Re: Can a computer virus kill the CPU?

Todd describes a condition that causes a crash; not hardware damage.
He also describes a manufacturer's design defect - not a conditon made
possible in the field. Motherboards, et al have been made, upgraded,
and enhanced from previous experiences so many times. Design is mostly
provided to motherboard manufacturers from Intel, AMD, and other high
tech companies. Design is repeated so often as to be well established.
What sells as cheap would cost massively if done from scratch. No,
those cheap motherboards have massive design and experience within
them. For example, today's Intel CPUs cannot overheat destructively
because of designs well proven in the 1980s.

Even circuits inside ICs are designed so that unintended logic states
cannot cause damage. But then also logic inside a microsequencer is
not accessible to a virus. Twice why virus cannot cause such damage.
In the rare case that a virus causes hardware damage, that manufacturer
would be scrambling to eliminate that problem since virus created
damage would mean we stop buying anything from that manufacturer; his
products are now considered all suspect. Any hardware damaged by a
virus - that manufacturer then has how many other defects?
Manufacturer would be flirting with bankruptcy - hardware damaged by
software is that unacceptable. Just another reason why viruses cannot
cause hardware damage.

As Todd says, it is an interesting question. But then designers
appreciate things that users would never even know. Let's take a
simple example. Take all outputs from a power supply. Short them all
together. Connect the +5 to the +3.3 to the +12 to the -12, etc. Now
turn on power supply? Is anything damaged? No - not if the power
supply contains functions that were even required 30+ years ago. Just
another legacy of so much experience that is now standard in today's
designs. Take some CMOS logic gates. Ties all those outputs together.
But the logic gate outputs at various and contradictory logic states.
Are any logic gates damage? Of course not. Again, a legacy of design
so many generations ago. It may sell cheap. But the massive knowledge
implemented in the design makes hardware damage unlikely and totally
unacceptable. The question begs you to grasp how expansive a cheap
design really is.

Todd H. wrote:
It'd be awfully hard to prove that such a thing would be impossible.

I'd instead pose the question of "has such a thing been reported on
any known platform/processor/OS combination." I know I haven't heard
of anything. As someone who's been on chip design teams and done
microsequencer design, I will say that it would be quite conceivable
for a microprocessor to have a design flaw whereby software could put
the microsequencer into a state that would lead to hardware failure.
And as you suggest, a flaw that allowed access to the BIOS could be
catastrophic that way, but with the plurality of BIOS implementations
out there, it'd be exceedingly difficult to create an attack that'd
take down a large number of targets.
...

Certainly possible. Likely? I have my doubts, and I'm not going to
worry about it a whole lot.
...

No. And even if it "should" we won't see it happen because the
enthusiast market has too many $'s in it such that mobo manufacturers
won't go back to the days of jumpered clock configs and such.

It's an interesting question though.

.