3  * Timer events oriented CPU idle governor
11  * wakeups from idle states.  Moreover, information about what happened in the
13  * idle state with target residency within the time to the closest timer is
14  * likely to be suitable for the upcoming idle time of the CPU and, if not, then
15  * which of the shallower idle states to choose.
18  * they can be covered by taking a few most recent idle time intervals of the
20  * idle duration values greater than the time till the closest timer, as the
24  * Thus this governor estimates whether or not the upcoming idle time of the CPU
26  * idle state for it in accordance with that, as follows:
28  * - Find an idle state on the basis of the sleep length and state statistics
31  *   o Find the deepest idle state whose target residency is less than or equal
34  *   o Select it if it matched both the sleep length and the observed idle
36  *     (i.e. the observed idle duration was significantly shorter than the sleep
42  * - If the majority of the most recent idle duration values are below the
43  *   target residency of the idle state selected so far, use those values to
44  *   compute the new expected idle duration and find an idle state matching it
62  * Number of the most recent idle duration values to take into consideration for
68  * struct teo_idle_state - Idle state data used by the TEO cpuidle governor.
73  * A CPU wakeup is "matched" by a given idle state if the idle duration measured
75  * residency of the next one (or if this is the deepest available idle state, it
76  * "matches" a CPU wakeup when the measured idle duration is at least equal to
79  * Also, from the TEO governor perspective, a CPU wakeup from idle is "early" if
81  * is, early enough to match an idle state shallower than the one matching the
86  * the time till the closest timer event used for idle state selection.
96  * @time_span_ns: Time between idle state selection and post-wakeup update.
98  * @states: Idle states data corresponding to this CPU.
99  * @interval_idx: Index of the most recent saved idle interval.
100  * @intervals: Saved idle duration values.
126 		 * enough to the closest timer event expected at the idle state  in teo_update()
135 		 * (saved) time till the next timer event and the measured idle  in teo_update()
155 	 * states matching the sleep length and the measured idle duration.  in teo_update()
171 	 * length.  If it matches the measured idle duration too, this is a hit,  in teo_update()
175 	 * matches the measured idle duration.  in teo_update()
197 	 * Save idle duration values corresponding to non-timer wakeups for  in teo_update()
211  * teo_find_shallower_state - Find shallower idle state matching given duration.
214  * @state_idx: Index of the capping idle state.
215  * @duration_ns: Idle duration value to match.
235  * teo_select - Selects the next idle state to enter.
274 			 * the anticipated idle duration.  in teo_select()
280 			 * This state is disabled, so the range of idle duration  in teo_select()
348 	 * If the "hits" metric of the idle state matching the sleep length is  in teo_select()
351 	 * idle duration observed after wakeup, so take the one with the maximum  in teo_select()
372 	 * shallower idle state than the one selected so far.  in teo_select()
384 		 * Count and sum the most recent idle duration values less than  in teo_select()
385 		 * the current expected idle duration value.  in teo_select()
398 		 * Give up unless the majority of the most recent idle duration  in teo_select()
405 			 * Avoid spending too much time in an idle state that  in teo_select()
419 	 * expected idle duration is shorter than the tick period length.  in teo_select()
450 	 * nets, assume that the CPU might have been idle for the entire sleep  in teo_reflect()