Merge branch 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull scheduler updates from Ingo Molnar:
 "The main changes in this cycle were:

   - Implement frequency/CPU invariance and OPP selection for
     SCHED_DEADLINE (Juri Lelli)

   - Tweak the task migration logic for better multi-tasking
     workload scalability (Mel Gorman)

   - Misc cleanups, fixes and improvements"

* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/deadline: Make bandwidth enforcement scale-invariant
  sched/cpufreq: Move arch_scale_{freq,cpu}_capacity() outside of #ifdef CONFIG_SMP
  sched/cpufreq: Remove arch_scale_freq_capacity()'s 'sd' parameter
  sched/cpufreq: Always consider all CPUs when deciding next freq
  sched/cpufreq: Split utilization signals
  sched/cpufreq: Change the worker kthread to SCHED_DEADLINE
  sched/deadline: Move CPU frequency selection triggering points
  sched/cpufreq: Use the DEADLINE utilization signal
  sched/deadline: Implement "runtime overrun signal" support
  sched/fair: Only immediately migrate tasks due to interrupts if prev and target CPUs share cache
  sched/fair: Correct obsolete comment about cpufreq_update_util()
  sched/fair: Remove impossible condition from find_idlest_group_cpu()
  sched/cpufreq: Don't pass flags to sugov_set_iowait_boost()
  sched/cpufreq: Initialize sg_cpu->flags to 0
  sched/fair: Consider RT/IRQ pressure in capacity_spare_wake()
  sched/fair: Use 'unsigned long' for utilization, consistently
  sched/core: Rework and clarify prepare_lock_switch()
  sched/fair: Remove unused 'curr' parameter from wakeup_gran
  sched/headers: Constify object_is_on_stack()

include/linux/arch_topology.h

@@ -27,7 +27,7 @@ void topology_set_cpu_scale(unsigned int cpu, unsigned long capacity);
 DECLARE_PER_CPU(unsigned long, freq_scale);
 
 static inline
-unsigned long topology_get_freq_scale(struct sched_domain *sd, int cpu)
+unsigned long topology_get_freq_scale(int cpu)
 {
 	return per_cpu(freq_scale, cpu);
 }

include/linux/sched.h

@@ -472,11 +472,15 @@ struct sched_dl_entity {
 	 * has not been executed yet. This flag is useful to avoid race
 	 * conditions between the inactive timer handler and the wakeup
 	 * code.
+	 *
+	 * @dl_overrun tells if the task asked to be informed about runtime
+	 * overruns.
 	 */
 	unsigned int			dl_throttled      : 1;
 	unsigned int			dl_boosted        : 1;
 	unsigned int			dl_yielded        : 1;
 	unsigned int			dl_non_contending : 1;
+	unsigned int			dl_overrun	  : 1;
 
 	/*
 	 * Bandwidth enforcement timer. Each -deadline task has its
@@ -1427,6 +1431,7 @@ extern int idle_cpu(int cpu);
 extern int sched_setscheduler(struct task_struct *, int, const struct sched_param *);
 extern int sched_setscheduler_nocheck(struct task_struct *, int, const struct sched_param *);
 extern int sched_setattr(struct task_struct *, const struct sched_attr *);
+extern int sched_setattr_nocheck(struct task_struct *, const struct sched_attr *);
 extern struct task_struct *idle_task(int cpu);
 
 /**

include/linux/sched/cpufreq.h

@@ -12,8 +12,6 @@
 #define SCHED_CPUFREQ_DL	(1U << 1)
 #define SCHED_CPUFREQ_IOWAIT	(1U << 2)
 
-#define SCHED_CPUFREQ_RT_DL	(SCHED_CPUFREQ_RT | SCHED_CPUFREQ_DL)
-
 #ifdef CONFIG_CPU_FREQ
 struct update_util_data {
        void (*func)(struct update_util_data *data, u64 time, unsigned int flags);

include/linux/sched/task_stack.h

@@ -78,7 +78,7 @@ static inline void put_task_stack(struct task_struct *tsk) {}
 #define task_stack_end_corrupted(task) \
 		(*(end_of_stack(task)) != STACK_END_MAGIC)
 
-static inline int object_is_on_stack(void *obj)
+static inline int object_is_on_stack(const void *obj)
 {
 	void *stack = task_stack_page(current);
 

include/linux/sched/topology.h

@@ -6,6 +6,12 @@
 
 #include <linux/sched/idle.h>
 
+/*
+ * Increase resolution of cpu_capacity calculations
+ */
+#define SCHED_CAPACITY_SHIFT	SCHED_FIXEDPOINT_SHIFT
+#define SCHED_CAPACITY_SCALE	(1L << SCHED_CAPACITY_SHIFT)
+
 /*
  * sched-domains (multiprocessor balancing) declarations:
  */
@@ -27,12 +33,6 @@
 #define SD_OVERLAP		0x2000	/* sched_domains of this level overlap */
 #define SD_NUMA			0x4000	/* cross-node balancing */
 
-/*
- * Increase resolution of cpu_capacity calculations
- */
-#define SCHED_CAPACITY_SHIFT	SCHED_FIXEDPOINT_SHIFT
-#define SCHED_CAPACITY_SCALE	(1L << SCHED_CAPACITY_SHIFT)
-
 #ifdef CONFIG_SCHED_SMT
 static inline int cpu_smt_flags(void)
 {

include/uapi/linux/sched.h

@@ -49,5 +49,10 @@
  */
 #define SCHED_FLAG_RESET_ON_FORK	0x01
 #define SCHED_FLAG_RECLAIM		0x02
+#define SCHED_FLAG_DL_OVERRUN		0x04
+
+#define SCHED_FLAG_ALL	(SCHED_FLAG_RESET_ON_FORK	| \
+			 SCHED_FLAG_RECLAIM		| \
+			 SCHED_FLAG_DL_OVERRUN)
 
 #endif /* _UAPI_LINUX_SCHED_H */

kernel/sched/core.c

@@ -2046,7 +2046,7 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
 	 * If the owning (remote) CPU is still in the middle of schedule() with
 	 * this task as prev, wait until its done referencing the task.
 	 *
-	 * Pairs with the smp_store_release() in finish_lock_switch().
+	 * Pairs with the smp_store_release() in finish_task().
 	 *
 	 * This ensures that tasks getting woken will be fully ordered against
 	 * their previous state and preserve Program Order.
@@ -2572,6 +2572,50 @@ fire_sched_out_preempt_notifiers(struct task_struct *curr,
 
 #endif /* CONFIG_PREEMPT_NOTIFIERS */
 
+static inline void prepare_task(struct task_struct *next)
+{
+#ifdef CONFIG_SMP
+	/*
+	 * Claim the task as running, we do this before switching to it
+	 * such that any running task will have this set.
+	 */
+	next->on_cpu = 1;
+#endif
+}
+
+static inline void finish_task(struct task_struct *prev)
+{
+#ifdef CONFIG_SMP
+	/*
+	 * After ->on_cpu is cleared, the task can be moved to a different CPU.
+	 * We must ensure this doesn't happen until the switch is completely
+	 * finished.
+	 *
+	 * In particular, the load of prev->state in finish_task_switch() must
+	 * happen before this.
+	 *
+	 * Pairs with the smp_cond_load_acquire() in try_to_wake_up().
+	 */
+	smp_store_release(&prev->on_cpu, 0);
+#endif
+}
+
+static inline void finish_lock_switch(struct rq *rq)
+{
+#ifdef CONFIG_DEBUG_SPINLOCK
+	/* this is a valid case when another task releases the spinlock */
+	rq->lock.owner = current;
+#endif
+	/*
+	 * If we are tracking spinlock dependencies then we have to
+	 * fix up the runqueue lock - which gets 'carried over' from
+	 * prev into current:
+	 */
+	spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_);
+
+	raw_spin_unlock_irq(&rq->lock);
+}
+
 /**
  * prepare_task_switch - prepare to switch tasks
  * @rq: the runqueue preparing to switch
@@ -2592,7 +2636,7 @@ prepare_task_switch(struct rq *rq, struct task_struct *prev,
 	sched_info_switch(rq, prev, next);
 	perf_event_task_sched_out(prev, next);
 	fire_sched_out_preempt_notifiers(prev, next);
-	prepare_lock_switch(rq, next);
+	prepare_task(next);
 	prepare_arch_switch(next);
 }
 
@@ -2647,7 +2691,7 @@ static struct rq *finish_task_switch(struct task_struct *prev)
 	 * the scheduled task must drop that reference.
 	 *
 	 * We must observe prev->state before clearing prev->on_cpu (in
-	 * finish_lock_switch), otherwise a concurrent wakeup can get prev
+	 * finish_task), otherwise a concurrent wakeup can get prev
 	 * running on another CPU and we could rave with its RUNNING -> DEAD
 	 * transition, resulting in a double drop.
 	 */
@@ -2664,7 +2708,8 @@ static struct rq *finish_task_switch(struct task_struct *prev)
 	 * to use.
 	 */
 	smp_mb__after_unlock_lock();
-	finish_lock_switch(rq, prev);
+	finish_task(prev);
+	finish_lock_switch(rq);
 	finish_arch_post_lock_switch();
 
 	fire_sched_in_preempt_notifiers(current);
@@ -4041,8 +4086,7 @@ recheck:
 			return -EINVAL;
 	}
 
-	if (attr->sched_flags &
-		~(SCHED_FLAG_RESET_ON_FORK | SCHED_FLAG_RECLAIM))
+	if (attr->sched_flags & ~(SCHED_FLAG_ALL | SCHED_FLAG_SUGOV))
 		return -EINVAL;
 
 	/*
@@ -4109,6 +4153,9 @@ recheck:
 	}
 
 	if (user) {
+		if (attr->sched_flags & SCHED_FLAG_SUGOV)
+			return -EINVAL;
+
 		retval = security_task_setscheduler(p);
 		if (retval)
 			return retval;
@@ -4164,7 +4211,8 @@ change:
 		}
 #endif
 #ifdef CONFIG_SMP
-		if (dl_bandwidth_enabled() && dl_policy(policy)) {
+		if (dl_bandwidth_enabled() && dl_policy(policy) &&
+				!(attr->sched_flags & SCHED_FLAG_SUGOV)) {
 			cpumask_t *span = rq->rd->span;
 
 			/*
@@ -4294,6 +4342,11 @@ int sched_setattr(struct task_struct *p, const struct sched_attr *attr)
 }
 EXPORT_SYMBOL_GPL(sched_setattr);
 
+int sched_setattr_nocheck(struct task_struct *p, const struct sched_attr *attr)
+{
+	return __sched_setscheduler(p, attr, false, true);
+}
+
 /**
  * sched_setscheduler_nocheck - change the scheduling policy and/or RT priority of a thread from kernelspace.
  * @p: the task in question.

kernel/sched/cpufreq_schedutil.c

@@ -60,7 +60,8 @@ struct sugov_cpu {
 	u64 last_update;
 
 	/* The fields below are only needed when sharing a policy. */
-	unsigned long util;
+	unsigned long util_cfs;
+	unsigned long util_dl;
 	unsigned long max;
 	unsigned int flags;
 
@@ -176,21 +177,28 @@ static unsigned int get_next_freq(struct sugov_policy *sg_policy,
 	return cpufreq_driver_resolve_freq(policy, freq);
 }
 
-static void sugov_get_util(unsigned long *util, unsigned long *max, int cpu)
+static void sugov_get_util(struct sugov_cpu *sg_cpu)
 {
-	struct rq *rq = cpu_rq(cpu);
-	unsigned long cfs_max;
+	struct rq *rq = cpu_rq(sg_cpu->cpu);
 
-	cfs_max = arch_scale_cpu_capacity(NULL, cpu);
+	sg_cpu->max = arch_scale_cpu_capacity(NULL, sg_cpu->cpu);
+	sg_cpu->util_cfs = cpu_util_cfs(rq);
+	sg_cpu->util_dl  = cpu_util_dl(rq);
+}
 
