I am using Bob Jenkin's one_at_a_time hash for that purpose. It is really simple and works with unaligned data. But it is not doing well in benchmarks. I wond

I am using Bob Jenkin's one_at_a_time hash for that purpose. It is really simple and works with unaligned data. But it is not doing well in benchmarks. I wonder if I should use another one