Merge tag 'sched-core-2022-12-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull scheduler updates from Ingo Molnar:

 - Implement persistent user-requested affinity: introduce
   affinity_context::user_mask and unconditionally preserve the
   user-requested CPU affinity masks, for long-lived tasks to better
   interact with cpusets & CPU hotplug events over longer timespans,
   without destroying the original affinity intent if the underlying
   topology changes.

 - Uclamp updates: fix relationship between uclamp and fits_capacity()

 - PSI fixes

 - Misc fixes & updates

* tag 'sched-core-2022-12-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched: Clear ttwu_pending after enqueue_task()
  sched/psi: Use task->psi_flags to clear in CPU migration
  sched/psi: Stop relying on timer_pending() for poll_work rescheduling
  sched/psi: Fix avgs_work re-arm in psi_avgs_work()
  sched/psi: Fix possible missing or delayed pending event
  sched: Always clear user_cpus_ptr in do_set_cpus_allowed()
  sched: Enforce user requested affinity
  sched: Always preserve the user requested cpumask
  sched: Introduce affinity_context
  sched: Add __releases annotations to affine_move_task()
  sched/fair: Check if prev_cpu has highest spare cap in feec()
  sched/fair: Consider capacity inversion in util_fits_cpu()
  sched/fair: Detect capacity inversion
  sched/uclamp: Cater for uclamp in find_energy_efficient_cpu()'s early exit condition
  sched/uclamp: Make cpu_overutilized() use util_fits_cpu()
  sched/uclamp: Make asym_fits_capacity() use util_fits_cpu()
  sched/uclamp: Make select_idle_capacity() use util_fits_cpu()
  sched/uclamp: Fix fits_capacity() check in feec()
  sched/uclamp: Make task_fits_capacity() use util_fits_cpu()
  sched/uclamp: Fix relationship between uclamp and migration margin

include/linux/psi_types.h

@@ -72,6 +72,9 @@ enum psi_states {
 /* Use one bit in the state mask to track TSK_ONCPU */
 #define PSI_ONCPU	(1 << NR_PSI_STATES)
 
+/* Flag whether to re-arm avgs_work, see details in get_recent_times() */
+#define PSI_STATE_RESCHEDULE	(1 << (NR_PSI_STATES + 1))
+
 enum psi_aggregators {
 	PSI_AVGS = 0,
 	PSI_POLL,
@@ -177,6 +180,7 @@ struct psi_group {
 	struct timer_list poll_timer;
 	wait_queue_head_t poll_wait;
 	atomic_t poll_wakeup;
+	atomic_t poll_scheduled;
 
 	/* Protects data used by the monitor */
 	struct mutex trigger_lock;

include/linux/sched.h

@@ -888,9 +888,6 @@ struct task_struct {
 	unsigned			sched_reset_on_fork:1;
 	unsigned			sched_contributes_to_load:1;
 	unsigned			sched_migrated:1;
-#ifdef CONFIG_PSI
-	unsigned			sched_psi_wake_requeue:1;
-#endif
 
 	/* Force alignment to the next boundary: */
 	unsigned			:0;

kernel/sched/deadline.c

@@ -2485,8 +2485,7 @@ static void task_woken_dl(struct rq *rq, struct task_struct *p)
 }
 
 static void set_cpus_allowed_dl(struct task_struct *p,
-				const struct cpumask *new_mask,
-				u32 flags)
+				struct affinity_context *ctx)
 {
 	struct root_domain *src_rd;
 	struct rq *rq;
@@ -2501,7 +2500,7 @@ static void set_cpus_allowed_dl(struct task_struct *p,
 	 * update. We already made space for us in the destination
 	 * domain (see cpuset_can_attach()).
 	 */
-	if (!cpumask_intersects(src_rd->span, new_mask)) {
+	if (!cpumask_intersects(src_rd->span, ctx->new_mask)) {
 		struct dl_bw *src_dl_b;
 
 		src_dl_b = dl_bw_of(cpu_of(rq));
@@ -2515,7 +2514,7 @@ static void set_cpus_allowed_dl(struct task_struct *p,
 		raw_spin_unlock(&src_dl_b->lock);
 	}
 
-	set_cpus_allowed_common(p, new_mask, flags);
+	set_cpus_allowed_common(p, ctx);
 }
 
