1 files changed, 123 insertions, 56 deletions

diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index b48baaba2fc2..ed2a47e4ddae 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -3,12 +3,8 @@
  * Real-Time Scheduling Class (mapped to the SCHED_FIFO and SCHED_RR
  * policies)
  */
-#include "sched.h"
-
-#include "pelt.h"
 
 int sched_rr_timeslice = RR_TIMESLICE;
-int sysctl_sched_rr_timeslice = (MSEC_PER_SEC / HZ) * RR_TIMESLICE;
 /* More than 4 hours if BW_SHIFT equals 20. */
 static const u64 max_rt_runtime = MAX_BW;
 
@@ -16,6 +12,57 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun);
 
 struct rt_bandwidth def_rt_bandwidth;
 
+/*
+ * period over which we measure -rt task CPU usage in us.
+ * default: 1s
+ */
+unsigned int sysctl_sched_rt_period = 1000000;
+
+/*
+ * part of the period that we allow rt tasks to run in us.
+ * default: 0.95s
+ */
+int sysctl_sched_rt_runtime = 950000;
+
+#ifdef CONFIG_SYSCTL
+static int sysctl_sched_rr_timeslice = (MSEC_PER_SEC / HZ) * RR_TIMESLICE;
+static int sched_rt_handler(struct ctl_table *table, int write, void *buffer,
+		size_t *lenp, loff_t *ppos);
+static int sched_rr_handler(struct ctl_table *table, int write, void *buffer,
+		size_t *lenp, loff_t *ppos);
+static struct ctl_table sched_rt_sysctls[] = {
+	{
+		.procname       = "sched_rt_period_us",
+		.data           = &sysctl_sched_rt_period,
+		.maxlen         = sizeof(unsigned int),
+		.mode           = 0644,
+		.proc_handler   = sched_rt_handler,
+	},
+	{
+		.procname       = "sched_rt_runtime_us",
+		.data           = &sysctl_sched_rt_runtime,
+		.maxlen         = sizeof(int),
+		.mode           = 0644,
+		.proc_handler   = sched_rt_handler,
+	},
+	{
+		.procname       = "sched_rr_timeslice_ms",
+		.data           = &sysctl_sched_rr_timeslice,
+		.maxlen         = sizeof(int),
+		.mode           = 0644,
+		.proc_handler   = sched_rr_handler,
+	},
+	{}
+};
+
+static int __init sched_rt_sysctl_init(void)
+{
+	register_sysctl_init("kernel", sched_rt_sysctls);
+	return 0;
+}
+late_initcall(sched_rt_sysctl_init);
+#endif
+
 static enum hrtimer_restart sched_rt_period_timer(struct hrtimer *timer)
 {
 	struct rt_bandwidth *rt_b =
@@ -52,11 +99,8 @@ void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime)
 	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;
@@ -75,6 +119,14 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b)
 	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)
 {
 	struct rt_prio_array *array;
@@ -266,8 +318,6 @@ int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent)
 
 #ifdef CONFIG_SMP
 
-static void pull_rt_task(struct rq *this_rq);
-
 static inline bool need_pull_rt_task(struct rq *rq, struct task_struct *prev)
 {
 	/* Try to pull RT tasks here if we lower this rq's prio */
@@ -360,8 +410,8 @@ static inline int has_pushable_tasks(struct rq *rq)
 	return !plist_head_empty(&rq->rt.pushable_tasks);
 }
 
-static DEFINE_PER_CPU(struct callback_head, rt_push_head);
-static DEFINE_PER_CPU(struct callback_head, rt_pull_head);
+static DEFINE_PER_CPU(struct balance_callback, rt_push_head);
+static DEFINE_PER_CPU(struct balance_callback, rt_pull_head);
 
 static void push_rt_tasks(struct rq *);
 static void pull_rt_task(struct rq *);
@@ -424,22 +474,13 @@ void dec_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 {
 }
 
