From: Ridong Chen <ridong.chen@linux.dev>
To: "Barry Song (Xiaomi)" <baohua@kernel.org>,
akpm@linux-foundation.org, linux-mm@kvack.org
Cc: axelrasmussen@google.com, baolin.wang@linux.alibaba.com,
baoquan.he@linux.dev, hannes@cmpxchg.org, hui.zhu@linux.dev,
kasong@tencent.com, kunwu.chan@gmail.com, liam@infradead.org,
lianux.mm@gmail.com, linux-kernel@vger.kernel.org,
liuxiang.277@bytedance.com, lyugaofei@xiaomi.com,
mhocko@kernel.org, mhocko@suse.com, qi.zheng@linux.dev,
rppt@kernel.org, ryncsn@gmail.com, shakeel.butt@linux.dev,
surenb@google.com, wangzicheng@honor.com, weixugc@google.com,
yuanchu@google.com, zhangbo56@xiaomi.com, zhuhui@kylinos.cn,
sidchintamaneni@gmail.com
Subject: Re: [PATCH v9 08/16] mm/mglru: lift youngest gen protection for higher priority
Date: Thu, 19 Nov 2026 18:57:40 +0800 [thread overview]
Message-ID: <c86a3b79-353d-48b8-ac7d-eb87920feb66@linux.dev> (raw)
In-Reply-To: <20261114200803.23129-9-baohua@kernel.org>
On 11/15/2026 4:07 AM, Barry Song (Xiaomi) wrote:
> From: Kairui Song <kasong@tencent.com>
>
> The forced protection of younger generations contradicts the
> goal of minimizing the I/O cost of reclaim. Under high memory
> pressure, MGLRU can repeatedly trigger premature OOMs, while
> the classic LRU does not.
> The problem is that aging may fail to keep up with reclaim.
> When this happens, reclaimers may all abort prematurely
> because they see only empty generations or just two
> generations remaining.
> Proactive aging does not solve this problem either. The
> number of threads performing proactive aging is limited,
> while the number of concurrent reclaimers can be much
> larger. This is a fundamental limitation of MGLRU. One
> possible solution is to use an aging lock. However, the
> aging lock introduces significant jitter under heavy memory
> pressure and has been reported to cause noticeable lags
> on Android.
> Furthermore, the generation-based reclaim model can
> conflict with the goal of swappiness, which is to minimize
> I/O cost rather than blindly follow folio hotness.
> The previous commit allowed reclaimers running at
> non-default priorities to reclaim up to `lrugen->max_seq`,
> while reclaimers at the default priority could reclaim only
> up to `lrugen->min_seq`.
> This patch further refines generation protection based on
> reclaim priority. At the default priority, reclaimers can
> reclaim generations down to `lrugen->max_seq - MIN_NR_GENS`.
> As reclaim priority increases, reclaimers are progressively
> allowed to reclaim younger generations.
> With this graduated protection, premature OOMs are no
> longer observed, while swappiness is also respected.
>
> Signed-off-by: Kairui Song <kasong@tencent.com>
> Signed-off-by: Barry Song (Xiaomi) <baohua@kernel.org>
> Reviewed-by: Kunwu Chan <kunwu.chan@gmail.com>
> ---
> mm/vmscan.c | 10 ++++++----
> 1 file changed, 6 insertions(+), 4 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 18cf1c084e78..b538abf02578 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -4772,12 +4772,14 @@ static int scan_folios(unsigned long nr_to_scan, struct lruvec *lruvec,
> VM_WARN_ON_ONCE(!list_empty(list));
>
> /*
> - * Respect the gen hotness distribution at low priority to avoid
> - * aging and scan cost, but prefer to satisfy the reclaimers request
> - * when under pressure to respect swappiness more and reduce IO cost.
> + * Respect hotness at default priority, and only reach into the
> + * newest generation when under severe pressure, so hot folios are
> + * not evicted as soon as priority is raised.
> */
> if (sc->priority == DEF_PRIORITY)
> - seq = lrugen->min_seq[type];
> + seq = lrugen->max_seq - MIN_NR_GENS;
> + else if (sc->priority > DEF_PRIORITY / 2)
> + seq = lrugen->max_seq - 1;
> else
> seq = lrugen->max_seq;
>
Reasonable.
Reviewed-by: Ridong Chen <ridong.chen@linux.dev>
Thanks.
