Merge branch 'pm-cpuidle'

Merge cpuidle updates for 6.20-rc1/7.0-rc1:

 - Add a command line option to adjust the C-states table in the
   intel_idle driver, remove the 'preferred_cstates' module parameter
   from it, add C-states validation to it and clean it up (Artem
   Bityutskiy)

 - Make the menu cpuidle governor always check the time till the closest
   timer event when the scheduler tick has been stopped to prevent it
   from mistakenly selecting the deepest available idle state (Rafael
   Wysocki)

 - Update the teo cpuidle governor to avoid making suboptimal decisions
   in certain corner cases and generally improve idle state selection
   accuracy (Rafael Wysocki)

 - Remove an unlikely() annotation on the early-return condition in
   menu_select() that leads to branch misprediction 100% of the time
   on systems with only 1 idle state enabled, like ARM64 servers (Breno
   Leitao)

 - Add Christian Loehle to MAINTAINERS as a cpuidle reviewer (Christian
   Loehle)

* pm-cpuidle:
  cpuidle: governors: teo: Refine intercepts-based idle state lookup
  cpuidle: governors: teo: Adjust the classification of wakeup events
  cpuidle: governors: teo: Refine tick_intercepts vs total events check
  cpuidle: governors: teo: Avoid fake intercepts produced by tick
  cpuidle: governors: teo: Avoid selecting states with zero-size bins
  cpuidle: governors: menu: Always check timers with tick stopped
  MAINTAINERS: Add myself as cpuidle reviewer
  cpuidle: menu: Remove incorrect unlikely() annotation
  intel_idle: Add C-states validation
  intel_idle: Add cmdline option to adjust C-states table
  intel_idle: Initialize sysfs after cpuidle driver initialization
  intel_idle: Remove the 'preferred_cstates' parameter
  intel_idle: Remove unused driver version constant

Author: rafaeljw

MAINTAINERS

@@ -6561,6 +6561,7 @@ F:	rust/kernel/cpu.rs
 CPU IDLE TIME MANAGEMENT FRAMEWORK
 M:	"Rafael J. Wysocki" <rafael@kernel.org>
 M:	Daniel Lezcano <daniel.lezcano@linaro.org>
+R:	Christian Loehle <christian.loehle@arm.com>
 L:	linux-pm@vger.kernel.org
 S:	Maintained
 B:	https://bugzilla.kernel.org

drivers/cpuidle/governors/menu.c

@@ -239,7 +239,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 
 	/* Find the shortest expected idle interval. */
 	predicted_ns = get_typical_interval(data) * NSEC_PER_USEC;
-	if (predicted_ns > RESIDENCY_THRESHOLD_NS) {
+	if (predicted_ns > RESIDENCY_THRESHOLD_NS || tick_nohz_tick_stopped()) {
 		unsigned int timer_us;
 
 		/* Determine the time till the closest timer. */
@@ -259,6 +259,16 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 				   RESOLUTION * DECAY * NSEC_PER_USEC);
 		/* Use the lowest expected idle interval to pick the idle state. */
 		predicted_ns = min((u64)timer_us * NSEC_PER_USEC, predicted_ns);
+		/*
+		 * If the tick is already stopped, the cost of possible short
+		 * idle duration misprediction is much higher, because the CPU
+		 * may be stuck in a shallow idle state for a long time as a
+		 * result of it.  In that case, say we might mispredict and use
+		 * the known time till the closest timer event for the idle
+		 * state selection.
+		 */
+		if (tick_nohz_tick_stopped() && predicted_ns < TICK_NSEC)
+			predicted_ns = data->next_timer_ns;
 	} else {
 		/*
 		 * Because the next timer event is not going to be determined
@@ -271,7 +281,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 		data->bucket = BUCKETS - 1;
 	}
 
-	if (unlikely(drv->state_count <= 1 || latency_req == 0) ||
+	if (drv->state_count <= 1 || latency_req == 0 ||
 	    ((data->next_timer_ns < drv->states[1].target_residency_ns ||
 	      latency_req < drv->states[1].exit_latency_ns) &&
 	     !dev->states_usage[0].disable)) {
@@ -284,16 +294,6 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 		return 0;
 	}
 
