* [PATCH] mm: bail out from partial cgroup_reclaim inside shrink_lruvec
@ 2026-02-05 6:45 zhaoyang.huang
2026-02-09 7:36 ` Zhaoyang Huang
0 siblings, 1 reply; 2+ messages in thread
From: zhaoyang.huang @ 2026-02-05 6:45 UTC (permalink / raw)
To: Andrew Morton, Yu Zhao, Michal Hocko, Rik van Riel, Shakeel Butt,
Roman Gushchin, Johannes Weiner, linux-mm, linux-kernel,
Zhaoyang Huang, steve.kang
From: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
Oversize reclaiming observed in "echo xxx > memory.reclaim". This commit
would like to sync the stop creteria in shrink_node_memcgs to
shrink_lruvec to have partial cgroup_reclaim bail out as early as it can
be.
Signed-off-by: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
---
mm/vmscan.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 670fe9fae5ba..03bda1094621 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -4832,8 +4832,8 @@ static bool should_abort_scan(struct lruvec *lruvec, struct scan_control *sc)
int i;
enum zone_watermarks mark;
- /* don't abort memcg reclaim to ensure fairness */
- if (!root_reclaim(sc))
+ /* don't abort full walk memcg reclaim to ensure fairness */
+ if (!root_reclaim(sc) && sc->memcg_full_walk)
return false;
if (sc->nr_reclaimed >= max(sc->nr_to_reclaim, compact_gap(sc->order)))
--
2.25.1
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH] mm: bail out from partial cgroup_reclaim inside shrink_lruvec
2026-02-05 6:45 [PATCH] mm: bail out from partial cgroup_reclaim inside shrink_lruvec zhaoyang.huang
@ 2026-02-09 7:36 ` Zhaoyang Huang
0 siblings, 0 replies; 2+ messages in thread
From: Zhaoyang Huang @ 2026-02-09 7:36 UTC (permalink / raw)
To: zhaoyang.huang
Cc: Andrew Morton, Yu Zhao, Michal Hocko, Rik van Riel, Shakeel Butt,
Roman Gushchin, Johannes Weiner, linux-mm, linux-kernel,
steve.kang
On Thu, Feb 5, 2026 at 2:47 PM zhaoyang.huang <zhaoyang.huang@unisoc.com> wrote:
>
> From: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
>
> Oversize reclaiming observed in "echo xxx > memory.reclaim". This commit
> would like to sync the stop criteria in shrink_node_memcgs to
> shrink_lruvec to have partial cgroup_reclaim bail out as early as it can
> be.
>
> Signed-off-by: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
> ---
> mm/vmscan.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 670fe9fae5ba..03bda1094621 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -4832,8 +4832,8 @@ static bool should_abort_scan(struct lruvec *lruvec, struct scan_control *sc)
> int i;
> enum zone_watermarks mark;
>
> - /* don't abort memcg reclaim to ensure fairness */
> - if (!root_reclaim(sc))
> + /* don't abort full walk memcg reclaim to ensure fairness */
> + if (!root_reclaim(sc) && sc->memcg_full_walk)
> return false;
>
> if (sc->nr_reclaimed >= max(sc->nr_to_reclaim, compact_gap(sc->order)))
> --
Talk more about motivation. Nowadays, android systems replace madivse
with memory.reclaim to implement user space memory management which
desires to reclaim a certain amount of memcg's memory. However,
oversized reclaiming is observed as there is no limitation within
try_to_shrink_lruvec when MGLRU enabled. This commit would like to fix
this by judging sc->memcg_full_walk.
> 2.25.1
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-02-09 7:36 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-02-05 6:45 [PATCH] mm: bail out from partial cgroup_reclaim inside shrink_lruvec zhaoyang.huang
2026-02-09 7:36 ` Zhaoyang Huang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox