sched/eevdf: More PELT vs DELAYED_DEQUEUE

Vincent and Dietmar noted that while
commit fc1892becd56 ("sched/eevdf: Fixup PELT vs DELAYED_DEQUEUE") fixes
the entity runnable stats, it does not adjust the cfs_rq runnable stats,
which are based off of h_nr_running.

Track h_nr_delayed such that we can discount those and adjust the
signal.

Fixes: fc1892becd56 ("sched/eevdf: Fixup PELT vs DELAYED_DEQUEUE")
Closes: https://lore.kernel.org/lkml/[e-mail]/
Closes: https://lore.kernel.org/lkml/CAKfTPtCNUvWE_GX5LyvTF-WdxUT=ZgvZZv-4t=[e-mail]/
[ Fixes checkpatch warnings and rebased ]
Signed-off-by: Peter Zijlstra (Intel) <[e-mail]>
Reported-by: Dietmar Eggemann <[e-mail]>
Reported-by: Vincent Guittot <[e-mail]>
Signed-off-by: "Peter Zijlstra (Intel)" <[e-mail]>
Signed-off-by: Vincent Guittot <[e-mail]>
Signed-off-by: Peter Zijlstra (Intel) <[e-mail]>
Reviewed-by: Dietmar Eggemann <[e-mail]>
Tested-by: K Prateek Nayak <[e-mail]>
Link: https://lore.kernel.org/r/[e-mail]

kernel/sched/debug.c

···
 	SEQ_printf(m, "  .%-30s: %Ld.%06ld\n", "spread", SPLIT_NS(spread));
 	SEQ_printf(m, "  .%-30s: %d\n", "nr_running", cfs_rq->nr_running);
 	SEQ_printf(m, "  .%-30s: %d\n", "h_nr_running", cfs_rq->h_nr_running);
+	SEQ_printf(m, "  .%-30s: %d\n", "h_nr_delayed", cfs_rq->h_nr_delayed);
 	SEQ_printf(m, "  .%-30s: %d\n", "idle_nr_running",
 			cfs_rq->idle_nr_running);
 	SEQ_printf(m, "  .%-30s: %d\n", "idle_h_nr_running",

kernel/sched/fair.c

···
 
 static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq);
 
-static inline void finish_delayed_dequeue_entity(struct sched_entity *se)
+static void set_delayed(struct sched_entity *se)
+{
+	se->sched_delayed = 1;
+	for_each_sched_entity(se) {
+		struct cfs_rq *cfs_rq = cfs_rq_of(se);
+
+		cfs_rq->h_nr_delayed++;
+		if (cfs_rq_throttled(cfs_rq))
+			break;
+	}
+}
+
+static void clear_delayed(struct sched_entity *se)
 {
 	se->sched_delayed = 0;
+	for_each_sched_entity(se) {
+		struct cfs_rq *cfs_rq = cfs_rq_of(se);
+
+		cfs_rq->h_nr_delayed--;
+		if (cfs_rq_throttled(cfs_rq))
+			break;
+	}
+}
+
+static inline void finish_delayed_dequeue_entity(struct sched_entity *se)
+{
+	clear_delayed(se);
 	if (sched_feat(DELAY_ZERO) && se->vlag > 0)
 		se->vlag = 0;
 }
···
 		if (sched_feat(DELAY_DEQUEUE) && delay &&
 		    !entity_eligible(cfs_rq, se)) {
 			update_load_avg(cfs_rq, se, 0);
-			se->sched_delayed = 1;
+			set_delayed(se);
 			return false;
 		}
 	}
···
 	struct rq *rq = rq_of(cfs_rq);
 	struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg);
 	struct sched_entity *se;
-	long task_delta, idle_task_delta, dequeue = 1;
+	long task_delta, idle_task_delta, delayed_delta, dequeue = 1;
 	long rq_h_nr_running = rq->cfs.h_nr_running;
 
 	raw_spin_lock(&cfs_b->lock);
···
 
 	task_delta = cfs_rq->h_nr_running;
 	idle_task_delta = cfs_rq->idle_h_nr_running;
+	delayed_delta = cfs_rq->h_nr_delayed;
 	for_each_sched_entity(se) {
 		struct cfs_rq *qcfs_rq = cfs_rq_of(se);
 		int flags;
···
 
 		qcfs_rq->h_nr_running -= task_delta;
 		qcfs_rq->idle_h_nr_running -= idle_task_delta;
+		qcfs_rq->h_nr_delayed -= delayed_delta;
 
 		if (qcfs_rq->load.weight) {
 			/* Avoid re-evaluating load for this entity: */
···
 
 		qcfs_rq->h_nr_running -= task_delta;
 		qcfs_rq->idle_h_nr_running -= idle_task_delta;
+		qcfs_rq->h_nr_delayed -= delayed_delta;
 	}
 
 	/* At this point se is NULL and we are at root level*/
···
 	struct rq *rq = rq_of(cfs_rq);
 	struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg);
 	struct sched_entity *se;
-	long task_delta, idle_task_delta;
+	long task_delta, idle_task_delta, delayed_delta;
 	long rq_h_nr_running = rq->cfs.h_nr_running;
 
 	se = cfs_rq->tg->se[cpu_of(rq)];
···
 
 	task_delta = cfs_rq->h_nr_running;
 	idle_task_delta = cfs_rq->idle_h_nr_running;
+	delayed_delta = cfs_rq->h_nr_delayed;
 	for_each_sched_entity(se) {
 		struct cfs_rq *qcfs_rq = cfs_rq_of(se);
 
···
 
 		qcfs_rq->h_nr_running += task_delta;
 		qcfs_rq->idle_h_nr_running += idle_task_delta;
+		qcfs_rq->h_nr_delayed += delayed_delta;
 
 		/* end evaluation on encountering a throttled cfs_rq */
 		if (cfs_rq_throttled(qcfs_rq))
···
 
 		qcfs_rq->h_nr_running += task_delta;
 		qcfs_rq->idle_h_nr_running += idle_task_delta;
+		qcfs_rq->h_nr_delayed += delayed_delta;
 
 		/* end evaluation on encountering a throttled cfs_rq */
 		if (cfs_rq_throttled(qcfs_rq))
···
 	}
 
 	update_load_avg(cfs_rq, se, 0);
-	se->sched_delayed = 0;
+	clear_delayed(se);
 }
 
 /*
···
 	struct cfs_rq *cfs_rq;
 	struct sched_entity *se = &p->se;
 	int idle_h_nr_running = task_has_idle_policy(p);
+	int h_nr_delayed = 0;
 	int task_new = !(flags & ENQUEUE_WAKEUP);
 	int rq_h_nr_running = rq->cfs.h_nr_running;
 	u64 slice = 0;
···
 	if (p->in_iowait)
 		cpufreq_update_util(rq, SCHED_CPUFREQ_IOWAIT);
 
+	if (task_new)
+		h_nr_delayed = !!se->sched_delayed;
+
 	for_each_sched_entity(se) {
 		if (se->on_rq) {
 			if (se->sched_delayed)
···
 
 		cfs_rq->h_nr_running++;
 		cfs_rq->idle_h_nr_running += idle_h_nr_running;
+		cfs_rq->h_nr_delayed += h_nr_delayed;
 
 		if (cfs_rq_is_idle(cfs_rq))
 			idle_h_nr_running = 1;
···
 
 		cfs_rq->h_nr_running++;
 		cfs_rq->idle_h_nr_running += idle_h_nr_running;
+		cfs_rq->h_nr_delayed += h_nr_delayed;
 
 		if (cfs_rq_is_idle(cfs_rq))
 			idle_h_nr_running = 1;
···
 	struct task_struct *p = NULL;
 	int idle_h_nr_running = 0;
 	int h_nr_running = 0;
+	int h_nr_delayed = 0;
 	struct cfs_rq *cfs_rq;
 	u64 slice = 0;
 
···
 		p = task_of(se);
 		h_nr_running = 1;
 		idle_h_nr_running = task_has_idle_policy(p);
+		if (!task_sleep && !task_delayed)
+			h_nr_delayed = !!se->sched_delayed;
 	} else {
 		cfs_rq = group_cfs_rq(se);
 		slice = cfs_rq_min_slice(cfs_rq);
···
 
 		cfs_rq->h_nr_running -= h_nr_running;
 		cfs_rq->idle_h_nr_running -= idle_h_nr_running;
+		cfs_rq->h_nr_delayed -= h_nr_delayed;
 
 		if (cfs_rq_is_idle(cfs_rq))
 			idle_h_nr_running = h_nr_running;
···
 
 		cfs_rq->h_nr_running -= h_nr_running;
 		cfs_rq->idle_h_nr_running -= idle_h_nr_running;
+		cfs_rq->h_nr_delayed -= h_nr_delayed;
 
 		if (cfs_rq_is_idle(cfs_rq))
 			idle_h_nr_running = h_nr_running;

kernel/sched/pelt.c

···
 {
 	if (___update_load_sum(now, &cfs_rq->avg,
 				scale_load_down(cfs_rq->load.weight),
-				cfs_rq->h_nr_running,
+				cfs_rq->h_nr_running - cfs_rq->h_nr_delayed,
 				cfs_rq->curr != NULL)) {
 
 		___update_load_avg(&cfs_rq->avg, 1);

kernel/sched/sched.h

···
 	unsigned int		h_nr_running;      /* SCHED_{NORMAL,BATCH,IDLE} */
 	unsigned int		idle_nr_running;   /* SCHED_IDLE */
 	unsigned int		idle_h_nr_running; /* SCHED_IDLE */
+	unsigned int		h_nr_delayed;
 
 	s64			avg_vruntime;
 	u64			avg_load;
···
 
 static inline void se_update_runnable(struct sched_entity *se)
 {
-	if (!entity_is_task(se))
-		se->runnable_weight = se->my_q->h_nr_running;
+	if (!entity_is_task(se)) {
+		struct cfs_rq *cfs_rq = se->my_q;
+
+		se->runnable_weight = cfs_rq->h_nr_running - cfs_rq->h_nr_delayed;
+	}
 }
 
 static inline long se_runnable(struct sched_entity *se)