-	/*
-	 * If the tick is already stopped, the cost of possible short idle
-	 * duration misprediction is much higher, because the CPU may be stuck
-	 * in a shallow idle state for a long time as a result of it.  In that
-	 * case, say we might mispredict and use the known time till the closest
-	 * timer event for the idle state selection.
-	 */
-	if (tick_nohz_tick_stopped() && predicted_ns < TICK_NSEC)
-		predicted_ns = data->next_timer_ns;
-
 	/*
 	 * Find the idle state with the lowest power while satisfying
 	 * our constraints.

drivers/cpuidle/governors/teo.c

@@ -48,12 +48,11 @@
  * in accordance with what happened last time.
  *
  * The "hits" metric reflects the relative frequency of situations in which the
- * sleep length and the idle duration measured after CPU wakeup fall into the
- * same bin (that is, the CPU appears to wake up "on time" relative to the sleep
- * length).  In turn, the "intercepts" metric reflects the relative frequency of
- * non-timer wakeup events for which the measured idle duration falls into a bin
- * that corresponds to an idle state shallower than the one whose bin is fallen
- * into by the sleep length (these events are also referred to as "intercepts"
+ * sleep length and the idle duration measured after CPU wakeup are close enough
+ * (that is, the CPU appears to wake up "on time" relative to the sleep length).
+ * In turn, the "intercepts" metric reflects the relative frequency of non-timer
+ * wakeup events for which the measured idle duration is significantly different
+ * from the sleep length (these events are also referred to as "intercepts"
  * below).
  *
  * The governor also counts "intercepts" with the measured idle duration below
@@ -75,12 +74,17 @@
  *      than the candidate one (it represents the cases in which the CPU was
  *      likely woken up by a non-timer wakeup source).
  *
+ *    Also find the idle state with the maximum intercepts metric (if there are
+ *    multiple states with the maximum intercepts metric, choose the one with
+ *    the highest index).
+ *
  * 2. If the second sum computed in step 1 is greater than a half of the sum of
  *    both metrics for the candidate state bin and all subsequent bins (if any),
  *    a shallower idle state is likely to be more suitable, so look for it.
  *
  *    - Traverse the enabled idle states shallower than the candidate one in the
- *      descending order.
+ *      descending order, starting at the state with the maximum intercepts
+ *      metric found in step 1.
  *
  *    - For each of them compute the sum of the "intercepts" metrics over all
  *      of the idle states between it and the candidate one (including the
@@ -167,6 +171,7 @@ static void teo_decay(unsigned int *metric)
  */
 static void teo_update(struct cpuidle_driver *drv, struct cpuidle_device *dev)
 {
+	s64 lat_ns = drv->states[dev->last_state_idx].exit_latency_ns;
 	struct teo_cpu *cpu_data = this_cpu_ptr(&teo_cpus);
 	int i, idx_timer = 0, idx_duration = 0;
 	s64 target_residency_ns, measured_ns;
@@ -182,8 +187,6 @@ static void teo_update(struct cpuidle_driver *drv, struct cpuidle_device *dev)
 		 */
 		measured_ns = S64_MAX;
 	} else {
-		s64 lat_ns = drv->states[dev->last_state_idx].exit_latency_ns;
-
 		measured_ns = dev->last_residency_ns;
 		/*
 		 * The delay between the wakeup and the first instruction
@@ -239,15 +242,31 @@ static void teo_update(struct cpuidle_driver *drv, struct cpuidle_device *dev)
 			cpu_data->state_bins[drv->state_count-1].hits += PULSE;
 			return;
 		}
+		/*
+		 * If intercepts within the tick period range are not frequent
+		 * enough, count this wakeup as a hit, since it is likely that
+		 * the tick has woken up the CPU because an expected intercept
+		 * was not there.  Otherwise, one of the intercepts may have
+		 * been incidentally preceded by the tick wakeup.
+		 */
+		if (3 * cpu_data->tick_intercepts < 2 * total) {
+			cpu_data->state_bins[idx_timer].hits += PULSE;
+			return;
+		}
 	}
 
