Re: FFI Bindings to Libraries using GMP

Peter Tanski wrote:

The one problem you will find with _all_ potential replacement libraries is incompatible behaviour for bitwise functions: they are implemented arithmetically in GMP but logically elsewhere (when they are implemented at all).

I don't fully understand this... I made sure my Haskell implementation's "Bits" was consistent with the existing (GMP) version, which you call "arithmetically"... it seemed that any other semantics would reveal (depend on) implementation details. and I was able to make it reasonably asymptotically-efficient anyway. Isaac