[Haskell-cafe] ANN: crypto-pubkey: all your public key crypto algorithms belong to us.

[Ertugrul Söylemez]

Vincent Hanquez <tab at snarc.org> wrote:

> Yes, the performance are terrible in term of integers. As the library
> is specific to public key algorithm, i just can't reasonable work on
> 64 bits integer :-), and multiprecision integers is the only way to
> go.
>
> I'm on-and-off working on some mutable mpi library to be able to
> define pure function that do the necessary stuff (i.e. expmod, mulmod,
> etc..)
>
> I'm hoping this could be reasonably competitive with a C mpi library,
> but time will tell.

It's a waste of time.  In my benchmarks Haskell Integer outperformed
equivalent (sane) C implementations using GMP's mpz_* interface.  You
would be reinventing GMP's mpn_* interface and a custom memory manager
to be able to match the speed of Integer.

The things that were slower than equivalent C code were not related to
Integer, but mostly to data structures like Set in my case, which was
the motivation for me to write the 'quickset' library.


Greets,
Ertugrul

-- 
Not to be or to be and (not to be or to be and (not to be or to be and
(not to be or to be and ... that is the list monad.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: not available
URL: <http://www.haskell.org/pipermail/haskell-cafe/attachments/20130115/428af449/attachment.pgp>