sched/fair: Skip LLC NOHZ logic for asymmetric systems

The LLC NOHZ condition will become true as soon as >=2 CPUs in a
single LLC domain are busy. On big.LITTLE systems, this translates to
two or more CPUs of a "cluster" (big or LITTLE) being busy.

Issuing a NOHZ kick in these conditions isn't desired for asymmetric
systems, as if the busy CPUs can provide enough compute capacity to
the running tasks, then we can leave the NOHZ CPUs in peace.

Skip the LLC NOHZ condition for asymmetric systems, and rely on
nr_running & capacity checks to trigger NOHZ kicks when the system
actually needs them.

Suggested-by: Morten Rasmussen <[email protected]>
Signed-off-by: Valentin Schneider <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: [email protected]
Cc: H. Peter Anvin <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index f0d2f8a352bf19360d7e5860e2124972e1d73cd7..51003e1c794d8d44e5013dc860280f4661b95ec2 100644 (file)
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9603,24 +9603,6 @@ static void nohz_balancer_kick(struct rq *rq)
        }
 
        rcu_read_lock();
-       sds = rcu_dereference(per_cpu(sd_llc_shared, cpu));
-       if (sds) {
-               /*
-                * If there is an imbalance between LLC domains (IOW we could
-                * increase the overall cache use), we need some less-loaded LLC
-                * domain to pull some load. Likewise, we may need to spread
-                * load within the current LLC domain (e.g. packed SMT cores but
-                * other CPUs are idle). We can't really know from here how busy
-                * the others are - so just get a nohz balance going if it looks
-                * like this LLC domain has tasks we could move.
-                */
-               nr_busy = atomic_read(&sds->nr_busy_cpus);
-               if (nr_busy > 1) {
-                       flags = NOHZ_KICK_MASK;
-                       goto unlock;
-               }
-
-       }
 
        sd = rcu_dereference(rq->sd);
        if (sd) {
@@ -9635,6 +9617,21 @@ static void nohz_balancer_kick(struct rq *rq)
                }
        }
 
+       sd = rcu_dereference(per_cpu(sd_asym_packing, cpu));
+       if (sd) {
+               /*
+                * When ASYM_PACKING; see if there's a more preferred CPU
+                * currently idle; in which case, kick the ILB to move tasks
+                * around.
+                */
+               for_each_cpu_and(i, sched_domain_span(sd), nohz.idle_cpus_mask) {
+                       if (sched_asym_prefer(i, cpu)) {
+                               flags = NOHZ_KICK_MASK;
+                               goto unlock;
+                       }
+               }
+       }
+
        sd = rcu_dereference(per_cpu(sd_asym_cpucapacity, cpu));
        if (sd) {
                /*
@@ -9645,20 +9642,32 @@ static void nohz_balancer_kick(struct rq *rq)
                        flags = NOHZ_KICK_MASK;
                        goto unlock;
                }
+
+               /*
+                * For asymmetric systems, we do not want to nicely balance
+                * cache use, instead we want to embrace asymmetry and only
+                * ensure tasks have enough CPU capacity.
+                *
+                * Skip the LLC logic because it's not relevant in that case.
+                */
+               goto unlock;
        }
 
-       sd = rcu_dereference(per_cpu(sd_asym_packing, cpu));
-       if (sd) {
+       sds = rcu_dereference(per_cpu(sd_llc_shared, cpu));
+       if (sds) {
                /*
-                * When ASYM_PACKING; see if there's a more preferred CPU
-                * currently idle; in which case, kick the ILB to move tasks
-                * around.
+                * If there is an imbalance between LLC domains (IOW we could
+                * increase the overall cache use), we need some less-loaded LLC
+                * domain to pull some load. Likewise, we may need to spread
+                * load within the current LLC domain (e.g. packed SMT cores but
+                * other CPUs are idle). We can't really know from here how busy
+                * the others are - so just get a nohz balance going if it looks
+                * like this LLC domain has tasks we could move.
                 */
-               for_each_cpu_and(i, sched_domain_span(sd), nohz.idle_cpus_mask) {
-                       if (sched_asym_prefer(i, cpu)) {
-                               flags = NOHZ_KICK_MASK;
-                               goto unlock;
-                       }
+               nr_busy = atomic_read(&sds->nr_busy_cpus);
+               if (nr_busy > 1) {
+                       flags = NOHZ_KICK_MASK;
+                       goto unlock;
                }
        }
 unlock: