On Feb 17, 12:18 pm, Geoff Beier <[email protected]> wrote:
> On Thu, Feb 17, 2011 at 12:10, Jeffrey Walton <[email protected]> wrote:
>
> > I'm not sure that it incorrect - maybe just non-standard??? Its
> > probably easiest to call the RSA decrypt function on the cipher text
> > (i.e., the hash) yourself.http://www.cryptopp.com/wiki/Raw_rsa
>
> Be *very* careful if you do this. It's an approach that invites
> serious, security-relevant errors. It sounds easy when you describe
> it, but I've seen extremely smart developers get this wrong in ways
> that led to acceptance of forged signatures.
Agreed. I don't recall seeing that advisory.

When the collisions were engineered using the extra data, what digest
was in play? MD2 or MD5? I would be surprised if someone could do it
with SHA-1, and bet against it with near certainty when using SHA-2 or
Whirlpool.

> MFSA 2006-60[1] is a good
> public example of a prominent case.
>
> It also marries your code pretty tightly to RSA, which may or may not
> be an issue for you.
>
> Geoff
>
> [1]http://www.mozilla.org/security/announce/2006/mfsa2006-60.html

-- 
You received this message because you are subscribed to the "Crypto++ Users" 
Google Group.
To unsubscribe, send an email to [email protected].
More information about Crypto++ and this group is available at 
http://www.cryptopp.com.

Reply via email to