sched/fair: Fix tg_set_cfs_bandwidth() deadlock on rq->lock

tg_set_cfs_bandwidth() sets cfs_b->timer_active to 0 to
force the period timer restart. It's not safe, because
can lead to deadlock, described in commit 927b54fccbf0:
"__start_cfs_bandwidth calls hrtimer_cancel while holding rq->lock,
waiting for the hrtimer to finish. However, if sched_cfs_period_timer
runs for another loop iteration, the hrtimer can attempt to take
rq->lock, resulting in deadlock."

Three CPUs must be involved:

  CPU0               CPU1                         CPU2
  take rq->lock      period timer fired
  ...                take cfs_b lock
  ...                ...                          tg_set_cfs_bandwidth()
  throttle_cfs_rq()  release cfs_b lock           take cfs_b lock
  ...                distribute_cfs_runtime()     timer_active = 0
  take cfs_b->lock   wait for rq->lock            ...
  __start_cfs_bandwidth()
  {wait for timer callback
   break if timer_active == 1}

So, CPU0 and CPU1 are deadlocked.

Instead of resetting cfs_b->timer_active, tg_set_cfs_bandwidth can
wait for period timer callbacks (ignoring cfs_b->timer_active) and
restart the timer explicitly.

Signed-off-by: Roman Gushchin <klamm@yandex-team.ru>
Reviewed-by: Ben Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/87wqdi9g8e.wl\%klamm@yandex-team.ru
Cc: pjt@google.com
Cc: chris.j.arges@canonical.com
Cc: gregkh@linuxfoundation.org
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

3 files changed, 6 insertions, 7 deletions

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 540542b93f9f..f3f48e7a2dda 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -7751,8 +7751,7 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota)
 	/* restart the period timer (if active) to handle new period expiry */
 	if (runtime_enabled && cfs_b->timer_active) {
 		/* force a reprogram */
-		cfs_b->timer_active = 0;
-		__start_cfs_bandwidth(cfs_b);
+		__start_cfs_bandwidth(cfs_b, true);
 	}
 	raw_spin_unlock_irq(&cfs_b->lock);
 
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index b4768c069392..8cbe2d2c16de 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3130,7 +3130,7 @@ static int assign_cfs_rq_runtime(struct cfs_rq *cfs_rq)
 		 */
 		if (!cfs_b->timer_active) {
 			__refill_cfs_bandwidth_runtime(cfs_b);
-			__start_cfs_bandwidth(cfs_b);
+			__start_cfs_bandwidth(cfs_b, false);
 		}
 
 		if (cfs_b->runtime > 0) {
@@ -3309,7 +3309,7 @@ static void throttle_cfs_rq(struct cfs_rq *cfs_rq)
 	raw_spin_lock(&cfs_b->lock);
 	list_add_tail_rcu(&cfs_rq->throttled_list, &cfs_b->throttled_cfs_rq);
 	if (!cfs_b->timer_active)
-		__start_cfs_bandwidth(cfs_b);
+		__start_cfs_bandwidth(cfs_b, false);
 	raw_spin_unlock(&cfs_b->lock);
 }
 
@@ -3691,7 +3691,7 @@ static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq)
 }
 
 /* requires cfs_b->lock, may release to reprogram timer */
-void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b)
+void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b, bool force)
 {
 	/*
 	 * The timer may be active because we're trying to set a new bandwidth
@@ -3706,7 +3706,7 @@ void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b)
 		cpu_relax();
 		raw_spin_lock(&cfs_b->lock);
 		/* if someone else restarted the timer then we're done */
-		if (cfs_b->timer_active)
+		if (!force && cfs_b->timer_active)
 			return;
 	}
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 456e492a3dca..369b4d663c42 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -278,7 +278,7 @@ extern void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b);
 extern int sched_group_set_shares(struct task_group *tg, unsigned long shares);
 
 extern void __refill_cfs_bandwidth_runtime(struct cfs_bandwidth *cfs_b);
-extern void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b);
+extern void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b, bool force);
 extern void unthrottle_cfs_rq(struct cfs_rq *cfs_rq);
 
 extern void free_rt_sched_group(struct task_group *tg);