Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:7 as permitted sender) client-ip=2620:137:e000::3:7;
MIME-Version: 1.0
References: <20231007130905.78554-1-jianlv@ebay.com> <87mswtkj8x.fsf@yhuang6-desk2.ccr.corp.intel.com>
In-Reply-To: <87mswtkj8x.fsf@yhuang6-desk2.ccr.corp.intel.com>
From:   Jianlin Lv <iecedge@gmail.com>
Date:   Sun, 8 Oct 2023 15:52:04 +0800
Message-ID: <CAFA-uR9ymPTktMbi96cb+smjQHB4Y=8SQfAqmsqTbniGbkGTLA@mail.gmail.com>
Subject: Re: [PATCH] memcg: add interface to force disable swap
To:     "Huang, Ying" <ying.huang@intel.com>
Cc:     tj@kernel.org, lizefan.x@bytedance.com, hannes@cmpxchg.org,
        corbet@lwn.net, mhocko@kernel.org, roman.gushchin@linux.dev,
        shakeelb@google.com, muchun.song@linux.dev,
        akpm@linux-foundation.org, yosryahmed@google.com,
        willy@infradead.org, linmiaohe@huawei.com,
        wangkefeng.wang@huawei.com, laoar.shao@gmail.com,
        yuzhao@google.com, wuyun.abel@bytedance.com, david@redhat.com,
        peterx@redhat.com, vishal.moola@gmail.com, hughd@google.com,
        cgroups@vger.kernel.org, linux-doc@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org, jianlv@ebay.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Sun, Oct 8, 2023 at 9:17=E2=80=AFAM Huang, Ying <ying.huang@intel.com> w=
rote:
>
> Jianlin Lv <iecedge@gmail.com> writes:
>
> > From: Jianlin Lv <iecedge@gmail.com>
> >
> > Global reclaim will swap even if swappiness is set to 0.
>
> Why?  Can you elaborate the situation?

We reproduced the issue of pages being swapped out even when swappiness is
set to 0 in the production environment through the following test program.
Not sure whether this program can reproduce the issue in any environment.

From the implementation of the get_scan_count code, it can be seen that,
based on the current runtime situation, memory reclamation will choose a
scanning method (SCAN_ANON/SCAN_FILE/SCAN_FRACT) to determine how
aggressively the anon and file LRU are scanned. However, this introduces
uncertainty.

For the JVM issue at hand, we expect deterministic SCAN_FILE scan to avoid
swapping out anon pages.


code::
#!/usr/bin/env python

import mmap
import os
import sys

def write_files():
    count =3D 1
    if not os.path.isdir(WRITE_DIR):
        os.mkdir(WRITE_DIR)

    while True:
        _, i =3D divmod(count, 6000)
        file =3D "{}/{}_{}.txt".format(WRITE_DIR, WRITE_FILE, i)

        with open(file, 'w') as f:
            # Write 100 MB to a file
            num_chars =3D  100 * 1024 * 1024
            f.write('0' * num_chars)
        count =3D count + 1

def create_read_file():
    with open(READ_FILE, 'wb') as f:
        num_chars =3D 10000 * 1024 * 1024
        f.write(b'0' * num_chars)

def read_file():
    with open(READ_FILE, mode=3D"r", encoding=3D"utf8") as f:
        mm =3D mmap.mmap(f.fileno(), length=3D0, access=3Dmmap.ACCESS_READ)
        text =3D mm.read()
        write_files()

WRITE_FILE =3D "file"
WRITE_DIR =3D "/tmp/rm_rf_me"
READ_FILE=3D"/tmp/10g_file_delete"
if not os.path.isfile(READ_FILE):
    create_read_file()
read_file()


Jianlin

>
> > In particular
> > case, users wish to be able to completely disable swap for specific
> > processes. One scenario is that if JVM memory pages falls into swap,
> > the performance will noticeably reduce and the GC pauses tend to increa=
se
> > to levels not tolerable by most applications.
> > If it's possible to only disable swap out for specific processes, it ca=
n
> > address the JVM GC pauses issues, and at the same time, memory reclaim
> > pressure is also manageable.
> >
> > This patch adds "memory.swap_force_disable" control file to support dis=
able
> > swap for non-root cgroup. When process is associated with a cgroup,
> > 'echo 1 > memory.swap_force_disable' will forbid anon pages be swapped =
out.
> > This patch also adds read and write handler of the control file.
>
> --
> Best Regards,
> Huang, Ying