 /* Assumes rq->lock is held */

kernel/sched/psi.c

@@ -189,6 +189,7 @@ static void group_init(struct psi_group *group)
 	INIT_DELAYED_WORK(&group->avgs_work, psi_avgs_work);
 	mutex_init(&group->avgs_lock);
 	/* Init trigger-related members */
+	atomic_set(&group->poll_scheduled, 0);
 	mutex_init(&group->trigger_lock);
 	INIT_LIST_HEAD(&group->triggers);
 	group->poll_min_period = U32_MAX;
@@ -242,6 +243,8 @@ static void get_recent_times(struct psi_group *group, int cpu,
 			     u32 *pchanged_states)
 {
 	struct psi_group_cpu *groupc = per_cpu_ptr(group->pcpu, cpu);
+	int current_cpu = raw_smp_processor_id();
+	unsigned int tasks[NR_PSI_TASK_COUNTS];
 	u64 now, state_start;
 	enum psi_states s;
 	unsigned int seq;
@@ -256,6 +259,8 @@ static void get_recent_times(struct psi_group *group, int cpu,
 		memcpy(times, groupc->times, sizeof(groupc->times));
 		state_mask = groupc->state_mask;
 		state_start = groupc->state_start;
+		if (cpu == current_cpu)
+			memcpy(tasks, groupc->tasks, sizeof(groupc->tasks));
 	} while (read_seqcount_retry(&groupc->seq, seq));
 
 	/* Calculate state time deltas against the previous snapshot */
@@ -280,6 +285,28 @@ static void get_recent_times(struct psi_group *group, int cpu,
 		if (delta)
 			*pchanged_states |= (1 << s);
 	}
+
+	/*
+	 * When collect_percpu_times() from the avgs_work, we don't want to
+	 * re-arm avgs_work when all CPUs are IDLE. But the current CPU running
+	 * this avgs_work is never IDLE, cause avgs_work can't be shut off.
+	 * So for the current CPU, we need to re-arm avgs_work only when
+	 * (NR_RUNNING > 1 || NR_IOWAIT > 0 || NR_MEMSTALL > 0), for other CPUs
+	 * we can just check PSI_NONIDLE delta.
+	 */
+	if (current_work() == &group->avgs_work.work) {
+		bool reschedule;
+
+		if (cpu == current_cpu)
+			reschedule = tasks[NR_RUNNING] +
+				     tasks[NR_IOWAIT] +
+				     tasks[NR_MEMSTALL] > 1;
+		else
+			reschedule = *pchanged_states & (1 << PSI_NONIDLE);
+
+		if (reschedule)
+			*pchanged_states |= PSI_STATE_RESCHEDULE;
+	}
 }
 
 static void calc_avgs(unsigned long avg[3], int missed_periods,
@@ -415,7 +442,6 @@ static void psi_avgs_work(struct work_struct *work)
 	struct delayed_work *dwork;
 	struct psi_group *group;
 	u32 changed_states;
-	bool nonidle;
 	u64 now;
 
 	dwork = to_delayed_work(work);
@@ -426,7 +452,6 @@ static void psi_avgs_work(struct work_struct *work)
 	now = sched_clock();
 
 	collect_percpu_times(group, PSI_AVGS, &changed_states);
-	nonidle = changed_states & (1 << PSI_NONIDLE);
 	/*
 	 * If there is task activity, periodically fold the per-cpu
 	 * times and feed samples into the running averages. If things
@@ -437,7 +462,7 @@ static void psi_avgs_work(struct work_struct *work)
 	if (now >= group->avg_next_update)
 		group->avg_next_update = update_averages(group, now);
 
-	if (nonidle) {
+	if (changed_states & PSI_STATE_RESCHEDULE) {
 		schedule_delayed_work(dwork, nsecs_to_jiffies(
 				group->avg_next_update - now) + 1);
 	}
@@ -539,10 +564,12 @@ static u64 update_triggers(struct psi_group *group, u64 now)
 
 			/* Calculate growth since last update */
 			growth = window_update(&t->win, now, total[t->state]);
-			if (growth < t->threshold)
-				continue;
+			if (!t->pending_event) {
+				if (growth < t->threshold)
+					continue;
 
-			t->pending_event = true;
+				t->pending_event = true;
+			}
 		}
 		/* Limit event signaling to once per window */
 		if (now < t->last_event_time + t->win.size)
@@ -563,18 +590,17 @@ static u64 update_triggers(struct psi_group *group, u64 now)
 	return now + group->poll_min_period;
 }
 
-/* Schedule polling if it's not already scheduled. */
-static void psi_schedule_poll_work(struct psi_group *group, unsigned long delay)
+/* Schedule polling if it's not already scheduled or forced. */
+static void psi_schedule_poll_work(struct psi_group *group, unsigned long delay,
+				   bool force)
 {
 	struct task_struct *task;
 
