sched: Reduce update_group_power() calls
authorPeter Zijlstra <a.p.zijlstra@chello.nl>
Fri, 9 Jul 2010 13:15:43 +0000 (15:15 +0200)
committerIngo Molnar <mingo@elte.hu>
Sat, 17 Jul 2010 10:05:14 +0000 (12:05 +0200)
commitbbc8cb5baead9607309583b20873ab0cc8d89eaf
tree425cfb2772ef50c1858a7d833408ad1bcc423f51
parent5343bdb8fd076f16edc9d113a9e35e2a1d1f4966
sched: Reduce update_group_power() calls

Currently we update cpu_power() too often, update_group_power() only
updates the local group's cpu_power but it gets called for all groups.

Furthermore, CPU_NEWLY_IDLE invocations will result in all cpus
calling it, even though a slow update of cpu_power is sufficient.

Therefore move the update under 'idle != CPU_NEWLY_IDLE &&
local_group' to reduce superfluous invocations.

Reported-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1278612989.1900.176.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/sched_fair.c