Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 78af2bab authored by Olav Haugan's avatar Olav Haugan Committed by Gerrit - the friendly Code Review server
Browse files

sched: Fix race condition with active balance 



There is a race condition between checking for whether an active load
balance request has been set and clearing the request. A cpu might have
an active load balance request set and queued but not executed yet.
Before the load balance request is executed the request flag might be
cleared by cpu isolation. Then subsequently the load balancer or tick
might try to do another active load balance.  This can cause the same
active load balance work to be queued twice causing report of list
corruption.

Fix this by moving the clearing of the request to the stopper thread and
ensuring that load balance will not try to queue a request on an
already isolated cpu.

Signed-off-by: default avatarOlav Haugan <ohaugan@codeaurora.org>
[markivx: fix minor conflict on 4.14]
Signed-off-by: default avatarVikram Mulukutla <markivx@codeaurora.org>

Change-Id: Idcdab548bf026681e005703614799f0ff39cf865
Signed-off-by: default avatarSatya Durga Srinivasu Prabhala <satyap@codeaurora.org>
parent 728bf6c8

kernel/sched/core.c

+9 −6
Original line number Diff line number Diff line
@@ -1835,7 +1835,7 @@ void scheduler_ipi(void) 
	/*

	 * Check if someone kicked us for doing the nohz idle load balance.

	 */

	if (unlikely(got_nohz_idle_kick())) {

	if (unlikely(got_nohz_idle_kick()) && !cpu_isolated(cpu)) {

		this_rq()->idle_balance = 1;

		raise_softirq_irqoff(SCHED_SOFTIRQ);

	}
@@ -5678,7 +5678,6 @@ void set_rq_offline(struct rq *rq); 


int do_isolation_work_cpu_stop(void *data)

{

	unsigned long flags;

	unsigned int cpu = smp_processor_id();

	struct rq *rq = cpu_rq(cpu);

	struct rq_flags rf;
@@ -5687,9 +5686,12 @@ int do_isolation_work_cpu_stop(void *data) 


	irq_migrate_all_off_this_cpu();



	local_irq_disable();



	sched_ttwu_pending();



	/* Update our root-domain */

	raw_spin_lock_irqsave(&rq->lock, flags);

	raw_spin_lock(&rq->lock);



	/*

	 * Temporarily mark the rq as offline. This will allow us to
@@ -5701,15 +5703,16 @@ int do_isolation_work_cpu_stop(void *data) 
	}



	migrate_tasks(rq, &rf, false);

	if (rq->rd)

		set_rq_online(rq);

	raw_spin_unlock_irqrestore(&rq->lock, flags);

	raw_spin_unlock(&rq->lock);



	/*

	 * We might have been in tickless state. Clear NOHZ flags to avoid

	 * us being kicked for helping out with balancing

	 */

	nohz_balance_clear_nohz_mask(cpu);



	clear_walt_request(cpu);

	local_irq_enable();

	return 0;

}

kernel/sched/fair.c

+17 −4
Original line number Diff line number Diff line
@@ -9840,6 +9840,15 @@ static int need_active_balance(struct lb_env *env) 
	return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2);

}



static int group_balance_cpu_not_isolated(struct sched_group *sg)

{

	cpumask_t cpus;



	cpumask_and(&cpus, sched_group_span(sg), group_balance_mask(sg));

	cpumask_andnot(&cpus, &cpus, cpu_isolated_mask);

	return cpumask_first(&cpus);

}



static int active_load_balance_cpu_stop(void *data);



static int should_we_balance(struct lb_env *env)
@@ -9863,7 +9872,7 @@ static int should_we_balance(struct lb_env *env) 


	/* Try to find first idle cpu */

	for_each_cpu_and(cpu, group_balance_mask(sg), env->cpus) {

		if (!idle_cpu(cpu))

		if (!idle_cpu(cpu) || cpu_isolated(cpu))

			continue;



		balance_cpu = cpu;
@@ -9871,7 +9880,7 @@ static int should_we_balance(struct lb_env *env) 
	}



	if (balance_cpu == -1)

		balance_cpu = group_balance_cpu(sg);

		balance_cpu = group_balance_cpu_not_isolated(sg);



	/*

	 * First idle cpu or the first cpu(busiest) in this sched group
@@ -10079,7 +10088,8 @@ static int load_balance(int this_cpu, struct rq *this_rq, 
			 * ->active_balance_work.  Once set, it's cleared

			 * only after active load balance is finished.

			 */

			if (!busiest->active_balance) {

			if (!busiest->active_balance &&

			    !cpu_isolated(cpu_of(busiest))) {

				busiest->active_balance = 1;

				busiest->push_cpu = this_cpu;

				active_balance = 1;
@@ -10685,6 +10695,7 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) 
	/* Earliest time when we have to do rebalance again */

	unsigned long next_balance = jiffies + 60*HZ;

	int update_next_balance = 0;

	cpumask_t cpus;



	if (idle != CPU_IDLE ||

	    !test_bit(NOHZ_BALANCE_KICK, nohz_flags(this_cpu)))
@@ -10707,7 +10718,9 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) 
		nohz.next_update = jiffies + msecs_to_jiffies(LOAD_AVG_PERIOD);

	rcu_read_unlock();



	for_each_cpu(balance_cpu, nohz.idle_cpus_mask) {

	cpumask_andnot(&cpus, nohz.idle_cpus_mask, cpu_isolated_mask);



	for_each_cpu(balance_cpu, &cpus) {

		if (balance_cpu == this_cpu || !idle_cpu(balance_cpu))

			continue;