 	/*
-	 * Do not reschedule if already scheduled.
-	 * Possible race with a timer scheduled after this check but before
-	 * mod_timer below can be tolerated because group->polling_next_update
-	 * will keep updates on schedule.
+	 * atomic_xchg should be called even when !force to provide a
+	 * full memory barrier (see the comment inside psi_poll_work).
 	 */
-	if (timer_pending(&group->poll_timer))
+	if (atomic_xchg(&group->poll_scheduled, 1) && !force)
 		return;
 
 	rcu_read_lock();
@@ -586,12 +612,15 @@ static void psi_schedule_poll_work(struct psi_group *group, unsigned long delay)
 	 */
 	if (likely(task))
 		mod_timer(&group->poll_timer, jiffies + delay);
+	else
+		atomic_set(&group->poll_scheduled, 0);
 
 	rcu_read_unlock();
 }
 
 static void psi_poll_work(struct psi_group *group)
 {
+	bool force_reschedule = false;
 	u32 changed_states;
 	u64 now;
 
@@ -599,6 +628,43 @@ static void psi_poll_work(struct psi_group *group)
 
 	now = sched_clock();
 
+	if (now > group->polling_until) {
+		/*
+		 * We are either about to start or might stop polling if no
+		 * state change was recorded. Resetting poll_scheduled leaves
+		 * a small window for psi_group_change to sneak in and schedule
+		 * an immediate poll_work before we get to rescheduling. One
+		 * potential extra wakeup at the end of the polling window
+		 * should be negligible and polling_next_update still keeps
+		 * updates correctly on schedule.
+		 */
+		atomic_set(&group->poll_scheduled, 0);
+		/*
+		 * A task change can race with the poll worker that is supposed to
+		 * report on it. To avoid missing events, ensure ordering between
+		 * poll_scheduled and the task state accesses, such that if the poll
+		 * worker misses the state update, the task change is guaranteed to
+		 * reschedule the poll worker:
+		 *
+		 * poll worker:
+		 *   atomic_set(poll_scheduled, 0)
+		 *   smp_mb()
+		 *   LOAD states
+		 *
+		 * task change:
+		 *   STORE states
+		 *   if atomic_xchg(poll_scheduled, 1) == 0:
+		 *     schedule poll worker
+		 *
+		 * The atomic_xchg() implies a full barrier.
+		 */
+		smp_mb();
+	} else {
+		/* Polling window is not over, keep rescheduling */
+		force_reschedule = true;
+	}
+
+
 	collect_percpu_times(group, PSI_POLL, &changed_states);
 
 	if (changed_states & group->poll_states) {
@@ -624,7 +690,8 @@ static void psi_poll_work(struct psi_group *group)
 		group->polling_next_update = update_triggers(group, now);
 
 	psi_schedule_poll_work(group,
-		nsecs_to_jiffies(group->polling_next_update - now) + 1);
+		nsecs_to_jiffies(group->polling_next_update - now) + 1,
+		force_reschedule);
 
 out:
 	mutex_unlock(&group->trigger_lock);
@@ -785,7 +852,7 @@ static void psi_group_change(struct psi_group *group, int cpu,
 	write_seqcount_end(&groupc->seq);
 
 	if (state_mask & group->poll_states)
-		psi_schedule_poll_work(group, 1);
+		psi_schedule_poll_work(group, 1, false);
 
 	if (wake_clock && !delayed_work_pending(&group->avgs_work))
 		schedule_delayed_work(&group->avgs_work, PSI_FREQ);
@@ -939,7 +1006,7 @@ void psi_account_irqtime(struct task_struct *task, u32 delta)
 		write_seqcount_end(&groupc->seq);
 
 		if (group->poll_states & (1 << PSI_IRQ_FULL))
-			psi_schedule_poll_work(group, 1);
+			psi_schedule_poll_work(group, 1, false);
 	} while ((group = group->parent));
 }
 #endif
@@ -1325,6 +1392,7 @@ void psi_trigger_destroy(struct psi_trigger *t)
 		 * can no longer be found through group->poll_task.
 		 */
 		kthread_stop(task_to_destroy);
+		atomic_set(&group->poll_scheduled, 0);
 	}
 	kfree(t);
 }

kernel/sched/sched.h

@@ -1041,6 +1041,7 @@ struct rq {
 
 	unsigned long		cpu_capacity;
 	unsigned long		cpu_capacity_orig;
+	unsigned long		cpu_capacity_inverted;
 
 	struct balance_callback *balance_callback;
 
