From: Shakeel Butt <shakeelb@google.com>
To: Johannes Weiner <hannes@cmpxchg.org>
Cc: Muchun Song <songmuchun@bytedance.com>,
Roman Gushchin <guro@fb.com>, Michal Hocko <mhocko@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Vladimir Davydov <vdavydov.dev@gmail.com>,
LKML <linux-kernel@vger.kernel.org>,
Linux Memory Management List <linux-mm@kvack.org>,
Xiongchun duan <duanxiongchun@bytedance.com>
Subject: Re: [External] Re: [PATCH v3 2/4] mm: memcontrol: make page_memcg{_rcu} only applicable for non-kmem page
Date: Fri, 12 Mar 2021 14:42:45 -0800 [thread overview]
Message-ID: <CALvZod5sEcsu9AHAzdkJyjxm4ws_aKXyyer99+f2rs-OnOQAMg@mail.gmail.com> (raw)
In-Reply-To: <YEu/pnK0n8nI2OBy@cmpxchg.org>
Hi Johannes,
On Fri, Mar 12, 2021 at 11:23 AM Johannes Weiner <hannes@cmpxchg.org> wrote:
>
[...]
>
> Longer term we most likely need it there anyway. The issue you are
> describing in the cover letter - allocations pinning memcgs for a long
> time - it exists at a larger scale and is causing recurring problems
> in the real world: page cache doesn't get reclaimed for a long time,
> or is used by the second, third, fourth, ... instance of the same job
> that was restarted into a new cgroup every time. Unreclaimable dying
> cgroups pile up, waste memory, and make page reclaim very inefficient.
>
For the scenario described above, do we really want to reparent the
page cache pages? Shouldn't we recharge the pages to the second,
third, fourth and so on, memcgs? My concern is that we will see a big
chunk of page cache pages charged to root and will only get reclaimed
on global pressure.
next prev parent reply other threads:[~2021-03-12 22:43 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-09 10:07 [PATCH v3 0/4] Use obj_cgroup APIs to charge kmem pages Muchun Song
2021-03-09 10:07 ` [PATCH v3 1/4] mm: memcontrol: introduce obj_cgroup_{un}charge_pages Muchun Song
2021-03-11 12:30 ` Johannes Weiner
2021-03-11 18:56 ` Shakeel Butt
2021-03-09 10:07 ` [PATCH v3 2/4] mm: memcontrol: make page_memcg{_rcu} only applicable for non-kmem page Muchun Song
2021-03-10 19:57 ` Roman Gushchin
2021-03-11 6:45 ` [External] " Muchun Song
2021-03-11 13:12 ` Johannes Weiner
2021-03-12 7:14 ` [External] " Muchun Song
2021-03-12 19:23 ` Johannes Weiner
2021-03-12 22:42 ` Shakeel Butt [this message]
2021-03-12 23:07 ` Johannes Weiner
2021-03-12 23:18 ` Shakeel Butt
2021-03-14 13:56 ` Muchun Song
2021-03-12 3:22 ` Shakeel Butt
2021-03-12 5:02 ` [External] " Muchun Song
2021-03-09 10:07 ` [PATCH v3 3/4] mm: memcontrol: use obj_cgroup APIs to charge kmem pages Muchun Song
2021-03-10 19:53 ` Roman Gushchin
2021-03-11 6:50 ` [External] " Muchun Song
2021-03-10 22:05 ` Johannes Weiner
2021-03-12 9:22 ` [External] " Muchun Song
2021-03-12 15:59 ` Johannes Weiner
2021-03-12 16:46 ` Muchun Song
2021-03-11 17:48 ` kernel test robot
2021-03-09 10:07 ` [PATCH v3 4/4] mm: memcontrol: move PageMemcgKmem to the scope of CONFIG_MEMCG_KMEM Muchun Song
2021-03-10 19:30 ` Roman Gushchin
2021-03-12 3:26 ` Shakeel Butt
2021-03-12 19:24 ` Johannes Weiner
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=CALvZod5sEcsu9AHAzdkJyjxm4ws_aKXyyer99+f2rs-OnOQAMg@mail.gmail.com \
--to=shakeelb@google.com \
--cc=akpm@linux-foundation.org \
--cc=duanxiongchun@bytedance.com \
--cc=guro@fb.com \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=songmuchun@bytedance.com \
--cc=vdavydov.dev@gmail.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox