sched/fair: Disable runtime_enabled on dying rq
authorKirill Tkhai <ktkhai@parallels.com>
Wed, 25 Jun 2014 08:19:42 +0000 (12:19 +0400)
committerIngo Molnar <mingo@kernel.org>
Sat, 5 Jul 2014 09:17:42 +0000 (11:17 +0200)
commit0e59bdaea75f12a7d7c03672f4ac22c0119a1bc0
treec0f74a1555354d622d64fb04b884c8a7f2bcb8d5
parenta22b4b012340b988dbe7a58461d6fcc582f34aa0
sched/fair: Disable runtime_enabled on dying rq

We kill rq->rd on the CPU_DOWN_PREPARE stage:

cpuset_cpu_inactive -> cpuset_update_active_cpus -> partition_sched_domains ->
-> cpu_attach_domain -> rq_attach_root -> set_rq_offline

This unthrottles all throttled cfs_rqs.

But the cpu is still able to call schedule() till

take_cpu_down->__cpu_disable()

is called from stop_machine.

This case the tasks from just unthrottled cfs_rqs are pickable
in a standard scheduler way, and they are picked by dying cpu.
The cfs_rqs becomes throttled again, and migrate_tasks()
in migration_call skips their tasks (one more unthrottle
in migrate_tasks()->CPU_DYING does not happen, because rq->rd
is already NULL).

Patch sets runtime_enabled to zero. This guarantees, the runtime
is not accounted, and the cfs_rqs won't exceed given
cfs_rq->runtime_remaining = 1, and tasks will be pickable
in migrate_tasks(). runtime_enabled is recalculated again
when rq becomes online again.

Ben Segall also noticed, we always enable runtime in
tg_set_cfs_bandwidth(). Actually, we should do that for online
cpus only. To prevent races with unthrottle_offline_cfs_rqs()
we take get_online_cpus() lock.

Reviewed-by: Ben Segall <bsegall@google.com>
Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Signed-off-by: Kirill Tkhai <ktkhai@parallels.com>
CC: Konstantin Khorenko <khorenko@parallels.com>
CC: Paul Turner <pjt@google.com>
CC: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1403684382.3462.42.camel@tkhai
Signed-off-by: Ingo Molnar <mingo@kernel.org>
kernel/sched/core.c
kernel/sched/fair.c