Ever since Alder Lake was launched, I have been curious about what I see as a potential security problem of efficiency cores. According to AnandTech, efficiency core clusters have a shared L2 cache. I am no computer engineer/scientist, but I feel like that could introduce a problem similar to Meltdown.
I assume that between Intel’s usage of these core clusters and since I haven’t heard of an actual vulnerability reported by now, that it is not a concern. So, why is this not a problem? Is CPU cache inaccessible to software? Is an attack like this simple improbable, or what?
L3 cache on most processors is generally shared for all cores, in general the last level cache for most cores is not private and shared (there are probably exceptions but I can’t think of any modern design that isn’t)
Wikipedia agrees:
In practice, the upper-level cache L1 (or sometimes L2) is implemented as private and lower-level caches are implemented as shared. This design provides high access rates for the high-level caches and low miss rates for the lower-level caches.
An even more extreme example is this since it shares cache not just for the whole chip but for multiple chips on one system
This means that if a cache line is evicted from the virtual L3 on one chip, it will go find another chip in the system to live on, and be marked as a virtual L4 cache line.