Hi Daniel,
On 02/11/2014 08:41 PM, Daniel Lezcano wrote:
The cpuidle_idle_call does nothing more than calling the three individuals
function and is no longer used by any arch specific code but only in the
cpuidle framework code.
We can move this function into the idle task code to ensure better
proximity to the scheduler code.
So my understanding of this patchset is that by moving
cpuidle_idle_call() under kernel/sched, we now have a way of calling
into the cpuidle governor and the cpuidle driver with additional
parameters like cpu_load(), idle_stamp etc.. so that we can expect the
governor and driver to take better decisions about entry and exit into
idle states. Is this the advantage we hope to begin with?