-	*util = min(rq->cfs.avg.util_avg, cfs_max);
-	*max = cfs_max;
+static unsigned long sugov_aggregate_util(struct sugov_cpu *sg_cpu)
+{
+	/*
+	 * Ideally we would like to set util_dl as min/guaranteed freq and
+	 * util_cfs + util_dl as requested freq. However, cpufreq is not yet
+	 * ready for such an interface. So, we only do the latter for now.
+	 */
+	return min(sg_cpu->util_cfs + sg_cpu->util_dl, sg_cpu->max);
 }
 
-static void sugov_set_iowait_boost(struct sugov_cpu *sg_cpu, u64 time,
-				   unsigned int flags)
+static void sugov_set_iowait_boost(struct sugov_cpu *sg_cpu, u64 time)
 {
-	if (flags & SCHED_CPUFREQ_IOWAIT) {
+	if (sg_cpu->flags & SCHED_CPUFREQ_IOWAIT) {
 		if (sg_cpu->iowait_boost_pending)
 			return;
 
@@ -264,7 +272,7 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
 	unsigned int next_f;
 	bool busy;
 
-	sugov_set_iowait_boost(sg_cpu, time, flags);
+	sugov_set_iowait_boost(sg_cpu, time);
 	sg_cpu->last_update = time;
 
 	if (!sugov_should_update_freq(sg_policy, time))
@@ -272,10 +280,12 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
 
 	busy = sugov_cpu_is_busy(sg_cpu);
 
-	if (flags & SCHED_CPUFREQ_RT_DL) {
+	if (flags & SCHED_CPUFREQ_RT) {
 		next_f = policy->cpuinfo.max_freq;
 	} else {
-		sugov_get_util(&util, &max, sg_cpu->cpu);
+		sugov_get_util(sg_cpu);
+		max = sg_cpu->max;
+		util = sugov_aggregate_util(sg_cpu);
 		sugov_iowait_boost(sg_cpu, &util, &max);
 		next_f = get_next_freq(sg_policy, util, max);
 		/*
@@ -305,23 +315,27 @@ static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, u64 time)
 		s64 delta_ns;
 
 		/*
-		 * If the CPU utilization was last updated before the previous
-		 * frequency update and the time elapsed between the last update
-		 * of the CPU utilization and the last frequency update is long
-		 * enough, don't take the CPU into account as it probably is
-		 * idle now (and clear iowait_boost for it).
+		 * If the CFS CPU utilization was last updated before the
+		 * previous frequency update and the time elapsed between the
+		 * last update of the CPU utilization and the last frequency
+		 * update is long enough, reset iowait_boost and util_cfs, as
+		 * they are now probably stale. However, still consider the
+		 * CPU contribution if it has some DEADLINE utilization
+		 * (util_dl).
 		 */
 		delta_ns = time - j_sg_cpu->last_update;
 		if (delta_ns > TICK_NSEC) {
 			j_sg_cpu->iowait_boost = 0;
 			j_sg_cpu->iowait_boost_pending = false;
-			continue;
+			j_sg_cpu->util_cfs = 0;
+			if (j_sg_cpu->util_dl == 0)
+				continue;
 		}
-		if (j_sg_cpu->flags & SCHED_CPUFREQ_RT_DL)
+		if (j_sg_cpu->flags & SCHED_CPUFREQ_RT)
 			return policy->cpuinfo.max_freq;
 
-		j_util = j_sg_cpu->util;
 		j_max = j_sg_cpu->max;
+		j_util = sugov_aggregate_util(j_sg_cpu);
 		if (j_util * max > j_max * util) {
 			util = j_util;
 			max = j_max;
@@ -338,22 +352,18 @@ static void sugov_update_shared(struct update_util_data *hook, u64 time,
 {
 	struct sugov_cpu *sg_cpu = container_of(hook, struct sugov_cpu, update_util);
 	struct sugov_policy *sg_policy = sg_cpu->sg_policy;
-	unsigned long util, max;
 	unsigned int next_f;
 
-	sugov_get_util(&util, &max, sg_cpu->cpu);
-
 	raw_spin_lock(&sg_policy->update_lock);
 
-	sg_cpu->util = util;
-	sg_cpu->max = max;
+	sugov_get_util(sg_cpu);
 	sg_cpu->flags = flags;
 
-	sugov_set_iowait_boost(sg_cpu, time, flags);
+	sugov_set_iowait_boost(sg_cpu, time);
 	sg_cpu->last_update = time;
 
 	if (sugov_should_update_freq(sg_policy, time)) {
-		if (flags & SCHED_CPUFREQ_RT_DL)
+		if (flags & SCHED_CPUFREQ_RT)
 			next_f = sg_policy->policy->cpuinfo.max_freq;
 		else
 			next_f = sugov_next_freq_shared(sg_cpu, time);
@@ -383,9 +393,9 @@ static void sugov_irq_work(struct irq_work *irq_work)
 	sg_policy = container_of(irq_work, struct sugov_policy, irq_work);
 
 	/*
-	 * For RT and deadline tasks, the schedutil governor shoots the
-	 * frequency to maximum. Special care must be taken to ensure that this
-	 * kthread doesn't result in the same behavior.
+	 * For RT tasks, the schedutil governor shoots the frequency to maximum.
+	 * Special care must be taken to ensure that this kthread doesn't result
+	 * in the same behavior.
 	 *
 	 * This is (mostly) guaranteed by the work_in_progress flag. The flag is
 	 * updated only at the end of the sugov_work() function and before that
@@ -470,7 +480,20 @@ static void sugov_policy_free(struct sugov_policy *sg_policy)
 static int sugov_kthread_create(struct sugov_policy *sg_policy)
 {
 	struct task_struct *thread;
-	struct sched_param param = { .sched_priority = MAX_USER_RT_PRIO / 2 };
+	struct sched_attr attr = {
+		.size = sizeof(struct sched_attr),
+		.sched_policy = SCHED_DEADLINE,
+		.sched_flags = SCHED_FLAG_SUGOV,
+		.sched_nice = 0,
+		.sched_priority = 0,
+		/*
+		 * Fake (unused) bandwidth; workaround to "fix"
+		 * priority inheritance.
+		 */
+		.sched_runtime	=  1000000,
+		.sched_deadline = 10000000,
+		.sched_period	= 10000000,
+	};
 	struct cpufreq_policy *policy = sg_policy->policy;
 	int ret;
 
@@ -488,10 +511,10 @@ static int sugov_kthread_create(struct sugov_policy *sg_policy)
 		return PTR_ERR(thread);
 	}
 
-	ret = sched_setscheduler_nocheck(thread, SCHED_FIFO, &param);
+	ret = sched_setattr_nocheck(thread, &attr);
 	if (ret) {
 		kthread_stop(thread);
-		pr_warn("%s: failed to set SCHED_FIFO\n", __func__);
+		pr_warn("%s: failed to set SCHED_DEADLINE\n", __func__);
 		return ret;
 	}
 
@@ -655,7 +678,7 @@ static int sugov_start(struct cpufreq_policy *policy)
 		memset(sg_cpu, 0, sizeof(*sg_cpu));
 		sg_cpu->cpu = cpu;
 		sg_cpu->sg_policy = sg_policy;
-		sg_cpu->flags = SCHED_CPUFREQ_RT;
+		sg_cpu->flags = 0;
 		sg_cpu->iowait_boost_max = policy->cpuinfo.max_freq;
 	}
 

kernel/sched/deadline.c

@@ -78,7 +78,7 @@ static inline int dl_bw_cpus(int i)
 #endif
 
 static inline
-void add_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
+void __add_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
 {
 	u64 old = dl_rq->running_bw;
 
@@ -86,10 +86,12 @@ void add_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
 	dl_rq->running_bw += dl_bw;
 	SCHED_WARN_ON(dl_rq->running_bw < old); /* overflow */
 	SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw);
+	/* kick cpufreq (see the comment in kernel/sched/sched.h). */
+	cpufreq_update_util(rq_of_dl_rq(dl_rq), SCHED_CPUFREQ_DL);
 }
 
 static inline
-void sub_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
+void __sub_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
 {
 	u64 old = dl_rq->running_bw;
 
@@ -98,10 +100,12 @@ void sub_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
 	SCHED_WARN_ON(dl_rq->running_bw > old); /* underflow */
 	if (dl_rq->running_bw > old)
 		dl_rq->running_bw = 0;
+	/* kick cpufreq (see the comment in kernel/sched/sched.h). */
+	cpufreq_update_util(rq_of_dl_rq(dl_rq), SCHED_CPUFREQ_DL);
 }
 
 static inline
-void add_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
+void __add_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
 {
 	u64 old = dl_rq->this_bw;
 
@@ -111,7 +115,7 @@ void add_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
 }
 
 static inline
-void sub_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
+void __sub_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
 {
 	u64 old = dl_rq->this_bw;
 
@@ -123,16 +127,46 @@ void sub_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
 	SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw);
 }
 
+static inline
+void add_rq_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
+{
+	if (!dl_entity_is_special(dl_se))
+		__add_rq_bw(dl_se->dl_bw, dl_rq);
+}
+
+static inline
+void sub_rq_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
+{
+	if (!dl_entity_is_special(dl_se))
+		__sub_rq_bw(dl_se->dl_bw, dl_rq);
+}
+
+static inline
+void add_running_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
+{
+	if (!dl_entity_is_special(dl_se))
+		__add_running_bw(dl_se->dl_bw, dl_rq);
+}
+
+static inline
+void sub_running_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
+{
+	if (!dl_entity_is_special(dl_se))
+		__sub_running_bw(dl_se->dl_bw, dl_rq);
+}
+
 void dl_change_utilization(struct task_struct *p, u64 new_bw)
 {
 	struct rq *rq;
 
+	BUG_ON(p->dl.flags & SCHED_FLAG_SUGOV);
+
 	if (task_on_rq_queued(p))
 		return;
 
 	rq = task_rq(p);
 	if (p->dl.dl_non_contending) {
-		sub_running_bw(p->dl.dl_bw, &rq->dl);
+		sub_running_bw(&p->dl, &rq->dl);
 		p->dl.dl_non_contending = 0;
 		/*
 		 * If the timer handler is currently running and the
@@ -144,8 +178,8 @@ void dl_change_utilization(struct task_struct *p, u64 new_bw)
 		if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1)
 			put_task_struct(p);
 	}
-	sub_rq_bw(p->dl.dl_bw, &rq->dl);
-	add_rq_bw(new_bw, &rq->dl);
+	__sub_rq_bw(p->dl.dl_bw, &rq->dl);
+	__add_rq_bw(new_bw, &rq->dl);
 }
 
 /*
@@ -217,6 +251,9 @@ static void task_non_contending(struct task_struct *p)
 	if (dl_se->dl_runtime == 0)
 		return;
 
+	if (dl_entity_is_special(dl_se))
+		return;
+
 	WARN_ON(hrtimer_active(&dl_se->inactive_timer));
 	WARN_ON(dl_se->dl_non_contending);
 
@@ -236,12 +273,12 @@ static void task_non_contending(struct task_struct *p)
 	 */
 	if (zerolag_time < 0) {
 		if (dl_task(p))
-			sub_running_bw(dl_se->dl_bw, dl_rq);
+			sub_running_bw(dl_se, dl_rq);
 		if (!dl_task(p) || p->state == TASK_DEAD) {
 			struct dl_bw *dl_b = dl_bw_of(task_cpu(p));
 
 			if (p->state == TASK_DEAD)
-				sub_rq_bw(p->dl.dl_bw, &rq->dl);
+				sub_rq_bw(&p->dl, &rq->dl);
 			raw_spin_lock(&dl_b->lock);
 			__dl_sub(dl_b, p->dl.dl_bw, dl_bw_cpus(task_cpu(p)));
 			__dl_clear_params(p);
@@ -268,7 +305,7 @@ static void task_contending(struct sched_dl_entity *dl_se, int flags)
 		return;
 
 	if (flags & ENQUEUE_MIGRATED)
-		add_rq_bw(dl_se->dl_bw, dl_rq);
+		add_rq_bw(dl_se, dl_rq);
 
 	if (dl_se->dl_non_contending) {
 		dl_se->dl_non_contending = 0;
@@ -289,7 +326,7 @@ static void task_contending(struct sched_dl_entity *dl_se, int flags)
 		 * when the "inactive timer" fired).
 		 * So, add it back.
 		 */
-		add_running_bw(dl_se->dl_bw, dl_rq);
+		add_running_bw(dl_se, dl_rq);
 	}
 }
 
@@ -1114,7 +1151,8 @@ static void update_curr_dl(struct rq *rq)
 {
 	struct task_struct *curr = rq->curr;
 	struct sched_dl_entity *dl_se = &curr->dl;
-	u64 delta_exec;
+	u64 delta_exec, scaled_delta_exec;
+	int cpu = cpu_of(rq);
 
 	if (!dl_task(curr) || !on_dl_rq(dl_se))
 		return;
@@ -1134,9 +1172,6 @@ static void update_curr_dl(struct rq *rq)
 		return;
 	}
 
-	/* kick cpufreq (see the comment in kernel/sched/sched.h). */
-	cpufreq_update_util(rq, SCHED_CPUFREQ_DL);
-
 	schedstat_set(curr->se.statistics.exec_max,
 		      max(curr->se.statistics.exec_max, delta_exec));
 
@@ -1148,13 +1183,39 @@ static void update_curr_dl(struct rq *rq)
 
 	sched_rt_avg_update(rq, delta_exec);
 
-	if (unlikely(dl_se->flags & SCHED_FLAG_RECLAIM))
-		delta_exec = grub_reclaim(delta_exec, rq, &curr->dl);
-	dl_se->runtime -= delta_exec;
+	if (dl_entity_is_special(dl_se))
+		return;
+
+	/*
+	 * For tasks that participate in GRUB, we implement GRUB-PA: the
+	 * spare reclaimed bandwidth is used to clock down frequency.
+	 *
+	 * For the others, we still need to scale reservation parameters
+	 * according to current frequency and CPU maximum capacity.
+	 */
+	if (unlikely(dl_se->flags & SCHED_FLAG_RECLAIM)) {
+		scaled_delta_exec = grub_reclaim(delta_exec,
+						 rq,
+						 &curr->dl);
+	} else {
+		unsigned long scale_freq = arch_scale_freq_capacity(cpu);
+		unsigned long scale_cpu = arch_scale_cpu_capacity(NULL, cpu);
+
+		scaled_delta_exec = cap_scale(delta_exec, scale_freq);
+		scaled_delta_exec = cap_scale(scaled_delta_exec, scale_cpu);
+	}
+
+	dl_se->runtime -= scaled_delta_exec;
 
 throttle:
 	if (dl_runtime_exceeded(dl_se) || dl_se->dl_yielded) {
 		dl_se->dl_throttled = 1;
+
+		/* If requested, inform the user about runtime overruns. */
+		if (dl_runtime_exceeded(dl_se) &&
+		    (dl_se->flags & SCHED_FLAG_DL_OVERRUN))
+			dl_se->dl_overrun = 1;
+
 		__dequeue_task_dl(rq, curr, 0);
 		if (unlikely(dl_se->dl_boosted || !start_dl_timer(curr)))
 			enqueue_task_dl(rq, curr, ENQUEUE_REPLENISH);
@@ -1204,8 +1265,8 @@ static enum hrtimer_restart inactive_task_timer(struct hrtimer *timer)
 		struct dl_bw *dl_b = dl_bw_of(task_cpu(p));
 
 		if (p->state == TASK_DEAD && dl_se->dl_non_contending) {
-			sub_running_bw(p->dl.dl_bw, dl_rq_of_se(&p->dl));
-			sub_rq_bw(p->dl.dl_bw, dl_rq_of_se(&p->dl));
+			sub_running_bw(&p->dl, dl_rq_of_se(&p->dl));
+			sub_rq_bw(&p->dl, dl_rq_of_se(&p->dl));
 			dl_se->dl_non_contending = 0;
 		}
 
@@ -1222,7 +1283,7 @@ static enum hrtimer_restart inactive_task_timer(struct hrtimer *timer)
 	sched_clock_tick();
 	update_rq_clock(rq);
 
-	sub_running_bw(dl_se->dl_bw, &rq->dl);
+	sub_running_bw(dl_se, &rq->dl);
 	dl_se->dl_non_contending = 0;
 unlock:
 	task_rq_unlock(rq, p, &rf);
@@ -1416,8 +1477,8 @@ static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags)
 		dl_check_constrained_dl(&p->dl);
 
 	if (p->on_rq == TASK_ON_RQ_MIGRATING || flags & ENQUEUE_RESTORE) {
-		add_rq_bw(p->dl.dl_bw, &rq->dl);
-		add_running_bw(p->dl.dl_bw, &rq->dl);
+		add_rq_bw(&p->dl, &rq->dl);
+		add_running_bw(&p->dl, &rq->dl);
 	}
 
 	/*
@@ -1457,8 +1518,8 @@ static void dequeue_task_dl(struct rq *rq, struct task_struct *p, int flags)
 	__dequeue_task_dl(rq, p, flags);
 
 	if (p->on_rq == TASK_ON_RQ_MIGRATING || flags & DEQUEUE_SAVE) {
-		sub_running_bw(p->dl.dl_bw, &rq->dl);
-		sub_rq_bw(p->dl.dl_bw, &rq->dl);
+		sub_running_bw(&p->dl, &rq->dl);
+		sub_rq_bw(&p->dl, &rq->dl);
 	}
 
 	/*
@@ -1564,7 +1625,7 @@ static void migrate_task_rq_dl(struct task_struct *p)
 	 */
 	raw_spin_lock(&rq->lock);
 	if (p->dl.dl_non_contending) {
-		sub_running_bw(p->dl.dl_bw, &rq->dl);
+		sub_running_bw(&p->dl, &rq->dl);
 		p->dl.dl_non_contending = 0;
 		/*
 		 * If the timer handler is currently running and the
@@ -1576,7 +1637,7 @@ static void migrate_task_rq_dl(struct task_struct *p)
 		if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1)
 			put_task_struct(p);
 	}
-	sub_rq_bw(p->dl.dl_bw, &rq->dl);
+	sub_rq_bw(&p->dl, &rq->dl);
 	raw_spin_unlock(&rq->lock);
 }
 
@@ -2019,11 +2080,11 @@ retry:
 	}
 
 	deactivate_task(rq, next_task, 0);
-	sub_running_bw(next_task->dl.dl_bw, &rq->dl);
-	sub_rq_bw(next_task->dl.dl_bw, &rq->dl);
+	sub_running_bw(&next_task->dl, &rq->dl);
+	sub_rq_bw(&next_task->dl, &rq->dl);
 	set_task_cpu(next_task, later_rq->cpu);
-	add_rq_bw(next_task->dl.dl_bw, &later_rq->dl);
-	add_running_bw(next_task->dl.dl_bw, &later_rq->dl);
+	add_rq_bw(&next_task->dl, &later_rq->dl);
+	add_running_bw(&next_task->dl, &later_rq->dl);
 	activate_task(later_rq, next_task, 0);
 	ret = 1;
 
@@ -2111,11 +2172,11 @@ static void pull_dl_task(struct rq *this_rq)
 			resched = true;
 
 			deactivate_task(src_rq, p, 0);
-			sub_running_bw(p->dl.dl_bw, &src_rq->dl);
-			sub_rq_bw(p->dl.dl_bw, &src_rq->dl);
+			sub_running_bw(&p->dl, &src_rq->dl);
+			sub_rq_bw(&p->dl, &src_rq->dl);
 			set_task_cpu(p, this_cpu);
-			add_rq_bw(p->dl.dl_bw, &this_rq->dl);
-			add_running_bw(p->dl.dl_bw, &this_rq->dl);
+			add_rq_bw(&p->dl, &this_rq->dl);
+			add_running_bw(&p->dl, &this_rq->dl);
 			activate_task(this_rq, p, 0);
 			dmin = p->dl.deadline;
 
@@ -2224,7 +2285,7 @@ static void switched_from_dl(struct rq *rq, struct task_struct *p)
 		task_non_contending(p);
 
 	if (!task_on_rq_queued(p))
-		sub_rq_bw(p->dl.dl_bw, &rq->dl);
+		sub_rq_bw(&p->dl, &rq->dl);
 
 	/*
 	 * We cannot use inactive_task_timer() to invoke sub_running_bw()
@@ -2256,7 +2317,7 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p)
 
 	/* If p is not queued we will update its parameters at next wakeup. */
 	if (!task_on_rq_queued(p)) {
-		add_rq_bw(p->dl.dl_bw, &rq->dl);
+		add_rq_bw(&p->dl, &rq->dl);
 
 		return;
 	}
@@ -2435,6 +2496,9 @@ int sched_dl_overflow(struct task_struct *p, int policy,
 	u64 new_bw = dl_policy(policy) ? to_ratio(period, runtime) : 0;
 	int cpus, err = -1;
 
+	if (attr->sched_flags & SCHED_FLAG_SUGOV)
+		return 0;
+
 	/* !deadline task may carry old deadline bandwidth */
 	if (new_bw == p->dl.dl_bw && task_has_dl_policy(p))
 		return 0;
@@ -2521,6 +2585,10 @@ void __getparam_dl(struct task_struct *p, struct sched_attr *attr)
  */
 bool __checkparam_dl(const struct sched_attr *attr)
 {
+	/* special dl tasks don't actually use any parameter */
+	if (attr->sched_flags & SCHED_FLAG_SUGOV)
+		return true;
+
 	/* deadline != 0 */
 	if (attr->sched_deadline == 0)
 		return false;
@@ -2566,6 +2634,7 @@ void __dl_clear_params(struct task_struct *p)
 	dl_se->dl_throttled = 0;
 	dl_se->dl_yielded = 0;
 	dl_se->dl_non_contending = 0;
+	dl_se->dl_overrun = 0;
 }
 
 bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr)

kernel/sched/fair.c

@@ -3020,9 +3020,7 @@ static inline void cfs_rq_util_change(struct cfs_rq *cfs_rq)
 		/*
 		 * There are a few boundary cases this might miss but it should
 		 * get called often enough that that should (hopefully) not be
-		 * a real problem -- added to that it only calls on the local
-		 * CPU, so if we enqueue remotely we'll miss an update, but
-		 * the next tick/schedule should update.
+		 * a real problem.
 		 *
 		 * It will not get called when we go idle, because the idle
 		 * thread is a different class (!fair), nor will the utilization
@@ -3091,8 +3089,6 @@ static u32 __accumulate_pelt_segments(u64 periods, u32 d1, u32 d3)
 	return c1 + c2 + c3;
 }
 
-#define cap_scale(v, s) ((v)*(s) >> SCHED_CAPACITY_SHIFT)
-
 /*
  * Accumulate the three separate parts of the sum; d1 the remainder
  * of the last (incomplete) period, d2 the span of full periods and d3
@@ -3122,7 +3118,7 @@ accumulate_sum(u64 delta, int cpu, struct sched_avg *sa,
 	u32 contrib = (u32)delta; /* p == 0 -> delta < 1024 */
 	u64 periods;
 