-static inline bool need_pull_rt_task(struct rq *rq, struct task_struct *prev)
-{
-	return false;
-}
-
-static inline void pull_rt_task(struct rq *this_rq)
-{
-}
-
 static inline void rt_queue_push_tasks(struct rq *rq)
 {
 }
 #endif /* CONFIG_SMP */
 
 static void enqueue_top_rt_rq(struct rt_rq *rt_rq);
-static void dequeue_top_rt_rq(struct rt_rq *rt_rq);
+static void dequeue_top_rt_rq(struct rt_rq *rt_rq, unsigned int count);
 
 static inline int on_rt_rq(struct sched_rt_entity *rt_se)
 {
@@ -468,7 +509,7 @@ static inline bool rt_task_fits_capacity(struct task_struct *p, int cpu)
 	unsigned int cpu_cap;
 
 	/* Only heterogeneous systems can benefit from this check */
-	if (!static_branch_unlikely(&sched_asym_cpucapacity))
+	if (!sched_asym_cpucap_active())
 		return true;
 
 	min_cap = uclamp_eff_value(p, UCLAMP_MIN);
@@ -560,7 +601,7 @@ static void sched_rt_rq_dequeue(struct rt_rq *rt_rq)
 	rt_se = rt_rq->tg->rt_se[cpu];
 
 	if (!rt_se) {
-		dequeue_top_rt_rq(rt_rq);
+		dequeue_top_rt_rq(rt_rq, rt_rq->rt_nr_running);
 		/* Kick cpufreq (see the comment in kernel/sched/sched.h). */
 		cpufreq_update_util(rq_of_rt_rq(rt_rq), 0);
 	}
@@ -646,7 +687,7 @@ static inline void sched_rt_rq_enqueue(struct rt_rq *rt_rq)
 
 static inline void sched_rt_rq_dequeue(struct rt_rq *rt_rq)
 {
-	dequeue_top_rt_rq(rt_rq);
+	dequeue_top_rt_rq(rt_rq, rt_rq->rt_nr_running);
 }
 
 static inline int rt_rq_throttled(struct rt_rq *rt_rq)
@@ -802,7 +843,7 @@ static void __disable_runtime(struct rq *rq)
 		 * We cannot be left wanting - that would mean some runtime
 		 * leaked out of the system.
 		 */
-		BUG_ON(want);
+		WARN_ON_ONCE(want);
 balanced:
 		/*
 		 * Disable all the borrow logic by pretending we have inf
@@ -880,6 +921,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
 		int enqueue = 0;
 		struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i);
 		struct rq *rq = rq_of_rt_rq(rt_rq);
+		struct rq_flags rf;
 		int skip;
 
 		/*
@@ -894,7 +936,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
 		if (skip)
 			continue;
 
-		raw_spin_rq_lock(rq);
+		rq_lock(rq, &rf);
 		update_rq_clock(rq);
 
 		if (rt_rq->rt_time) {
@@ -932,7 +974,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
 
 		if (enqueue)
 			sched_rt_rq_enqueue(rt_rq);
-		raw_spin_rq_unlock(rq);
+		rq_unlock(rq, &rf);
 	}
 
 	if (!throttled && (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF))
@@ -1020,30 +1062,30 @@ static void update_curr_rt(struct rq *rq)
 
 	trace_sched_stat_runtime(curr, delta_exec, 0);
 
-	curr->se.sum_exec_runtime += delta_exec;
-	account_group_exec_runtime(curr, delta_exec);
-
-	curr->se.exec_start = now;
-	cgroup_account_cputime(curr, delta_exec);
+	update_current_exec_runtime(curr, now, delta_exec);
 
 	if (!rt_bandwidth_enabled())
 		return;
 
 	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
-dequeue_top_rt_rq(struct rt_rq *rt_rq)
+dequeue_top_rt_rq(struct rt_rq *rt_rq, unsigned int count)
 {
 	struct rq *rq = rq_of_rt_rq(rt_rq);
 
