[openssl-users] Stand alone AES-CTR module

Jeffrey Walton noloader at gmail.com
Mon May 11 17:24:24 UTC 2015


On Sun, May 10, 2015 at 5:58 PM, Mike Mohr <akihana at gmail.com> wrote:
> The task of implementing AES should not be undertaken by a novice
> programmer.  Please save the world another heartbleed and pick something
> more in line with your skill level.
Maybe I am not parsing it correctly.... It sounds like he wants to use
something like AES/CTR from an existing library like OpenSSL; and not
implement AES from the ground up.

Also, OpenSSL is susceptible to side channel attacks due to its table
driven implementation. So its not clear to me how he could do much
worse since a break is a break. You are either pregnant or you are
not. There's no degree's of pregnancy, like being half pregnant.

(There's some hand waiving above since it depends on the threat model.
But you usually don't exclude threats because they are inconvenient or
don't meet your efficiency goals. That's similar to what Browsers do
in their threat models - they exclude inconvenient stuff or stuff that
conflicts with their usability goals. That's one of the reasons
browser security is such a joke).

Jeff


More information about the openssl-users mailing list