3 ms·
These are really timing attacks. It's not a matter of directly reading the data, it's a matter of being able to deduce what's there based on how long it takes t
by zrm 7y ago
These are really timing attacks. It's not a matter of directly reading the data, it's a matter of being able to deduce what's there based on how long it takes to do it.
- JoeAltmaier 7y agoYes; but that would be irrelevant if they didn't depend on secrecy for security, right?
- zrm 7y agoAh, you're proposing a solution for Meltdown rather than Spectre. The difference is that with Meltdown there is speculative access to memory that isn't even mapped to the original process. Memory encryption might work as a way to solve that, but we already know how to solve that -- do what AMD and others have been doing the whole time and don't continue speculative execution through a page fault.
- JoeAltmaier 7y ago...until somebody finds another vector to reveal the value of Kernel memory. The mantra is, secrecy is not security. The User/Kernel dichotomy flagrantly ignores this principle.