Merge tag 'sched_core_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull scheduler updates from Borislav Petkov:
"Mostly minor things this time; some highlights:

- core-sched: Add 'Forced Idle' accounting; this allows to track how
much CPU time is 'lost' due to core scheduling constraints.

- psi: Fix for MEM_FULL; a task running reclaim would be counted as a
runnable task and prevent MEM_FULL from being reported.

- cpuacct: Long standing fixes for some cgroup accounting issues.

- rt: Bandwidth timer could, under unusual circumstances, be failed
to armed, leading to indefinite throttling."

[ Description above by Peter Zijlstra ]

* tag 'sched_core_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Replace CFS internal cpu_util() with cpu_util_cfs()
sched/fair: Cleanup task_util and capacity type
sched/rt: Try to restart rt period timer when rt runtime exceeded
sched/fair: Document the slow path and fast path in select_task_rq_fair
sched/fair: Fix per-CPU kthread and wakee stacking for asym CPU capacity
sched/fair: Fix detection of per-CPU kthreads waking a task
sched/cpuacct: Make user/system times in cpuacct.stat more precise
sched/cpuacct: Fix user/system in shown cpuacct.usage*
cpuacct: Convert BUG_ON() to WARN_ON_ONCE()
cputime, cpuacct: Include guest time in user time in cpuacct.stat
psi: Fix PSI_MEM_FULL state when tasks are in memstall and doing reclaim
sched/core: Forced idle accounting
psi: Add a missing SPDX license header
psi: Remove repeated verbose comment

include/linux/psi.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef _LINUX_PSI_H
 #define _LINUX_PSI_H
 

include/linux/psi_types.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
 #ifndef _LINUX_PSI_TYPES_H
 #define _LINUX_PSI_TYPES_H
 
···
 	 * don't have to special case any state tracking for it.
 	 */
 	NR_ONCPU,
-	NR_PSI_TASK_COUNTS = 4,
+	/*
+	 * For IO and CPU stalls the presence of running/oncpu tasks
+	 * in the domain means a partial rather than a full stall.
+	 * For memory it's not so simple because of page reclaimers:
+	 * they are running/oncpu while representing a stall. To tell
+	 * whether a domain has productivity left or not, we need to
+	 * distinguish between regular running (i.e. productive)
+	 * threads and memstall ones.
+	 */
+	NR_MEMSTALL_RUNNING,
+	NR_PSI_TASK_COUNTS = 5,
 };
 
 /* Task state bitmasks */
···
 #define TSK_MEMSTALL	(1 << NR_MEMSTALL)
 #define TSK_RUNNING	(1 << NR_RUNNING)
 #define TSK_ONCPU	(1 << NR_ONCPU)
+#define TSK_MEMSTALL_RUNNING	(1 << NR_MEMSTALL_RUNNING)
 
 /* Resources that workloads could be stalled on */
 enum psi_res {

include/linux/sched.h

···
 	u64				nr_wakeups_affine_attempts;
 	u64				nr_wakeups_passive;
 	u64				nr_wakeups_idle;
+
+#ifdef CONFIG_SCHED_CORE
+	u64				core_forceidle_sum;
 #endif
+#endif /* CONFIG_SCHEDSTATS */
 } ____cacheline_aligned;
 
 struct sched_entity {

kernel/sched/core.c

···
 		return false;
 
 	/* flip prio, so high prio is leftmost */
-	if (prio_less(b, a, task_rq(a)->core->core_forceidle))
+	if (prio_less(b, a, !!task_rq(a)->core->core_forceidle_count))
 		return true;
 
 	return false;
···
 	rb_add(&p->core_node, &rq->core_tree, rb_sched_core_less);
 }
 
-void sched_core_dequeue(struct rq *rq, struct task_struct *p)
+void sched_core_dequeue(struct rq *rq, struct task_struct *p, int flags)
 {
 	rq->core->core_task_seq++;
 
-	if (!sched_core_enqueued(p))
-		return;
+	if (sched_core_enqueued(p)) {
+		rb_erase(&p->core_node, &rq->core_tree);
+		RB_CLEAR_NODE(&p->core_node);
+	}
 
-	rb_erase(&p->core_node, &rq->core_tree);
-	RB_CLEAR_NODE(&p->core_node);
+	/*
+	 * Migrating the last task off the cpu, with the cpu in forced idle
+	 * state. Reschedule to create an accounting edge for forced idle,
+	 * and re-examine whether the core is still in forced idle state.
+	 */
+	if (!(flags & DEQUEUE_SAVE) && rq->nr_running == 1 &&
+	    rq->core->core_forceidle_count && rq->curr == rq->idle)
+		resched_curr(rq);
 }
 
 /*
···
 		for_each_cpu(t, smt_mask)
 			cpu_rq(t)->core_enabled = enabled;
 
+		cpu_rq(cpu)->core->core_forceidle_start = 0;
+
 		sched_core_unlock(cpu, &flags);
 
 		cpumask_andnot(&sched_core_mask, &sched_core_mask, smt_mask);
···
 #else /* !CONFIG_SCHED_CORE */
 
 static inline void sched_core_enqueue(struct rq *rq, struct task_struct *p) { }
-static inline void sched_core_dequeue(struct rq *rq, struct task_struct *p) { }
+static inline void
+sched_core_dequeue(struct rq *rq, struct task_struct *p, int flags) { }
 
 #endif /* CONFIG_SCHED_CORE */
 
···
 static inline void dequeue_task(struct rq *rq, struct task_struct *p, int flags)
 {
 	if (sched_core_enabled(rq))
-		sched_core_dequeue(rq, p);
+		sched_core_dequeue(rq, p, flags);
 
 	if (!(flags & DEQUEUE_NOCLOCK))
 		update_rq_clock(rq);
···
 	if (sched_feat(LATENCY_WARN))
 		resched_latency = cpu_resched_latency(rq);
 	calc_global_load_tick(rq);
+	sched_core_tick(rq);
 
 	rq_unlock(rq, &rf);
 
···
 	struct task_struct *next, *p, *max = NULL;
 	const struct cpumask *smt_mask;
 	bool fi_before = false;
+	bool core_clock_updated = (rq == rq->core);
 	unsigned long cookie;
 	int i, cpu, occ = 0;
 	struct rq *rq_i;
···
 
 	/* reset state */
 	rq->core->core_cookie = 0UL;
-	if (rq->core->core_forceidle) {
+	if (rq->core->core_forceidle_count) {
+		if (!core_clock_updated) {
+			update_rq_clock(rq->core);
+			core_clock_updated = true;
+		}
+		sched_core_account_forceidle(rq);
+		/* reset after accounting force idle */
+		rq->core->core_forceidle_start = 0;
+		rq->core->core_forceidle_count = 0;
+		rq->core->core_forceidle_occupation = 0;
 		need_sync = true;
 		fi_before = true;
-		rq->core->core_forceidle = false;
 	}
 
 	/*
···
 	for_each_cpu_wrap(i, smt_mask, cpu) {
 		rq_i = cpu_rq(i);
 
-		if (i != cpu)
+		/*
+		 * Current cpu always has its clock updated on entrance to
+		 * pick_next_task(). If the current cpu is not the core,
+		 * the core may also have been updated above.
+		 */
+		if (i != cpu && (rq_i != rq->core || !core_clock_updated))
 			update_rq_clock(rq_i);
 
 		p = rq_i->core_pick = pick_task(rq_i);
···
 
 		if (p == rq_i->idle) {
 			if (rq_i->nr_running) {
-				rq->core->core_forceidle = true;
+				rq->core->core_forceidle_count++;
 				if (!fi_before)
 					rq->core->core_forceidle_seq++;
 			}
 		} else {
 			occ++;
 		}
+	}
+
+	if (schedstat_enabled() && rq->core->core_forceidle_count) {
+		if (cookie)
+			rq->core->core_forceidle_start = rq_clock(rq->core);
+		rq->core->core_forceidle_occupation = occ;
 	}
 
 	rq->core->core_pick_seq = rq->core->core_task_seq;
···
 		 *  1            0       1
 		 *  1            1       0
 		 */
-		if (!(fi_before && rq->core->core_forceidle))
-			task_vruntime_update(rq_i, rq_i->core_pick, rq->core->core_forceidle);
+		if (!(fi_before && rq->core->core_forceidle_count))
+			task_vruntime_update(rq_i, rq_i->core_pick, !!rq->core->core_forceidle_count);
 
 		rq_i->core_pick->core_occupation = occ;
 
···
 		goto unlock;
 
 	/* copy the shared state to the new leader */
-	core_rq->core_task_seq      = rq->core_task_seq;
-	core_rq->core_pick_seq      = rq->core_pick_seq;
-	core_rq->core_cookie        = rq->core_cookie;
-	core_rq->core_forceidle     = rq->core_forceidle;
-	core_rq->core_forceidle_seq = rq->core_forceidle_seq;
+	core_rq->core_task_seq             = rq->core_task_seq;
+	core_rq->core_pick_seq             = rq->core_pick_seq;
+	core_rq->core_cookie               = rq->core_cookie;
+	core_rq->core_forceidle_count      = rq->core_forceidle_count;
+	core_rq->core_forceidle_seq        = rq->core_forceidle_seq;
+	core_rq->core_forceidle_occupation = rq->core_forceidle_occupation;
+
+	/*
+	 * Accounting edge for forced idle is handled in pick_next_task().
+	 * Don't need another one here, since the hotplug thread shouldn't
+	 * have a cookie.
+	 */
+	core_rq->core_forceidle_start = 0;
 
 	/* install new leader */
 	for_each_cpu(t, smt_mask) {
···
 
 unsigned long sched_cpu_util(int cpu, unsigned long max)
 {
-	return effective_cpu_util(cpu, cpu_util_cfs(cpu_rq(cpu)), max,
+	return effective_cpu_util(cpu, cpu_util_cfs(cpu), max,
 				  ENERGY_UTIL, NULL);
 }
 #endif /* CONFIG_SMP */
···
 		rq->core_pick = NULL;
 		rq->core_enabled = 0;
 		rq->core_tree = RB_ROOT;
-		rq->core_forceidle = false;
+		rq->core_forceidle_count = 0;
+		rq->core_forceidle_occupation = 0;
+		rq->core_forceidle_start = 0;
 
 		rq->core_cookie = 0UL;
 #endif

kernel/sched/core_sched.c

···
 
 	enqueued = sched_core_enqueued(p);
 	if (enqueued)
-		sched_core_dequeue(rq, p);
+		sched_core_dequeue(rq, p, DEQUEUE_SAVE);
 
 	old_cookie = p->core_cookie;
 	p->core_cookie = cookie;
···
 	 * If task is currently running, it may not be compatible anymore after
 	 * the cookie change, so enter the scheduler on its CPU to schedule it
 	 * away.
+	 *
+	 * Note that it is possible that as a result of this cookie change, the
+	 * core has now entered/left forced idle state. Defer accounting to the
+	 * next scheduling edge, rather than always forcing a reschedule here.
 	 */
 	if (task_running(rq, p))
 		resched_curr(rq);
···
 	return err;
 }
 
