sched: Trivial forced-newidle balancer

When a sibling is forced-idle to match the core-cookie; search for
matching tasks to fill the core.

rcu_read_unlock() can incur an infrequent deadlock in
sched_core_balance(). Fix this by using the RCU-sched flavor instead.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: NDon Hiatt <dhiatt@digitalocean.com>
Tested-by: NHongyu Ning <hongyu.ning@linux.intel.com>
Tested-by: NVincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20210422123308.800048269@infradead.org

include/linux/sched.h

@@ -705,6 +705,7 @@ struct task_struct {
 #ifdef CONFIG_SCHED_CORE
 	struct rb_node			core_node;
 	unsigned long			core_cookie;
+	unsigned int			core_occupation;
 #endif
 
 #ifdef CONFIG_CGROUP_SCHED

kernel/sched/core.c

@@ -204,6 +204,21 @@ static struct task_struct *sched_core_find(struct rq *rq, unsigned long cookie)
 	return __node_2_sc(node);
 }
 
+static struct task_struct *sched_core_next(struct task_struct *p, unsigned long cookie)
+{
+	struct rb_node *node = &p->core_node;
+
+	node = rb_next(node);
+	if (!node)
+		return NULL;
+
+	p = container_of(node, struct task_struct, core_node);
+	if (p->core_cookie != cookie)
+		return NULL;
+
+	return p;
+}
+
 /*
  * Magic required such that:
  *
@@ -5389,8 +5404,8 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 	const struct sched_class *class;
 	const struct cpumask *smt_mask;
 	bool fi_before = false;
+	int i, j, cpu, occ = 0;
 	bool need_sync;
-	int i, j, cpu;
 
 	if (!sched_core_enabled(rq))
 		return __pick_next_task(rq, prev, rf);
@@ -5512,6 +5527,9 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 			if (!p)
 				continue;
 
+			if (!is_task_rq_idle(p))
+				occ++;
+
 			rq_i->core_pick = p;
 			if (rq_i->idle == p && rq_i->nr_running) {
 				rq->core->core_forceidle = true;
@@ -5543,6 +5561,7 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 
 						cpu_rq(j)->core_pick = NULL;
 					}
+					occ = 1;
 					goto again;
 				}
 			}
@@ -5588,6 +5607,8 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 		if (!(fi_before && rq->core->core_forceidle))
 			task_vruntime_update(rq_i, rq_i->core_pick, rq->core->core_forceidle);
 
+		rq_i->core_pick->core_occupation = occ;
+
 		if (i == cpu) {
 			rq_i->core_pick = NULL;
 			continue;
@@ -5609,6 +5630,113 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 	return next;
 }
 
+static bool try_steal_cookie(int this, int that)
+{
+	struct rq *dst = cpu_rq(this), *src = cpu_rq(that);
+	struct task_struct *p;
+	unsigned long cookie;
+	bool success = false;
+
+	local_irq_disable();
+	double_rq_lock(dst, src);
+
+	cookie = dst->core->core_cookie;
+	if (!cookie)
+		goto unlock;
+
+	if (dst->curr != dst->idle)
+		goto unlock;
+
+	p = sched_core_find(src, cookie);
+	if (p == src->idle)
+		goto unlock;
+
+	do {
+		if (p == src->core_pick || p == src->curr)
+			goto next;
+
+		if (!cpumask_test_cpu(this, &p->cpus_mask))
+			goto next;
+
+		if (p->core_occupation > dst->idle->core_occupation)
+			goto next;
+
+		p->on_rq = TASK_ON_RQ_MIGRATING;
+		deactivate_task(src, p, 0);
+		set_task_cpu(p, this);
+		activate_task(dst, p, 0);
+		p->on_rq = TASK_ON_RQ_QUEUED;
+
+		resched_curr(dst);
+
+		success = true;
+		break;
+
+next:
+		p = sched_core_next(p, cookie);
+	} while (p);
+
+unlock:
+	double_rq_unlock(dst, src);
+	local_irq_enable();
+
+	return success;
+}
+
+static bool steal_cookie_task(int cpu, struct sched_domain *sd)
+{
+	int i;
+
+	for_each_cpu_wrap(i, sched_domain_span(sd), cpu) {
+		if (i == cpu)
+			continue;
+
+		if (need_resched())
+			break;
+
+		if (try_steal_cookie(cpu, i))
+			return true;
+	}
+
+	return false;
+}
+
+static void sched_core_balance(struct rq *rq)
+{
+	struct sched_domain *sd;
+	int cpu = cpu_of(rq);
+
+	preempt_disable();
+	rcu_read_lock();
+	raw_spin_rq_unlock_irq(rq);
+	for_each_domain(cpu, sd) {
+		if (need_resched())
+			break;
+
+		if (steal_cookie_task(cpu, sd))
+			break;
+	}
+	raw_spin_rq_lock_irq(rq);
+	rcu_read_unlock();
+	preempt_enable();
+}
+
+static DEFINE_PER_CPU(struct callback_head, core_balance_head);
+
+void queue_core_balance(struct rq *rq)
+{
+	if (!sched_core_enabled(rq))
+		return;
+
+	if (!rq->core->core_cookie)
+		return;
+
+	if (!rq->nr_running) /* not forced idle */
+		return;
+
+	queue_balance_callback(rq, &per_cpu(core_balance_head, rq->cpu), sched_core_balance);
+}
+
 static inline void sched_core_cpu_starting(unsigned int cpu)
 {
 	const struct cpumask *smt_mask = cpu_smt_mask(cpu);

kernel/sched/idle.c

@@ -437,6 +437,7 @@ static void set_next_task_idle(struct rq *rq, struct task_struct *next, bool fir
 {
 	update_idle_core(rq);
 	schedstat_inc(rq->sched_goidle);
+	queue_core_balance(rq);
 }
 
 #ifdef CONFIG_SMP

kernel/sched/sched.h

@@ -1170,6 +1170,8 @@ static inline raw_spinlock_t *__rq_lockp(struct rq *rq)
 
 bool cfs_prio_less(struct task_struct *a, struct task_struct *b, bool fi);
 
+extern void queue_core_balance(struct rq *rq);
+
 #else /* !CONFIG_SCHED_CORE */
 
 static inline bool sched_core_enabled(struct rq *rq)
@@ -1192,6 +1194,10 @@ static inline raw_spinlock_t *__rq_lockp(struct rq *rq)
 	return &rq->__lock;
 }
 
+static inline void queue_core_balance(struct rq *rq)
+{
+}
+
 #endif /* CONFIG_SCHED_CORE */
 
 static inline void lockdep_assert_rq_held(struct rq *rq)