Ah. So this Atom patcher is a bit pointless, then? Pity. Would have been a neat solution without having to bug those that understand kernel development better to come up with an updated kernel for each point release.
It's not pointless at all (well, this one might be), in fact I have a kernel patcher that goes and reads the correct addresses + patches everything appropriately (it's the same codebase as the KernelPatcher module for chameleon)