--
Best regards
Ridong
next prev parent reply other threads:[~2026-11-19 10:57 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-14 20:07 [PATCH v9 00/16] mm: mglru: fix swappiness behavior Barry Song (Xiaomi)
2026-11-14 20:07 ` [PATCH v9 01/16] mm/mglru: report evictable sizes per type Barry Song (Xiaomi)
2026-11-16 11:29 ` Ridong Chen
2026-11-14 20:07 ` [PATCH v9 02/16] mm/mglru: extract lruvec_populated_min_seq() helper Barry Song (Xiaomi)
2026-11-16 11:58 ` Ridong Chen
2026-11-14 20:07 ` [PATCH v9 03/16] mm/mglru: reclaim from the oldest populated gen Barry Song (Xiaomi)
2026-11-16 12:27 ` Ridong Chen
2026-11-14 20:07 ` [PATCH v9 04/16] mm/mglru: lift the forced protection of reclaimer Barry Song (Xiaomi)
2026-11-18 10:50 ` Ridong Chen
2026-11-14 20:07 ` [PATCH v9 05/16] mm/mglru: skip zones above reclaim_idx instead of rotating them Barry Song (Xiaomi)
2026-11-18 11:22 ` Ridong Chen
2026-11-14 20:07 ` [PATCH v9 06/16] mm/mglru: return for unavailable anon reclaim earlier Barry Song (Xiaomi)
2026-11-18 11:23 ` Ridong Chen
2026-11-14 20:07 ` [PATCH v9 07/16] mm/mglru: balance anon and file scan budget by refault rate Barry Song (Xiaomi)
2026-11-19 9:48 ` Ridong Chen
2026-11-14 20:07 ` [PATCH v9 08/16] mm/mglru: lift youngest gen protection for higher priority Barry Song (Xiaomi)
2026-11-19 10:57 ` Ridong Chen [this message]
2026-11-14 20:07 ` [PATCH v9 09/16] mm/mglru: Avoid placing rejected folios below populated gen Barry Song (Xiaomi)
2026-11-15 15:15 ` Kairui Song
2026-11-14 20:07 ` [PATCH v9 10/16] mm/mglru: Run aging when pages are rare outside the youngest gen Barry Song (Xiaomi)
2026-11-15 15:54 ` Kairui Song
2026-11-16 4:25 ` Barry Song (Xiaomi)
2026-11-16 4:50 ` Andrew Morton
2026-11-16 5:29 ` Barry Song
2026-11-14 20:07 ` [PATCH v9 11/16] mm/mglru: Run aging when one type is drained Barry Song (Xiaomi)
2026-11-14 20:50 ` sashiko-bot
2026-11-14 21:44 ` Barry Song (Xiaomi)
2026-11-15 18:07 ` Kairui Song
2026-11-15 22:00 ` Barry Song (Xiaomi)
2026-11-14 20:07 ` [PATCH v9 12/16] mm/mglru: Restrict eviction accounting to eligible zones Barry Song (Xiaomi)
2026-11-14 20:08 ` [PATCH v9 13/16] mm/mglru: Reduce LRU lock contention by shrinking folio batches in inc_min_seq Barry Song (Xiaomi)
2026-11-15 15:52 ` Kairui Song
2026-11-14 20:08 ` [PATCH v9 14/16] mm/mglru: Avoid prolonged anon reclaim with swappiness=200 Barry Song (Xiaomi)
2026-11-14 20:08 ` [PATCH v9 15/16] mm/mglru: Make swappiness=1 file reclaim more aggressive Barry Song (Xiaomi)
2026-11-14 20:08 ` [PATCH v9 16/16] mm/mglru: Avoid scan anon while swapcache is low for GFP_NOIO allocation Barry Song (Xiaomi)
2026-11-14 22:09 ` [PATCH v9 00/16] mm: mglru: fix swappiness behavior Andrew Morton
2026-11-15 1:10 ` Barry Song
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=c86a3b79-353d-48b8-ac7d-eb87920feb66@linux.dev \
--to=ridong.chen@linux.dev \
--cc=akpm@linux-foundation.org \
--cc=axelrasmussen@google.com \
--cc=baohua@kernel.org \
--cc=baolin.wang@linux.alibaba.com \
--cc=baoquan.he@linux.dev \
--cc=hannes@cmpxchg.org \
--cc=hui.zhu@linux.dev \
--cc=kasong@tencent.com \
--cc=kunwu.chan@gmail.com \
--cc=liam@infradead.org \
--cc=lianux.mm@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=liuxiang.277@bytedance.com \
--cc=lyugaofei@xiaomi.com \
--cc=mhocko@kernel.org \
--cc=mhocko@suse.com \
--cc=qi.zheng@linux.dev \
--cc=rppt@kernel.org \
--cc=ryncsn@gmail.com \
--cc=shakeel.butt@linux.dev \
--cc=sidchintamaneni@gmail.com \
--cc=surenb@google.com \
--cc=wangzicheng@honor.com \
--cc=weixugc@google.com \
--cc=yuanchu@google.com \
--cc=zhangbo56@xiaomi.com \
--cc=zhuhui@kylinos.cn \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox