The processor issue, who knew

‘It Can’t Be True.’ Inside the Semiconductor Industry’s Meltdown

It was becoming an issue, and not widely shared.  Some big Linux companies had already been submitting proposals to secure the kernel before the big news hit.

I didn't suspect the pipeline issue, but I knew all along that the kernel has to keep some secure memory that is not memory mapped, only available via kernel calls, or better an internal processor instruction.  My conclusion was simply based on the idea of warping a 'secure' network back such that all security was at the endpoints of a network.  Essentially, one ends up needing secret keys , at the leaves, that are counterfeit proof.

The lesson here is, we don't care.  We only need to control the leaks so the processor 'leak' becomes a run time processor 'optional setting' on new processors.