[147900] in cryptography@c2.net mail archive
Re: [Cryptography] Randomness from network hardware?
daemon@ATHENA.MIT.EDU (Bill Stewart)
Wed Oct 30 14:54:05 2013
X-Original-To: cryptography@metzdowd.com
Date: Tue, 29 Oct 2013 15:09:06 -0700
To: Jerry Leichter <leichter@lrw.com>
From: Bill Stewart <bill.stewart@pobox.com>
In-Reply-To: <B3AABCE3-5541-4AE6-9A2B-E31C1FE600DD@lrw.com>
Cc: Cryptography List <cryptography@metzdowd.com>
Errors-To: cryptography-bounces+crypto.discuss=bloom-picayune.mit.edu@metzdowd.com
At 01:50 PM 10/28/2013, Jerry Leichter wrote:
>Ethernet is supposed to back off randomly after a collision.
...
>Do current interfaces, perhaps in some test mode
>(which a special driver could get at during boot),
>provide access to anything that could be used as (part of) a random seed?
I'd be surprised if most hardware or drivers makes that visible,
especially since "ethernet" is full-duplex with switches
rather than half-duplex with CSMA/CD these days.
It's not something you'd normally want the CPU to know.
Maybe some wifi gear makes it more visible,
but if you had deeply debuggable wifi there'd be a lot more sources
of random noise available.
_______________________________________________
The cryptography mailing list
cryptography@metzdowd.com
http://www.metzdowd.com/mailman/listinfo/cryptography
