Merge branch 'pm-cpuidle'

* pm-cpuidle:
  cpuidle: menu: Avoid computations when result will be discarded
  cpuidle: menu: Drop redundant comparison
  cpuidle: menu: Simplify checks related to the polling state
  cpuidle: poll_state: Revise loop termination condition
  cpuidle: menu: Move the latency_req == 0 special case check
  cpuidle: menu: Avoid computations for very close timers
  cpuidle: menu: Do not update last_state_idx in menu_select()
  cpuidle: menu: Get rid of first_idx from menu_select()
  cpuidle: menu: Compute first_idx when latency_req is known
  cpuidle: menu: Fix wakeup statistics updates for polling state
  cpuidle: menu: Replace data->predicted_us with local variable
  cpuidle: enter_state: Don't needlessly calculate diff time
  cpuidle: Remove unnecessary wrapper cpuidle_get_last_residency()
  intel_idle: Get rid of custom ICPU() macro

drivers/cpuidle/cpuidle.c

@@ -247,17 +247,17 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv,
 	if (!cpuidle_state_is_coupled(drv, index))
 		local_irq_enable();
 
-	diff = ktime_us_delta(time_end, time_start);
-	if (diff > INT_MAX)
-		diff = INT_MAX;
-
-	dev->last_residency = (int) diff;
-
 	if (entered_state >= 0) {
-		/* Update cpuidle counters */
-		/* This can be moved to within driver enter routine
+		/*
+		 * Update cpuidle counters
+		 * This can be moved to within driver enter routine,
 		 * but that results in multiple copies of same code.
 		 */
+		diff = ktime_us_delta(time_end, time_start);
+		if (diff > INT_MAX)
+			diff = INT_MAX;
+
+		dev->last_residency = (int)diff;
 		dev->states_usage[entered_state].time += dev->last_residency;
 		dev->states_usage[entered_state].usage++;
 	} else {

drivers/cpuidle/governors/ladder.c

@@ -80,7 +80,7 @@ static int ladder_select_state(struct cpuidle_driver *drv,
 
 	last_state = &ldev->states[last_idx];
 
-	last_residency = cpuidle_get_last_residency(dev) - drv->states[last_idx].exit_latency;
+	last_residency = dev->last_residency - drv->states[last_idx].exit_latency;
 
 	/* consider promotion */
 	if (last_idx < drv->state_count - 1 &&

drivers/cpuidle/governors/menu.c

@@ -124,7 +124,6 @@ struct menu_device {
 	int             tick_wakeup;
 
 	unsigned int	next_timer_us;
-	unsigned int	predicted_us;
 	unsigned int	bucket;
 	unsigned int	correction_factor[BUCKETS];
 	unsigned int	intervals[INTERVALS];
@@ -197,10 +196,11 @@ static void menu_update(struct cpuidle_driver *drv, struct cpuidle_device *dev);
  * of points is below a threshold. If it is... then use the
  * average of these 8 points as the estimated value.
  */
-static unsigned int get_typical_interval(struct menu_device *data)
+static unsigned int get_typical_interval(struct menu_device *data,
+					 unsigned int predicted_us)
 {
 	int i, divisor;
-	unsigned int max, thresh, avg;
+	unsigned int min, max, thresh, avg;
 	uint64_t sum, variance;
 
 	thresh = UINT_MAX; /* Discard outliers above this value */
@@ -208,6 +208,7 @@ static unsigned int get_typical_interval(struct menu_device *data)
 again:
 
 	/* First calculate the average of past intervals */
+	min = UINT_MAX;
 	max = 0;
 	sum = 0;
 	divisor = 0;
@@ -218,8 +219,19 @@ static unsigned int get_typical_interval(struct menu_device *data)
 			divisor++;
 			if (value > max)
 				max = value;
+
+			if (value < min)
+				min = value;
 		}
 	}
+
+	/*
+	 * If the result of the computation is going to be discarded anyway,
+	 * avoid the computation altogether.
+	 */
+	if (min >= predicted_us)
+		return UINT_MAX;
+
 	if (divisor == INTERVALS)
 		avg = sum >> INTERVAL_SHIFT;
 	else
@@ -286,10 +298,9 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 	struct menu_device *data = this_cpu_ptr(&menu_devices);
 	int latency_req = cpuidle_governor_latency_req(dev->cpu);
 	int i;
-	int first_idx;
 	int idx;
 	unsigned int interactivity_req;
-	unsigned int expected_interval;
+	unsigned int predicted_us;
 	unsigned long nr_iowaiters, cpu_load;
 	ktime_t delta_next;
 
@@ -298,50 +309,36 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 		data->needs_update = 0;
 	}
 
-	/* Special case when user has set very strict latency requirement */
-	if (unlikely(latency_req == 0)) {
-		*stop_tick = false;
-		return 0;
-	}
-
 	/* determine the expected residency time, round up */
 	data->next_timer_us = ktime_to_us(tick_nohz_get_sleep_length(&delta_next));
 
 	get_iowait_load(&nr_iowaiters, &cpu_load);
 	data->bucket = which_bucket(data->next_timer_us, nr_iowaiters);
 
+	if (unlikely(drv->state_count <= 1 || latency_req == 0) ||
+	    ((data->next_timer_us < drv->states[1].target_residency ||
+	      latency_req < drv->states[1].exit_latency) &&
+	     !drv->states[0].disabled && !dev->states_usage[0].disable)) {
+		/*
+		 * In this case state[0] will be used no matter what, so return
+		 * it right away and keep the tick running.
+		 */
+		*stop_tick = false;
+		return 0;
+	}
+
 	/*
 	 * Force the result of multiplication to be 64 bits even if both
 	 * operands are 32 bits.
 	 * Make sure to round up for half microseconds.
 	 */
-	data->predicted_us = DIV_ROUND_CLOSEST_ULL((uint64_t)data->next_timer_us *
+	predicted_us = DIV_ROUND_CLOSEST_ULL((uint64_t)data->next_timer_us *
 					 data->correction_factor[data->bucket],
 					 RESOLUTION * DECAY);
-
-	expected_interval = get_typical_interval(data);
-	expected_interval = min(expected_interval, data->next_timer_us);
-
-	first_idx = 0;
-	if (drv->states[0].flags & CPUIDLE_FLAG_POLLING) {
-		struct cpuidle_state *s = &drv->states[1];
-		unsigned int polling_threshold;
-
-		/*
-		 * Default to a physical idle state, not to busy polling, unless
-		 * a timer is going to trigger really really soon.
-		 */
-		polling_threshold = max_t(unsigned int, 20, s->target_residency);
-		if (data->next_timer_us > polling_threshold &&
-		    latency_req > s->exit_latency && !s->disabled &&
-		    !dev->states_usage[1].disable)
-			first_idx = 1;
-	}
-
 	/*
 	 * Use the lowest expected idle interval to pick the idle state.
 	 */
-	data->predicted_us = min(data->predicted_us, expected_interval);
+	predicted_us = min(predicted_us, get_typical_interval(data, predicted_us));
 
 	if (tick_nohz_tick_stopped()) {
 		/*
@@ -352,34 +349,46 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 		 * the known time till the closest timer event for the idle
 		 * state selection.
 		 */
-		if (data->predicted_us < TICK_USEC)
-			data->predicted_us = ktime_to_us(delta_next);
+		if (predicted_us < TICK_USEC)
+			predicted_us = ktime_to_us(delta_next);
 	} else {
 		/*
 		 * Use the performance multiplier and the user-configurable
 		 * latency_req to determine the maximum exit latency.
 		 */
-		interactivity_req = data->predicted_us / performance_multiplier(nr_iowaiters, cpu_load);
+		interactivity_req = predicted_us / performance_multiplier(nr_iowaiters, cpu_load);
 		if (latency_req > interactivity_req)
 			latency_req = interactivity_req;
 	}
 
-	expected_interval = data->predicted_us;
 	/*
 	 * Find the idle state with the lowest power while satisfying
 	 * our constraints.
 	 */
 	idx = -1;
-	for (i = first_idx; i < drv->state_count; i++) {
+	for (i = 0; i < drv->state_count; i++) {
 		struct cpuidle_state *s = &drv->states[i];
 		struct cpuidle_state_usage *su = &dev->states_usage[i];
 
