On my PC the low-level version was 20-30% faster than the *C* version (probably thank to the malloc managed by the GC), but on his PC things are very different:

http://shootout.alioth.debian.org/gp4sandbox/benchmark.php?test=knucleotide&lang=all

While the simple_hash is here: https://alioth.debian.org/tracker/download.php/30402/411646/310541/2355/simple_hash.d

Bye,
bearophile