+#ifdef CONFIG_SCHEDSTATS
+
+/* REQUIRES: rq->core's clock recently updated. */
+void __sched_core_account_forceidle(struct rq *rq)
+{
+	const struct cpumask *smt_mask = cpu_smt_mask(cpu_of(rq));
+	u64 delta, now = rq_clock(rq->core);
+	struct rq *rq_i;
+	struct task_struct *p;
+	int i;
+
+	lockdep_assert_rq_held(rq);
+
+	WARN_ON_ONCE(!rq->core->core_forceidle_count);
+
+	if (rq->core->core_forceidle_start == 0)
+		return;
+
+	delta = now - rq->core->core_forceidle_start;
+	if (unlikely((s64)delta <= 0))
+		return;
+
+	rq->core->core_forceidle_start = now;
+
+	if (WARN_ON_ONCE(!rq->core->core_forceidle_occupation)) {
+		/* can't be forced idle without a running task */
+	} else if (rq->core->core_forceidle_count > 1 ||
+		   rq->core->core_forceidle_occupation > 1) {
+		/*
+		 * For larger SMT configurations, we need to scale the charged
+		 * forced idle amount since there can be more than one forced
+		 * idle sibling and more than one running cookied task.
+		 */
+		delta *= rq->core->core_forceidle_count;
+		delta = div_u64(delta, rq->core->core_forceidle_occupation);
+	}
+
+	for_each_cpu(i, smt_mask) {
+		rq_i = cpu_rq(i);
+		p = rq_i->core_pick ?: rq_i->curr;
+
+		if (!p->core_cookie)
+			continue;
+
+		__schedstat_add(p->stats.core_forceidle_sum, delta);
+	}
+}
+
+void __sched_core_tick(struct rq *rq)
+{
+	if (!rq->core->core_forceidle_count)
+		return;
+
+	if (rq != rq->core)
+		update_rq_clock(rq->core);
+
+	__sched_core_account_forceidle(rq);
+}
+
+#endif /* CONFIG_SCHEDSTATS */

kernel/sched/cpuacct.c

···
 	[CPUACCT_STAT_SYSTEM] = "system",
 };
 
-struct cpuacct_usage {
-	u64	usages[CPUACCT_STAT_NSTATS];
-};
-
 /* track CPU usage of a group of tasks and its child groups */
 struct cpuacct {
 	struct cgroup_subsys_state	css;
 	/* cpuusage holds pointer to a u64-type object on every CPU */
-	struct cpuacct_usage __percpu	*cpuusage;
+	u64 __percpu	*cpuusage;
 	struct kernel_cpustat __percpu	*cpustat;
 };
 
···
 	return css_ca(ca->css.parent);
 }
 
-static DEFINE_PER_CPU(struct cpuacct_usage, root_cpuacct_cpuusage);
+static DEFINE_PER_CPU(u64, root_cpuacct_cpuusage);
 static struct cpuacct root_cpuacct = {
 	.cpustat	= &kernel_cpustat,
 	.cpuusage	= &root_cpuacct_cpuusage,
···
 	if (!ca)
 		goto out;
 
-	ca->cpuusage = alloc_percpu(struct cpuacct_usage);
+	ca->cpuusage = alloc_percpu(u64);
 	if (!ca->cpuusage)
 		goto out_free_ca;
 
···
 static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu,
 				 enum cpuacct_stat_index index)
 {
-	struct cpuacct_usage *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
+	u64 *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
+	u64 *cpustat = per_cpu_ptr(ca->cpustat, cpu)->cpustat;
 	u64 data;
 
 	/*
 	 * We allow index == CPUACCT_STAT_NSTATS here to read
 	 * the sum of usages.
 	 */
-	BUG_ON(index > CPUACCT_STAT_NSTATS);
+	if (WARN_ON_ONCE(index > CPUACCT_STAT_NSTATS))
+		return 0;
 
 #ifndef CONFIG_64BIT
 	/*
···
 	raw_spin_rq_lock_irq(cpu_rq(cpu));
 #endif
 
-	if (index == CPUACCT_STAT_NSTATS) {
-		int i = 0;
-
-		data = 0;
-		for (i = 0; i < CPUACCT_STAT_NSTATS; i++)
-			data += cpuusage->usages[i];
-	} else {
-		data = cpuusage->usages[index];
+	switch (index) {
+	case CPUACCT_STAT_USER:
+		data = cpustat[CPUTIME_USER] + cpustat[CPUTIME_NICE];
+		break;
+	case CPUACCT_STAT_SYSTEM:
+		data = cpustat[CPUTIME_SYSTEM] + cpustat[CPUTIME_IRQ] +
+			cpustat[CPUTIME_SOFTIRQ];
+		break;
+	case CPUACCT_STAT_NSTATS:
+		data = *cpuusage;
+		break;
 	}
 
 #ifndef CONFIG_64BIT
···
 	return data;
 }
 
-static void cpuacct_cpuusage_write(struct cpuacct *ca, int cpu, u64 val)
+static void cpuacct_cpuusage_write(struct cpuacct *ca, int cpu)
 {
-	struct cpuacct_usage *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
-	int i;
+	u64 *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
+	u64 *cpustat = per_cpu_ptr(ca->cpustat, cpu)->cpustat;
+
+	/* Don't allow to reset global kernel_cpustat */
+	if (ca == &root_cpuacct)
+		return;
 
 #ifndef CONFIG_64BIT
 	/*
···
 	 */
 	raw_spin_rq_lock_irq(cpu_rq(cpu));
 #endif
-
-	for (i = 0; i < CPUACCT_STAT_NSTATS; i++)
-		cpuusage->usages[i] = val;
+	*cpuusage = 0;
+	cpustat[CPUTIME_USER] = cpustat[CPUTIME_NICE] = 0;
+	cpustat[CPUTIME_SYSTEM] = cpustat[CPUTIME_IRQ] = 0;
+	cpustat[CPUTIME_SOFTIRQ] = 0;
 
 #ifndef CONFIG_64BIT
 	raw_spin_rq_unlock_irq(cpu_rq(cpu));
···
 		return -EINVAL;
 
 	for_each_possible_cpu(cpu)
-		cpuacct_cpuusage_write(ca, cpu, 0);
+		cpuacct_cpuusage_write(ca, cpu);
 
 	return 0;
 }
···
 	seq_puts(m, "\n");
 
 	for_each_possible_cpu(cpu) {
-		struct cpuacct_usage *cpuusage = per_cpu_ptr(ca->cpuusage, cpu);
-
 		seq_printf(m, "%d", cpu);
-
-		for (index = 0; index < CPUACCT_STAT_NSTATS; index++) {
-#ifndef CONFIG_64BIT
-			/*
-			 * Take rq->lock to make 64-bit read safe on 32-bit
-			 * platforms.
-			 */
-			raw_spin_rq_lock_irq(cpu_rq(cpu));
-#endif
-
-			seq_printf(m, " %llu", cpuusage->usages[index]);
-
-#ifndef CONFIG_64BIT
-			raw_spin_rq_unlock_irq(cpu_rq(cpu));
-#endif
-		}
+		for (index = 0; index < CPUACCT_STAT_NSTATS; index++)
+			seq_printf(m, " %llu",
+				   cpuacct_cpuusage_read(ca, cpu, index));
 		seq_puts(m, "\n");
 	}
 	return 0;
···
 static int cpuacct_stats_show(struct seq_file *sf, void *v)
 {
 	struct cpuacct *ca = css_ca(seq_css(sf));
-	s64 val[CPUACCT_STAT_NSTATS];
+	struct task_cputime cputime;
+	u64 val[CPUACCT_STAT_NSTATS];
 	int cpu;
 	int stat;
 
-	memset(val, 0, sizeof(val));
+	memset(&cputime, 0, sizeof(cputime));
 	for_each_possible_cpu(cpu) {
 		u64 *cpustat = per_cpu_ptr(ca->cpustat, cpu)->cpustat;
 
-		val[CPUACCT_STAT_USER]   += cpustat[CPUTIME_USER];
-		val[CPUACCT_STAT_USER]   += cpustat[CPUTIME_NICE];
-		val[CPUACCT_STAT_SYSTEM] += cpustat[CPUTIME_SYSTEM];
-		val[CPUACCT_STAT_SYSTEM] += cpustat[CPUTIME_IRQ];
-		val[CPUACCT_STAT_SYSTEM] += cpustat[CPUTIME_SOFTIRQ];
+		cputime.utime += cpustat[CPUTIME_USER];
+		cputime.utime += cpustat[CPUTIME_NICE];
+		cputime.stime += cpustat[CPUTIME_SYSTEM];
+		cputime.stime += cpustat[CPUTIME_IRQ];
+		cputime.stime += cpustat[CPUTIME_SOFTIRQ];
+
+		cputime.sum_exec_runtime += *per_cpu_ptr(ca->cpuusage, cpu);
 	}
 
+	cputime_adjust(&cputime, &seq_css(sf)->cgroup->prev_cputime,
+		&val[CPUACCT_STAT_USER], &val[CPUACCT_STAT_SYSTEM]);
+
 	for (stat = 0; stat < CPUACCT_STAT_NSTATS; stat++) {
-		seq_printf(sf, "%s %lld\n",
-			   cpuacct_stat_desc[stat],
-			   (long long)nsec_to_clock_t(val[stat]));
+		seq_printf(sf, "%s %llu\n", cpuacct_stat_desc[stat],
+			nsec_to_clock_t(val[stat]));
 	}
 
 	return 0;
···
 void cpuacct_charge(struct task_struct *tsk, u64 cputime)
 {
 	struct cpuacct *ca;
-	int index = CPUACCT_STAT_SYSTEM;
-	struct pt_regs *regs = get_irq_regs() ? : task_pt_regs(tsk);
-
-	if (regs && user_mode(regs))
-		index = CPUACCT_STAT_USER;
 
 	rcu_read_lock();
 
 	for (ca = task_ca(tsk); ca; ca = parent_ca(ca))
-		__this_cpu_add(ca->cpuusage->usages[index], cputime);
+		__this_cpu_add(*ca->cpuusage, cputime);
 
 	rcu_read_unlock();
 }

kernel/sched/cpufreq_schedutil.c

···
 
 	sg_cpu->max = max;
 	sg_cpu->bw_dl = cpu_bw_dl(rq);
-	sg_cpu->util = effective_cpu_util(sg_cpu->cpu, cpu_util_cfs(rq), max,
+	sg_cpu->util = effective_cpu_util(sg_cpu->cpu, cpu_util_cfs(sg_cpu->cpu), max,
 					  FREQUENCY_UTIL, NULL);
 }
 

kernel/sched/cputime.c

···
 
 	/* Add guest time to cpustat. */
 	if (task_nice(p) > 0) {
-		cpustat[CPUTIME_NICE] += cputime;
+		task_group_account_field(p, CPUTIME_NICE, cputime);
 		cpustat[CPUTIME_GUEST_NICE] += cputime;
 	} else {
-		cpustat[CPUTIME_USER] += cputime;
+		task_group_account_field(p, CPUTIME_USER, cputime);
 		cpustat[CPUTIME_GUEST] += cputime;
 	}
 }

kernel/sched/debug.c

···
 
 		__PN(avg_atom);
 		__PN(avg_per_cpu);
+
+#ifdef CONFIG_SCHED_CORE
+		PN_SCHEDSTAT(core_forceidle_sum);
+#endif
 	}
 
 	__P(nr_switches);

kernel/sched/fair.c

···
 
 static unsigned long cpu_load(struct rq *rq);
 static unsigned long cpu_runnable(struct rq *rq);
-static unsigned long cpu_util(int cpu);
 static inline long adjust_numa_imbalance(int imbalance,
 					int dst_running, int dst_weight);
 
···
 
 		ns->load += cpu_load(rq);
 		ns->runnable += cpu_runnable(rq);
-		ns->util += cpu_util(cpu);
+		ns->util += cpu_util_cfs(cpu);
 		ns->nr_running += rq->cfs.h_nr_running;
 		ns->compute_capacity += capacity_of(cpu);
 
···
 		 * As is, the util number is not freq-invariant (we'd have to
 		 * implement arch_scale_freq_capacity() for that).
 		 *
-		 * See cpu_util().
+		 * See cpu_util_cfs().
 		 */
 		cpufreq_update_util(rq, flags);
 	}
···
 	trace_sched_util_est_se_tp(&p->se);
 }
 
-static inline int task_fits_capacity(struct task_struct *p, long capacity)
+static inline int task_fits_capacity(struct task_struct *p,
+				     unsigned long capacity)
 {
 	return fits_capacity(uclamp_task_util(p), capacity);
 }
···
 #endif
 
 #ifdef CONFIG_SMP
-static inline unsigned long cpu_util(int cpu);
-
 static inline bool cpu_overutilized(int cpu)
 {
-	return !fits_capacity(cpu_util(cpu), capacity_of(cpu));
+	return !fits_capacity(cpu_util_cfs(cpu), capacity_of(cpu));
 }
 
 static inline void update_overutilized_status(struct rq *rq)
···
 	return best_cpu;
 }
 
-static inline bool asym_fits_capacity(int task_util, int cpu)
+static inline bool asym_fits_capacity(unsigned long task_util, int cpu)
 {
 	if (static_branch_unlikely(&sched_asym_cpucapacity))
 		return fits_capacity(task_util, capacity_of(cpu));
···
 	 * pattern is IO completions.
 	 */
 	if (is_per_cpu_kthread(current) &&
+	    in_task() &&
 	    prev == smp_processor_id() &&
-	    this_rq()->nr_running <= 1) {
+	    this_rq()->nr_running <= 1 &&
+	    asym_fits_capacity(task_util, prev)) {
 		return prev;
 	}
 
···
 	return target;
 }
 
-/**
- * cpu_util - Estimates the amount of capacity of a CPU used by CFS tasks.
- * @cpu: the CPU to get the utilization of
- *
- * The unit of the return value must be the one of capacity so we can compare
- * the utilization with the capacity of the CPU that is available for CFS task
- * (ie cpu_capacity).
- *
- * cfs_rq.avg.util_avg is the sum of running time of runnable tasks plus the
- * recent utilization of currently non-runnable tasks on a CPU. It represents
- * the amount of utilization of a CPU in the range [0..capacity_orig] where
- * capacity_orig is the cpu_capacity available at the highest frequency
- * (arch_scale_freq_capacity()).
- * The utilization of a CPU converges towards a sum equal to or less than the
- * current capacity (capacity_curr <= capacity_orig) of the CPU because it is
- * the running time on this CPU scaled by capacity_curr.
- *
- * The estimated utilization of a CPU is defined to be the maximum between its
- * cfs_rq.avg.util_avg and the sum of the estimated utilization of the tasks
- * currently RUNNABLE on that CPU.
- * This allows to properly represent the expected utilization of a CPU which
- * has just got a big task running since a long sleep period. At the same time
- * however it preserves the benefits of the "blocked utilization" in
- * describing the potential for other tasks waking up on the same CPU.
- *
- * Nevertheless, cfs_rq.avg.util_avg can be higher than capacity_curr or even
- * higher than capacity_orig because of unfortunate rounding in
- * cfs.avg.util_avg or just after migrating tasks and new task wakeups until
- * the average stabilizes with the new running time. We need to check that the
- * utilization stays within the range of [0..capacity_orig] and cap it if
- * necessary. Without utilization capping, a group could be seen as overloaded
- * (CPU0 utilization at 121% + CPU1 utilization at 80%) whereas CPU1 has 20% of
- * available capacity. We allow utilization to overshoot capacity_curr (but not
- * capacity_orig) as it useful for predicting the capacity required after task
- * migrations (scheduler-driven DVFS).
- *
- * Return: the (estimated) utilization for the specified CPU
- */
-static inline unsigned long cpu_util(int cpu)
-{
-	struct cfs_rq *cfs_rq;
-	unsigned int util;
-
-	cfs_rq = &cpu_rq(cpu)->cfs;
-	util = READ_ONCE(cfs_rq->avg.util_avg);
-
-	if (sched_feat(UTIL_EST))
-		util = max(util, READ_ONCE(cfs_rq->avg.util_est.enqueued));
-
-	return min_t(unsigned long, util, capacity_orig_of(cpu));
-}
-
 /*
  * cpu_util_without: compute cpu utilization without any contributions from *p
  * @cpu: the CPU which utilization is requested
···
 
 	/* Task has no contribution or is new */
 	if (cpu != task_cpu(p) || !READ_ONCE(p->se.avg.last_update_time))
-		return cpu_util(cpu);
+		return cpu_util_cfs(cpu);
 
 	cfs_rq = &cpu_rq(cpu)->cfs;
 	util = READ_ONCE(cfs_rq->avg.util_avg);
···
 	/*
 	 * Utilization (estimated) can exceed the CPU capacity, thus let's
 	 * clamp to the maximum CPU capacity to ensure consistency with
-	 * the cpu_util call.
+	 * cpu_util.
 	 */
 	return min_t(unsigned long, util, capacity_orig_of(cpu));
 }
···
 		 * During wake-up, the task isn't enqueued yet and doesn't
 		 * appear in the cfs_rq->avg.util_est.enqueued of any rq,
 		 * so just add it (if needed) to "simulate" what will be
-		 * cpu_util() after the task has been enqueued.
+		 * cpu_util after the task has been enqueued.
 		 */
 		if (dst_cpu == cpu)
 			util_est += _task_util_est(p);
···
 			break;
 		}
 
+		/*
+		 * Usually only true for WF_EXEC and WF_FORK, as sched_domains
+		 * usually do not have SD_BALANCE_WAKE set. That means wakeup
+		 * will usually go to the fast path.
+		 */
 		if (tmp->flags & sd_flag)
 			sd = tmp;
 		else if (!want_affine)
···
 		struct rq *rq = cpu_rq(i);
 
 		sgs->group_load += cpu_load(rq);
-		sgs->group_util += cpu_util(i);
+		sgs->group_util += cpu_util_cfs(i);
 		sgs->group_runnable += cpu_runnable(rq);
 		sgs->sum_h_nr_running += rq->cfs.h_nr_running;
 
···
 			break;
 
 		case migrate_util:
-			util = cpu_util(cpu_of(rq));
+			util = cpu_util_cfs(i);
 
 			/*
 			 * Don't try to pull utilization from a CPU with one
···
 	 * MIN_NR_TASKS_DURING_FORCEIDLE - 1 tasks and use that to check
 	 * if we need to give up the CPU.
 	 */
-	if (rq->core->core_forceidle && rq->cfs.nr_running == 1 &&
+	if (rq->core->core_forceidle_count && rq->cfs.nr_running == 1 &&
 	    __entity_slice_used(&curr->se, MIN_NR_TASKS_DURING_FORCEIDLE))
 		resched_curr(rq);
 }