 		if (s->disabled || su->disable)
 			continue;
+
 		if (idx == -1)
 			idx = i; /* first enabled state */
-		if (s->target_residency > data->predicted_us) {
-			if (data->predicted_us < TICK_USEC)
+
+		if (s->target_residency > predicted_us) {
+			/*
+			 * Use a physical idle state, not busy polling, unless
+			 * a timer is going to trigger soon enough.
+			 */
+			if ((drv->states[idx].flags & CPUIDLE_FLAG_POLLING) &&
+			    s->exit_latency <= latency_req &&
+			    s->target_residency <= data->next_timer_us) {
+				predicted_us = s->target_residency;
+				idx = i;
+				break;
+			}
+			if (predicted_us < TICK_USEC)
 				break;
 
 			if (!tick_nohz_tick_stopped()) {
@@ -389,7 +398,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 				 * tick in that case and let the governor run
 				 * again in the next iteration of the loop.
 				 */
-				expected_interval = drv->states[idx].target_residency;
+				predicted_us = drv->states[idx].target_residency;
 				break;
 			}
 
@@ -403,7 +412,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 			    s->target_residency <= ktime_to_us(delta_next))
 				idx = i;
 
-			goto out;
+			return idx;
 		}
 		if (s->exit_latency > latency_req) {
 			/*
@@ -412,7 +421,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 			 * expected idle duration so that the tick is retained
 			 * as long as that target residency is low enough.
 			 */
-			expected_interval = drv->states[idx].target_residency;
+			predicted_us = drv->states[idx].target_residency;
 			break;
 		}
 		idx = i;
@@ -426,7 +435,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 	 * expected idle duration is shorter than the tick period length.
 	 */
 	if (((drv->states[idx].flags & CPUIDLE_FLAG_POLLING) ||
-	     expected_interval < TICK_USEC) && !tick_nohz_tick_stopped()) {
+	     predicted_us < TICK_USEC) && !tick_nohz_tick_stopped()) {
 		unsigned int delta_next_us = ktime_to_us(delta_next);
 
 		*stop_tick = false;
@@ -450,10 +459,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 		}
 	}
 
-out:
-	data->last_state_idx = idx;
-
-	return data->last_state_idx;
+	return idx;
 }
 
 /**
@@ -512,9 +518,19 @@ static void menu_update(struct cpuidle_driver *drv, struct cpuidle_device *dev)
 		 * duration predictor do a better job next time.
 		 */
 		measured_us = 9 * MAX_INTERESTING / 10;
+	} else if ((drv->states[last_idx].flags & CPUIDLE_FLAG_POLLING) &&
+		   dev->poll_time_limit) {
+		/*
+		 * The CPU exited the "polling" state due to a time limit, so
+		 * the idle duration prediction leading to the selection of that
+		 * state was inaccurate.  If a better prediction had been made,
+		 * the CPU might have been woken up from idle by the next timer.
+		 * Assume that to be the case.
+		 */
+		measured_us = data->next_timer_us;
 	} else {
 		/* measured value */
-		measured_us = cpuidle_get_last_residency(dev);
+		measured_us = dev->last_residency;
 
 		/* Deduct exit latency */
 		if (measured_us > 2 * target->exit_latency)

drivers/cpuidle/poll_state.c

@@ -9,7 +9,6 @@
 #include <linux/sched/clock.h>
 #include <linux/sched/idle.h>
 
-#define POLL_IDLE_TIME_LIMIT	(TICK_NSEC / 16)
 #define POLL_IDLE_RELAX_COUNT	200
 
 static int __cpuidle poll_idle(struct cpuidle_device *dev,
@@ -17,8 +16,11 @@ static int __cpuidle poll_idle(struct cpuidle_device *dev,
 {
 	u64 time_start = local_clock();
 
+	dev->poll_time_limit = false;
+
 	local_irq_enable();
 	if (!current_set_polling_and_test()) {
+		u64 limit = (u64)drv->states[1].target_residency * NSEC_PER_USEC;
 		unsigned int loop_count = 0;
 
 		while (!need_resched()) {
@@ -27,8 +29,10 @@ static int __cpuidle poll_idle(struct cpuidle_device *dev,
 				continue;
 
 			loop_count = 0;
-			if (local_clock() - time_start > POLL_IDLE_TIME_LIMIT)
+			if (local_clock() - time_start > limit) {
+				dev->poll_time_limit = true;
 				break;
+			}
 		}
 	}
 	current_clr_polling();

drivers/idle/intel_idle.c

@@ -1066,46 +1066,43 @@ static const struct idle_cpu idle_cpu_dnv = {
 	.disable_promotion_to_c1e = true,
 };
 
-#define ICPU(model, cpu) \
-	{ X86_VENDOR_INTEL, 6, model, X86_FEATURE_ANY, (unsigned long)&cpu }
-
 static const struct x86_cpu_id intel_idle_ids[] __initconst = {
-	ICPU(INTEL_FAM6_NEHALEM_EP,		idle_cpu_nehalem),
-	ICPU(INTEL_FAM6_NEHALEM,		idle_cpu_nehalem),
-	ICPU(INTEL_FAM6_NEHALEM_G,		idle_cpu_nehalem),
-	ICPU(INTEL_FAM6_WESTMERE,		idle_cpu_nehalem),
-	ICPU(INTEL_FAM6_WESTMERE_EP,		idle_cpu_nehalem),
-	ICPU(INTEL_FAM6_NEHALEM_EX,		idle_cpu_nehalem),
-	ICPU(INTEL_FAM6_ATOM_PINEVIEW,		idle_cpu_atom),
-	ICPU(INTEL_FAM6_ATOM_LINCROFT,		idle_cpu_lincroft),
-	ICPU(INTEL_FAM6_WESTMERE_EX,		idle_cpu_nehalem),
-	ICPU(INTEL_FAM6_SANDYBRIDGE,		idle_cpu_snb),
-	ICPU(INTEL_FAM6_SANDYBRIDGE_X,		idle_cpu_snb),
-	ICPU(INTEL_FAM6_ATOM_CEDARVIEW,		idle_cpu_atom),
-	ICPU(INTEL_FAM6_ATOM_SILVERMONT1,	idle_cpu_byt),
-	ICPU(INTEL_FAM6_ATOM_MERRIFIELD,	idle_cpu_tangier),
-	ICPU(INTEL_FAM6_ATOM_AIRMONT,		idle_cpu_cht),
-	ICPU(INTEL_FAM6_IVYBRIDGE,		idle_cpu_ivb),
-	ICPU(INTEL_FAM6_IVYBRIDGE_X,		idle_cpu_ivt),
-	ICPU(INTEL_FAM6_HASWELL_CORE,		idle_cpu_hsw),
-	ICPU(INTEL_FAM6_HASWELL_X,		idle_cpu_hsw),
-	ICPU(INTEL_FAM6_HASWELL_ULT,		idle_cpu_hsw),
-	ICPU(INTEL_FAM6_HASWELL_GT3E,		idle_cpu_hsw),
-	ICPU(INTEL_FAM6_ATOM_SILVERMONT2,	idle_cpu_avn),
-	ICPU(INTEL_FAM6_BROADWELL_CORE,		idle_cpu_bdw),
-	ICPU(INTEL_FAM6_BROADWELL_GT3E,		idle_cpu_bdw),
-	ICPU(INTEL_FAM6_BROADWELL_X,		idle_cpu_bdw),
-	ICPU(INTEL_FAM6_BROADWELL_XEON_D,	idle_cpu_bdw),
-	ICPU(INTEL_FAM6_SKYLAKE_MOBILE,		idle_cpu_skl),
-	ICPU(INTEL_FAM6_SKYLAKE_DESKTOP,	idle_cpu_skl),
-	ICPU(INTEL_FAM6_KABYLAKE_MOBILE,	idle_cpu_skl),
-	ICPU(INTEL_FAM6_KABYLAKE_DESKTOP,	idle_cpu_skl),
-	ICPU(INTEL_FAM6_SKYLAKE_X,		idle_cpu_skx),
-	ICPU(INTEL_FAM6_XEON_PHI_KNL,		idle_cpu_knl),
-	ICPU(INTEL_FAM6_XEON_PHI_KNM,		idle_cpu_knl),
-	ICPU(INTEL_FAM6_ATOM_GOLDMONT,		idle_cpu_bxt),
-	ICPU(INTEL_FAM6_ATOM_GEMINI_LAKE,	idle_cpu_bxt),
-	ICPU(INTEL_FAM6_ATOM_DENVERTON,		idle_cpu_dnv),
+	INTEL_CPU_FAM6(NEHALEM_EP,		idle_cpu_nehalem),
+	INTEL_CPU_FAM6(NEHALEM,			idle_cpu_nehalem),
+	INTEL_CPU_FAM6(NEHALEM_G,		idle_cpu_nehalem),
+	INTEL_CPU_FAM6(WESTMERE,		idle_cpu_nehalem),
+	INTEL_CPU_FAM6(WESTMERE_EP,		idle_cpu_nehalem),
+	INTEL_CPU_FAM6(NEHALEM_EX,		idle_cpu_nehalem),
