Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1449489

[PATCH v3 13/13] sched/fair: Avoid pulling tasks from non-overloaded higher capacity groups

Show all headers | View raw

For asymmetric cpu capacity systems it is counter-productive for
throughput if low capacity cpus are pulling tasks from non-overloaded
cpus with higher capacity. The assumption is that higher cpu capacity is
preferred over running alone in a group with lower cpu capacity.

This patch rejects higher cpu capacity groups with one or less task per
cpu as potential busiest group which could otherwise lead to a series of
failing load-balancing attempts leading to a force-migration.

cc: Ingo Molnar <mingo@redhat.com>
cc: Peter Zijlstra <peterz@infradead.org>

Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
---
 kernel/sched/fair.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index f3019b85b6e3..df07235f743d 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6895,6 +6895,17 @@ group_is_overloaded(struct lb_env *env, struct sg_lb_stats *sgs)
 	return false;
 }
 
+/*
+ * group_smaller_cpu_capacity: Returns true if sched_group sg has smaller
+ * per-cpu capacity than sched_group ref.
+ */
+static inline bool
+group_smaller_cpu_capacity(struct sched_group *sg, struct sched_group *ref)
+{
+	return sg->sgc->min_capacity * capacity_margin <
+						ref->sgc->min_capacity * 1024;
+}
+
 static inline enum
 group_type group_classify(struct sched_group *group,
 			  struct sg_lb_stats *sgs)
@@ -6998,6 +7009,20 @@ static bool update_sd_pick_busiest(struct lb_env *env,
 	if (sgs->avg_load <= busiest->avg_load)
 		return false;
 
+	if (!(env->sd->flags & SD_ASYM_CPUCAPACITY))
+		goto asym_packing;
+
+	/*
+	 * Candidate sg has no more than one task per cpu and
+	 * has higher per-cpu capacity. Migrating tasks to less
+	 * capable cpus may harm throughput. Maximize throughput,
+	 * power/energy consequences are not considered.
+	 */
+	if (sgs->sum_nr_running <= sgs->group_weight &&
+	    group_smaller_cpu_capacity(sds->local, sg))
+		return false;
+
+asym_packing:
 	/* This is the busiest node in its class. */
 	if (!(env->sd->flags & SD_ASYM_PACKING))
 		return true;
-- 
1.9.1

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH v3 00/13] sched: Clean-ups and asymmetric cpu capacity support Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-25 15:40 +0200
  [PATCH v3 09/13] sched/fair: Let asymmetric cpu configurations balance at wake-up Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-25 15:40 +0200
  [PATCH v3 13/13] sched/fair: Avoid pulling tasks from non-overloaded higher capacity groups Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-25 15:40 +0200
  [PATCH v3 06/13] sched/core: Pass child domain into sd_init Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-25 15:40 +0200
  [PATCH v3 05/13] sched: Introduce SD_ASYM_CPUCAPACITY sched_domain topology flag Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-25 15:40 +0200
  [PATCH v3 08/13] sched: Store maximum per-cpu capacity in root domain Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-25 15:40 +0200
  [PATCH v3 10/13] sched/fair: Compute task/cpu utilization at wake-up more correctly Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-25 15:40 +0200
  [PATCH v3 04/13] sched/core: Remove unnecessary null-pointer check Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-25 15:40 +0200
  [PATCH v3 03/13] sched/fair: Optimize find_idlest_cpu() when there is no choice Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-25 15:40 +0200

csiph-web