Doug / smarc-fsl-linux-kernel

Download as
Browse Code ยป

Commit 7abc63b1bd412f7655b62ef3e35c3c11c5134636

Authored by Peter Zijlstra
Committed by Ingo Molnar
1 parent 42c62a589f
Exists in smarc-l5.0.0_1.0.0-ga and in 5 other branches imx_3.10.17_1.0.1_ga, imx_3.10.53_1.1.0_ga, imx_3.14.28_1.0.0_ga, smarc-imx_3.10.53_1.1.0_ga, smarc-imx_3.14.28_1.0.0_ga

sched/rt: Do not throttle when PI boosting 

When a runqueue has rt_runtime_us = 0 then the only way it can
accumulate rt_time is via PI boosting. That causes the runqueue
to be throttled and replenishing does not change anything due to
rt_runtime_us = 0. So avoid that situation by clearing rt_time and
skip the throttling alltogether.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
[ Changelog ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/n/tip-7x70cypsotjb4jvcor3edctk@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Showing 1 changed file with 18 additions and 2 deletions Side-by-side Diff

... ... @@ -857,8 +857,24 @@
857 857 return 0;
858 858  
859 859 if (rt_rq->rt_time > runtime) {
860   - rt_rq->rt_throttled = 1;
861   - printk_once(KERN_WARNING "sched: RT throttling activated\n");
  860 + struct rt_bandwidth *rt_b = sched_rt_bandwidth(rt_rq);
  861 +
  862 + /*
  863 + * Don't actually throttle groups that have no runtime assigned
  864 + * but accrue some time due to boosting.
  865 + */
  866 + if (likely(rt_b->rt_runtime)) {
  867 + rt_rq->rt_throttled = 1;
  868 + printk_once(KERN_WARNING "sched: RT throttling activated\n");
  869 + } else {
  870 + /*
  871 + * In case we did anyway, make it go away,
  872 + * replenishment is a joke, since it will replenish us
  873 + * with exactly 0 ns.
  874 + */
  875 + rt_rq->rt_time = 0;
  876 + }
  877 +
862 878 if (rt_rq_throttled(rt_rq)) {
863 879 sched_rt_rq_dequeue(rt_rq);
864 880 return 1;