+	INTEL_CPU_FAM6(ATOM_PINEVIEW,		idle_cpu_atom),
+	INTEL_CPU_FAM6(ATOM_LINCROFT,		idle_cpu_lincroft),
+	INTEL_CPU_FAM6(WESTMERE_EX,		idle_cpu_nehalem),
+	INTEL_CPU_FAM6(SANDYBRIDGE,		idle_cpu_snb),
+	INTEL_CPU_FAM6(SANDYBRIDGE_X,		idle_cpu_snb),
+	INTEL_CPU_FAM6(ATOM_CEDARVIEW,		idle_cpu_atom),
+	INTEL_CPU_FAM6(ATOM_SILVERMONT1,	idle_cpu_byt),
+	INTEL_CPU_FAM6(ATOM_MERRIFIELD,		idle_cpu_tangier),
+	INTEL_CPU_FAM6(ATOM_AIRMONT,		idle_cpu_cht),
+	INTEL_CPU_FAM6(IVYBRIDGE,		idle_cpu_ivb),
+	INTEL_CPU_FAM6(IVYBRIDGE_X,		idle_cpu_ivt),
+	INTEL_CPU_FAM6(HASWELL_CORE,		idle_cpu_hsw),
+	INTEL_CPU_FAM6(HASWELL_X,		idle_cpu_hsw),
+	INTEL_CPU_FAM6(HASWELL_ULT,		idle_cpu_hsw),
+	INTEL_CPU_FAM6(HASWELL_GT3E,		idle_cpu_hsw),
+	INTEL_CPU_FAM6(ATOM_SILVERMONT2,	idle_cpu_avn),
+	INTEL_CPU_FAM6(BROADWELL_CORE,		idle_cpu_bdw),
+	INTEL_CPU_FAM6(BROADWELL_GT3E,		idle_cpu_bdw),
+	INTEL_CPU_FAM6(BROADWELL_X,		idle_cpu_bdw),
+	INTEL_CPU_FAM6(BROADWELL_XEON_D,	idle_cpu_bdw),
+	INTEL_CPU_FAM6(SKYLAKE_MOBILE,		idle_cpu_skl),
+	INTEL_CPU_FAM6(SKYLAKE_DESKTOP,		idle_cpu_skl),
+	INTEL_CPU_FAM6(KABYLAKE_MOBILE,		idle_cpu_skl),
+	INTEL_CPU_FAM6(KABYLAKE_DESKTOP,	idle_cpu_skl),
+	INTEL_CPU_FAM6(SKYLAKE_X,		idle_cpu_skx),
+	INTEL_CPU_FAM6(XEON_PHI_KNL,		idle_cpu_knl),
+	INTEL_CPU_FAM6(XEON_PHI_KNM,		idle_cpu_knl),
+	INTEL_CPU_FAM6(ATOM_GOLDMONT,		idle_cpu_bxt),
+	INTEL_CPU_FAM6(ATOM_GEMINI_LAKE,	idle_cpu_bxt),
+	INTEL_CPU_FAM6(ATOM_DENVERTON,		idle_cpu_dnv),
 	{}
 };
 

include/linux/cpuidle.h

@@ -81,6 +81,7 @@ struct cpuidle_device {
 	unsigned int		registered:1;
 	unsigned int		enabled:1;
 	unsigned int		use_deepest_state:1;
+	unsigned int		poll_time_limit:1;
 	unsigned int		cpu;
 
 	int			last_residency;
@@ -99,16 +100,6 @@ struct cpuidle_device {
 DECLARE_PER_CPU(struct cpuidle_device *, cpuidle_devices);
 DECLARE_PER_CPU(struct cpuidle_device, cpuidle_dev);
 
-/**
- * cpuidle_get_last_residency - retrieves the last state's residency time
- * @dev: the target CPU
- */
-static inline int cpuidle_get_last_residency(struct cpuidle_device *dev)
-{
-	return dev->last_residency;
-}
-
-
 /****************************
  * CPUIDLE DRIVER INTERFACE *
  ****************************/