Re: [PATCH encrypted swsusp 1/3] core functionality

[Pavel Machek <pavel@ucw.cz>]

On Čt 14-04-05 18:08:37, Herbert Xu wrote:
> On Thu, Apr 14, 2005 at 08:51:25AM +0200, Pavel Machek wrote:
> >
> > > This solution is all wrong.
> > > 
> > > If you want security of the suspend image while "suspended", encrypt
> > > with dm-crypt. If you want security of the swap image after resume,
> > > zero out the portion of swap used. If you want both, do both.
> 
> Pavel, you're not answering our questions.
> 
> How is the proposed patch any more secure compared to swsusp over
> dmcrypt?

It is not "more secure". It solves completely different problem.

> In fact if anything it is less secure.  If I understand correctly the
> proposal is to store the key used to encrypt the swsusp image in the
> swap device.  This means that anybody who gains access to the swap
> device can trivially decrypt it.

Yes. It also means that key is gone after resume.

> Compare this to the properly setup dmcrypt case where the swap
> device can only be decrypted with a passphrase obtained from the
> user at resume time.

Solution above does not require passphrase (so users will actually use
it) and dmcrypt with passphrase does not destroy the key after resume,
so data can still be recovered.

They are orthogonal. You want both.

If something is still unclear, we can talk on irc somewhere, if you
agree to write FAQ entry afterwards ;-).
								Pavel
-- 
Boycott Kodak -- for their patent abuse against Java.