FNV_PRIME and efficiency

[OhGodAPet]

I may be wrong, but I believe this code may result in an inferred multiplier, which in turn may be mapped to a DSP48E2 slice? I was not able to build it and check: https://github.com/Xilinx/blockchainacceleration/blob/main/hw/include/ethashV8return.hpp#L33

If this is indeed the case, that... would be unfortunate, seeing as it can be done in 101 LUTs with 8 CARRY8s, (I think - I would have to test again.)