sched: Ensure cpu_power periodic update

With a lot of small tasks, the softirq sched is nearly never called
when no_hz is enabled. In this case load_balance() is mainly called
with the newly_idle mode which doesn't update the cpu_power.

Add a next_update field which ensure a maximum update period when
there is short activity.

Having stale cpu_power information can skew the load-balancing
decisions, this is cured by the guaranteed update.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1323717668-2143-1-git-send-email-vincent.guittot@linaro.org

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 0e1959568836f2b6328da27b2f492e90f0742c08..92313a3f6f771a5659ee1351f4a2b76bf2f8f7a8 100644 (file)
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -905,6 +905,7 @@ struct sched_group_power {
         * single CPU.
         */
        unsigned int power, power_orig;
+       unsigned long next_update;
        /*
         * Number of busy cpus in this group.
         */

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 7c6414fc669de4f09dc03fc763aaddd8040f98bf..8e77a6bd597bafc0857b5478eea160d1c4fabe46 100644 (file)
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -215,6 +215,8 @@ calc_delta_mine(unsigned long delta_exec, unsigned long weight,
 
 const struct sched_class fair_sched_class;
 
+static unsigned long __read_mostly max_load_balance_interval = HZ/10;
+
 /**************************************************************
  * CFS operations on generic schedulable entities:
  */
@@ -3776,6 +3778,11 @@ void update_group_power(struct sched_domain *sd, int cpu)
        struct sched_domain *child = sd->child;
        struct sched_group *group, *sdg = sd->groups;
        unsigned long power;
+       unsigned long interval;
+
+       interval = msecs_to_jiffies(sd->balance_interval);
+       interval = clamp(interval, 1UL, max_load_balance_interval);
+       sdg->sgp->next_update = jiffies + interval;
 
        if (!child) {
                update_cpu_power(sd, cpu);
@@ -3883,12 +3890,15 @@ static inline void update_sg_lb_stats(struct sched_domain *sd,
         * domains. In the newly idle case, we will allow all the cpu's
         * to do the newly idle load balance.
         */
-       if (idle != CPU_NEWLY_IDLE && local_group) {
-               if (balance_cpu != this_cpu) {
-                       *balance = 0;
-                       return;
-               }
-               update_group_power(sd, this_cpu);
+       if (local_group) {
+               if (idle != CPU_NEWLY_IDLE) {
+                       if (balance_cpu != this_cpu) {
+                               *balance = 0;
+                               return;
+                       }
+                       update_group_power(sd, this_cpu);
+               } else if (time_after_eq(jiffies, group->sgp->next_update))
+                       update_group_power(sd, this_cpu);
        }
 
        /* Adjust by relative CPU power of the group */
@@ -4945,8 +4955,6 @@ static int __cpuinit sched_ilb_notifier(struct notifier_block *nfb,
 
 static DEFINE_SPINLOCK(balancing);
 
-static unsigned long __read_mostly max_load_balance_interval = HZ/10;
-
 /*
  * Scale the max load_balance interval with the number of CPUs in the system.
  * This trades load-balance latency on larger machines for less cross talk.