Add x4 batched keccak implementation using MVE
This MR includes 6 files:
- A symbolic assembly implementation of x4 batched keccak
- A C harness for calling into the x4 batched keccak including a demo that runs the batched keccak with a single state, and interleave and deinterleave functions for converting between 4 individual states and the interleaved state used by the assembly code.
- A set of load offsets and round constants used by the assembly
- The output of SLOTHY register assignment applied to the symbolic assembly
- The output of SLOTHY optimisation over the register-assigned assembly
- A readme describing the code
Edited by Brendan Moran