cgroup/rstat: Reduce cpu_lock hold time in cgroup_rstat_flush_locked()
JIRA: https://issues.redhat.com/browse/RHEL-21091
MR: !3564 (merged)
The occasional latency spike in the execution of cgroup_rstat_updated() is likely caused by percpu cpu_lock contention with cgroup_rstat_flush_locked(). This MR includes upstream commits that reduce the cpu_lock hold time so to reduce the maximum latency spike to cgroup_rstat_updated().
Signed-off-by: Waiman Long longman@redhat.com
Edited by Waiman Long