From: Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
To: Muchun Song <songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
Cc: mhocko-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
roman.gushchin-fxUVXftIFDnyG1zEObXtfA@public.gmane.org,
shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
duanxiongchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org,
longman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org
Subject: Re: [PATCH v4 01/11] mm: memcontrol: prepare objcg API for non-kmem usage
Date: Tue, 24 May 2022 15:01:25 -0400 [thread overview]
Message-ID: <Yo0rhX2LFNPTv47b@cmpxchg.org> (raw)
In-Reply-To: <20220524060551.80037-2-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
On Tue, May 24, 2022 at 02:05:41PM +0800, Muchun Song wrote:
> Pagecache pages are charged at the allocation time and holding a
> reference to the original memory cgroup until being reclaimed.
> Depending on the memory pressure, specific patterns of the page
> sharing between different cgroups and the cgroup creation and
> destruction rates, a large number of dying memory cgroups can be
> pinned by pagecache pages. It makes the page reclaim less efficient
> and wastes memory.
>
> We can convert LRU pages and most other raw memcg pins to the objcg
> direction to fix this problem, and then the page->memcg will always
> point to an object cgroup pointer.
>
> Therefore, the infrastructure of objcg no longer only serves
> CONFIG_MEMCG_KMEM. In this patch, we move the infrastructure of the
> objcg out of the scope of the CONFIG_MEMCG_KMEM so that the LRU pages
> can reuse it to charge pages.
>
> We know that the LRU pages are not accounted at the root level. But
> the page->memcg_data points to the root_mem_cgroup. So the
> page->memcg_data of the LRU pages always points to a valid pointer.
> But the root_mem_cgroup dose not have an object cgroup. If we use
> obj_cgroup APIs to charge the LRU pages, we should set the
> page->memcg_data to a root object cgroup. So we also allocate an
> object cgroup for the root_mem_cgroup.
>
> Signed-off-by: Muchun Song <songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
Acked-by: Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
Looks good to me. Also gets rid of some use_hierarchy cruft.
WARNING: multiple messages have this Message-ID (diff)
From: Johannes Weiner <hannes@cmpxchg.org>
To: Muchun Song <songmuchun@bytedance.com>
Cc: mhocko@kernel.org, roman.gushchin@linux.dev, shakeelb@google.com,
cgroups@vger.kernel.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, duanxiongchun@bytedance.com,
longman@redhat.com
Subject: Re: [PATCH v4 01/11] mm: memcontrol: prepare objcg API for non-kmem usage
Date: Tue, 24 May 2022 15:01:25 -0400 [thread overview]
Message-ID: <Yo0rhX2LFNPTv47b@cmpxchg.org> (raw)
In-Reply-To: <20220524060551.80037-2-songmuchun@bytedance.com>
On Tue, May 24, 2022 at 02:05:41PM +0800, Muchun Song wrote:
> Pagecache pages are charged at the allocation time and holding a
> reference to the original memory cgroup until being reclaimed.
> Depending on the memory pressure, specific patterns of the page
> sharing between different cgroups and the cgroup creation and
> destruction rates, a large number of dying memory cgroups can be
> pinned by pagecache pages. It makes the page reclaim less efficient
> and wastes memory.
>
> We can convert LRU pages and most other raw memcg pins to the objcg
> direction to fix this problem, and then the page->memcg will always
> point to an object cgroup pointer.
>
> Therefore, the infrastructure of objcg no longer only serves
> CONFIG_MEMCG_KMEM. In this patch, we move the infrastructure of the
> objcg out of the scope of the CONFIG_MEMCG_KMEM so that the LRU pages
> can reuse it to charge pages.
>
> We know that the LRU pages are not accounted at the root level. But
> the page->memcg_data points to the root_mem_cgroup. So the
> page->memcg_data of the LRU pages always points to a valid pointer.
> But the root_mem_cgroup dose not have an object cgroup. If we use
> obj_cgroup APIs to charge the LRU pages, we should set the
> page->memcg_data to a root object cgroup. So we also allocate an
> object cgroup for the root_mem_cgroup.
