The sad thing is, the finding isn't asinine. It's truly interesting (though practically irrelevant, except that it appears to have rendered the ref implementation of bcrypt incompatible with Mazieres' paper).

But instead of spending time talking about the interesting finding, this blogger clubbed their own reputation to death on it. I'd like to presume they were just spectacularly bored, but to steal some of the author's own words, that's "vastly too big a discrepancy to be explainable by a simple inadvertent bug".