cryptoanarchy.wiki

Arise, you have nothing to lose but your barbed wire fences!

1995-12-12 - Re: Timing Cryptanalysis Attack

Header Data

From: “Perry E. Metzger” <perry@piermont.com>
To: Matt Blaze <mab@crypto.com>
Message Hash: 5b7ee6c4bde08335de0e27a98222ed7209f5ca9def27bf70d0d506703212cb00
Message ID: <199512111944.OAA02490@jekyll.piermont.com>
Reply To: <199512111906.OAA01139@crypto.com>
UTC Datetime: 1995-12-12 21:53:55 UTC
Raw Date: Wed, 13 Dec 1995 05:53:55 +0800

Raw message

From: "Perry E. Metzger" <perry@piermont.com>
Date: Wed, 13 Dec 1995 05:53:55 +0800
To: Matt Blaze <mab@crypto.com>
Subject: Re: Timing Cryptanalysis Attack
In-Reply-To: <199512111906.OAA01139@crypto.com>
Message-ID: <199512111944.OAA02490@jekyll.piermont.com>
MIME-Version: 1.0
Content-Type: text/plain



Matt Blaze writes:
> >The trivial way to handle this is simply to check user time with the
> >right system calls and make sure it always comes out the same with an
> >apropriate number of sleeps.
> 
> Of course, this works against a remote adversary, but not against one
> on the same machine who can look at actual CPU consumption (which doesn't
> increase when the target is blocked).

True enough, but using busy loops could handle that. However, I must
admit to being far more interested in handling the remote case
efficiently, especially given concerns people have about using
Photuris like systems on heavily pounded servers.

Perry

Thread