Re: LibTomMath vs MIRACL
From: Phil Carmody (thefatphil_demunge_at_yahoo.co.uk)
Date: 07/30/03
- Next message: Arthur J. O'Dwyer: "Re: Flirting with cryptography problem"
- Previous message: Phil Carmody: "Re: random data"
- In reply to: Peter Gutmann: "Re: LibTomMath vs MIRACL"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Wed, 30 Jul 2003 23:09:34 +0300
On Wed, 30 Jul 2003 14:54:15 +0000, Peter Gutmann wrote:
> Paul Rubin <http://phr.cx@NOSPAM.invalid> writes:
>
>>You might also take a look at GMP, which is under the Gnu LGPL, and
>>(like MIRACL) includes fast assembly code. There is also the bn
>>library which is part of OpenSSL. I believe bn is generally pretty
>>fast, but not quite as fast as GMP or MIRACL.
>
> The advantage of the OpenSSL bignum code is that it's targeted specifically at
> doing crypto, while GMP is a general-purpose numeric package. If your goal is
> just to implement the usual suspects { RSA, DH, DSA/Elgamal }, you need to do
> quite a bit more work with GMP than with OpenSSL bignum.
The Lenstra ('maintained' by Paul Leyland now) LIP package has a few
things that are aimed at PK crypto apps too. e.g. that thing where you do
more than one expmod at a time (I forget what it's called, it's in HAC).
Phil
- Next message: Arthur J. O'Dwyer: "Re: Flirting with cryptography problem"
- Previous message: Phil Carmody: "Re: random data"
- In reply to: Peter Gutmann: "Re: LibTomMath vs MIRACL"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|
|
