-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 07/28/2014 01:51 PM, Vincent Guittot wrote:
If the CPU is used for handling lot of IRQs, trig a load balance to check if it's worth moving its tasks on another CPU that has more capacity.
As a sidenote, this will note generate more spurious ilb because we already trig an ilb if there is more than 1 busy cpu. If this cpu is the only one that has a task, we will trig the ilb once for migrating the task.
The nohz_kick_needed function has been cleaned up a bit while adding the new test
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org --- kernel/sched/fair.c | 69 +++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 49 insertions(+), 20 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 6843016..1cde8dd 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5969,6 +5969,14 @@ static bool update_sd_pick_busiest(struct lb_env *env, return true; }
- /* + * The group capacity is reduced probably because of
activity from other + * sched class or interrupts which use part of the available capacity + */ + if ((sg->sgc->capacity_orig * 100) > (sgs->group_capacity * + env->sd->imbalance_pct)) + return true; + return false; }
Isn't this already reflected in avg_load, by having avg_load increase due to the capacity decreasing when a cpu is busy with non-CFS loads?
Also, this part of update_sd_pick_busiest will not be reached in the !SD_ASYM_PACKING case once my patch is applied, so this is a small conflict between our series :)
- -- All rights reversed