-	scale_freq = arch_scale_freq_capacity(NULL, cpu);
+	scale_freq = arch_scale_freq_capacity(cpu);
 	scale_cpu = arch_scale_cpu_capacity(NULL, cpu);
 
 	delta += sa->period_contrib;
@@ -5689,8 +5685,8 @@ static int wake_wide(struct task_struct *p)
  * soonest. For the purpose of speed we only consider the waking and previous
  * CPU.
  *
- * wake_affine_idle() - only considers 'now', it check if the waking CPU is (or
- *			will be) idle.
+ * wake_affine_idle() - only considers 'now', it check if the waking CPU is
+ *			cache-affine and is (or	will be) idle.
  *
  * wake_affine_weight() - considers the weight to reflect the average
  *			  scheduling latency of the CPUs. This seems to work
@@ -5701,7 +5697,13 @@ static bool
 wake_affine_idle(struct sched_domain *sd, struct task_struct *p,
 		 int this_cpu, int prev_cpu, int sync)
 {
-	if (idle_cpu(this_cpu))
+	/*
+	 * If this_cpu is idle, it implies the wakeup is from interrupt
+	 * context. Only allow the move if cache is shared. Otherwise an
+	 * interrupt intensive workload could force all tasks onto one
+	 * node depending on the IO topology or IRQ affinity settings.
+	 */
+	if (idle_cpu(this_cpu) && cpus_share_cache(this_cpu, prev_cpu))
 		return true;
 
 	if (sync && cpu_rq(this_cpu)->nr_running == 1)
@@ -5765,12 +5767,12 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p,
 	return affine;
 }
 
-static inline int task_util(struct task_struct *p);
-static int cpu_util_wake(int cpu, struct task_struct *p);
+static inline unsigned long task_util(struct task_struct *p);
+static unsigned long cpu_util_wake(int cpu, struct task_struct *p);
 
 static unsigned long capacity_spare_wake(int cpu, struct task_struct *p)
 {
-	return capacity_orig_of(cpu) - cpu_util_wake(cpu, p);
+	return max_t(long, capacity_of(cpu) - cpu_util_wake(cpu, p), 0);
 }
 
 /*
@@ -5950,7 +5952,7 @@ find_idlest_group_cpu(struct sched_group *group, struct task_struct *p, int this
 			}
 		} else if (shallowest_idle_cpu == -1) {
 			load = weighted_cpuload(cpu_rq(i));
-			if (load < min_load || (load == min_load && i == this_cpu)) {
+			if (load < min_load) {
 				min_load = load;
 				least_loaded_cpu = i;
 			}
@@ -6247,7 +6249,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
  * capacity_orig) as it useful for predicting the capacity required after task
  * migrations (scheduler-driven DVFS).
  */
-static int cpu_util(int cpu)
+static unsigned long cpu_util(int cpu)
 {
 	unsigned long util = cpu_rq(cpu)->cfs.avg.util_avg;
 	unsigned long capacity = capacity_orig_of(cpu);
@@ -6255,7 +6257,7 @@ static int cpu_util(int cpu)
 	return (util >= capacity) ? capacity : util;
 }
 
-static inline int task_util(struct task_struct *p)
+static inline unsigned long task_util(struct task_struct *p)
 {
 	return p->se.avg.util_avg;
 }
@@ -6264,7 +6266,7 @@ static inline int task_util(struct task_struct *p)
  * cpu_util_wake: Compute cpu utilization with any contributions from
  * the waking task p removed.
  */
-static int cpu_util_wake(int cpu, struct task_struct *p)
+static unsigned long cpu_util_wake(int cpu, struct task_struct *p)
 {
 	unsigned long util, capacity;
 
@@ -6449,8 +6451,7 @@ static void task_dead_fair(struct task_struct *p)
 }
 #endif /* CONFIG_SMP */
 
-static unsigned long
-wakeup_gran(struct sched_entity *curr, struct sched_entity *se)
+static unsigned long wakeup_gran(struct sched_entity *se)
 {
 	unsigned long gran = sysctl_sched_wakeup_granularity;
 
@@ -6492,7 +6493,7 @@ wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se)
 	if (vdiff <= 0)
 		return -1;
 
-	gran = wakeup_gran(curr, se);
+	gran = wakeup_gran(se);
 	if (vdiff > gran)
 		return 1;
 

kernel/sched/sched.h

@@ -156,13 +156,39 @@ static inline int task_has_dl_policy(struct task_struct *p)
 	return dl_policy(p->policy);
 }
 
+#define cap_scale(v, s) ((v)*(s) >> SCHED_CAPACITY_SHIFT)
+
+/*
+ * !! For sched_setattr_nocheck() (kernel) only !!
+ *
+ * This is actually gross. :(
+ *
+ * It is used to make schedutil kworker(s) higher priority than SCHED_DEADLINE
+ * tasks, but still be able to sleep. We need this on platforms that cannot
+ * atomically change clock frequency. Remove once fast switching will be
+ * available on such platforms.
+ *
+ * SUGOV stands for SchedUtil GOVernor.
+ */
+#define SCHED_FLAG_SUGOV	0x10000000
+
+static inline bool dl_entity_is_special(struct sched_dl_entity *dl_se)
+{
+#ifdef CONFIG_CPU_FREQ_GOV_SCHEDUTIL
+	return unlikely(dl_se->flags & SCHED_FLAG_SUGOV);
+#else
+	return false;
+#endif
+}
+
 /*
  * Tells if entity @a should preempt entity @b.
  */
 static inline bool
 dl_entity_preempt(struct sched_dl_entity *a, struct sched_dl_entity *b)
 {
-	return dl_time_before(a->deadline, b->deadline);
+	return dl_entity_is_special(a) ||
+	       dl_time_before(a->deadline, b->deadline);
 }
 
 /*
@@ -1328,47 +1354,6 @@ static inline int task_on_rq_migrating(struct task_struct *p)
 # define finish_arch_post_lock_switch()	do { } while (0)
 #endif
 
