Thread (66 messages) 66 messages, 8 authors, 2014-07-18

[PATCH v3 08/12] sched: move cfs task on a CPU with higher capacity

From: peterz@infradead.org (Peter Zijlstra)
Date: 2014-07-10 11:31:24
Also in: lkml

On Mon, Jun 30, 2014 at 06:05:39PM +0200, Vincent Guittot wrote:

You 'forgot' to update the comment that goes with nohz_kick_needed().
quoted hunk ↗ jump to hunk
@@ -7233,9 +7253,10 @@ static inline int nohz_kick_needed(struct rq *rq)
 	struct sched_domain *sd;
 	struct sched_group_capacity *sgc;
 	int nr_busy, cpu = rq->cpu;
+	bool kick = false;
 
 	if (unlikely(rq->idle_balance))
+		return false;
 
        /*
 	* We may be recently in ticked or tickless idle mode. At the first
@@ -7249,38 +7270,41 @@ static inline int nohz_kick_needed(struct rq *rq)
 	 * balancing.
 	 */
 	if (likely(!atomic_read(&nohz.nr_cpus)))
+		return false;
 
 	if (time_before(now, nohz.next_balance))
+		return false;
 
 	if (rq->nr_running >= 2)
+		return true;
 
 	rcu_read_lock();
 	sd = rcu_dereference(per_cpu(sd_busy, cpu));
 	if (sd) {
 		sgc = sd->groups->sgc;
 		nr_busy = atomic_read(&sgc->nr_busy_cpus);
 
+		if (nr_busy > 1) {
+			kick = true;
+			goto unlock;
+		}
+
+		if ((rq->cfs.h_nr_running >= 1)
+		 && ((rq->cpu_capacity * sd->imbalance_pct) <
+					(rq->cpu_capacity_orig * 100))) {
+			kick = true;
+			goto unlock;
+		}
Again, why only for shared caches?
 	}
 
 	sd = rcu_dereference(per_cpu(sd_asym, cpu));
 	if (sd && (cpumask_first_and(nohz.idle_cpus_mask,
 				  sched_domain_span(sd)) < cpu))
+		kick = true;
 
+unlock:
 	rcu_read_unlock();
+	return kick;
 }
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help