Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
From:   Pierre Gondois <pierre.gondois@arm.com>
To:     linux-kernel@vger.kernel.org
Cc:     Qais Yousef <qyousef@layalina.io>,
        Pierre Gondois <pierre.gondois@arm.com>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Juri Lelli <juri.lelli@redhat.com>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        Valentin Schneider <vschneid@redhat.com>
Subject: [PATCH v2] sched/fair: Use all little CPUs for CPU-bound workload
Date:   Fri, 24 Nov 2023 16:33:23 +0100
Message-Id: <20231124153323.3202444-1-pierre.gondois@arm.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Running n CPU-bound tasks on an n CPUs platform:
- with asymmetric CPU capacity
- not having SD_SHARE_PKG_RESOURCES flag set at the DIE
  sched domain level (i.e. not DynamIQ systems)
might result in a task placement where two tasks run on a big CPU
and none on a little CPU. This placement could be more optimal by
using all CPUs.

Testing platform:
Juno-r2:
- 2 big CPUs (1-2), maximum capacity of 1024
- 4 little CPUs (0,3-5), maximum capacity of 383

Testing workload ([1]):
Spawn 6 CPU-bound tasks. During the first 100ms (step 1), each tasks
is affine to a CPU, except for:
- one little CPU which is left idle.
- one big CPU which has 2 tasks affine.
After the 100ms (step 2), remove the cpumask affinity.

Before patch:
During step 2, the load balancer running from the idle CPU tags sched
domains as:
- little CPUs: 'group_has_spare'. Indeed, 3 CPU-bound tasks run on a
  4 CPUs sched-domain, and the idle CPU provides enough spare
  capacity.
- big CPUs: 'group_overloaded'. Indeed, 3 tasks run on a 2 CPUs
  sched-domain, so the following path is used:
  group_is_overloaded()
  \-if (sgs->sum_nr_running <= sgs->group_weight) return true;

  The following path which would change the migration type to
  'migrate_task' is not taken:
  calculate_imbalance()
  \-if (env->idle != CPU_NOT_IDLE && env->imbalance == 0)
  as the local group has some spare capacity, so the imbalance
  is not 0.

The migration type requested is 'migrate_util' and the busiest
runqueue is the big CPU's runqueue having 2 tasks (each having a
utilization of 512). The idle little CPU cannot pull one of these
task as its capacity is too small for the task. The following path
is used:
detach_tasks()
\-case migrate_util:
  \-if (util > env->imbalance) goto next;

After patch:
As the number of failed balancing attempts grows (with
'nr_balance_failed'), progressively make it easier to migrate
a big task to the idling little CPU. A similar mechanism is
used for the 'migrate_load' migration type.

Improvement:
Running the testing workload [1] with the step 2 representing
a ~10s load for a big CPU:
Before patch: ~19.3s
After patch: ~18s (-6.7%)

Similar issue reported at:
https://lore.kernel.org/lkml/20230716014125.139577-1-qyousef@layalina.io/

v1:
https://lore.kernel.org/all/20231110125902.2152380-1-pierre.gondois@arm.com/

Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
---
 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 df348aa55d3c..53c18fd23ae7 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -8907,7 +8907,7 @@ static int detach_tasks(struct lb_env *env)
 		case migrate_util:
 			util = task_util_est(p);
 
-			if (util > env->imbalance)
+			if (shr_bound(util, env->sd->nr_balance_failed) > env->imbalance)
 				goto next;
 
 			env->imbalance -= util;
-- 
2.25.1