x86 - AshokBhat/notes GitHub Wiki
About
- Architecture features of x86-64 architecture
Vector Neural Network Instructions (VNNI)
- For accelerating convolutional neural network on x86
- Four new instructions
- First available in Intel Cascade Lake series for servers
Instruction | Description |
---|---|
VPDPBUSD | Multiply and add unsigned and signed bytes |
VPDPBUSDS | Multiply and add unsigned and signed bytes with saturation |
VPDPWSSD | Multiply and add signed word integers |
VPDPWSSDS | Multiply and add word integers with saturation |
See also
- [Intel]] ](/AshokBhat/notes/wiki/[[-AMD)
- Arm