Piotr Luc of Intel submitted a patch to the Linux kernel, adding ‘deep learning’ instrutions for future Xeon processors.
[PATCH] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features.
AVX512_4VNNIW – Vector instructions for deep learning enhanced word variable precision.
AVX512_4FMAPS – Vector instructions for deep learning floating-point single precision.
The new instructions are to be used in future Intel Xeon & Xeon Phi processors.
The spec can be found in Intel Software Developer Manual or in Instruction Set Extensions Programming Reference.