@@ -1054,7 +1096,7 @@ dequeue_top_rt_rq(struct rt_rq *rt_rq)
 
 	BUG_ON(!rq->nr_running);
 
-	sub_nr_running(rq, rt_rq->rt_nr_running);
+	sub_nr_running(rq, count);
 	rt_rq->rt_queued = 0;
 
 }
@@ -1440,18 +1482,21 @@ static void __dequeue_rt_entity(struct sched_rt_entity *rt_se, unsigned int flag
 static void dequeue_rt_stack(struct sched_rt_entity *rt_se, unsigned int flags)
 {
 	struct sched_rt_entity *back = NULL;
+	unsigned int rt_nr_running;
 
 	for_each_sched_rt_entity(rt_se) {
 		rt_se->back = back;
 		back = rt_se;
 	}
 
-	dequeue_top_rt_rq(rt_rq_of_se(back));
+	rt_nr_running = rt_rq_of_se(back)->rt_nr_running;
 
 	for (rt_se = back; rt_se; rt_se = rt_se->back) {
 		if (on_rt_rq(rt_se))
 			__dequeue_rt_entity(rt_se, flags);
 	}
+
+	dequeue_top_rt_rq(rt_rq_of_se(back), rt_nr_running);
 }
 
 static void enqueue_rt_entity(struct sched_rt_entity *rt_se, unsigned int flags)
@@ -1721,8 +1766,7 @@ static inline void set_next_task_rt(struct rq *rq, struct task_struct *p, bool f
 	rt_queue_push_tasks(rq);
 }
 
-static struct sched_rt_entity *pick_next_rt_entity(struct rq *rq,
-						   struct rt_rq *rt_rq)
+static struct sched_rt_entity *pick_next_rt_entity(struct rt_rq *rt_rq)
 {
 	struct rt_prio_array *array = &rt_rq->active;
 	struct sched_rt_entity *next = NULL;
@@ -1744,7 +1788,7 @@ static struct task_struct *_pick_next_task_rt(struct rq *rq)
 	struct rt_rq *rt_rq  = &rq->rt;
 
 	do {
-		rt_se = pick_next_rt_entity(rq, rt_rq);
+		rt_se = pick_next_rt_entity(rt_rq);
 		BUG_ON(!rt_se);
 		rt_rq = group_rt_rq(rt_se);
 	} while (rt_rq);
@@ -1801,7 +1845,7 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p)
 
 static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu)
 {
-	if (!task_running(rq, p) &&
+	if (!task_on_cpu(rq, p) &&
 	    cpumask_test_cpu(cpu, &p->cpus_mask))
 		return 1;
 
@@ -1849,7 +1893,7 @@ static int find_lowest_rq(struct task_struct *task)
 	 * If we're on asym system ensure we consider the different capacities
 	 * of the CPUs when searching for the lowest_mask.
 	 */
-	if (static_branch_unlikely(&sched_asym_cpucapacity)) {
+	if (sched_asym_cpucap_active()) {
 
 		ret = cpupri_find_fitness(&task_rq(task)->rd->cpupri,
 					  task, lowest_mask,
@@ -1956,7 +2000,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 			 */
 			if (unlikely(task_rq(task) != rq ||
 				     !cpumask_test_cpu(lowest_rq->cpu, &task->cpus_mask) ||
-				     task_running(rq, task) ||
+				     task_on_cpu(rq, task) ||
 				     !rt_task(task) ||
 				     !task_on_rq_queued(task))) {
 
@@ -2017,6 +2061,16 @@ static int push_rt_task(struct rq *rq, bool pull)
 		return 0;
 
 retry:
+	/*
+	 * It's possible that the next_task slipped in of
+	 * higher priority than current. If that's the case
+	 * just reschedule current.
+	 */
+	if (unlikely(next_task->prio < rq->curr->prio)) {
+		resched_curr(rq);
+		return 0;
+	}
+
 	if (is_migration_disabled(next_task)) {
 		struct task_struct *push_task = NULL;
 		int cpu;
@@ -2024,6 +2078,18 @@ retry:
 		if (!pull || rq->push_busy)
 			return 0;
 
