News for hackers. Stuff that matters.
Re:Hertzbleed Attack by stingraycharles on Tuesday June 14, 2022 @02:19PM (#31743637)
This is probably a naive question, but could this be mitigated by fencing a part of code by some “frequency fence” of some sorts? This is of course a long-term mitigation as it may require compiler support, may affect performance and other threads and whatnot, but I wonder what a proper solution would look like.
Re:Hertzbleed Attack by mlor on Tuesday June 14, 2022 @06:19PM (#31746741)
I'm curious whether this type of thing would work as well. It sounds like you're suggesting to be able to wrap sections of code in compiler-specific declarations (e.g., UNSAFE blocks in C#) that force the underlying hardware to operate at a constant frequency.

I have PRECISELY no idea whether that is coherent or makes sense. It's just interesting at a glance.

Re:Hertzbleed Attack by tumetab1 on Wednesday June 15, 2022 @03:54AM (#31750254)
Does not seem practical to me.

A server accepting TLS connections would have almost always some thread doing “frequency fenced” code so the CPU would always be frequency locked. It's much more practical to just disable Turbo-boost.

A more simpler mitigation is just add noise, perform random computations so the whole algorithm is constant time + random time. Greatly increases the difficulty of gathering timing data.