-static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next)
-{
-#ifdef CONFIG_SMP
-	/*
-	 * We can optimise this out completely for !SMP, because the
-	 * SMP rebalancing from interrupt is the only thing that cares
-	 * here.
-	 */
-	next->on_cpu = 1;
-#endif
-}
-
-static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev)
-{
-#ifdef CONFIG_SMP
-	/*
-	 * After ->on_cpu is cleared, the task can be moved to a different CPU.
-	 * We must ensure this doesn't happen until the switch is completely
-	 * finished.
-	 *
-	 * In particular, the load of prev->state in finish_task_switch() must
-	 * happen before this.
-	 *
-	 * Pairs with the smp_cond_load_acquire() in try_to_wake_up().
-	 */
-	smp_store_release(&prev->on_cpu, 0);
-#endif
-#ifdef CONFIG_DEBUG_SPINLOCK
-	/* this is a valid case when another task releases the spinlock */
-	rq->lock.owner = current;
-#endif
-	/*
-	 * If we are tracking spinlock dependencies then we have to
-	 * fix up the runqueue lock - which gets 'carried over' from
-	 * prev into current:
-	 */
-	spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_);
-
-	raw_spin_unlock_irq(&rq->lock);
-}
-
 /*
  * wake flags
  */
@@ -1687,17 +1672,17 @@ static inline int hrtick_enabled(struct rq *rq)
 
 #endif /* CONFIG_SCHED_HRTICK */
 
-#ifdef CONFIG_SMP
-extern void sched_avg_update(struct rq *rq);
-
 #ifndef arch_scale_freq_capacity
 static __always_inline
-unsigned long arch_scale_freq_capacity(struct sched_domain *sd, int cpu)
+unsigned long arch_scale_freq_capacity(int cpu)
 {
 	return SCHED_CAPACITY_SCALE;
 }
 #endif
 
+#ifdef CONFIG_SMP
+extern void sched_avg_update(struct rq *rq);
+
 #ifndef arch_scale_cpu_capacity
 static __always_inline
 unsigned long arch_scale_cpu_capacity(struct sched_domain *sd, int cpu)
@@ -1711,10 +1696,17 @@ unsigned long arch_scale_cpu_capacity(struct sched_domain *sd, int cpu)
 
 static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta)
 {
-	rq->rt_avg += rt_delta * arch_scale_freq_capacity(NULL, cpu_of(rq));
+	rq->rt_avg += rt_delta * arch_scale_freq_capacity(cpu_of(rq));
 	sched_avg_update(rq);
 }
 #else
+#ifndef arch_scale_cpu_capacity
+static __always_inline
+unsigned long arch_scale_cpu_capacity(void __always_unused *sd, int cpu)
+{
+	return SCHED_CAPACITY_SCALE;
+}
+#endif
 static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { }
 static inline void sched_avg_update(struct rq *rq) { }
 #endif
@@ -2096,14 +2088,14 @@ DECLARE_PER_CPU(struct update_util_data *, cpufreq_update_util_data);
  * The way cpufreq is currently arranged requires it to evaluate the CPU
  * performance state (frequency/voltage) on a regular basis to prevent it from
  * being stuck in a completely inadequate performance level for too long.
- * That is not guaranteed to happen if the updates are only triggered from CFS,
- * though, because they may not be coming in if RT or deadline tasks are active
- * all the time (or there are RT and DL tasks only).
+ * That is not guaranteed to happen if the updates are only triggered from CFS
+ * and DL, though, because they may not be coming in if only RT tasks are
+ * active all the time (or there are RT tasks only).
  *
- * As a workaround for that issue, this function is called by the RT and DL
- * sched classes to trigger extra cpufreq updates to prevent it from stalling,
+ * As a workaround for that issue, this function is called periodically by the
+ * RT sched class to trigger extra cpufreq updates to prevent it from stalling,
  * but that really is a band-aid.  Going forward it should be replaced with
- * solutions targeted more specifically at RT and DL tasks.
+ * solutions targeted more specifically at RT tasks.
  */
 static inline void cpufreq_update_util(struct rq *rq, unsigned int flags)
 {
@@ -2125,3 +2117,17 @@ static inline void cpufreq_update_util(struct rq *rq, unsigned int flags) {}
 #else /* arch_scale_freq_capacity */
 #define arch_scale_freq_invariant()	(false)
 #endif
+
+#ifdef CONFIG_CPU_FREQ_GOV_SCHEDUTIL
+
+static inline unsigned long cpu_util_dl(struct rq *rq)
+{
+	return (rq->dl.running_bw * SCHED_CAPACITY_SCALE) >> BW_SHIFT;
+}
+
+static inline unsigned long cpu_util_cfs(struct rq *rq)
+{
+	return rq->cfs.avg.util_avg;
+}
+
+#endif

kernel/time/posix-cpu-timers.c

@@ -14,6 +14,7 @@
 #include <linux/tick.h>
 #include <linux/workqueue.h>
 #include <linux/compat.h>
+#include <linux/sched/deadline.h>
 
 #include "posix-timers.h"
 
@@ -791,6 +792,14 @@ check_timers_list(struct list_head *timers,
 	return 0;
 }
 
+static inline void check_dl_overrun(struct task_struct *tsk)
+{
+	if (tsk->dl.dl_overrun) {
+		tsk->dl.dl_overrun = 0;
+		__group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk);
+	}
+}
+
 /*
  * Check for any per-thread CPU timers that have fired and move them off
  * the tsk->cpu_timers[N] list onto the firing list.  Here we update the
@@ -804,6 +813,9 @@ static void check_thread_timers(struct task_struct *tsk,
 	u64 expires;
 	unsigned long soft;
 
+	if (dl_task(tsk))
+		check_dl_overrun(tsk);
+
 	/*
 	 * If cputime_expires is zero, then there are no active
 	 * per thread CPU timers.
@@ -906,6 +918,9 @@ static void check_process_timers(struct task_struct *tsk,
 	struct task_cputime cputime;
 	unsigned long soft;
 
+	if (dl_task(tsk))
+		check_dl_overrun(tsk);
+
 	/*
 	 * If cputimer is not running, then there are no active
 	 * process wide timers (POSIX 1.b, itimers, RLIMIT_CPU).
@@ -1111,6 +1126,9 @@ static inline int fastpath_timer_check(struct task_struct *tsk)
 			return 1;
 	}
 
+	if (dl_task(tsk) && tsk->dl.dl_overrun)
+		return 1;
+
 	return 0;
 }