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

Commit 699140ba authored by Tejun Heo's avatar Tejun Heo
Browse files

cpuset: drop async_rebuild_sched_domains() 



In general, we want to make cgroup_mutex one of the outermost locks
and be able to use get_online_cpus() and friends from cgroup methods.
With cpuset hotplug made async, get_online_cpus() can now be nested
inside cgroup_mutex.

Currently, cpuset avoids nesting get_online_cpus() inside cgroup_mutex
by bouncing sched_domain rebuilding to a work item.  As such nesting
is allowed now, remove the workqueue bouncing code and always rebuild
sched_domains synchronously.  This also nests sched_domains_mutex
inside cgroup_mutex, which is intended and should be okay.

Signed-off-by: default avatarTejun Heo <tj@kernel.org>
Acked-by: default avatarLi Zefan <lizefan@huawei.com>
parent 3a5a6d0c

kernel/cpuset.c

+16 −60
Original line number Original line Diff line number Diff line
@@ -60,14 +60,6 @@ 
#include <linux/workqueue.h>

#include <linux/workqueue.h>

#include <linux/cgroup.h>

#include <linux/cgroup.h>





/*

 * Workqueue for cpuset related tasks.

 *

 * Using kevent workqueue may cause deadlock when memory_migrate

 * is set. So we create a separate workqueue thread for cpuset.

 */

static struct workqueue_struct *cpuset_wq;



/*

/*

 * Tracks how many cpusets are currently defined in system.

 * Tracks how many cpusets are currently defined in system.

 * When there is only one cpuset (the root cpuset) we can

 * When there is only one cpuset (the root cpuset) we can
@@ -753,25 +745,25 @@ static int generate_sched_domains(cpumask_var_t **domains, 
/*

/*

 * Rebuild scheduler domains.

 * Rebuild scheduler domains.

 *

 *

 * Call with neither cgroup_mutex held nor within get_online_cpus().

 * If the flag 'sched_load_balance' of any cpuset with non-empty

 * Takes both cgroup_mutex and get_online_cpus().

 * 'cpus' changes, or if the 'cpus' allowed changes in any cpuset

 * which has that flag enabled, or if any cpuset with a non-empty

 * 'cpus' is removed, then call this routine to rebuild the

 * scheduler's dynamic sched domains.

 *

 *

 * Cannot be directly called from cpuset code handling changes

 * Call with cgroup_mutex held.  Takes get_online_cpus().

 * to the cpuset pseudo-filesystem, because it cannot be called

 * from code that already holds cgroup_mutex.

 */

 */

static void do_rebuild_sched_domains(struct work_struct *unused)

static void rebuild_sched_domains_locked(void)

{

{

	struct sched_domain_attr *attr;

	struct sched_domain_attr *attr;

	cpumask_var_t *doms;

	cpumask_var_t *doms;

	int ndoms;

	int ndoms;





	WARN_ON_ONCE(!cgroup_lock_is_held());

	get_online_cpus();

	get_online_cpus();





	/* Generate domain masks and attrs */

	/* Generate domain masks and attrs */

	cgroup_lock();

	ndoms = generate_sched_domains(&doms, &attr);

	ndoms = generate_sched_domains(&doms, &attr);

	cgroup_unlock();





	/* Have scheduler rebuild the domains */

	/* Have scheduler rebuild the domains */

	partition_sched_domains(ndoms, doms, attr);

	partition_sched_domains(ndoms, doms, attr);
@@ -779,7 +771,7 @@ static void do_rebuild_sched_domains(struct work_struct *unused) 
	put_online_cpus();

	put_online_cpus();

}

}

#else /* !CONFIG_SMP */

#else /* !CONFIG_SMP */

static void do_rebuild_sched_domains(struct work_struct *unused)

static void rebuild_sched_domains_locked(void)

{

{

}

}
@@ -791,44 +783,11 @@ static int generate_sched_domains(cpumask_var_t **domains, 
}

}

#endif /* CONFIG_SMP */

#endif /* CONFIG_SMP */





static DECLARE_WORK(rebuild_sched_domains_work, do_rebuild_sched_domains);



/*

 * Rebuild scheduler domains, asynchronously via workqueue.

