The quilt patch titled
Subject: mm/memcontrol.c: remove the redundant updating of stats_flush_threshold
has been removed from the -mm tree. Its filename was
mm-remove-the-redundant-updating-of-stats_flush_threshold.patch
This patch was dropped because it was merged into the mm-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
------------------------------------------------------
From: Jiebin Sun <jiebin.sun@xxxxxxxxx>
Subject: mm/memcontrol.c: remove the redundant updating of stats_flush_threshold
Date: Sat, 23 Jul 2022 00:49:49 +0800
Remove the redundant updating of stats_flush_threshold. If the global var
stats_flush_threshold has exceeded the trigger value for
__mem_cgroup_flush_stats, further increment is unnecessary.
Apply the patch and test the pts/hackbench-1.0.0 Count:4 (160 threads).
Score gain: 1.95x
Reduce CPU cycles in __mod_memcg_lruvec_state (44.88% -> 0.12%)
CPU: ICX 8380 x 2 sockets
Core number: 40 x 2 physical cores
Benchmark: pts/hackbench-1.0.0 Count:4 (160 threads)
Link: https://lkml.kernel.org/r/20220722164949.47760-1-jiebin.sun@xxxxxxxxx
Signed-off-by: Jiebin Sun <jiebin.sun@xxxxxxxxx>
Acked-by: Shakeel Butt <shakeelb@xxxxxxxxxx>
Reviewed-by: Roman Gushchin <roman.gushchin@xxxxxxxxx>
Reviewed-by: Tim Chen <tim.c.chen@xxxxxxxxxxxxxxx>
Acked-by: Muchun Song <songmuchun@xxxxxxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxxxx>
Cc: "Huang, Ying" <ying.huang@xxxxxxxxx>
Cc: Amadeusz Sawiski <amadeuszx.slawinski@xxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/memcontrol.c | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)
--- a/mm/memcontrol.c~mm-remove-the-redundant-updating-of-stats_flush_threshold
+++ a/mm/memcontrol.c
@@ -626,7 +626,14 @@ static inline void memcg_rstat_updated(s
x = __this_cpu_add_return(stats_updates, abs(val));
if (x > MEMCG_CHARGE_BATCH) {
- atomic_add(x / MEMCG_CHARGE_BATCH, &stats_flush_threshold);
+ /*
+ * If stats_flush_threshold exceeds the threshold
+ * (>num_online_cpus()), cgroup stats update will be triggered
+ * in __mem_cgroup_flush_stats(). Increasing this var further
+ * is redundant and simply adds overhead in atomic update.
+ */
+ if (atomic_read(&stats_flush_threshold) <= num_online_cpus())
+ atomic_add(x / MEMCG_CHARGE_BATCH, &stats_flush_threshold);
__this_cpu_write(stats_updates, 0);
}
}
_
Patches currently in -mm which might be from jiebin.sun@xxxxxxxxx are