>
> Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Looks good to me. Also gets rid of some use_hierarchy cruft.
next prev parent reply other threads:[~2022-05-24 19:01 UTC|newest]
Thread overview: 93+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-24 6:05 [PATCH v4 00/11] Use obj_cgroup APIs to charge the LRU pages Muchun Song
2022-05-24 6:05 ` Muchun Song
[not found] ` <20220524060551.80037-1-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
2022-05-24 6:05 ` [PATCH v4 01/11] mm: memcontrol: prepare objcg API for non-kmem usage Muchun Song
2022-05-24 6:05 ` Muchun Song
[not found] ` <20220524060551.80037-2-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
2022-05-24 19:01 ` Johannes Weiner [this message]
2022-05-24 19:01 ` Johannes Weiner
[not found] ` <Yo0rhX2LFNPTv47b-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-05-25 8:46 ` Muchun Song
2022-05-25 8:46 ` Muchun Song
2022-05-25 2:36 ` Roman Gushchin
2022-05-25 2:36 ` Roman Gushchin
2022-05-25 7:57 ` Muchun Song
2022-05-25 7:57 ` Muchun Song
[not found] ` <Yo3hXe8bF3boWx5C-t1y1lxtqHnnw2QbUemf3bixXY32XiHfO@public.gmane.org>
2022-05-25 12:37 ` Johannes Weiner
2022-05-25 12:37 ` Johannes Weiner
[not found] ` <Yo4jJjBksFVeJHxA-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-05-25 13:08 ` Muchun Song
2022-05-25 13:08 ` Muchun Song
2022-05-24 6:05 ` [PATCH v4 02/11] mm: memcontrol: introduce compact_folio_lruvec_lock_irqsave Muchun Song
2022-05-24 6:05 ` Muchun Song
[not found] ` <20220524060551.80037-3-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
2022-05-24 19:22 ` Johannes Weiner
2022-05-24 19:22 ` Johannes Weiner
[not found] ` <Yo0wj9OAPptmUoWM-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-05-25 9:38 ` Muchun Song
2022-05-25 9:38 ` Muchun Song
2022-05-24 6:05 ` [PATCH v4 03/11] mm: memcontrol: make lruvec lock safe when LRU pages are reparented Muchun Song
2022-05-24 6:05 ` Muchun Song
2022-05-24 19:23 ` Waiman Long
[not found] ` <f55976e6-d209-32c2-504d-f73a9b504511-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2022-05-25 10:20 ` Muchun Song
2022-05-25 10:20 ` Muchun Song
[not found] ` <Yo4C6T56O7EBADbX-t1y1lxtqHnkkxugUQmjJ3WroNbXEc4rFQQ4Iyu8u01E@public.gmane.org>
2022-05-25 14:59 ` Waiman Long
2022-05-25 14:59 ` Waiman Long
[not found] ` <20220524060551.80037-4-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
2022-05-24 19:27 ` Johannes Weiner
2022-05-24 19:27 ` Johannes Weiner
[not found] ` <Yo0xmKOkBkhRy+bq-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-05-25 9:53 ` Muchun Song
2022-05-25 9:53 ` Muchun Song
[not found] ` <Yo38mlkMBFz2h+yP-t1y1lxtqHnkkxugUQmjJ3WroNbXEc4rFQQ4Iyu8u01E@public.gmane.org>
2022-05-25 12:30 ` Johannes Weiner
2022-05-25 12:30 ` Johannes Weiner
[not found] ` <Yo4hVw7B+bUlMzLX-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-05-25 13:03 ` Muchun Song
2022-05-25 13:03 ` Muchun Song
[not found] ` <Yo4pPw+IHPBZvZUv-t1y1lxtqHnkkxugUQmjJ3WroNbXEc4rFQQ4Iyu8u01E@public.gmane.org>
2022-05-25 14:48 ` Johannes Weiner
2022-05-25 14:48 ` Johannes Weiner
[not found] ` <Yo5B1tLcYPUoaACS-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-05-25 15:38 ` Muchun Song
2022-05-25 15:38 ` Muchun Song
[not found] ` <Yo5NdncOsqL0xP8Q-t1y1lxtqHnkkxugUQmjJ3WroNbXEc4rFQQ4Iyu8u01E@public.gmane.org>
2022-05-26 20:17 ` Waiman Long
2022-05-26 20:17 ` Waiman Long
[not found] ` <9fe57cf7-9d21-3f91-ef27-e046b426c219-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2022-05-27 2:55 ` Muchun Song
2022-05-27 2:55 ` Muchun Song
2022-05-24 6:05 ` [PATCH v4 04/11] mm: vmscan: rework move_pages_to_lru() Muchun Song
2022-05-24 6:05 ` Muchun Song
[not found] ` <20220524060551.80037-5-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
2022-05-24 19:38 ` Johannes Weiner
2022-05-24 19:38 ` Johannes Weiner
[not found] ` <Yo00Svsy/N8cJVmf-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-05-25 11:38 ` Muchun Song
2022-05-25 11:38 ` Muchun Song
2022-05-24 19:52 ` Waiman Long
2022-05-24 19:52 ` Waiman Long
[not found] ` <78de6197-7de6-9fe7-9567-1321c06c6e9b-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2022-05-25 11:43 ` Muchun Song
2022-05-25 11:43 ` Muchun Song
2022-05-25 2:43 ` Roman Gushchin
2022-05-25 2:43 ` Roman Gushchin
2022-05-25 11:41 ` Muchun Song
2022-05-25 11:41 ` Muchun Song
2022-05-24 6:05 ` [PATCH v4 05/11] mm: thp: introduce folio_split_queue_lock{_irqsave}() Muchun Song
2022-05-24 6:05 ` Muchun Song
2022-05-24 6:05 ` [PATCH v4 06/11] mm: thp: make split queue lock safe when LRU pages are reparented Muchun Song
2022-05-24 6:05 ` Muchun Song
[not found] ` <20220524060551.80037-7-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
2022-05-25 2:54 ` Roman Gushchin
2022-05-25 2:54 ` Roman Gushchin
2022-05-25 11:44 ` Muchun Song
2022-05-25 11:44 ` Muchun Song
2022-05-24 6:05 ` [PATCH v4 07/11] mm: memcontrol: make all the callers of {folio,page}_memcg() safe Muchun Song
2022-05-24 6:05 ` Muchun Song
[not found] ` <20220524060551.80037-8-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
2022-05-25 3:03 ` Roman Gushchin
2022-05-25 3:03 ` Roman Gushchin
2022-05-25 11:51 ` Muchun Song
2022-05-25 11:51 ` Muchun Song
2022-05-24 6:05 ` [PATCH v4 08/11] mm: memcontrol: introduce memcg_reparent_ops Muchun Song
2022-05-24 6:05 ` Muchun Song
2022-05-24 6:05 ` [PATCH v4 09/11] mm: memcontrol: use obj_cgroup APIs to charge the LRU pages Muchun Song
2022-05-24 6:05 ` Muchun Song
2022-05-24 12:29 ` kernel test robot
[not found] ` <20220524060551.80037-10-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
2022-05-24 18:16 ` kernel test robot
2022-05-24 18:16 ` kernel test robot
2022-05-25 7:14 ` [mm] bec0ae1210: WARNING:possible_recursive_locking_detected kernel test robot
2022-05-25 7:14 ` kernel test robot
2022-05-24 6:05 ` [PATCH v4 10/11] mm: lru: add VM_BUG_ON_FOLIO to lru maintenance function Muchun Song
2022-05-24 6:05 ` Muchun Song
[not found] ` <20220524060551.80037-11-songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>
2022-05-24 19:44 ` Johannes Weiner
2022-05-24 19:44 ` Johannes Weiner
[not found] ` <Yo01ghDEu4KcYKpH-druUgvl0LCNAfugRpC6u6w@public.gmane.org>
2022-05-25 11:59 ` Muchun Song
2022-05-25 11:59 ` Muchun Song
2022-05-25 2:40 ` Roman Gushchin
2022-05-25 2:40 ` Roman Gushchin
2022-05-25 11:58 ` Muchun Song
2022-05-24 6:05 ` [PATCH v4 11/11] mm: lru: use lruvec lock to serialize memcg changes Muchun Song
2022-05-24 6:05 ` Muchun 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=Yo0rhX2LFNPTv47b@cmpxchg.org \
--to=hannes-druugvl0lcnafugrpc6u6w@public.gmane.org \
--cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=duanxiongchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
--cc=longman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=mhocko-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=roman.gushchin-fxUVXftIFDnyG1zEObXtfA@public.gmane.org \
--cc=shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=songmuchun-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org \
/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.