Description of problem: Backport SPR new instructions AVX512_FP16 and AVX_VNNI to Anolis kernel. AVX512_FP16 is provided on Intel new platforms. It could gain better performance compared with FP32 if the precision or magnitude requirements are met. AVX version of the Vector Neural Network Instruction(AVX_VNNI) is new instruction on Alder Lake and Sapphire Rapids. The following instructions are included when the feature is available: 1. VPDPBUS: Multiply and Add Unsigned and Signed Bytes 2. VPDPBUSDS: Multiply and Add Unsigned and Signed Bytes with Saturation 3. VPDPWSSD: Multiply and Add Signed Word Integers 4. VPDPWSSDS: Multiply and Add Signed Integers with Saturation Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: AVX512_FP16's availability is indicated by CPUID.(EAX=7,ECX=0):EDX[bit 23]. It could be tested on guest by: lscpu | grep -i flags | grep avx512_fp16 cpuid --leaf=7 -r -i -1 AVX_VNNI's availability is indicated by CPUID.(EAC=7,ECX=1):EAX[bit 4]. It could be tested on guest by: lscpu | grep -i flags | grep avx_vnni cpuid --leaf=7 --subleaf=1 -r -i -1 Additional info: