xref: /linux/drivers/cpuidle/governors/teo.c (revision 24168c5e6dfbdd5b414f048f47f75d64533296ca)
1 // SPDX-License-Identifier: GPL-2.0
2 /*
3  * Timer events oriented CPU idle governor
4  *
5  * TEO governor:
6  * Copyright (C) 2018 - 2021 Intel Corporation
7  * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
8  *
9  * Util-awareness mechanism:
10  * Copyright (C) 2022 Arm Ltd.
11  * Author: Kajetan Puchalski <kajetan.puchalski@arm.com>
12  */
13 
14 /**
15  * DOC: teo-description
16  *
17  * The idea of this governor is based on the observation that on many systems
18  * timer events are two or more orders of magnitude more frequent than any
19  * other interrupts, so they are likely to be the most significant cause of CPU
20  * wakeups from idle states.  Moreover, information about what happened in the
21  * (relatively recent) past can be used to estimate whether or not the deepest
22  * idle state with target residency within the (known) time till the closest
23  * timer event, referred to as the sleep length, is likely to be suitable for
24  * the upcoming CPU idle period and, if not, then which of the shallower idle
25  * states to choose instead of it.
26  *
27  * Of course, non-timer wakeup sources are more important in some use cases
28  * which can be covered by taking a few most recent idle time intervals of the
29  * CPU into account.  However, even in that context it is not necessary to
30  * consider idle duration values greater than the sleep length, because the
31  * closest timer will ultimately wake up the CPU anyway unless it is woken up
32  * earlier.
33  *
34  * Thus this governor estimates whether or not the prospective idle duration of
35  * a CPU is likely to be significantly shorter than the sleep length and selects
36  * an idle state for it accordingly.
37  *
38  * The computations carried out by this governor are based on using bins whose
39  * boundaries are aligned with the target residency parameter values of the CPU
40  * idle states provided by the %CPUIdle driver in the ascending order.  That is,
41  * the first bin spans from 0 up to, but not including, the target residency of
42  * the second idle state (idle state 1), the second bin spans from the target
43  * residency of idle state 1 up to, but not including, the target residency of
44  * idle state 2, the third bin spans from the target residency of idle state 2
45  * up to, but not including, the target residency of idle state 3 and so on.
46  * The last bin spans from the target residency of the deepest idle state
47  * supplied by the driver to infinity.
48  *
49  * Two metrics called "hits" and "intercepts" are associated with each bin.
50  * They are updated every time before selecting an idle state for the given CPU
51  * in accordance with what happened last time.
52  *
53  * The "hits" metric reflects the relative frequency of situations in which the
54  * sleep length and the idle duration measured after CPU wakeup fall into the
55  * same bin (that is, the CPU appears to wake up "on time" relative to the sleep
56  * length).  In turn, the "intercepts" metric reflects the relative frequency of
57  * situations in which the measured idle duration is so much shorter than the
58  * sleep length that the bin it falls into corresponds to an idle state
59  * shallower than the one whose bin is fallen into by the sleep length (these
60  * situations are referred to as "intercepts" below).
61  *
62  * In addition to the metrics described above, the governor counts recent
63  * intercepts (that is, intercepts that have occurred during the last
64  * %NR_RECENT invocations of it for the given CPU) for each bin.
65  *
66  * In order to select an idle state for a CPU, the governor takes the following
67  * steps (modulo the possible latency constraint that must be taken into account
68  * too):
69  *
70  * 1. Find the deepest CPU idle state whose target residency does not exceed
71  *    the current sleep length (the candidate idle state) and compute 3 sums as
72  *    follows:
73  *
74  *    - The sum of the "hits" and "intercepts" metrics for the candidate state
75  *      and all of the deeper idle states (it represents the cases in which the
76  *      CPU was idle long enough to avoid being intercepted if the sleep length
77  *      had been equal to the current one).
78  *
79  *    - The sum of the "intercepts" metrics for all of the idle states shallower
80  *      than the candidate one (it represents the cases in which the CPU was not
81  *      idle long enough to avoid being intercepted if the sleep length had been
82  *      equal to the current one).
83  *
84  *    - The sum of the numbers of recent intercepts for all of the idle states
85  *      shallower than the candidate one.
86  *
87  * 2. If the second sum is greater than the first one or the third sum is
88  *    greater than %NR_RECENT / 2, the CPU is likely to wake up early, so look
89  *    for an alternative idle state to select.
90  *
91  *    - Traverse the idle states shallower than the candidate one in the
92  *      descending order.
93  *
94  *    - For each of them compute the sum of the "intercepts" metrics and the sum
95  *      of the numbers of recent intercepts over all of the idle states between
96  *      it and the candidate one (including the former and excluding the
97  *      latter).
98  *
99  *    - If each of these sums that needs to be taken into account (because the
100  *      check related to it has indicated that the CPU is likely to wake up
101  *      early) is greater than a half of the corresponding sum computed in step
102  *      1 (which means that the target residency of the state in question had
103  *      not exceeded the idle duration in over a half of the relevant cases),
104  *      select the given idle state instead of the candidate one.
105  *
106  * 3. By default, select the candidate state.
107  *
108  * Util-awareness mechanism:
109  *
110  * The idea behind the util-awareness extension is that there are two distinct
111  * scenarios for the CPU which should result in two different approaches to idle
112  * state selection - utilized and not utilized.
113  *
114  * In this case, 'utilized' means that the average runqueue util of the CPU is
115  * above a certain threshold.
116  *
117  * When the CPU is utilized while going into idle, more likely than not it will
118  * be woken up to do more work soon and so a shallower idle state should be
119  * selected to minimise latency and maximise performance. When the CPU is not
120  * being utilized, the usual metrics-based approach to selecting the deepest
121  * available idle state should be preferred to take advantage of the power
122  * saving.
123  *
124  * In order to achieve this, the governor uses a utilization threshold.
125  * The threshold is computed per-CPU as a percentage of the CPU's capacity
126  * by bit shifting the capacity value. Based on testing, the shift of 6 (~1.56%)
127  * seems to be getting the best results.
128  *
129  * Before selecting the next idle state, the governor compares the current CPU
130  * util to the precomputed util threshold. If it's below, it defaults to the
131  * TEO metrics mechanism. If it's above, the closest shallower idle state will
132  * be selected instead, as long as is not a polling state.
133  */
134 
135 #include <linux/cpuidle.h>
136 #include <linux/jiffies.h>
137 #include <linux/kernel.h>
138 #include <linux/sched.h>
139 #include <linux/sched/clock.h>
140 #include <linux/sched/topology.h>
141 #include <linux/tick.h>
142 
143 #include "gov.h"
144 
145 /*
146  * The number of bits to shift the CPU's capacity by in order to determine
147  * the utilized threshold.
148  *
149  * 6 was chosen based on testing as the number that achieved the best balance
150  * of power and performance on average.
151  *
152  * The resulting threshold is high enough to not be triggered by background
153  * noise and low enough to react quickly when activity starts to ramp up.
154  */
155 #define UTIL_THRESHOLD_SHIFT 6
156 
157 /*
158  * The PULSE value is added to metrics when they grow and the DECAY_SHIFT value
159  * is used for decreasing metrics on a regular basis.
160  */
161 #define PULSE		1024
162 #define DECAY_SHIFT	3
163 
164 /*
165  * Number of the most recent idle duration values to take into consideration for
166  * the detection of recent early wakeup patterns.
167  */
168 #define NR_RECENT	9
169 
170 /**
171  * struct teo_bin - Metrics used by the TEO cpuidle governor.
172  * @intercepts: The "intercepts" metric.
173  * @hits: The "hits" metric.
174  * @recent: The number of recent "intercepts".
175  */
176 struct teo_bin {
177 	unsigned int intercepts;
178 	unsigned int hits;
179 	unsigned int recent;
180 };
181 
182 /**
183  * struct teo_cpu - CPU data used by the TEO cpuidle governor.
184  * @time_span_ns: Time between idle state selection and post-wakeup update.
185  * @sleep_length_ns: Time till the closest timer event (at the selection time).
186  * @state_bins: Idle state data bins for this CPU.
187  * @total: Grand total of the "intercepts" and "hits" metrics for all bins.
188  * @next_recent_idx: Index of the next @recent_idx entry to update.
189  * @recent_idx: Indices of bins corresponding to recent "intercepts".
190  * @tick_hits: Number of "hits" after TICK_NSEC.
191  * @util_threshold: Threshold above which the CPU is considered utilized
192  */
193 struct teo_cpu {
194 	s64 time_span_ns;
195 	s64 sleep_length_ns;
196 	struct teo_bin state_bins[CPUIDLE_STATE_MAX];
197 	unsigned int total;
198 	int next_recent_idx;
199 	int recent_idx[NR_RECENT];
200 	unsigned int tick_hits;
201 	unsigned long util_threshold;
202 };
203 
204 static DEFINE_PER_CPU(struct teo_cpu, teo_cpus);
205 
206 /**
207  * teo_cpu_is_utilized - Check if the CPU's util is above the threshold
208  * @cpu: Target CPU
209  * @cpu_data: Governor CPU data for the target CPU
210  */
211 #ifdef CONFIG_SMP
212 static bool teo_cpu_is_utilized(int cpu, struct teo_cpu *cpu_data)
213 {
214 	return sched_cpu_util(cpu) > cpu_data->util_threshold;
215 }
216 #else
217 static bool teo_cpu_is_utilized(int cpu, struct teo_cpu *cpu_data)
218 {
219 	return false;
220 }
221 #endif
222 
223 /**
224  * teo_update - Update CPU metrics after wakeup.
225  * @drv: cpuidle driver containing state data.
226  * @dev: Target CPU.
227  */
228 static void teo_update(struct cpuidle_driver *drv, struct cpuidle_device *dev)
229 {
230 	struct teo_cpu *cpu_data = per_cpu_ptr(&teo_cpus, dev->cpu);
231 	int i, idx_timer = 0, idx_duration = 0;
232 	s64 target_residency_ns;
233 	u64 measured_ns;
234 
235 	if (cpu_data->time_span_ns >= cpu_data->sleep_length_ns) {
236 		/*
237 		 * One of the safety nets has triggered or the wakeup was close
238 		 * enough to the closest timer event expected at the idle state
239 		 * selection time to be discarded.
240 		 */
241 		measured_ns = U64_MAX;
242 	} else {
243 		u64 lat_ns = drv->states[dev->last_state_idx].exit_latency_ns;
244 
245 		/*
246 		 * The computations below are to determine whether or not the
247 		 * (saved) time till the next timer event and the measured idle
248 		 * duration fall into the same "bin", so use last_residency_ns
249 		 * for that instead of time_span_ns which includes the cpuidle
250 		 * overhead.
251 		 */
252 		measured_ns = dev->last_residency_ns;
253 		/*
254 		 * The delay between the wakeup and the first instruction
255 		 * executed by the CPU is not likely to be worst-case every
256 		 * time, so take 1/2 of the exit latency as a very rough
257 		 * approximation of the average of it.
258 		 */
259 		if (measured_ns >= lat_ns)
260 			measured_ns -= lat_ns / 2;
261 		else
262 			measured_ns /= 2;
263 	}
264 
265 	cpu_data->total = 0;
266 
267 	/*
268 	 * Decay the "hits" and "intercepts" metrics for all of the bins and
269 	 * find the bins that the sleep length and the measured idle duration
270 	 * fall into.
271 	 */
272 	for (i = 0; i < drv->state_count; i++) {
273 		struct teo_bin *bin = &cpu_data->state_bins[i];
274 
275 		bin->hits -= bin->hits >> DECAY_SHIFT;
276 		bin->intercepts -= bin->intercepts >> DECAY_SHIFT;
277 
278 		cpu_data->total += bin->hits + bin->intercepts;
279 
280 		target_residency_ns = drv->states[i].target_residency_ns;
281 
282 		if (target_residency_ns <= cpu_data->sleep_length_ns) {
283 			idx_timer = i;
284 			if (target_residency_ns <= measured_ns)
285 				idx_duration = i;
286 		}
287 	}
288 
289 	i = cpu_data->next_recent_idx++;
290 	if (cpu_data->next_recent_idx >= NR_RECENT)
291 		cpu_data->next_recent_idx = 0;
292 
293 	if (cpu_data->recent_idx[i] >= 0)
294 		cpu_data->state_bins[cpu_data->recent_idx[i]].recent--;
295 
296 	/*
297 	 * If the deepest state's target residency is below the tick length,
298 	 * make a record of it to help teo_select() decide whether or not
299 	 * to stop the tick.  This effectively adds an extra hits-only bin
300 	 * beyond the last state-related one.
301 	 */
302 	if (target_residency_ns < TICK_NSEC) {
303 		cpu_data->tick_hits -= cpu_data->tick_hits >> DECAY_SHIFT;
304 
305 		cpu_data->total += cpu_data->tick_hits;
306 
307 		if (TICK_NSEC <= cpu_data->sleep_length_ns) {
308 			idx_timer = drv->state_count;
309 			if (TICK_NSEC <= measured_ns) {
310 				cpu_data->tick_hits += PULSE;
311 				goto end;
312 			}
313 		}
314 	}
315 
316 	/*
317 	 * If the measured idle duration falls into the same bin as the sleep
318 	 * length, this is a "hit", so update the "hits" metric for that bin.
319 	 * Otherwise, update the "intercepts" metric for the bin fallen into by
320 	 * the measured idle duration.
321 	 */
322 	if (idx_timer == idx_duration) {
323 		cpu_data->state_bins[idx_timer].hits += PULSE;
324 		cpu_data->recent_idx[i] = -1;
325 	} else {
326 		cpu_data->state_bins[idx_duration].intercepts += PULSE;
327 		cpu_data->state_bins[idx_duration].recent++;
328 		cpu_data->recent_idx[i] = idx_duration;
329 	}
330 
331 end:
332 	cpu_data->total += PULSE;
333 }
334 
335 static bool teo_state_ok(int i, struct cpuidle_driver *drv)
336 {
337 	return !tick_nohz_tick_stopped() ||
338 		drv->states[i].target_residency_ns >= TICK_NSEC;
339 }
340 
341 /**
342  * teo_find_shallower_state - Find shallower idle state matching given duration.
343  * @drv: cpuidle driver containing state data.
344  * @dev: Target CPU.
345  * @state_idx: Index of the capping idle state.
346  * @duration_ns: Idle duration value to match.
347  * @no_poll: Don't consider polling states.
348  */
349 static int teo_find_shallower_state(struct cpuidle_driver *drv,
350 				    struct cpuidle_device *dev, int state_idx,
351 				    s64 duration_ns, bool no_poll)
352 {
353 	int i;
354 
355 	for (i = state_idx - 1; i >= 0; i--) {
356 		if (dev->states_usage[i].disable ||
357 				(no_poll && drv->states[i].flags & CPUIDLE_FLAG_POLLING))
358 			continue;
359 
360 		state_idx = i;
361 		if (drv->states[i].target_residency_ns <= duration_ns)
362 			break;
363 	}
364 	return state_idx;
365 }
366 
367 /**
368  * teo_select - Selects the next idle state to enter.
369  * @drv: cpuidle driver containing state data.
370  * @dev: Target CPU.
371  * @stop_tick: Indication on whether or not to stop the scheduler tick.
372  */
373 static int teo_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
374 		      bool *stop_tick)
375 {
376 	struct teo_cpu *cpu_data = per_cpu_ptr(&teo_cpus, dev->cpu);
377 	s64 latency_req = cpuidle_governor_latency_req(dev->cpu);
378 	ktime_t delta_tick = TICK_NSEC / 2;
379 	unsigned int tick_intercept_sum = 0;
380 	unsigned int idx_intercept_sum = 0;
381 	unsigned int intercept_sum = 0;
382 	unsigned int idx_recent_sum = 0;
383 	unsigned int recent_sum = 0;
384 	unsigned int idx_hit_sum = 0;
385 	unsigned int hit_sum = 0;
386 	int constraint_idx = 0;
387 	int idx0 = 0, idx = -1;
388 	bool alt_intercepts, alt_recent;
389 	bool cpu_utilized;
390 	s64 duration_ns;
391 	int i;
392 
393 	if (dev->last_state_idx >= 0) {
394 		teo_update(drv, dev);
395 		dev->last_state_idx = -1;
396 	}
397 
398 	cpu_data->time_span_ns = local_clock();
399 	/*
400 	 * Set the expected sleep length to infinity in case of an early
401 	 * return.
402 	 */
403 	cpu_data->sleep_length_ns = KTIME_MAX;
404 
405 	/* Check if there is any choice in the first place. */
406 	if (drv->state_count < 2) {
407 		idx = 0;
408 		goto out_tick;
409 	}
410 
411 	if (!dev->states_usage[0].disable)
412 		idx = 0;
413 
414 	cpu_utilized = teo_cpu_is_utilized(dev->cpu, cpu_data);
415 	/*
416 	 * If the CPU is being utilized over the threshold and there are only 2
417 	 * states to choose from, the metrics need not be considered, so choose
418 	 * the shallowest non-polling state and exit.
419 	 */
420 	if (drv->state_count < 3 && cpu_utilized) {
421 		/*
422 		 * If state 0 is enabled and it is not a polling one, select it
423 		 * right away unless the scheduler tick has been stopped, in
424 		 * which case care needs to be taken to leave the CPU in a deep
425 		 * enough state in case it is not woken up any time soon after
426 		 * all.  If state 1 is disabled, though, state 0 must be used
427 		 * anyway.
428 		 */
429 		if ((!idx && !(drv->states[0].flags & CPUIDLE_FLAG_POLLING) &&
430 		    teo_state_ok(0, drv)) || dev->states_usage[1].disable) {
431 			idx = 0;
432 			goto out_tick;
433 		}
434 		/* Assume that state 1 is not a polling one and use it. */
435 		idx = 1;
436 		duration_ns = drv->states[1].target_residency_ns;
437 		goto end;
438 	}
439 
440 	/* Compute the sums of metrics for early wakeup pattern detection. */
441 	for (i = 1; i < drv->state_count; i++) {
442 		struct teo_bin *prev_bin = &cpu_data->state_bins[i-1];
443 		struct cpuidle_state *s = &drv->states[i];
444 
445 		/*
446 		 * Update the sums of idle state mertics for all of the states
447 		 * shallower than the current one.
448 		 */
449 		intercept_sum += prev_bin->intercepts;
450 		hit_sum += prev_bin->hits;
451 		recent_sum += prev_bin->recent;
452 
453 		if (dev->states_usage[i].disable)
454 			continue;
455 
456 		if (idx < 0)
457 			idx0 = i; /* first enabled state */
458 
459 		idx = i;
460 
461 		if (s->exit_latency_ns <= latency_req)
462 			constraint_idx = i;
463 
464 		/* Save the sums for the current state. */
465 		idx_intercept_sum = intercept_sum;
466 		idx_hit_sum = hit_sum;
467 		idx_recent_sum = recent_sum;
468 	}
469 
470 	/* Avoid unnecessary overhead. */
471 	if (idx < 0) {
472 		idx = 0; /* No states enabled, must use 0. */
473 		goto out_tick;
474 	}
475 
476 	if (idx == idx0) {
477 		/*
478 		 * Only one idle state is enabled, so use it, but do not
479 		 * allow the tick to be stopped it is shallow enough.
480 		 */
481 		duration_ns = drv->states[idx].target_residency_ns;
482 		goto end;
483 	}
484 
485 	tick_intercept_sum = intercept_sum +
486 			cpu_data->state_bins[drv->state_count-1].intercepts;
487 
488 	/*
489 	 * If the sum of the intercepts metric for all of the idle states
490 	 * shallower than the current candidate one (idx) is greater than the
491 	 * sum of the intercepts and hits metrics for the candidate state and
492 	 * all of the deeper states, or the sum of the numbers of recent
493 	 * intercepts over all of the states shallower than the candidate one
494 	 * is greater than a half of the number of recent events taken into
495 	 * account, a shallower idle state is likely to be a better choice.
496 	 */
497 	alt_intercepts = 2 * idx_intercept_sum > cpu_data->total - idx_hit_sum;
498 	alt_recent = idx_recent_sum > NR_RECENT / 2;
499 	if (alt_recent || alt_intercepts) {
500 		int first_suitable_idx = idx;
501 
502 		/*
503 		 * Look for the deepest idle state whose target residency had
504 		 * not exceeded the idle duration in over a half of the relevant
505 		 * cases (both with respect to intercepts overall and with
506 		 * respect to the recent intercepts only) in the past.
507 		 *
508 		 * Take the possible duration limitation present if the tick
509 		 * has been stopped already into account.
510 		 */
511 		intercept_sum = 0;
512 		recent_sum = 0;
513 
514 		for (i = idx - 1; i >= 0; i--) {
515 			struct teo_bin *bin = &cpu_data->state_bins[i];
516 
517 			intercept_sum += bin->intercepts;
518 			recent_sum += bin->recent;
519 
520 			if ((!alt_recent || 2 * recent_sum > idx_recent_sum) &&
521 			    (!alt_intercepts ||
522 			     2 * intercept_sum > idx_intercept_sum)) {
523 				/*
524 				 * Use the current state unless it is too
525 				 * shallow or disabled, in which case take the
526 				 * first enabled state that is deep enough.
527 				 */
528 				if (teo_state_ok(i, drv) &&
529 				    !dev->states_usage[i].disable)
530 					idx = i;
531 				else
532 					idx = first_suitable_idx;
533 
534 				break;
535 			}
536 
537 			if (dev->states_usage[i].disable)
538 				continue;
539 
540 			if (!teo_state_ok(i, drv)) {
541 				/*
542 				 * The current state is too shallow, but if an
543 				 * alternative candidate state has been found,
544 				 * it may still turn out to be a better choice.
545 				 */
546 				if (first_suitable_idx != idx)
547 					continue;
548 
549 				break;
550 			}
551 
552 			first_suitable_idx = i;
553 		}
554 	}
555 
556 	/*
557 	 * If there is a latency constraint, it may be necessary to select an
558 	 * idle state shallower than the current candidate one.
559 	 */
560 	if (idx > constraint_idx)
561 		idx = constraint_idx;
562 
563 	/*
564 	 * If the CPU is being utilized over the threshold, choose a shallower
565 	 * non-polling state to improve latency, unless the scheduler tick has
566 	 * been stopped already and the shallower state's target residency is
567 	 * not sufficiently large.
568 	 */
569 	if (cpu_utilized) {
570 		i = teo_find_shallower_state(drv, dev, idx, KTIME_MAX, true);
571 		if (teo_state_ok(i, drv))
572 			idx = i;
573 	}
574 
575 	/*
576 	 * Skip the timers check if state 0 is the current candidate one,
577 	 * because an immediate non-timer wakeup is expected in that case.
578 	 */
579 	if (!idx)
580 		goto out_tick;
581 
582 	/*
583 	 * If state 0 is a polling one, check if the target residency of
584 	 * the current candidate state is low enough and skip the timers
585 	 * check in that case too.
586 	 */
587 	if ((drv->states[0].flags & CPUIDLE_FLAG_POLLING) &&
588 	    drv->states[idx].target_residency_ns < RESIDENCY_THRESHOLD_NS)
589 		goto out_tick;
590 
591 	duration_ns = tick_nohz_get_sleep_length(&delta_tick);
592 	cpu_data->sleep_length_ns = duration_ns;
593 
594 	/*
595 	 * If the closest expected timer is before the terget residency of the
596 	 * candidate state, a shallower one needs to be found.
597 	 */
598 	if (drv->states[idx].target_residency_ns > duration_ns) {
599 		i = teo_find_shallower_state(drv, dev, idx, duration_ns, false);
600 		if (teo_state_ok(i, drv))
601 			idx = i;
602 	}
603 
604 	/*
605 	 * If the selected state's target residency is below the tick length
606 	 * and intercepts occurring before the tick length are the majority of
607 	 * total wakeup events, do not stop the tick.
608 	 */
609 	if (drv->states[idx].target_residency_ns < TICK_NSEC &&
610 	    tick_intercept_sum > cpu_data->total / 2 + cpu_data->total / 8)
611 		duration_ns = TICK_NSEC / 2;
612 
613 end:
614 	/*
615 	 * Allow the tick to be stopped unless the selected state is a polling
616 	 * one or the expected idle duration is shorter than the tick period
617 	 * length.
618 	 */
619 	if ((!(drv->states[idx].flags & CPUIDLE_FLAG_POLLING) &&
620 	    duration_ns >= TICK_NSEC) || tick_nohz_tick_stopped())
621 		return idx;
622 
623 	/*
624 	 * The tick is not going to be stopped, so if the target residency of
625 	 * the state to be returned is not within the time till the closest
626 	 * timer including the tick, try to correct that.
627 	 */
628 	if (idx > idx0 &&
629 	    drv->states[idx].target_residency_ns > delta_tick)
630 		idx = teo_find_shallower_state(drv, dev, idx, delta_tick, false);
631 
632 out_tick:
633 	*stop_tick = false;
634 	return idx;
635 }
636 
637 /**
638  * teo_reflect - Note that governor data for the CPU need to be updated.
639  * @dev: Target CPU.
640  * @state: Entered state.
641  */
642 static void teo_reflect(struct cpuidle_device *dev, int state)
643 {
644 	struct teo_cpu *cpu_data = per_cpu_ptr(&teo_cpus, dev->cpu);
645 
646 	dev->last_state_idx = state;
647 	/*
648 	 * If the wakeup was not "natural", but triggered by one of the safety
649 	 * nets, assume that the CPU might have been idle for the entire sleep
650 	 * length time.
651 	 */
652 	if (dev->poll_time_limit ||
653 	    (tick_nohz_idle_got_tick() && cpu_data->sleep_length_ns > TICK_NSEC)) {
654 		dev->poll_time_limit = false;
655 		cpu_data->time_span_ns = cpu_data->sleep_length_ns;
656 	} else {
657 		cpu_data->time_span_ns = local_clock() - cpu_data->time_span_ns;
658 	}
659 }
660 
661 /**
662  * teo_enable_device - Initialize the governor's data for the target CPU.
663  * @drv: cpuidle driver (not used).
664  * @dev: Target CPU.
665  */
666 static int teo_enable_device(struct cpuidle_driver *drv,
667 			     struct cpuidle_device *dev)
668 {
669 	struct teo_cpu *cpu_data = per_cpu_ptr(&teo_cpus, dev->cpu);
670 	unsigned long max_capacity = arch_scale_cpu_capacity(dev->cpu);
671 	int i;
672 
673 	memset(cpu_data, 0, sizeof(*cpu_data));
674 	cpu_data->util_threshold = max_capacity >> UTIL_THRESHOLD_SHIFT;
675 
676 	for (i = 0; i < NR_RECENT; i++)
677 		cpu_data->recent_idx[i] = -1;
678 
679 	return 0;
680 }
681 
682 static struct cpuidle_governor teo_governor = {
683 	.name =		"teo",
684 	.rating =	19,
685 	.enable =	teo_enable_device,
686 	.select =	teo_select,
687 	.reflect =	teo_reflect,
688 };
689 
690 static int __init teo_governor_init(void)
691 {
692 	return cpuidle_register_governor(&teo_governor);
693 }
694 
695 postcore_initcall(teo_governor_init);
696