> [[email protected] - Wed Jun 02 22:12:07 2010]:
> 
> I'm currently working on ECDSA, and use OpenSSL for tests. I also use 
> BouncyCastle to verify the validity of PKCS10 request.
> I have been able to generate a PKCS10 with a HSM that BouncyCastle 
> rejects but OpenSSL verifies. At first I thought it was BouncyCastle 
> that was mistaken, but after talking with David Hook from BouncyCastle, 
> it seems that there might be a problem with the INTEGER Structure from 
> ASN.1 in the ecdsa-sig-value from OpenSSL.
> In fact the PKCS10 request generated have the s integer first bit to 1, 
> which should mean that it is negative. BouncyCastle interprets it as 
> negative which is why it does not verify the validity of the request, 
> whereas OpenSSL interprets as an unsigned integer and therefore verifies 
> the signature. According to David Hook the ASN.1 Structure INTEGER is 
> signed, therefore the problem is maybe in OpenSSL.
> I have tested the file with OpenSSL 0.9.8h and OpenSSL 1.0.0, both 
> verify the enclosed file.
> 

OpenSSL's behaviour is expected: it is a bug workaround for some broken
software that encodes an INTEGER incorrectly. 

OpenSSL should never *produce* such a broken encoding but it will
tolerate it.

Steve.
-- 
Dr Stephen N. Henson. OpenSSL project core developer.
Commercial tech support now available see: http://www.openssl.org

______________________________________________________________________
OpenSSL Project                                 http://www.openssl.org
Development Mailing List                       [email protected]
Automated List Manager                           [email protected]

Reply via email to