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

torvalds · torvalds · commit e6e7214fbbda · 2016-08-12T13:51:52.000-07:00
Pull scheduler fixes from Ingo Molnar:
 "Misc fixes: cputime fixes, two deadline scheduler fixes and a cgroups
  scheduling fix"

* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/cputime: Fix omitted ticks passed in parameter
  sched/cputime: Fix steal time accounting
  sched/deadline: Fix lock pinning warning during CPU hotplug
  sched/cputime: Mitigate performance regression in times()/clock_gettime()
  sched/fair: Fix typo in sync_throttle()
  sched/deadline: Fix wrap-around in DL heap
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
@@ -74,6 +74,7 @@
 #include <linux/context_tracking.h>
 #include <linux/compiler.h>
 #include <linux/frame.h>
+#include <linux/prefetch.h>
 
 #include <asm/switch_to.h>
 #include <asm/tlb.h>
@@ -2971,6 +2972,23 @@ DEFINE_PER_CPU(struct kernel_cpustat, kernel_cpustat);
 EXPORT_PER_CPU_SYMBOL(kstat);
 EXPORT_PER_CPU_SYMBOL(kernel_cpustat);
 
+/*
+ * The function fair_sched_class.update_curr accesses the struct curr
+ * and its field curr->exec_start; when called from task_sched_runtime(),
+ * we observe a high rate of cache misses in practice.
+ * Prefetching this data results in improved performance.
+ */
+static inline void prefetch_curr_exec_start(struct task_struct *p)
+{
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	struct sched_entity *curr = (&p->se)->cfs_rq->curr;
+#else
+	struct sched_entity *curr = (&task_rq(p)->cfs)->curr;
+#endif
+	prefetch(curr);
+	prefetch(&curr->exec_start);
+}
+
 /*
  * Return accounted runtime for the task.
  * In case the task is currently running, return the runtime plus current's
@@ -3005,6 +3023,7 @@ unsigned long long task_sched_runtime(struct task_struct *p)
 	 * thread, breaking clock_gettime().
 	 */
 	if (task_current(rq, p) && task_on_rq_queued(p)) {
+		prefetch_curr_exec_start(p);
 		update_rq_clock(rq);
 		p->sched_class->update_curr(rq);
 	}
diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
@@ -168,7 +168,7 @@ void cpudl_set(struct cpudl *cp, int cpu, u64 dl, int is_valid)
 
 	if (old_idx == IDX_INVALID) {
 		cp->size++;
-		cp->elements[cp->size - 1].dl = 0;
+		cp->elements[cp->size - 1].dl = dl;
 		cp->elements[cp->size - 1].cpu = cpu;
 		cp->elements[cpu].idx = cp->size - 1;
 		cpudl_change_key(cp, cp->size - 1, dl);
diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
@@ -508,13 +508,21 @@ void account_process_tick(struct task_struct *p, int user_tick)
  */
 void account_idle_ticks(unsigned long ticks)
 {
+	cputime_t cputime, steal;
 
 	if (sched_clock_irqtime) {
 		irqtime_account_idle_ticks(ticks);
 		return;
 	}
 
-	account_idle_time(jiffies_to_cputime(ticks));
+	cputime = jiffies_to_cputime(ticks);
+	steal = steal_account_process_time(cputime);
+
+	if (steal >= cputime)
+		return;
+
+	cputime -= steal;
+	account_idle_time(cputime);
 }
 
 /*
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
@@ -658,8 +658,11 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
 	 *
 	 * XXX figure out if select_task_rq_dl() deals with offline cpus.
 	 */
-	if (unlikely(!rq->online))
+	if (unlikely(!rq->online)) {
+		lockdep_unpin_lock(&rq->lock, rf.cookie);
 		rq = dl_task_offline_migration(rq, p);
+		rf.cookie = lockdep_pin_lock(&rq->lock);
+	}
 
 	/*
 	 * Queueing this task back might have overloaded rq, check if we need
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
@@ -4269,7 +4269,7 @@ static void sync_throttle(struct task_group *tg, int cpu)
 	pcfs_rq = tg->parent->cfs_rq[cpu];
 
 	cfs_rq->throttle_count = pcfs_rq->throttle_count;
-	pcfs_rq->throttled_clock_task = rq_clock_task(cpu_rq(cpu));
+	cfs_rq->throttled_clock_task = rq_clock_task(cpu_rq(cpu));
 }
 
 /* conditionally throttle active cfs_rq's from put_prev_entity() */

Original file line number	Diff line number	Diff line change
`@@ -4269,7 +4269,7 @@ static void sync_throttle(struct task_group *tg, int cpu)`
`4269`	`4269`	`pcfs_rq = tg->parent->cfs_rq[cpu];`
`4270`	`4270`
`4271`	`4271`	`cfs_rq->throttle_count = pcfs_rq->throttle_count;`
`4272`		`- pcfs_rq->throttled_clock_task = rq_clock_task(cpu_rq(cpu));`
	`4272`	`+ cfs_rq->throttled_clock_task = rq_clock_task(cpu_rq(cpu));`
`4273`	`4273`	`}`
`4274`	`4274`
`4275`	`4275`	`/* conditionally throttle active cfs_rq's from put_prev_entity() */`