Visible to Intel only — GUID: GUID-2A249606-4D3C-4A00-B7EA-6E6F3EA0951B
Visible to Intel only — GUID: GUID-2A249606-4D3C-4A00-B7EA-6E6F3EA0951B
enum dnnl::cpu_isa
Overview
CPU instruction set flags. More…
#include <dnnl.hpp>
enum cpu_isa
{
isa_default = dnnl_cpu_isa_default,
sse41 = dnnl_cpu_isa_sse41,
avx = dnnl_cpu_isa_avx,
avx2 = dnnl_cpu_isa_avx2,
avx2_vnni = dnnl_cpu_isa_avx2_vnni,
avx2_vnni_2 = dnnl_cpu_isa_avx2_vnni_2,
avx512_core = dnnl_cpu_isa_avx512_core,
avx512_core_vnni = dnnl_cpu_isa_avx512_core_vnni,
avx512_core_bf16 = dnnl_cpu_isa_avx512_core_bf16,
avx512_core_fp16 = dnnl_cpu_isa_avx512_core_fp16,
avx512_core_amx = dnnl_cpu_isa_avx512_core_amx,
avx512_core_amx_fp16 = dnnl_cpu_isa_avx512_core_amx_fp16,
};
Detailed Documentation
CPU instruction set flags.
Enum Values
isa_default
Library choice of ISA (excepting those listed as initial support)
sse41
Intel Streaming SIMD Extensions 4.1 (Intel SSE4.1)
avx
Intel Advanced Vector Extensions (Intel AVX)
avx2
Intel Advanced Vector Extensions 2 (Intel AVX2)
avx2_vnni
Intel AVX2 and Intel Deep Learning Boost (Intel DL Boost) support.
avx2_vnni_2
Intel AVX2 and Intel Deep Learning Boost (Intel DL Boost) with 8-bit integer, float16 and bfloat16 support.
avx512_core
Intel AVX-512 subset for Intel Xeon Scalable processor family and Intel Core processor family.
avx512_core_vnni
Intel AVX-512 and Intel Deep Learning Boost (Intel DL Boost) support for Intel Xeon Scalable processor family and Intel Core processor family.
avx512_core_bf16
Intel AVX-512, Intel DL Boost and bfloat16 support for Intel Xeon Scalable processor family and Intel Core processor family.
avx512_core_fp16
Intel AVX-512 with float16, Intel DL Boost and bfloat16 support for Intel Xeon Scalable processor family and Intel Core processor family.
avx512_core_amx
Intel AVX-512 with float16, Intel DL Boost and bfloat16 support and Intel AMX with 8-bit integer and bfloat16 support.
avx512_core_amx_fp16
Intel AVX-512 with float16, Intel DL Boost and bfloat16 support and Intel AMX with 8-bit integer, bfloat16 and float16 support.