[$] Controlling realtime priorities in kernel threads

The realtime scheduler classes are intended to allow a developer to statewhich tasks have the highest priorities with the assurance that, at anygiven time, the highest-priority task will have unimpeded access to theCPU. The kernel itself carries out a number of tasks that have tight timeconstraints, so it is natural to want to assign realtime priorities tokernel threads carrying out those tasks. But, as Peter Zijlstra arguesin a new patch set, it makes little sense for the kernel to be assigningsuch priorities; to put an end to that practice, he is proposing to takeaway most of the kernel's ability to prioritize its own threads.