 *

 * If the flag 'sched_load_balance' of any cpuset with non-empty

 * 'cpus' changes, or if the 'cpus' allowed changes in any cpuset

 * which has that flag enabled, or if any cpuset with a non-empty

 * 'cpus' is removed, then call this routine to rebuild the

 * scheduler's dynamic sched domains.

 *

 * The rebuild_sched_domains() and partition_sched_domains()

 * routines must nest cgroup_lock() inside get_online_cpus(),

 * but such cpuset changes as these must nest that locking the

 * other way, holding cgroup_lock() for much of the code.

 *

 * So in order to avoid an ABBA deadlock, the cpuset code handling

 * these user changes delegates the actual sched domain rebuilding

 * to a separate workqueue thread, which ends up processing the

 * above do_rebuild_sched_domains() function.

 */

static void async_rebuild_sched_domains(void)

{

	queue_work(cpuset_wq, &rebuild_sched_domains_work);

}



/*

 * Accomplishes the same scheduler domain rebuild as the above

 * async_rebuild_sched_domains(), however it directly calls the

 * rebuild routine synchronously rather than calling it via an

 * asynchronous work thread.

 *

 * This can only be called from code that is not holding

 * cgroup_mutex (not nested in a cgroup_lock() call.)

 */

void rebuild_sched_domains(void)

void rebuild_sched_domains(void)

{

{

	do_rebuild_sched_domains(NULL);

	cgroup_lock();

	rebuild_sched_domains_locked();

	cgroup_unlock();

}

}





/**

/**
@@ -948,7 +907,7 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs, 
	heap_free(&heap);

	heap_free(&heap);





	if (is_load_balanced)

	if (is_load_balanced)

		async_rebuild_sched_domains();

		rebuild_sched_domains_locked();

	return 0;

	return 0;

}

}
@@ -1196,7 +1155,7 @@ static int update_relax_domain_level(struct cpuset *cs, s64 val) 
		cs->relax_domain_level = val;

		cs->relax_domain_level = val;

		if (!cpumask_empty(cs->cpus_allowed) &&

		if (!cpumask_empty(cs->cpus_allowed) &&

		    is_sched_load_balance(cs))

		    is_sched_load_balance(cs))

			async_rebuild_sched_domains();

			rebuild_sched_domains_locked();

	}

	}





	return 0;

	return 0;
@@ -1288,7 +1247,7 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, 
	mutex_unlock(&callback_mutex);

	mutex_unlock(&callback_mutex);





	if (!cpumask_empty(trialcs->cpus_allowed) && balance_flag_changed)

	if (!cpumask_empty(trialcs->cpus_allowed) && balance_flag_changed)

		async_rebuild_sched_domains();

		rebuild_sched_domains_locked();





	if (spread_flag_changed)

	if (spread_flag_changed)

		update_tasks_flags(cs, &heap);

		update_tasks_flags(cs, &heap);
@@ -1925,7 +1884,7 @@ static void cpuset_css_offline(struct cgroup *cgrp) 
/*

/*

 * If the cpuset being removed has its flag 'sched_load_balance'

 * If the cpuset being removed has its flag 'sched_load_balance'

 * enabled, then simulate turning sched_load_balance off, which

 * enabled, then simulate turning sched_load_balance off, which

 * will call async_rebuild_sched_domains().

 * will call rebuild_sched_domains_locked().

 */

 */





static void cpuset_css_free(struct cgroup *cont)

static void cpuset_css_free(struct cgroup *cont)
@@ -2237,9 +2196,6 @@ void __init cpuset_init_smp(void) 
	top_cpuset.mems_allowed = node_states[N_MEMORY];

	top_cpuset.mems_allowed = node_states[N_MEMORY];





	hotplug_memory_notifier(cpuset_track_online_nodes, 10);

	hotplug_memory_notifier(cpuset_track_online_nodes, 10);



	cpuset_wq = create_singlethread_workqueue("cpuset");

	BUG_ON(!cpuset_wq);

}

}





/**

/**