Hi,
It appears that your sample values are given in little-endian format (
big-endian is more usual and BN_bn2hex use it for example). I did check
the computation with various mathematical toolbox and they all gave the
same result as OpenSSL BN_mod_exp which is
66694e28fdca0bb1f34fe13e3b38088180766f8cc4283ef6a9c1e427f364d7e2 in
big-endian or
E2D764F327E4C1A9F63E28C48C6F76808108383B3EE14FF3B10BCAFD284E6966 in
little endian.as you mention it in your message.
So, it's your "expected" result that is false. How did you compute this
value?
Cheers,
--
Mounir IDRASSI
IDRIX
http://www.idrix.fr
Sargeras wrote:
I'm working in a client for authentication of SRP6 and I have a problem with
the function BN_mod_exp.
I'd appreciate some guidance
The values that I use with this function perfectly matches the ones
required.
It's written in C++ and here are the parts of the code that I think are
important.
BigNumber BigNumber::ModExp(const BigNumber &bn1, const BigNumber &bn2)
{
BigNumber ret;
BN_CTX *bnctx;
bnctx = BN_CTX_new();
BN_mod_exp(ret._bn, _bn, bn1._bn, bn2._bn, bnctx);
BN_CTX_free(bnctx);
return ret;
}
BigNumber temp;
temp = g.ModExp(x, N);
Where g, x and N are instances of the class BigNumber.
Here are my output checkings, where I print the calculated value and the
expected.
N (calculated):
B79B3E2A87823CAB8F5EBFBF8EB10108535006298B5BADBD5B53E1895E644B89
N (expected):
B79B3E2A87823CAB8F5EBFBF8EB10108535006298B5BADBD5B53E1895E644B89
g (calculated): 07
g (expected): 07
x (calculated): CBFE739D7B671EE5EF8BD304A670459ED445C6CB
x (expected): CBFE739D7B671EE5EF8BD304A670459ED445C6CB
temp (calculated):
E2D764F327E4C1A9F63E28C48C6F76808108383B3EE14FF3B10BCAFD284E6966
temp (expected):
F965C5F89FF0175FA73E2DFCDB28CEAEEFD69999BE88CC015E3356DCE95FC265
Any other piece of code will be given if asked.
Thank you all for your time.
______________________________________________________________________
OpenSSL Project http://www.openssl.org
User Support Mailing List openssl-users@openssl.org
Automated List Manager majord...@openssl.org