@@ -1150,6 +1151,9 @@ struct rq {
 	unsigned int		core_forceidle_occupation;
 	u64			core_forceidle_start;
 #endif
+
+	/* Scratch cpumask to be temporarily used under rq_lock */
+	cpumask_var_t		scratch_mask;
 };
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
@@ -1877,6 +1881,13 @@ static inline void dirty_sched_domain_sysctl(int cpu)
 #endif
 
 extern int sched_update_scaling(void);
+
+static inline const struct cpumask *task_user_cpus(struct task_struct *p)
+{
+	if (!p->user_cpus_ptr)
+		return cpu_possible_mask; /* &init_task.cpus_mask */
+	return p->user_cpus_ptr;
+}
 #endif /* CONFIG_SMP */
 
 #include "stats.h"
@@ -2144,6 +2155,12 @@ extern const u32		sched_prio_to_wmult[40];
 
 #define RETRY_TASK		((void *)-1UL)
 
+struct affinity_context {
+	const struct cpumask *new_mask;
+	struct cpumask *user_mask;
+	unsigned int flags;
+};
+
 struct sched_class {
 
 #ifdef CONFIG_UCLAMP_TASK
@@ -2172,9 +2189,7 @@ struct sched_class {
 
 	void (*task_woken)(struct rq *this_rq, struct task_struct *task);
 
-	void (*set_cpus_allowed)(struct task_struct *p,
-				 const struct cpumask *newmask,
-				 u32 flags);
+	void (*set_cpus_allowed)(struct task_struct *p, struct affinity_context *ctx);
 
 	void (*rq_online)(struct rq *rq);
 	void (*rq_offline)(struct rq *rq);
@@ -2285,7 +2300,7 @@ extern void update_group_capacity(struct sched_domain *sd, int cpu);
 
 extern void trigger_load_balance(struct rq *rq);
 
-extern void set_cpus_allowed_common(struct task_struct *p, const struct cpumask *new_mask, u32 flags);
+extern void set_cpus_allowed_common(struct task_struct *p, struct affinity_context *ctx);
 
 static inline struct task_struct *get_push_task(struct rq *rq)
 {
@@ -2878,6 +2893,24 @@ static inline unsigned long capacity_orig_of(int cpu)
 	return cpu_rq(cpu)->cpu_capacity_orig;
 }
 
+/*
+ * Returns inverted capacity if the CPU is in capacity inversion state.
+ * 0 otherwise.
+ *
+ * Capacity inversion detection only considers thermal impact where actual
+ * performance points (OPPs) gets dropped.
+ *
+ * Capacity inversion state happens when another performance domain that has
+ * equal or lower capacity_orig_of() becomes effectively larger than the perf
+ * domain this CPU belongs to due to thermal pressure throttling it hard.
+ *
+ * See comment in update_cpu_capacity().
+ */
+static inline unsigned long cpu_in_capacity_inversion(int cpu)
+{
+	return cpu_rq(cpu)->cpu_capacity_inverted;
+}
+
 /**
  * enum cpu_util_type - CPU utilization type
  * @FREQUENCY_UTIL:	Utilization used to select frequency
@@ -2979,6 +3012,23 @@ static inline unsigned long cpu_util_rt(struct rq *rq)
 #ifdef CONFIG_UCLAMP_TASK
 unsigned long uclamp_eff_value(struct task_struct *p, enum uclamp_id clamp_id);
 
+static inline unsigned long uclamp_rq_get(struct rq *rq,
+					  enum uclamp_id clamp_id)
+{
+	return READ_ONCE(rq->uclamp[clamp_id].value);
+}
+
+static inline void uclamp_rq_set(struct rq *rq, enum uclamp_id clamp_id,
+				 unsigned int value)
+{
+	WRITE_ONCE(rq->uclamp[clamp_id].value, value);
+}
+
+static inline bool uclamp_rq_is_idle(struct rq *rq)
+{
+	return rq->uclamp_flags & UCLAMP_FLAG_IDLE;
+}
+
 /**
  * uclamp_rq_util_with - clamp @util with @rq and @p effective uclamp values.
  * @rq:		The rq to clamp against. Must not be NULL.
@@ -3014,12 +3064,12 @@ unsigned long uclamp_rq_util_with(struct rq *rq, unsigned long util,
 		 * Ignore last runnable task's max clamp, as this task will
 		 * reset it. Similarly, no need to read the rq's min clamp.
 		 */
-		if (rq->uclamp_flags & UCLAMP_FLAG_IDLE)
+		if (uclamp_rq_is_idle(rq))
 			goto out;
 	}
 