kernel/sched/psi.c

···
+// SPDX-License-Identifier: GPL-2.0
 /*
  * Pressure stall information for CPU, memory and IO
  *
···
  * delayed on that resource such that nobody is advancing and the CPU
  * goes idle. This leaves both workload and CPU unproductive.
  *
- * Naturally, the FULL state doesn't exist for the CPU resource at the
- * system level, but exist at the cgroup level, means all non-idle tasks
- * in a cgroup are delayed on the CPU resource which used by others outside
- * of the cgroup or throttled by the cgroup cpu.max configuration.
- *
  *	SOME = nr_delayed_tasks != 0
- *	FULL = nr_delayed_tasks != 0 && nr_running_tasks == 0
+ *	FULL = nr_delayed_tasks != 0 && nr_productive_tasks == 0
+ *
+ * What it means for a task to be productive is defined differently
+ * for each resource. For IO, productive means a running task. For
+ * memory, productive means a running task that isn't a reclaimer. For
+ * CPU, productive means an oncpu task.
+ *
+ * Naturally, the FULL state doesn't exist for the CPU resource at the
+ * system level, but exist at the cgroup level. At the cgroup level,
+ * FULL means all non-idle tasks in the cgroup are delayed on the CPU
+ * resource which is being used by others outside of the cgroup or
+ * throttled by the cgroup cpu.max configuration.
  *
  * The percentage of wallclock time spent in those compound stall
  * states gives pressure numbers between 0 and 100 for each resource,
···
  *
  *	threads = min(nr_nonidle_tasks, nr_cpus)
  *	   SOME = min(nr_delayed_tasks / threads, 1)
- *	   FULL = (threads - min(nr_running_tasks, threads)) / threads
+ *	   FULL = (threads - min(nr_productive_tasks, threads)) / threads
  *
  * For the 257 number crunchers on 256 CPUs, this yields:
  *
  *	threads = min(257, 256)
  *	   SOME = min(1 / 256, 1)             = 0.4%
- *	   FULL = (256 - min(257, 256)) / 256 = 0%
+ *	   FULL = (256 - min(256, 256)) / 256 = 0%
  *
  * For the 1 out of 4 memory-delayed tasks, this yields:
  *
···
  * For each runqueue, we track:
  *
  *	   tSOME[cpu] = time(nr_delayed_tasks[cpu] != 0)
- *	   tFULL[cpu] = time(nr_delayed_tasks[cpu] && !nr_running_tasks[cpu])
+ *	   tFULL[cpu] = time(nr_delayed_tasks[cpu] && !nr_productive_tasks[cpu])
  *	tNONIDLE[cpu] = time(nr_nonidle_tasks[cpu] != 0)
  *
  * and then periodically aggregate:
···
 	case PSI_MEM_SOME:
 		return unlikely(tasks[NR_MEMSTALL]);
 	case PSI_MEM_FULL:
-		return unlikely(tasks[NR_MEMSTALL] && !tasks[NR_RUNNING]);
+		return unlikely(tasks[NR_MEMSTALL] &&
+			tasks[NR_RUNNING] == tasks[NR_MEMSTALL_RUNNING]);
 	case PSI_CPU_SOME:
 		return unlikely(tasks[NR_RUNNING] > tasks[NR_ONCPU]);
 	case PSI_CPU_FULL:
···
 		if (groupc->tasks[t]) {
 			groupc->tasks[t]--;
 		} else if (!psi_bug) {
-			printk_deferred(KERN_ERR "psi: task underflow! cpu=%d t=%d tasks=[%u %u %u %u] clear=%x set=%x\n",
+			printk_deferred(KERN_ERR "psi: task underflow! cpu=%d t=%d tasks=[%u %u %u %u %u] clear=%x set=%x\n",
 					cpu, t, groupc->tasks[0],
 					groupc->tasks[1], groupc->tasks[2],
-					groupc->tasks[3], clear, set);
+					groupc->tasks[3], groupc->tasks[4],
+					clear, set);
 			psi_bug = 1;
 		}
 	}
···
 		/*
 		 * When switching between tasks that have an identical
 		 * runtime state, the cgroup that contains both tasks
-		 * runtime state, the cgroup that contains both tasks
 		 * we reach the first common ancestor. Iterate @next's
 		 * ancestors only until we encounter @prev's ONCPU.
 		 */
···
 		int clear = TSK_ONCPU, set = 0;
 
 		/*
-		 * When we're going to sleep, psi_dequeue() lets us handle
-		 * TSK_RUNNING and TSK_IOWAIT here, where we can combine it
-		 * with TSK_ONCPU and save walking common ancestors twice.
+		 * When we're going to sleep, psi_dequeue() lets us
+		 * handle TSK_RUNNING, TSK_MEMSTALL_RUNNING and
+		 * TSK_IOWAIT here, where we can combine it with
+		 * TSK_ONCPU and save walking common ancestors twice.
 		 */
 		if (sleep) {
 			clear |= TSK_RUNNING;
+			if (prev->in_memstall)
+				clear |= TSK_MEMSTALL_RUNNING;
 			if (prev->in_iowait)
 				set |= TSK_IOWAIT;
 		}
···
 	rq = this_rq_lock_irq(&rf);
 
 	current->in_memstall = 1;
-	psi_task_change(current, 0, TSK_MEMSTALL);
+	psi_task_change(current, 0, TSK_MEMSTALL | TSK_MEMSTALL_RUNNING);
 
 	rq_unlock_irq(rq, &rf);
 }
···
 	rq = this_rq_lock_irq(&rf);
 
 	current->in_memstall = 0;
-	psi_task_change(current, TSK_MEMSTALL, 0);
+	psi_task_change(current, TSK_MEMSTALL | TSK_MEMSTALL_RUNNING, 0);
 
 	rq_unlock_irq(rq, &rf);
 }

kernel/sched/rt.c

···
 	rt_b->rt_period_timer.function = sched_rt_period_timer;
 }
 
-static void start_rt_bandwidth(struct rt_bandwidth *rt_b)
+static inline void do_start_rt_bandwidth(struct rt_bandwidth *rt_b)
 {
-	if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF)
-		return;
-
 	raw_spin_lock(&rt_b->rt_runtime_lock);
 	if (!rt_b->rt_period_active) {
 		rt_b->rt_period_active = 1;
···
 				      HRTIMER_MODE_ABS_PINNED_HARD);
 	}
 	raw_spin_unlock(&rt_b->rt_runtime_lock);
+}
+
+static void start_rt_bandwidth(struct rt_bandwidth *rt_b)
+{
+	if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF)
+		return;
+
+	do_start_rt_bandwidth(rt_b);
 }
 
 void init_rt_rq(struct rt_rq *rt_rq)
···
 
 	for_each_sched_rt_entity(rt_se) {
 		struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
+		int exceeded;
 
 		if (sched_rt_runtime(rt_rq) != RUNTIME_INF) {
 			raw_spin_lock(&rt_rq->rt_runtime_lock);
 			rt_rq->rt_time += delta_exec;
-			if (sched_rt_runtime_exceeded(rt_rq))
+			exceeded = sched_rt_runtime_exceeded(rt_rq);
+			if (exceeded)
 				resched_curr(rq);
 			raw_spin_unlock(&rt_rq->rt_runtime_lock);
+			if (exceeded)
+				do_start_rt_bandwidth(sched_rt_bandwidth(rt_rq));
 		}
 	}
 }
···
 
 static void sched_rt_do_global(void)
 {
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags);
 	def_rt_bandwidth.rt_runtime = global_rt_runtime();
 	def_rt_bandwidth.rt_period = ns_to_ktime(global_rt_period());
+	raw_spin_unlock_irqrestore(&def_rt_bandwidth.rt_runtime_lock, flags);
 }
 
 int sched_rt_handler(struct ctl_table *table, int write, void *buffer,

kernel/sched/sched.h

···
 	unsigned int		core_task_seq;
 	unsigned int		core_pick_seq;
 	unsigned long		core_cookie;
-	unsigned char		core_forceidle;
+	unsigned int		core_forceidle_count;
 	unsigned int		core_forceidle_seq;
+	unsigned int		core_forceidle_occupation;
+	u64			core_forceidle_start;
 #endif
 };
 
