diff mbox series

[v3,09/24] sched/fair: Use IPCC stats to break ties between fully_busy SMT groups

Message ID 20230207051105.11575-10-ricardo.neri-calderon@linux.intel.com
State New
Headers
Series sched: Introduce classes of tasks for load balance |

Commit Message

Ricardo Neri Feb. 7, 2023, 5:10 a.m. UTC 
  IPCC statistics are used during idle load balancing. After balancing one
of the siblings of an SMT core will become idle. The rest of the busy
siblings will enjoy increased throughput. The IPCC statistics provide
a measure of the increased throughput. Use them to pick a busiest group
from otherwise identical fully_busy scheduling groups (of which the
avg_load is equal - and zero).

Using IPCC scores to break ties with non-SMT fully_busy sched groups
is not necessary. SMT sched groups always need more help.

Add a stub sched_asym_ipcc_prefer() for !CONFIG_IPC_CLASSES.

Cc: Ben Segall <bsegall@google.com>
Cc: Daniel Bristot de Oliveira <bristot@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Ionela Voinescu <ionela.voinescu@arm.com>
Cc: Joel Fernandes (Google) <joel@joelfernandes.org>
Cc: Len Brown <len.brown@intel.com>
Cc: Lukasz Luba <lukasz.luba@arm.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tim C. Chen <tim.c.chen@intel.com>
Cc: Valentin Schneider <vschneid@redhat.com>
Cc: x86@kernel.org
Cc: linux-pm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
---
Changes since v2:
 * Introduced this patch.

Changes since v1:
 * N/A
---
 kernel/sched/fair.c | 23 ++++++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)
diff mbox series

Patch

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 841927b9b192..72d88270b320 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9415,6 +9415,12 @@  static void update_sg_lb_stats_scores(struct sg_lb_stats *sgs,
 {
 }
 
+static bool sched_asym_ipcc_prefer(struct sg_lb_stats *a,
+				   struct sg_lb_stats *b)
+{
+	return false;
+}
+
 static bool sched_asym_ipcc_pick(struct sched_group *a,
 				 struct sched_group *b,
 				 struct sg_lb_stats *a_stats,
@@ -9698,10 +9704,21 @@  static bool update_sd_pick_busiest(struct lb_env *env,
 		if (sgs->avg_load == busiest->avg_load) {
 			/*
 			 * SMT sched groups need more help than non-SMT groups.
-			 * If @sg happens to also be SMT, either choice is good.
 			 */
-			if (sds->busiest->flags & SD_SHARE_CPUCAPACITY)
-				return false;
+			if (sds->busiest->flags & SD_SHARE_CPUCAPACITY) {
+				if (!(sg->flags & SD_SHARE_CPUCAPACITY))
+					return false;
+
+				/*
+				 * Between two SMT groups, use IPCC scores to pick the
+				 * one that would improve throughput the most (only
+				 * asym_packing uses IPCC scores for now).
+				 */
+				if (sched_ipcc_enabled() &&
+				    env->sd->flags & SD_ASYM_PACKING &&
+				    sched_asym_ipcc_prefer(busiest, sgs))
+					return false;
+			}
 		}
 
 		break;