Alright look with all the side channel discourse happening again I just need to say it: modern silicon design is fucking hard, as is security in general. At the risk of sounding like an Intel shill: chill the hell out. There is a lot more to this than a lot of you realize.
-
Show this thread
-
Replying to @hedgeberg
Have you seen "The Thirty Million Line Problem" by
@cmuratori? It's partly about "overengineering" but also abt ways to make hardware (designs) simpler, which'd lead to better security. I'm leaving out a lot, so if you've got a free hour (+40min Q&A) here:https://www.youtube.com/watch?v=kZRE7HIO3vk&t=0s …1 reply 0 retweets 1 like -
Replying to @JaceCear @hedgeberg
Your posts just reminded me of it, so maybe it'll provide some new viewpoints for you?
1 reply 0 retweets 0 likes -
Replying to @JaceCear
i mean, I used to do silicon-level circuit design and layout, and architecture design. both at separate points, mind you, but i know the problem pretty well. the issue is that simple hardware is slow hardware, and asking for a complete architecture redesign is no small thing.
3 replies 0 retweets 0 likes -
Replying to @hedgeberg @JaceCear
Like, it's hard to overstate the benefit of modern architecture changes. pipeline improvements, specex, etc, have made hardware /drastically/ faster by reducing fundamental bottlenecks. It's a tall order to ask people to go back to that level of simplicity.
2 replies 0 retweets 1 like -
Replying to @hedgeberg @JaceCear
That being said, my "ideal" solution to architecture problems in general isn't to fix architecture problems, its to have a physically separate introspection processor. One which is cheap, simple, open source, and replacable, which runs a watchdog firmware that monitors the CPU.
1 reply 0 retweets 2 likes -
Replying to @hedgeberg @JaceCear
It's the easiest way to fix these flaws: have a fundamentally separate SoC with a small attack surface in a purely management role, which is responsible for confirming main CPU's behavior is in line with expectations. Like ME, but external and auditable.
1 reply 0 retweets 3 likes -
Replying to @hedgeberg @JaceCear
How would this work? If it has to monitor everything the main CPU does, it has to be as powerful as the main CPU. If it doesn't, then you could make the main CPU do evil things that it wouldn't catch.
2 replies 0 retweets 4 likes -
It wouldn’t be to different from monitoring software currently used. The goal is detection of malicious activity, so if you can do classification and identification fast enough you don’t need to replicate the CPU.
2 replies 0 retweets 0 likes -
I mean, we all know antivirus doesn't work. Reactive technologies are great and all, but how would you fundamentally distinguish malicious behavior from normal operation besides known signatures?
2 replies 0 retweets 1 like
I mean, if this idea worked, and leaving aside CPU side channels for a second, we'd all be running our OSes inside VMs with a hypervisor looking for compromise already. It would have similar visibility into what the guest is doing.
-
-
Replying to @marcan42 @siriusfox and
Antivirus does not work because programmers don't code for it. If control loops were inherently part of the execution/data plane, developers would just provide the required parametrization. See app sandboxes and web API authentication: alive and kicking.
0 replies 0 retweets 0 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.