···
 }
 
 extern void sched_core_enqueue(struct rq *rq, struct task_struct *p);
-extern void sched_core_dequeue(struct rq *rq, struct task_struct *p);
+extern void sched_core_dequeue(struct rq *rq, struct task_struct *p, int flags);
 
 extern void sched_core_get(void);
 extern void sched_core_put(void);
···
 
 #include "stats.h"
 #include "autogroup.h"
+
+#if defined(CONFIG_SCHED_CORE) && defined(CONFIG_SCHEDSTATS)
+
+extern void __sched_core_account_forceidle(struct rq *rq);
+
+static inline void sched_core_account_forceidle(struct rq *rq)
+{
+	if (schedstat_enabled())
+		__sched_core_account_forceidle(rq);
+}
+
+extern void __sched_core_tick(struct rq *rq);
+
+static inline void sched_core_tick(struct rq *rq)
+{
+	if (sched_core_enabled(rq) && schedstat_enabled())
+		__sched_core_tick(rq);
+}
+
+#else
+
+static inline void sched_core_account_forceidle(struct rq *rq) {}
+
+static inline void sched_core_tick(struct rq *rq) {}
+
+#endif /* CONFIG_SCHED_CORE && CONFIG_SCHEDSTATS */
 
 #ifdef CONFIG_CGROUP_SCHED
 
···
 	return READ_ONCE(rq->avg_dl.util_avg);
 }
 
-static inline unsigned long cpu_util_cfs(struct rq *rq)
+/**
+ * cpu_util_cfs() - Estimates the amount of CPU capacity used by CFS tasks.
+ * @cpu: the CPU to get the utilization for.
+ *
+ * The unit of the return value must be the same as the one of CPU capacity
+ * so that CPU utilization can be compared with CPU capacity.
+ *
+ * CPU utilization is the sum of running time of runnable tasks plus the
+ * recent utilization of currently non-runnable tasks on that CPU.
+ * It represents the amount of CPU capacity currently used by CFS tasks in
+ * the range [0..max CPU capacity] with max CPU capacity being the CPU
+ * capacity at f_max.
+ *
+ * The estimated CPU utilization is defined as the maximum between CPU
+ * utilization and sum of the estimated utilization of the currently
+ * runnable tasks on that CPU. It preserves a utilization "snapshot" of
+ * previously-executed tasks, which helps better deduce how busy a CPU will
+ * be when a long-sleeping task wakes up. The contribution to CPU utilization
+ * of such a task would be significantly decayed at this point of time.
+ *
+ * CPU utilization can be higher than the current CPU capacity
+ * (f_curr/f_max * max CPU capacity) or even the max CPU capacity because
+ * of rounding errors as well as task migrations or wakeups of new tasks.
+ * CPU utilization has to be capped to fit into the [0..max CPU capacity]
+ * range. Otherwise a group of CPUs (CPU0 util = 121% + CPU1 util = 80%)
+ * could be seen as over-utilized even though CPU1 has 20% of spare CPU
+ * capacity. CPU utilization is allowed to overshoot current CPU capacity
+ * though since this is useful for predicting the CPU capacity required
+ * after task migrations (scheduler-driven DVFS).
+ *
+ * Return: (Estimated) utilization for the specified CPU.
+ */
+static inline unsigned long cpu_util_cfs(int cpu)
 {
-	unsigned long util = READ_ONCE(rq->cfs.avg.util_avg);
+	struct cfs_rq *cfs_rq;
+	unsigned long util;
+
+	cfs_rq = &cpu_rq(cpu)->cfs;
+	util = READ_ONCE(cfs_rq->avg.util_avg);
 
 	if (sched_feat(UTIL_EST)) {
 		util = max_t(unsigned long, util,
-			     READ_ONCE(rq->cfs.avg.util_est.enqueued));
+			     READ_ONCE(cfs_rq->avg.util_est.enqueued));
 	}
 
-	return util;
+	return min(util, capacity_orig_of(cpu));
 }
 
 static inline unsigned long cpu_util_rt(struct rq *rq)

kernel/sched/stats.h

···
 	if (static_branch_likely(&psi_disabled))
 		return;
 
+	if (p->in_memstall)
+		set |= TSK_MEMSTALL_RUNNING;
+
 	if (!wakeup || p->sched_psi_wake_requeue) {
 		if (p->in_memstall)
 			set |= TSK_MEMSTALL;
···
 		return;
 
 	if (p->in_memstall)
-		clear |= TSK_MEMSTALL;
+		clear |= (TSK_MEMSTALL | TSK_MEMSTALL_RUNNING);
 
 	psi_task_change(p, clear, 0);
 }