Answer by vnp for Arbitrary large unsigned integers

Multiplication cries for optimization (e.g. Karatzuba or Toom-Cook)

Division also can be optimized; given a fast multiplication a most straightforward approach is Newton-Raphson.

The semantics of hugeint_sub and hugeint_decrement is questionable, or, if I may say so, surprising.

hugeint_ladd_cutoverflow assumes that the carry fits in an unsigned int. It is not necessarily a case.

I don't think it is a good idea to interpret a position being NULL as an implicit shift by 1.