At Wed, 10 Jan 2007 23:36:54 +0100,
Pierre THIERRY wrote:
> Scribit Neal H. Walfield dies 10/01/2007 hora 20:20:
> > The application requests the schedule which the resource manager
> > considers in light of the current policy configuration.  When it is
> > not longer possible to meet the requested schedule, the resource
> > manager will send the application a fault when it is next scheduled to
> > run.
> 
> In the general case, isn't this an opportunity to create covert
> channels? I had understood from previous discussions (here or on the
> Coyotos list) that application should in general not be aware of
> system-wide information such as memory pressure because of this issue.
> 
> Of course, I understand there are exceptions. I'm happy that my video
> player is able to detect how much frames it drops because of scheduling
> difficulties, and suggest me to solve the external problem while it
> stops to play.

Sure.  If we get to point where the least expensive attack (and that
is, I think, what attackers are looking for) is via covert channels,
I'll feel that we'll have really accomplished something.

Neal


_______________________________________________
L4-hurd mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/l4-hurd

Reply via email to