/*
* This might appear controversial, but the fact is that generic
* prime method was observed to deliver better performance even
/*
* This might appear controversial, but the fact is that generic
* prime method was observed to deliver better performance even
* arguable, because the dependency of improvement coefficient
* from key length is not a "monotone" curve. For example while
* 571-bit result is 23% on ARM, 384-bit one is -1%. But it's
* arguable, because the dependency of improvement coefficient
* from key length is not a "monotone" curve. For example while
* 571-bit result is 23% on ARM, 384-bit one is -1%. But it's
* "tolerably" slower... What effectively happens is that loop
* with bn_mul_add_words is put against bn_mul_mont, and the
* latter "wins" on short vectors. Correct solution should be
* "tolerably" slower... What effectively happens is that loop
* with bn_mul_add_words is put against bn_mul_mont, and the
* latter "wins" on short vectors. Correct solution should be