Comment by archangel_one

I agree that it's conceptually nice to support non-power of two word sizes, but how often do those come up these days in practice? If none of the architectures they're targeting actually have such a word size, it seems like it might be a useful simplifying assumption.