Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210421070059.69361-1-songmuchun@bytedance.com>
 <20210421070059.69361-2-songmuchun@bytedance.com> <YK0lckLSLxhZMk3S@carbon.dhcp.thefacebook.com>
In-Reply-To: <YK0lckLSLxhZMk3S@carbon.dhcp.thefacebook.com>
From:   Muchun Song <songmuchun@bytedance.com>
Date:   Wed, 26 May 2021 10:46:34 +0800
Message-ID: <CAMZfGtWWHHVL1hten=tGBcYUye19i+PK0_3o8=p0uYJ_w=o1cQ@mail.gmail.com>
Subject: Re: [External] Re: [RFC PATCH v3 01/12] mm: memcontrol: move the
 objcg infrastructure out of CONFIG_MEMCG_KMEM
To:     Roman Gushchin <guro@fb.com>
Cc:     Johannes Weiner <hannes@cmpxchg.org>,
        Michal Hocko <mhocko@kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Shakeel Butt <shakeelb@google.com>,
        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>,
        fam.zheng@bytedance.com, "Singh, Balbir" <bsingharora@gmail.com>,
        Yang Shi <shy828301@gmail.com>,
        Alex Shi <alex.shi@linux.alibaba.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, May 26, 2021 at 12:27 AM Roman Gushchin <guro@fb.com> wrote:
>
> On Wed, Apr 21, 2021 at 03:00:48PM +0800, Muchun Song wrote:
> > Because memory 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.
> >
> > 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.
>
> Overall the patch looks very good to me. There are few small things to enhance:
>
> 1) I'd rename it. Looking at the title I expect a trivial code move,
>    however the patch is doing more than this: e.g. allocating an objcg
>    for the root memcg. Something like "prepare objcg API for non-kmem usage".

OK. I will rename it.

> 2) How about obj_cgroup_release_kmem() instead of obj_cgroup_release_uncharge()?

LGTM. Will use this name.

> 3) The first paragraph of the commit log looks a bit vague: which allocations
>    pinning memcgs? How about something like this?
>
> 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.

More clear. I would love to use this.

Thanks Roman.

>
>
> Thanks!