, {er} EVEX.512 -mavx512fp16

_mm512_mask_fmsub_ph __m512h _mm512_mask_fmsub_ph(__m512h __A, __mmask32 __U, __m512h __B, __m512h __C)

VFMSUB213PH zmm {k}, zmm, m512/m16bcst EVEX.512 -mavx512fp16when __C comes from memory

_mm512_maskz_fmsub_ph __m512h _mm512_maskz_fmsub_ph(__mmask32 __U, __m512h __A, __m512h __B, __m512h __C)

VFMSUB213PH zmm {k}{z}, zmm, zmm EVEX.512 -mavx512fp16

_mm512_maskz_fmsub_round_ph __m512h _mm512_maskz_fmsub_round_ph(__mmask32 U, __m512h A, __m512h B, __m512h C, const int R)

VFMSUB213PH zmm {k}{z}, zmm, zmm, {er} EVEX.512 -mavx512fp16

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.

CPUID requirements and processors of the XED ISA sets above. Alternatives are separated by “or”; processors follow XED’s chip model, which covers AMD processors up to Zen 2.
ISA setCPUIDProcessors in XED
AVX512_FP16_128

AVX10.1
CPUID.(EAX=07H,ECX=1):EDX[19] AVX10
CPUID.(EAX=24H,ECX=0):EBX[7:0] ≥ 1 AVX10 version

or

AVX512_FP16 + AVX512VL
CPUID.(EAX=07H,ECX=0):EDX[23] AVX512_FP16
CPUID.(EAX=07H,ECX=0):EBX[31] AVX512VL

Nova Lake, Sapphire Rapids, Emerald Rapids, Granite Rapids, Diamond Rapids
AVX512_FP16_256

AVX10.1
CPUID.(EAX=07H,ECX=1):EDX[19] AVX10
CPUID.(EAX=24H,ECX=0):EBX[7:0] ≥ 1 AVX10 version

or

AVX512_FP16 + AVX512VL
CPUID.(EAX=07H,ECX=0):EDX[23] AVX512_FP16
CPUID.(EAX=07H,ECX=0):EBX[31] AVX512VL

Nova Lake, Sapphire Rapids, Emerald Rapids, Granite Rapids, Diamond Rapids
AVX512_FP16_512

AVX10.1
CPUID.(EAX=07H,ECX=1):EDX[19] AVX10
CPUID.(EAX=24H,ECX=0):EBX[7:0] ≥ 1 AVX10 version

or

AVX512_FP16
CPUID.(EAX=07H,ECX=0):EDX[23] AVX512_FP16

Nova Lake, Sapphire Rapids, Emerald Rapids, Granite Rapids, Diamond Rapids

More AVX10.1 instructions

All 686 AVX10.1 instructions

Sources

  1. Intel XED v2026.08.23 (commit 0bcb6237345c): forms, encodings, ISA sets, CPUID bits and chip model.
  2. Clang's x86 intrinsic headers, LLVM 23.1.2 (commit 85ac56026243): the intrinsics, their signatures and doc comments, and the Clang of that release, which compiled each of them.
  3. Intel Advanced Vector Extensions 10.2 Architecture Specification, revision 6.0 (361050-006, January 2026)

The tables are derived from Intel XED, Copyright Intel Corporation, licensed under the Apache License 2.0; x86-64.net converted and reformatted them. The intrinsics and their signatures come from Clang, part of the LLVM Project, licensed under the Apache License 2.0 with LLVM Exceptions. The text is our own.