-	min_util = max_t(unsigned long, min_util, READ_ONCE(rq->uclamp[UCLAMP_MIN].value));
-	max_util = max_t(unsigned long, max_util, READ_ONCE(rq->uclamp[UCLAMP_MAX].value));
+	min_util = max_t(unsigned long, min_util, uclamp_rq_get(rq, UCLAMP_MIN));
+	max_util = max_t(unsigned long, max_util, uclamp_rq_get(rq, UCLAMP_MAX));
 out:
 	/*
 	 * Since CPU's {min,max}_util clamps are MAX aggregated considering
@@ -3060,6 +3110,15 @@ static inline bool uclamp_is_used(void)
 	return static_branch_likely(&sched_uclamp_used);
 }
 #else /* CONFIG_UCLAMP_TASK */
+static inline unsigned long uclamp_eff_value(struct task_struct *p,
+					     enum uclamp_id clamp_id)
+{
+	if (clamp_id == UCLAMP_MIN)
+		return 0;
+
+	return SCHED_CAPACITY_SCALE;
+}
+
 static inline
 unsigned long uclamp_rq_util_with(struct rq *rq, unsigned long util,
 				  struct task_struct *p)
@@ -3073,6 +3132,25 @@ static inline bool uclamp_is_used(void)
 {
 	return false;
 }
+
+static inline unsigned long uclamp_rq_get(struct rq *rq,
+					  enum uclamp_id clamp_id)
+{
+	if (clamp_id == UCLAMP_MIN)
+		return 0;
+
+	return SCHED_CAPACITY_SCALE;
+}
+
+static inline void uclamp_rq_set(struct rq *rq, enum uclamp_id clamp_id,
+				 unsigned int value)
+{
+}
+
+static inline bool uclamp_rq_is_idle(struct rq *rq)
+{
+	return false;
+}
 #endif /* CONFIG_UCLAMP_TASK */
 
 #ifdef CONFIG_HAVE_SCHED_AVG_IRQ

kernel/sched/stats.h

@@ -128,11 +128,9 @@ static inline void psi_enqueue(struct task_struct *p, bool wakeup)
 	if (p->in_memstall)
 		set |= TSK_MEMSTALL_RUNNING;
 
-	if (!wakeup || p->sched_psi_wake_requeue) {
+	if (!wakeup) {
 		if (p->in_memstall)
 			set |= TSK_MEMSTALL;
-		if (p->sched_psi_wake_requeue)
-			p->sched_psi_wake_requeue = 0;
 	} else {
 		if (p->in_iowait)
 			clear |= TSK_IOWAIT;
@@ -143,8 +141,6 @@ static inline void psi_enqueue(struct task_struct *p, bool wakeup)
 
 static inline void psi_dequeue(struct task_struct *p, bool sleep)
 {
-	int clear = TSK_RUNNING;
-
 	if (static_branch_likely(&psi_disabled))
 		return;
 
@@ -157,10 +153,7 @@ static inline void psi_dequeue(struct task_struct *p, bool sleep)
 	if (sleep)
 		return;
 
-	if (p->in_memstall)
-		clear |= (TSK_MEMSTALL | TSK_MEMSTALL_RUNNING);
-
-	psi_task_change(p, clear, 0);
+	psi_task_change(p, p->psi_flags, 0);
 }
 
 static inline void psi_ttwu_dequeue(struct task_struct *p)
@@ -172,19 +165,12 @@ static inline void psi_ttwu_dequeue(struct task_struct *p)
 	 * deregister its sleep-persistent psi states from the old
 	 * queue, and let psi_enqueue() know it has to requeue.
 	 */
-	if (unlikely(p->in_iowait || p->in_memstall)) {
+	if (unlikely(p->psi_flags)) {
 		struct rq_flags rf;
 		struct rq *rq;
-		int clear = 0;
-
-		if (p->in_iowait)
-			clear |= TSK_IOWAIT;
-		if (p->in_memstall)
-			clear |= TSK_MEMSTALL;
 
 		rq = __task_rq_lock(p, &rf);
-		psi_task_change(p, clear, 0);
-		p->sched_psi_wake_requeue = 1;
+		psi_task_change(p, p->psi_flags, 0);
 		__task_rq_unlock(rq, &rf);
 	}
 }