* Re: [PATCH v2] mm: vmscan: don't count per-node proactive reclaim as memory pressure
2026-10-12 2:29 [PATCH v2] mm: vmscan: don't count per-node proactive reclaim as memory pressure Vernon Yang
@ 2026-10-12 3:05 ` Barry Song
2026-10-14 2:48 ` Ridong Chen
2026-10-14 3:35 ` Baolin Wang
2 siblings, 0 replies; 4+ messages in thread
From: Barry Song @ 2026-10-12 3:05 UTC (permalink / raw)
To: Vernon Yang
Cc: akpm, david, kasong, qi.zheng, shakeel.butt, axelrasmussen,
yuanchu, weixugc, baoquan.he, hannes, mhocko, ljs, roman.gushchin,
dave, linux-mm, linux-kernel, Vernon Yang, stable
On Mon, Oct 12, 2026 at 10:29 AM Vernon Yang <vernon2gm@gmail.com> wrote:
>
> From: Vernon Yang <yanglincheng@kylinos.cn>
>
> When the cgroup has no memory pressure at all, writing to
> /sys/devices/system/nodeX/reclaim triggers proactive reclaim on
> NUMA node, causing increase in the writer cgroup's memory PSI.
>
> Due to this reclaim is performed in the context of the write(),
> accounted as memory pressure on the writer, like
> commit e22c6ed90aa9 ("mm: memcontrol: don't count limit-setting reclaim
> as memory pressure"). This is unexpected, the similar scenarios (e.g.
> senpai) will continue to occur, only this time it is for the
> /sys/devices/system/nodeX/reclaim knob.
>
> The Documentation/ABI/stable/sysfs-devices-node documentation also
> notes that "This interface is equivalent to the memcg variant."
>
> This patch unifies the semantics of the memcg and node interfaces:
> per-node proactive reclaim is no longer counted as memory pressure,
> and the per-node proactive reclaim interface no longer produces
> phantom pressure.
>
> I ran demo[1] that performs per-node proactive reclaim 10000 times
> in qemu, writer cgroup memory.pressure as follows:
>
> without patch:
>
> some avg10=31.53 avg60=13.42 avg300=3.31 total=10602985
> full avg10=31.53 avg60=13.42 avg300=3.31 total=10602985
>
> with patch:
>
> some avg10=9.59 avg60=3.41 avg300=0.81 total=2686221
> full avg10=9.59 avg60=3.41 avg300=0.81 total=2686221
>
> [1] https://github.com/vernon2gh/app_and_module/tree/main/reclaim_node_psi
>
> Fixes: b980077899ea ("mm: introduce per-node proactive reclaim interface")
> Cc: stable@vger.kernel.org
> Signed-off-by: Vernon Yang <yanglincheng@kylinos.cn>
> ---
Reviewed-by: Barry Song <baohua@kernel.org>
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH v2] mm: vmscan: don't count per-node proactive reclaim as memory pressure
2026-10-12 2:29 [PATCH v2] mm: vmscan: don't count per-node proactive reclaim as memory pressure Vernon Yang
2026-10-12 3:05 ` Barry Song
@ 2026-10-14 2:48 ` Ridong Chen
2026-10-14 3:35 ` Baolin Wang
2 siblings, 0 replies; 4+ messages in thread
From: Ridong Chen @ 2026-10-14 2:48 UTC (permalink / raw)
To: Vernon Yang, akpm, david, kasong, qi.zheng, shakeel.butt, baohua,
axelrasmussen, yuanchu, weixugc, baoquan.he, hannes, mhocko, ljs,
roman.gushchin, dave
Cc: linux-mm, linux-kernel, Vernon Yang, stable
On 10/12/2026 10:29 AM, Vernon Yang wrote:
> From: Vernon Yang <yanglincheng@kylinos.cn>
>
> When the cgroup has no memory pressure at all, writing to
> /sys/devices/system/nodeX/reclaim triggers proactive reclaim on
> NUMA node, causing increase in the writer cgroup's memory PSI.
>
> Due to this reclaim is performed in the context of the write(),
> accounted as memory pressure on the writer, like
> commit e22c6ed90aa9 ("mm: memcontrol: don't count limit-setting reclaim
> as memory pressure"). This is unexpected, the similar scenarios (e.g.
> senpai) will continue to occur, only this time it is for the
> /sys/devices/system/nodeX/reclaim knob.
>
> The Documentation/ABI/stable/sysfs-devices-node documentation also
> notes that "This interface is equivalent to the memcg variant."
>
> This patch unifies the semantics of the memcg and node interfaces:
> per-node proactive reclaim is no longer counted as memory pressure,
> and the per-node proactive reclaim interface no longer produces
> phantom pressure.
