I added some details to help flesh out the #Spectre and #Meltdown attacks. CPUs are complicated.
https://www.kb.cert.org/vuls/id/584653
My understanding is this: eBPF JIT lives in kernel land. So the user/kernel boundary is not crossed by the Spectre attack itself. But rather they leveraged Spectre using code that's already living in kernel space.
-
-
ok, I see your view "Spectre intrinsically doesn't allow user/kernel boundary bypass, but P0 attack technique allowed the bypass". Since worst case scenario here is "bypass" as shown by P0, maybe VU should say that depending on attack, "bypass" is possible?
-
I considered it, but I'd like to avoid conflating the issues at hand. Once you start considering the aspects of vulnerability chaining, things like impact get murky very quickly. The eBPF JIT engine provides a mechanism for userspace to be able to execute some code in the kernel
- 4 more replies
New conversation -
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.