News for hackers. Stuff that matters.

A comment on Hertzbleed Attack. Parent comment

Re:Hertzbleed Attack by nickelpro on Tuesday June 14, 2022 @05:48PM (#31746485)
Fully concur, although now that I've read some of the white paper some of this doesn't even appear to be a real issue? Like the claimed "remote" attacks against, "Cloudflare’s Interoperable Reusable Cryptographic Library (CIRCL) [28], written in Go, and Microsoft’s PQCrypto-SIDH [65], written in C ... [are] meant to run in constant time"

But they just straight up don't run in constant time, so they're vulnerable to a timing attack across the network. That's clearly just a library bug? Like surely the dumbest part of a "constant time" algorithm is double checking that you ran for a constant wall clock amount of time?

[ Parent | on HN ]
Re:Hertzbleed Attack by Syonyk on Tuesday June 14, 2022 @06:27PM (#31746822)
> But they just straight up don't run in constant time, so they're vulnerable to a timing attack across the network. That's clearly just a library bug? Like surely the dumbest part of a "constant time" algorithm is double checking that you ran for a constant wall clock amount of time?

It's... hard. A lot of the "constant cache behavior" and "constant time behavior" algorithms were written back in the day when the CPU speeds didn't change randomly on you, or at worst toggled between "idle" and "running hard." Think... oh, even the Core 2 days, really. They didn't switch that fast.

And then the hardware behavior changed out from under the algorithms, and nobody noticed. Now the throttling is far more rapid. So they may still be "constant instruction count," but that no longer implies constant time.

It's... complicated. :( And what's worse, even the people in charge of managing the complexity don't understand all the details anymore. When stuff like this surprises Intel, we've got problems.

[ Parent | on HN ]
Re:Hertzbleed Attack by nickelpro on Tuesday June 14, 2022 @07:57PM (#31747540)
Sure, but you can just check the high precision wall-clock timer at the end of your computation and makes sure you took at least X nanoseconds, and pad that out so that X is always greater than the amount of wall-clock nanoseconds the actual computation takes. Then, following a computation, you sleep until X.

While this won't fool timing attacks that are operating on the same machine as your process, the computation time becomes completely opaque to the network which is what the "remote" attacks are built on.

[ Parent | on HN ]