Using Chip Malfunction To Leak Private Keys

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
Overheating or undervolting a processor used to hack a Linux box? :eek:

"We basically made the hardware temporarily fail and through that the system gives incorrect signed messages from which we can then extract a private key," says Valeria Bertacco, associate professor of electrical engineering and computer science at University of Michigan, summing up the work she's presenting on at Black Hat.
 
who thinks up this stuff?! Oh, let's cause a fault in a chip to extract part of a security key. Geesh. There are definitely brighter people in the room...
 
Their network has been hacked so many times I'm surprised one can't walk into any WalMart and buy an off-the-shelf box full of user IDs and passwords.
 
If someone has physical access to the computer, then no amount of security is going to prevent them from hijacking it.

That is why data centers for financial institutions have armed guards, multiple layers of security including biometric, and strict surveillance.
 
If someone has physical access to the computer, then no amount of security is going to prevent them from hijacking it.

That is why data centers for financial institutions have armed guards, multiple layers of security including biometric, and strict surveillance.

Physical Access and People are the 2-hardest things to secure. People of course are generally the easiest to break.
 
In general a HW failure could do almost anything and you might not like the results. Airplanes have mostly solved this problem a while ago.
 
This is nothing new... people used a similar technique to dump/reprogram Hu cards on DirecTV many years back. Basically running them in a lower voltage state allowed access to the encrypted keys...
 
This is nothing new... people used a similar technique to dump/reprogram Hu cards on DirecTV many years back. Basically running them in a lower voltage state allowed access to the encrypted keys...

When I was teaching Electronics Maintenance in the Navy, I tried to get across to the guys that it's not always a LL0 or a LL1 that triggers a circuit. In many cases, Digital Electronics is about transitions, not necessarily Logic Levels.

We had NAND gates all over the place that were sensitive to the inputs transitions, and the transitions hitting at the correct time to produce the desired output.

It was not beyond me (bwahahaha) to tweak the voltage of a power supply down just a tad. Just enough so the transition of the inputs of these gates would "miss" each other and the gate would not fire.

They'd hook up the scope, see the LL0 or LL1 or whatever and say, "Well, the inputs are good, the output is bad, so the card with that gate on it is bad". I'd give them a replacement and of course it would not fix it.

I'd let them flounder for a few minutes then remind them of what I taught them earlier in the lesson for that cabinet :) I love it when you see the lightbulbs in a students eyes come on. A Volt meter, a tweak and 5 minutes later and they were finally finished :)
 
Back
Top