[PATCH v4 0/3] sched/fair: some fixes for asym_packing

[PATCH v4 0/3] sched/fair: some fixes for asym_packing

[Vincent Guittot]

During the review of misfit task patchset, Morten and Valentin raised some
problems with the use of SD_ASYM_PACKING flag on asymmetric system like
hikey960 arm64 big/LITTLE platform. The study of the use cases has shown
some problems that can happen for every systems that use the flag.

The 3 patches fixes the problems raised for lmbench and the rt-app UC that
creates 2 tasks that start as small tasks and then become suddenly always
running tasks. (I can provide the rt-app json if needed)

Changes since v3:
- simplify imbalance computation 

Changes since v2:
- include other active balance reasons
- set imbalance to avg_load as suggested by Valentin

Changes since v1:
- rebase on tip/sched/core
- changes asym_active_balance() as suggested by Peter

Vincent Guittot (3):
  sched/fair: fix rounding issue for asym packing
  sched/fair: trigger asym_packing during idle load balance
  sched/fair: fix unnecessary increase of balance interval

 kernel/sched/fair.c | 44 ++++++++++++++++++++++++++++----------------
 1 file changed, 28 insertions(+), 16 deletions(-)

-- 
2.7.4

[PATCH v4 1/3] sched/fair: fix rounding issue for asym packing

[Vincent Guittot]

When check_asym_packing() is triggered, the imbalance is set to :
  busiest_stat.avg_load * busiest_stat.group_capacity / SCHED_CAPACITY_SCALE
But busiest_stat.avg_load equals :
  sgs->group_load *SCHED_CAPACITY_SCALE / sgs->group_capacity
These divisions can generate a rounding that will make imbalance slightly
lower than the weighted load of the cfs_rq.
But this is enough to skip the rq in find_busiest_queue and prevents asym
migration to happen.

Directly set imbalance to busiest's sgs->group_load to remove the rounding.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
---
 kernel/sched/fair.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index ca46964..1e4bed4 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -8476,9 +8476,7 @@ static int check_asym_packing(struct lb_env *env, struct sd_lb_stats *sds)
 	if (sched_asym_prefer(busiest_cpu, env->dst_cpu))
 		return 0;
 
-	env->imbalance = DIV_ROUND_CLOSEST(
-		sds->busiest_stat.avg_load * sds->busiest_stat.group_capacity,
-		SCHED_CAPACITY_SCALE);
+	env->imbalance = sds->busiest_stat.group_load;
 
 	return 1;
 }
-- 
2.7.4

[PATCH v4 2/3] sched/fair: trigger asym_packing during idle load balance

[Vincent Guittot]

newly idle load balance is not always triggered when a cpu becomes idle.
This prevent the scheduler to get a chance to migrate task for asym packing.
Enable active migration because of asym packing during idle load balance too.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
---
 kernel/sched/fair.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 1e4bed4..bd3b0ac 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -8853,7 +8853,7 @@ static int need_active_balance(struct lb_env *env)
 {
 	struct sched_domain *sd = env->sd;
 
-	if (env->idle == CPU_NEWLY_IDLE) {
+	if (env->idle != CPU_NOT_IDLE) {
 
 		/*
 		 * ASYM_PACKING needs to force migrate tasks from busy but
-- 
2.7.4

[PATCH v4 3/3] sched/fair: fix unnecessary increase of balance interval

[Vincent Guittot]

In case of active balance, we increase the balance interval to cover
pinned tasks cases not covered by all_pinned logic. Neverthless, the
active migration triggered by asym packing should be treated as the normal
unbalanced case and reset the interval to default value otherwise active
migration for asym_packing can be easily delayed for hundreds of ms
because of this pinned task detection mecanism.
The same happen to other conditions tested in need_active_balance() like
mistfit task and when the capacity of src_cpu is reduced compared to
dst_cpu (see comments in need_active_balance() for details).

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
---
 kernel/sched/fair.c | 40 +++++++++++++++++++++++++++-------------
 1 file changed, 27 insertions(+), 13 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index bd3b0ac..0e17991 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -8849,21 +8849,25 @@ static struct rq *find_busiest_queue(struct lb_env *env,
  */
 #define MAX_PINNED_INTERVAL	512
 
-static int need_active_balance(struct lb_env *env)
+static inline bool
+asym_active_balance(struct lb_env *env)
 {
-	struct sched_domain *sd = env->sd;
+	/*
+	 * ASYM_PACKING needs to force migrate tasks from busy but
+	 * lower priority CPUs in order to pack all tasks in the
+	 * highest priority CPUs.
+	 */
+	return env->idle != CPU_NOT_IDLE && (env->sd->flags & SD_ASYM_PACKING) &&
+	       sched_asym_prefer(env->dst_cpu, env->src_cpu);
+}
 
-	if (env->idle != CPU_NOT_IDLE) {
+static inline bool
+voluntary_active_balance(struct lb_env *env)
+{
+	struct sched_domain *sd = env->sd;
 
-		/*
-		 * ASYM_PACKING needs to force migrate tasks from busy but
-		 * lower priority CPUs in order to pack all tasks in the
-		 * highest priority CPUs.
-		 */
-		if ((sd->flags & SD_ASYM_PACKING) &&
-		    sched_asym_prefer(env->dst_cpu, env->src_cpu))
-			return 1;
-	}
+	if (asym_active_balance(env))
+		return 1;
 
 	/*
 	 * The dst_cpu is idle and the src_cpu CPU has only 1 CFS task.
@@ -8881,6 +8885,16 @@ static int need_active_balance(struct lb_env *env)
 	if (env->src_grp_type == group_misfit_task)
 		return 1;
 
+	return 0;
+}
+
+static int need_active_balance(struct lb_env *env)
+{
+	struct sched_domain *sd = env->sd;
+
+	if (voluntary_active_balance(env))
+		return 1;
+
 	return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2);
 }
 
@@ -9142,7 +9156,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 	} else
 		sd->nr_balance_failed = 0;
 
-	if (likely(!active_balance)) {
+	if (likely(!active_balance) || voluntary_active_balance(&env)) {
 		/* We were unbalanced, so reset the balancing interval */
 		sd->balance_interval = sd->min_interval;
 	} else {
-- 
2.7.4

[PATCH v4 1/3] sched/fair: fix rounding issue for asym packing

[Vincent Guittot]

When check_asym_packing() is triggered, the imbalance is set to :
  busiest_stat.avg_load * busiest_stat.group_capacity / SCHED_CAPACITY_SCALE
But busiest_stat.avg_load equals :
  sgs->group_load *SCHED_CAPACITY_SCALE / sgs->group_capacity
These divisions can generate a rounding that will make imbalance slightly
lower than the weighted load of the cfs_rq.
But this is enough to skip the rq in find_busiest_queue and prevents asym
migration to happen.

Directly set imbalance to busiest's sgs->group_load to remove the rounding.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
---
 kernel/sched/fair.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index ca46964..1e4bed4 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -8476,9 +8476,7 @@ static int check_asym_packing(struct lb_env *env, struct sd_lb_stats *sds)
 	if (sched_asym_prefer(busiest_cpu, env->dst_cpu))
 		return 0;
 
-	env->imbalance = DIV_ROUND_CLOSEST(
-		sds->busiest_stat.avg_load * sds->busiest_stat.group_capacity,
-		SCHED_CAPACITY_SCALE);
+	env->imbalance = sds->busiest_stat.group_load;
 
 	return 1;
 }
-- 
2.7.4