On Fri, Jan 28, 2011 at 04:05:38PM +0000, Peter Gutmann wrote:

[Disclaimer: I work for Simtec and worked on the Entropy Key]

> Does anyone know more about the hardware they're using to communicate via USB?
> If it's something like a standard PL2303 it shouldn't be too hard to talk to
> on systems other than the few Linux ones they support

The USB device in question is the microcontroller which also does all of the
crypto and entropy analysis on the streams.  This is so that it isn't possible
to interface to the data stream before it is encrypted and to ensure that you
cannot substitute another data stream in the device easily.

> the only problem will be the complex protocol they use, which uses MAC'd
> packets implementing a custom comms protocol instead of just opening a serial
> port and reading the output, definitely not a good design decision because it
> limits use to systems running their (from the webpage) not very stable
> drivers.

Note that, as the webpage explains, the device is a USB CDC device providing an
asychronous serial port.  We chose this particular method of delivering the
data stream because it was well supported in most UNIX systems, along with Mac
OS X and Windows, without the need to write kernel drivers which would cause
more concern among the target market. 

The issue with driver stability was limited to a particular set of Linux
kernels, which two years ago were prevalent, but are now becoming less common.
Indeed the driver issue was purely related to the Linux TTY layer which has had
numerous issues in the past.

The custom protocol which is implemented on top of the serial stream is
entirely there to ensure various security measures.  It is not possible
(without the shared keys) to pretend to be a valid device and provide known
data to the ekeyd for inclusion into the kernel.  It is not possible to record
the USB data stream and then (without the shared keys) work out what data was
actually passed to the kernel.  Also, the MACs allow us to be certain no
packets got tampered with or corrupted en-route.

Regards,

Daniel.

-- 
Daniel Silverstone                         http://www.digital-scurf.org/
PGP mail accepted and encouraged.            Key Id: 3CCE BABE 206C 3B69
_______________________________________________
cryptography mailing list
[email protected]
http://lists.randombit.net/mailman/listinfo/cryptography

Reply via email to