+		/*
+		 * Invoking find_lowest_rq() on anything but an RT task doesn't
+		 * make sense. Per the above priority check, curr has to
+		 * be of higher priority than next_task, so no need to
+		 * reschedule when bailing out.
+		 *
+		 * Note that the stoppers are masqueraded as SCHED_FIFO
+		 * (cf. sched_set_stop_task()), so we can't rely on rt_task().
+		 */
+		if (rq->curr->sched_class != &rt_sched_class)
+			return 0;
+
 		cpu = find_lowest_rq(rq->curr);
 		if (cpu == -1 || cpu == rq->cpu)
 			return 0;
@@ -2048,16 +2114,6 @@ retry:
 	if (WARN_ON(next_task == rq->curr))
 		return 0;
 
-	/*
-	 * It's possible that the next_task slipped in of
-	 * higher priority than current. If that's the case
-	 * just reschedule current.
-	 */
-	if (unlikely(next_task->prio < rq->curr->prio)) {
-		resched_curr(rq);
-		return 0;
-	}
-
 	/* We might release rq lock */
 	get_task_struct(next_task);
 
@@ -2402,7 +2458,7 @@ skip:
  */
 static void task_woken_rt(struct rq *rq, struct task_struct *p)
 {
-	bool need_to_push = !task_running(rq, p) &&
+	bool need_to_push = !task_on_cpu(rq, p) &&
 			    !test_tsk_need_resched(rq->curr) &&
 			    p->nr_cpus_allowed > 1 &&
 			    (dl_task(rq->curr) || rt_task(rq->curr)) &&
@@ -2855,6 +2911,7 @@ long sched_group_rt_period(struct task_group *tg)
 	return rt_period_us;
 }
 
+#ifdef CONFIG_SYSCTL
 static int sched_rt_global_constraints(void)
 {
 	int ret = 0;
@@ -2865,6 +2922,7 @@ static int sched_rt_global_constraints(void)
 
 	return ret;
 }
+#endif /* CONFIG_SYSCTL */
 
 int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk)
 {
@@ -2876,6 +2934,8 @@ int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk)
 }
 
 #else /* !CONFIG_RT_GROUP_SCHED */
+
+#ifdef CONFIG_SYSCTL
 static int sched_rt_global_constraints(void)
 {
 	unsigned long flags;
@@ -2893,8 +2953,10 @@ static int sched_rt_global_constraints(void)
 
 	return 0;
 }
+#endif /* CONFIG_SYSCTL */
 #endif /* CONFIG_RT_GROUP_SCHED */
 
+#ifdef CONFIG_SYSCTL
 static int sched_rt_global_validate(void)
 {
 	if (sysctl_sched_rt_period <= 0)
@@ -2911,11 +2973,15 @@ static int sched_rt_global_validate(void)
 
 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,
+static int sched_rt_handler(struct ctl_table *table, int write, void *buffer,
 		size_t *lenp, loff_t *ppos)
 {
 	int old_period, old_runtime;
@@ -2954,7 +3020,7 @@ undo:
 	return ret;
 }
 
-int sched_rr_handler(struct ctl_table *table, int write, void *buffer,
+static int sched_rr_handler(struct ctl_table *table, int write, void *buffer,
 		size_t *lenp, loff_t *ppos)
 {
 	int ret;
@@ -2975,6 +3041,7 @@ int sched_rr_handler(struct ctl_table *table, int write, void *buffer,
 
 	return ret;
 }
+#endif /* CONFIG_SYSCTL */
 
 #ifdef CONFIG_SCHED_DEBUG
 void print_rt_stats(struct seq_file *m, int cpu)