[RESEND PATCH v2 1/2] sched/rt: add utilization tracking

[Vincent Guittot]

schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs
tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks
are preempted by rt tasks and in this case util_avg reflects the remaining
capacity that is used by cfs tasks but not what cfs tasks want to use. In such
case, schedutil can select a lower OPP when cfs task runs whereas the CPU is
overloaded. In order to have a more accurate view of the utilization of the
CPU, we track the utilization that is used by RT tasks.
DL tasks are not taken into account as they have their own utilization
tracking mecanism.

We don't use rt_avg which doesn't have the same dynamic as PELT and which
can include IRQ time.

Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>

---

Change since v1:
- rebase on tip/sched/core

There were several comments on v1:
- As raised by Peter for v1, if IRQ time is taken into account in
  rt_avg, it will not be accounted in rq->clock_task. This means that cfs
  utilization is not affected by some extra contributions or decays
  because of IRQ.  
- Regading the sync of rt and cfs utilization, both cfs and rt use the same
  rq->clock_task. Then, we have the same issue than cfs regarding blocked value.
  The utilization of idle cfs/rt rqs are not updated regularly but only when a
  load_balance is triggered (more precisely a call to update_blocked_average).
  I'd like to fix this issue for both cfs and rt with a separate patch that
  will ensure that utilization (and load) are updated regularly even for
  idle CPUs
- One last open question is the location of rt utilization function in fair.c
  file. PELT related funtions should probably move in a dedicated pelt.c file.
  This would also help to address one comment about having a place to update
  metrics of NOHZ idle CPUs. Thought ?

 kernel/sched/fair.c  | 21 +++++++++++++++++++++
 kernel/sched/rt.c    |  9 +++++++++
 kernel/sched/sched.h |  3 +++
 3 files changed, 33 insertions(+)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 008c514..50fe0a2 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3012,6 +3012,17 @@ __update_load_avg_cfs_rq(u64 now, int cpu, struct cfs_rq *cfs_rq)
 			cfs_rq->curr != NULL, cfs_rq);
 }
 
+int update_rt_rq_load_avg(u64 now, int cpu, struct rt_rq *rt_rq, int running)
+{
+	int ret;
+
+	ret = ___update_load_avg(now, cpu, &rt_rq->avg, 0, running, NULL);
+
+
+	return ret;
+}
+
+
 /*
  * Signed add and clamp on underflow.
  *
@@ -3539,6 +3550,11 @@ update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq, bool update_freq)
 	return 0;
 }
 
+int update_rt_rq_load_avg(u64 now, int cpu, struct rt_rq *rt_rq, int running)
+{
+	return 0;
+}
+
 #define UPDATE_TG	0x0
 #define SKIP_AGE_LOAD	0x0
 
@@ -6932,6 +6948,10 @@ static void update_blocked_averages(int cpu)
 		if (cfs_rq_is_decayed(cfs_rq))
 			list_del_leaf_cfs_rq(cfs_rq);
 	}
+
+	update_rt_rq_load_avg(rq_clock_task(rq), cpu, &rq->rt, 0);
+
+
 	rq_unlock_irqrestore(rq, &rf);
 }
 
@@ -6991,6 +7011,7 @@ static inline void update_blocked_averages(int cpu)
 	rq_lock_irqsave(rq, &rf);
 	update_rq_clock(rq);
 	update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq, true);
+	update_rt_rq_load_avg(rq_clock_task(rq), cpu, &rq->rt, 0);
 	rq_unlock_irqrestore(rq, &rf);
 }
 
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index 45caf93..e72d572 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -1534,6 +1534,8 @@ static struct task_struct *_pick_next_task_rt(struct rq *rq)
 	return p;
 }
 
+extern int update_rt_rq_load_avg(u64 now, int cpu, struct rt_rq *rt_rq, int running);
+
 static struct task_struct *
 pick_next_task_rt(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 {
@@ -1579,6 +1581,10 @@ pick_next_task_rt(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 
 	queue_push_tasks(rq);
 
+	if (p)
+		update_rt_rq_load_avg(rq_clock_task(rq), cpu_of(rq), rt_rq,
+				rq->curr->sched_class == &rt_sched_class);
+
 	return p;
 }
 
@@ -1586,6 +1592,8 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p)
 {
 	update_curr_rt(rq);
 
+	update_rt_rq_load_avg(rq_clock_task(rq), cpu_of(rq), &rq->rt, 1);
+
 	/*
 	 * The previous task needs to be made eligible for pushing
 	 * if it is still active
@@ -2368,6 +2376,7 @@ static void task_tick_rt(struct rq *rq, struct task_struct *p, int queued)
 	struct sched_rt_entity *rt_se = &p->rt;
 
 	update_curr_rt(rq);
+	update_rt_rq_load_avg(rq_clock_task(rq), cpu_of(rq), &rq->rt, 1);
 
 	watchdog(rq, p);
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index eeef1a3..e7ee5b0 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -524,6 +524,9 @@ struct rt_rq {
 	unsigned long rt_nr_total;
 	int overloaded;
 	struct plist_head pushable_tasks;
+
+	struct sched_avg avg;
+
 #ifdef HAVE_RT_PUSH_IPI
 	int push_flags;
 	int push_cpu;
-- 
2.7.4