From: Baoquan He <baoquan.he@linux.dev>
To: "Barry Song (Xiaomi)" <baohua@kernel.org>
Cc: akpm@linux-foundation.org, linux-mm@kvack.org,
axelrasmussen@google.com, baolin.wang@linux.alibaba.com,
chenridong@xiaomi.com, david@kernel.org, hannes@cmpxchg.org,
kasong@tencent.com, lianux.mm@gmail.com,
linux-kernel@vger.kernel.org, ljs@kernel.org,
lyugaofei@xiaomi.com, mhocko@kernel.org, qi.zheng@linux.dev,
shakeel.butt@linux.dev, stevensd@chromium.org,
wangzicheng@honor.com, weixugc@google.com, yuanchu@google.com,
zhangbo56@xiaomi.com
Subject: Re: [PATCH 2/6] mm/mglru: batch update lrugen->protected in inc_min_seq()
Date: Wed, 26 Aug 2026 17:10:47 +0800 [thread overview]
Message-ID: <ao6tlwWNN1MH29w7@fedora> (raw)
In-Reply-To: <20260821102538.22642-3-baohua@kernel.org>
On 08/21/26 at 06:25pm, Barry Song (Xiaomi) wrote:
> Avoid updating lrugen->protected with WRITE_ONCE() for each folio,
> which may prevent potential compiler optimizations. Accumulate the
> updates locally and apply them in a batch instead.
Wondering how much efficiency this can bring, is there a number for this
standalone patch?
>
> Signed-off-by: Barry Song (Xiaomi) <baohua@kernel.org>
> ---
> mm/vmscan.c | 8 +++++---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 0d74fc00abd3..99ee3c833d54 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -3931,7 +3931,7 @@ static bool inc_min_seq(struct lruvec *lruvec, int type, int swappiness)
> /* prevent cold/hot inversion if the type is evictable */
> for (zone = 0; zone < MAX_NR_ZONES; zone++) {
> struct list_head *head = &lrugen->folios[old_gen][type][zone];
> - unsigned long delta = 0;
> + unsigned long protected[MAX_NR_TIERS] = {}, delta = 0;
>
> while (!list_empty(head)) {
> struct folio *folio = lru_to_folio(head);
> @@ -3953,8 +3953,7 @@ static bool inc_min_seq(struct lruvec *lruvec, int type, int swappiness)
> if (refs + workingset != BIT(LRU_REFS_WIDTH) + 1) {
> int tier = lru_tier_from_refs(refs, workingset);
>
> - WRITE_ONCE(lrugen->protected[hist][type][tier],
> - lrugen->protected[hist][type][tier] + nr_pages);
> + protected[tier] += nr_pages;
> }
>
> if (!--remaining)
> @@ -3964,6 +3963,9 @@ static bool inc_min_seq(struct lruvec *lruvec, int type, int swappiness)
> lrugen->nr_pages[old_gen][type][zone] - delta);
> WRITE_ONCE(lrugen->nr_pages[target_gen][type][zone],
> lrugen->nr_pages[target_gen][type][zone] + delta);
> + for (int tier = 0; tier < MAX_NR_TIERS; tier++)
> + WRITE_ONCE(lrugen->protected[hist][type][tier],
> + lrugen->protected[hist][type][tier] + protected[tier]);
> if (!remaining)
> return false;
> }
> --
> 2.34.1
>
next prev parent reply other threads:[~2026-08-26 9:11 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-21 10:25 [PATCH 0/6] mm/mglru: speed up inc_min_seq() and fix cold/hot inversions Barry Song (Xiaomi)
2026-08-21 10:25 ` [PATCH 1/6] mm/mglru: batch update lrugen->nr_pages in inc_min_seq() Barry Song (Xiaomi)
2026-08-22 1:42 ` Lian Wang (ProcessMission)
2026-08-25 21:38 ` Barry Song
2026-08-26 8:23 ` Baoquan He
2026-08-27 3:20 ` Kairui Song
2026-08-27 11:21 ` Barry Song
2026-08-27 11:30 ` Kairui Song
2026-08-21 10:25 ` [PATCH 2/6] mm/mglru: batch update lrugen->protected " Barry Song (Xiaomi)
2026-08-26 9:10 ` Baoquan He [this message]
2026-08-27 5:09 ` Barry Song
2026-08-27 12:14 ` Xueyuan Chen
2026-08-21 10:25 ` [PATCH 3/6] mm/mglru: enhance cold/hot inversion handling " Barry Song (Xiaomi)
2026-08-26 8:56 ` Baoquan He
2026-08-26 21:43 ` Barry Song
2026-08-27 0:46 ` Baoquan He
2026-08-27 1:24 ` Barry Song
2026-08-27 2:14 ` Baoquan He
2026-08-27 2:19 ` Baoquan He
2026-08-27 4:30 ` Kairui Song
2026-08-27 6:13 ` Baoquan He
2026-08-27 4:37 ` Kairui Song
2026-08-21 10:25 ` [PATCH 4/6] mm/mglru: exclude folios promoted by aging from protected " Barry Song (Xiaomi)
2026-08-26 8:57 ` Baoquan He
2026-08-21 10:25 ` [PATCH 5/6] mm/mglru: move folios from oldest gen to second-oldest gen from head to tail Barry Song (Xiaomi)
2026-08-22 5:45 ` Kairui Song
2026-08-25 21:32 ` Barry Song
2026-08-26 9:06 ` Baoquan He
2026-08-21 10:25 ` [PATCH 6/6] mm/mglru: batch move folios to the second-oldest gen's LRU Barry Song (Xiaomi)
2026-08-26 9:34 ` Baoquan He
2026-08-27 3:54 ` [PATCH 0/6] mm/mglru: speed up inc_min_seq() and fix cold/hot inversions Xueyuan Chen
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=ao6tlwWNN1MH29w7@fedora \
--to=baoquan.he@linux.dev \
--cc=akpm@linux-foundation.org \
--cc=axelrasmussen@google.com \
--cc=baohua@kernel.org \
--cc=baolin.wang@linux.alibaba.com \
--cc=chenridong@xiaomi.com \
--cc=david@kernel.org \
--cc=hannes@cmpxchg.org \
--cc=kasong@tencent.com \
--cc=lianux.mm@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=ljs@kernel.org \
--cc=lyugaofei@xiaomi.com \
--cc=mhocko@kernel.org \
--cc=qi.zheng@linux.dev \
--cc=shakeel.butt@linux.dev \
--cc=stevensd@chromium.org \
--cc=wangzicheng@honor.com \
--cc=weixugc@google.com \
--cc=yuanchu@google.com \
--cc=zhangbo56@xiaomi.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.