Re: Reverse engineering of Intel branch predictors

Stefan Monnier <monnier@iro.umontreal.ca> wrote:

In case of branch predictor itself it means delay feedback by some
number of clocks, which looks like minor cost.

 
You can still make your next predictions based on "architectural state
+ pending predictions" if the pending predictions themselves only
depend ultimately on the architectural state.
 

OTOH delaying fetches from speculatively fetched addresses will
increase latency on critical path, possibly leading to
significant slowdown.

 
I think you can similarly perform eagerly the fetches from speculatively
fetched addresses but only if you can ensure that these will leave no
trace if the speculation happens to fail.

It looks extremaly hard if not impossible.

So whether and how you can do it depends the definition of "leave no
trace".  E.g. Mitch argues you can do it if you can refrain from putting
that info into the normal cache (where it would have to displace
something else, thus leaving a trace) and instead have to keep it in
what we could call a "speculative cache" but would likely be just some
sort of load buffer.

Alone that is clearly insufficient.

If "leave no trace" includes not slowing down other concurrent memory
accesses (e.g. from other CPUs), it might require some kind of
priority scheme.

First, one needs to ensure that the CPU performing speculative
fetch will not slown down due to say resource contention.  If you
put some arbitrary limit like one or two speculative fetches in
flight, that is likely to be detectable by the attacker and may
leak information.  If you want several ("arbitrarily many") speculative
fetches without slowing down normal execution, that would mean highly
overprovisioned machine.

If you're worried about, say, a Spectre-like attack measuring the
temperature or the power consumption of the CPU to try and guess what
operations were performed (or what addresses were accessed, ...)
speculatively, then you'd have to be yet more careful.

I am mostly concerned with remote attacks.  To block them it should
be enough to ensure that machine never goes into thermal throttling
(I consider adversary who is not able to directly monitor power
or temperature, so only thing remaining is thermal throttling and
its effect on execution time).

--
                              Waldek Hebisch