LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Roman Gushchin <guro@fb.com>
To: Muchun Song <songmuchun@bytedance.com>
Cc: <hannes@cmpxchg.org>, <mhocko@kernel.org>,
	<akpm@linux-foundation.org>, <shakeelb@google.com>,
	<vdavydov.dev@gmail.com>, <linux-kernel@vger.kernel.org>,
	<linux-mm@kvack.org>, <duanxiongchun@bytedance.com>,
	<fam.zheng@bytedance.com>, <bsingharora@gmail.com>,
	<shy828301@gmail.com>, <alexs@kernel.org>, <smuchun@gmail.com>,
	<zhengqi.arch@bytedance.com>
Subject: Re: [PATCH v1 01/12] mm: memcontrol: prepare objcg API for non-kmem usage
Date: Tue, 17 Aug 2021 20:01:36 -0700	[thread overview]
Message-ID: <YRx4EM0MWSjtutPD@carbon.dhcp.thefacebook.com> (raw)
In-Reply-To: <20210814052519.86679-2-songmuchun@bytedance.com>

On Sat, Aug 14, 2021 at 01:25:08PM +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>

I like the "move objcg stuff to memcg/css level" part.

I'm less convinced about making byte-sized charging kmem-specific
(both naming and #ifdef CONFIG_MEMCG_KMEM). Do we really win a lot?

I understand why we might wanna compile out some checks from the
hot allocation path, but few bytes in struct objcg will not make a big
difference, as well as few lines of code in cgroup creation/removal paths.

Also it might be useful for byte-sized accounting outside kmem, e.g. zswap.
So, I'd remove this dependency and rename to something like
obj_cgroup_release_bytes().

In the long run we might wanna to eliminate CONFIG_MEMCG_KMEM completely,
so let's at least not add new dependencies.

Thanks!

  parent reply	other threads:[~2021-08-18  3:02 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-14  5:25 [PATCH v1 00/12] Use obj_cgroup APIs to charge the LRU pages Muchun Song
2021-08-14  5:25 ` [PATCH v1 01/12] mm: memcontrol: prepare objcg API for non-kmem usage Muchun Song
2021-08-14 22:23   ` kernel test robot
2021-08-18  3:01   ` Roman Gushchin [this message]
2021-08-20  6:44     ` Muchun Song
2021-08-14  5:25 ` [PATCH v1 02/12] mm: memcontrol: introduce compact_folio_lruvec_lock_irqsave Muchun Song
2021-08-14  5:25 ` [PATCH v1 03/12] mm: memcontrol: make lruvec lock safe when LRU pages are reparented Muchun Song
2021-08-18  3:18   ` Roman Gushchin
2021-08-18  4:28     ` Muchun Song
2021-08-18  4:47       ` Roman Gushchin
2021-08-14  5:25 ` [PATCH v1 04/12] mm: vmscan: rework move_pages_to_lru() Muchun Song
2021-08-14  5:25 ` [PATCH v1 05/12] mm: thp: introduce folio_split_queue_lock{_irqsave}() Muchun Song
2021-08-14  8:22   ` kernel test robot
2021-08-14 10:38   ` kernel test robot
2021-08-14  5:25 ` [PATCH v1 06/12] mm: thp: make split queue lock safe when LRU pages are reparented Muchun Song
2021-08-14  5:25 ` [PATCH v1 07/12] mm: memcontrol: make all the callers of {folio,page}_memcg() safe Muchun Song
2021-08-14  5:25 ` [PATCH v1 08/12] mm: memcontrol: introduce memcg_reparent_ops Muchun Song
2021-08-14  5:25 ` [PATCH v1 09/12] mm: memcontrol: use obj_cgroup APIs to charge the LRU pages Muchun Song
2021-08-14 14:08   ` kernel test robot
2021-08-14  5:25 ` [PATCH v1 10/12] mm: memcontrol: rename {un}lock_page_memcg() to {un}lock_page_objcg() Muchun Song
2021-08-14  5:25 ` [PATCH v1 11/12] mm: lru: add VM_BUG_ON_FOLIO to lru maintenance function Muchun Song
2021-08-14  5:25 ` [PATCH v1 12/12] mm: lru: use lruvec lock to serialize memcg changes 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=YRx4EM0MWSjtutPD@carbon.dhcp.thefacebook.com \
    --to=guro@fb.com \
    --cc=akpm@linux-foundation.org \
    --cc=alexs@kernel.org \
    --cc=bsingharora@gmail.com \
    --cc=duanxiongchun@bytedance.com \
    --cc=fam.zheng@bytedance.com \
    --cc=hannes@cmpxchg.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@kernel.org \
    --cc=shakeelb@google.com \
    --cc=shy828301@gmail.com \
    --cc=smuchun@gmail.com \
    --cc=songmuchun@bytedance.com \
    --cc=vdavydov.dev@gmail.com \
    --cc=zhengqi.arch@bytedance.com \
    --subject='Re: [PATCH v1 01/12] mm: memcontrol: prepare objcg API for non-kmem usage' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).