>
> I ran demo[1] that performs per-node proactive reclaim 10000 times
> in qemu, writer cgroup memory.pressure as follows:
>
> without patch:
>
> some avg10=31.53 avg60=13.42 avg300=3.31 total=10602985
> full avg10=31.53 avg60=13.42 avg300=3.31 total=10602985
>
> with patch:
>
> some avg10=9.59 avg60=3.41 avg300=0.81 total=2686221
> full avg10=9.59 avg60=3.41 avg300=0.81 total=2686221
>
> [1] https://github.com/vernon2gh/app_and_module/tree/main/reclaim_node_psi
>
> Fixes: b980077899ea ("mm: introduce per-node proactive reclaim interface")
> Cc: stable@vger.kernel.org
> Signed-off-by: Vernon Yang <yanglincheng@kylinos.cn>
> ---
> V1 -> V2:
> - Check `sc->proactive` to differentiate between proactive reclaim or
> page allocation. (Thanks Barry)
>
> V1 : https://lore.kernel.org/linux-mm/20261005062236.564210-1-vernon2gm@gmail.com/
> ---
> mm/vmscan.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 91295070ca33..2e9ca5e4c741 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -7994,7 +7994,8 @@ static unsigned long __node_reclaim(struct pglist_data *pgdat,
> sc->gfp_mask);
>
> cond_resched();
> - psi_memstall_enter(&pflags);
> + if (!sc->proactive)
> + psi_memstall_enter(&pflags);
> delayacct_freepages_start();
> fs_reclaim_acquire(sc->gfp_mask);
> /*
> @@ -8011,7 +8012,8 @@ static unsigned long __node_reclaim(struct pglist_data *pgdat,
> memalloc_noreclaim_restore(noreclaim_flag);
> fs_reclaim_release(sc->gfp_mask);
> delayacct_freepages_end();
> - psi_memstall_leave(&pflags);
> + if (!sc->proactive)
> + psi_memstall_leave(&pflags);
>
> trace_mm_vmscan_node_reclaim_end(sc->nr_reclaimed, NULL);
>
LGTM.
Reviewed-by: Ridong Chen <ridong.chen@linux.dev>
Thanks.
--
Best regards
Ridong
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH v2] mm: vmscan: don't count per-node proactive reclaim as memory pressure
2026-10-12 2:29 [PATCH v2] mm: vmscan: don't count per-node proactive reclaim as memory pressure Vernon Yang
2026-10-12 3:05 ` Barry Song
2026-10-14 2:48 ` Ridong Chen
@ 2026-10-14 3:35 ` Baolin Wang
2 siblings, 0 replies; 4+ messages in thread
From: Baolin Wang @ 2026-10-14 3:35 UTC (permalink / raw)
To: Vernon Yang, akpm, david, kasong, qi.zheng, shakeel.butt, baohua,
axelrasmussen, yuanchu, weixugc, baoquan.he, hannes, mhocko, ljs,
roman.gushchin, dave
Cc: linux-mm, linux-kernel, Vernon Yang, stable
On 10/12/26 10:29 AM, Vernon Yang wrote:
> From: Vernon Yang <yanglincheng@kylinos.cn>
>
> When the cgroup has no memory pressure at all, writing to
> /sys/devices/system/nodeX/reclaim triggers proactive reclaim on
> NUMA node, causing increase in the writer cgroup's memory PSI.
>
> Due to this reclaim is performed in the context of the write(),
> accounted as memory pressure on the writer, like
> commit e22c6ed90aa9 ("mm: memcontrol: don't count limit-setting reclaim
> as memory pressure"). This is unexpected, the similar scenarios (e.g.
> senpai) will continue to occur, only this time it is for the
> /sys/devices/system/nodeX/reclaim knob.
>
> The Documentation/ABI/stable/sysfs-devices-node documentation also
> notes that "This interface is equivalent to the memcg variant."
>
> This patch unifies the semantics of the memcg and node interfaces:
> per-node proactive reclaim is no longer counted as memory pressure,
> and the per-node proactive reclaim interface no longer produces
> phantom pressure.
>
> I ran demo[1] that performs per-node proactive reclaim 10000 times
> in qemu, writer cgroup memory.pressure as follows:
>
> without patch:
>
> some avg10=31.53 avg60=13.42 avg300=3.31 total=10602985
> full avg10=31.53 avg60=13.42 avg300=3.31 total=10602985
>
> with patch:
>
> some avg10=9.59 avg60=3.41 avg300=0.81 total=2686221
> full avg10=9.59 avg60=3.41 avg300=0.81 total=2686221
>
> [1] https://github.com/vernon2gh/app_and_module/tree/main/reclaim_node_psi
>
> Fixes: b980077899ea ("mm: introduce per-node proactive reclaim interface")
> Cc: stable@vger.kernel.org
> Signed-off-by: Vernon Yang <yanglincheng@kylinos.cn>
> ---
LGTM.
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
^ permalink raw reply [flat|nested] 4+ messages in thread