 	/*
-	 * If the measured idle duration falls into the same bin as the sleep
-	 * length, this is a "hit", so update the "hits" metric for that bin.
+	 * If the measured idle duration (adjusted for the entered state exit
+	 * latency) falls into the same bin as the sleep length and the latter
+	 * is less than the "raw" measured idle duration (so the wakeup appears
+	 * to have occurred after the anticipated timer event), this is a "hit",
+	 * so update the "hits" metric for that bin.
+	 *
 	 * Otherwise, update the "intercepts" metric for the bin fallen into by
 	 * the measured idle duration.
 	 */
-	if (idx_timer == idx_duration) {
+	if (idx_timer == idx_duration &&
+	    cpu_data->sleep_length_ns - measured_ns < lat_ns / 2) {
 		cpu_data->state_bins[idx_timer].hits += PULSE;
 	} else {
 		cpu_data->state_bins[idx_duration].intercepts += PULSE;
@@ -294,8 +313,10 @@ static int teo_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 	ktime_t delta_tick = TICK_NSEC / 2;
 	unsigned int idx_intercept_sum = 0;
 	unsigned int intercept_sum = 0;
+	unsigned int intercept_max = 0;
 	unsigned int idx_hit_sum = 0;
 	unsigned int hit_sum = 0;
+	int intercept_max_idx = -1;
 	int constraint_idx = 0;
 	int idx0 = 0, idx = -1;
 	s64 duration_ns;
@@ -326,17 +347,32 @@ static int teo_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 	if (!dev->states_usage[0].disable)
 		idx = 0;
 
-	/* Compute the sums of metrics for early wakeup pattern detection. */
+	/*
+	 * Compute the sums of metrics for early wakeup pattern detection and
+	 * look for the state bin with the maximum intercepts metric below the
+	 * deepest enabled one (if there are multiple states with the maximum
+	 * intercepts metric, choose the one with the highest index).
+	 */
 	for (i = 1; i < drv->state_count; i++) {
 		struct teo_bin *prev_bin = &cpu_data->state_bins[i-1];
+		unsigned int prev_intercepts = prev_bin->intercepts;
 		struct cpuidle_state *s = &drv->states[i];
 
 		/*
 		 * Update the sums of idle state metrics for all of the states
 		 * shallower than the current one.
 		 */
-		intercept_sum += prev_bin->intercepts;
 		hit_sum += prev_bin->hits;
+		intercept_sum += prev_intercepts;
+		/*
+		 * Check if this is the bin with the maximum number of
+		 * intercepts so far and in that case update the index of
+		 * the state with the maximum intercepts metric.
+		 */
+		if (prev_intercepts >= intercept_max) {
+			intercept_max = prev_intercepts;
+			intercept_max_idx = i - 1;
+		}
 
 		if (dev->states_usage[i].disable)
 			continue;
@@ -388,12 +424,34 @@ static int teo_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 			while (min_idx < idx &&
 			       drv->states[min_idx].target_residency_ns < TICK_NSEC)
 				min_idx++;
+
+			/*
+			 * Avoid selecting a state with a lower index, but with
+			 * the same target residency as the current candidate
+			 * one.
+			 */
+			if (drv->states[min_idx].target_residency_ns ==
+					drv->states[idx].target_residency_ns)
+				goto constraint;
+		}
+
+		/*
+		 * If the minimum state index is greater than or equal to the
+		 * index of the state with the maximum intercepts metric and
+		 * the corresponding state is enabled, there is no need to look
+		 * at the deeper states.
+		 */
+		if (min_idx >= intercept_max_idx &&
+		    !dev->states_usage[min_idx].disable) {
+			idx = min_idx;
+			goto constraint;
 		}
 
 		/*
-		 * Look for the deepest idle state whose target residency had
-		 * not exceeded the idle duration in over a half of the relevant
-		 * cases in the past.
+		 * Look for the deepest enabled idle state, at most as deep as
+		 * the one with the maximum intercepts metric, whose target
+		 * residency had not been greater than the idle duration in over
+		 * a half of the relevant cases in the past.
 		 *
 		 * Take the possible duration limitation present if the tick
 		 * has been stopped already into account.
@@ -405,11 +463,13 @@ static int teo_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 				continue;
 
 			idx = i;
-			if (2 * intercept_sum > idx_intercept_sum)
+			if (2 * intercept_sum > idx_intercept_sum &&
+			    i <= intercept_max_idx)
 				break;
 		}
 	}
 
+constraint:
 	/*
 	 * If there is a latency constraint, it may be necessary to select an
 	 * idle state shallower than the current candidate one.
@@ -464,7 +524,7 @@ static int teo_select(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 	 * total wakeup events, do not stop the tick.
 	 */
 	if (drv->states[idx].target_residency_ns < TICK_NSEC &&
-	    cpu_data->tick_intercepts > cpu_data->total / 2 + cpu_data->total / 8)
+	    3 * cpu_data->tick_intercepts >= 2 * cpu_data->total)
 		duration_ns = TICK_NSEC / 2;
 
 end: