[question/feature request] Process multiple bits per step

[Nic30]

Hello,

I am working on hardware implementation of Toeplitz hash. One of of requirements is to process multiple bits from the input per clock cycle.

For this it is required to unwind a loops in main algorithm and construct matrix which will contain information about which bit of input should be xor-ed with which input or key bit.

I did similar thing for CRC hash recently. Even If I had working partial solutions it took lot of time to put it all things together.

Do you know about any implementation of Toeplitz hash which processes multiple bits of input per step/HW implementation.
Can you write any?