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: <20210121134529.1201249-1-daeho43@gmail.com> <6e4996f8-fdf0-93dd-8253-9e8f5e6fca06@huawei.com>
In-Reply-To: <6e4996f8-fdf0-93dd-8253-9e8f5e6fca06@huawei.com>
From:   Daeho Jeong <daeho43@gmail.com>
Date:   Fri, 22 Jan 2021 10:36:58 +0900
Message-ID: <CACOAw_yDYWw-3f2q_3EoHu8M196KEi=4aNsbbZqXYra=d7ebng@mail.gmail.com>
Subject: Re: [f2fs-dev] [PATCH v5 1/2] f2fs: introduce checkpoint=merge mount option
To:     Chao Yu <yuchao0@huawei.com>
Cc:     linux-kernel@vger.kernel.org,
        linux-f2fs-devel@lists.sourceforge.net, kernel-team@android.com,
        Sungjong Seo <sj1557.seo@samsung.com>,
        Daeho Jeong <daehojeong@google.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Got it~! :)

2021=EB=85=84 1=EC=9B=94 22=EC=9D=BC (=EA=B8=88) =EC=98=A4=EC=A0=84 9:55, C=
hao Yu <yuchao0@huawei.com>=EB=8B=98=EC=9D=B4 =EC=9E=91=EC=84=B1:
>
> If there is no update in v5, it will be better to include my Reviewed-by
> tag, please note that. :)
>
> Thanks,
>
> On 2021/1/21 21:45, Daeho Jeong wrote:
> > From: Daeho Jeong<daehojeong@google.com>
> >
> > We've added a new mount option "checkpoint=3Dmerge", which creates a
> > kernel daemon and makes it to merge concurrent checkpoint requests as
> > much as possible to eliminate redundant checkpoint issues. Plus, we
> > can eliminate the sluggish issue caused by slow checkpoint operation
> > when the checkpoint is done in a process context in a cgroup having
> > low i/o budget and cpu shares. To make this do better, we set the
> > default i/o priority of the kernel daemon to "3", to give one higher
> > priority than other kernel threads. The below verification result
> > explains this.
> > The basic idea has come fromhttps://opensource.samsung.com.
> >
> > [Verification]
> > Android Pixel Device(ARM64, 7GB RAM, 256GB UFS)
> > Create two I/O cgroups (fg w/ weight 100, bg w/ wight 20)
> > Set "strict_guarantees" to "1" in BFQ tunables
> >
> > In "fg" cgroup,
> > - thread A =3D> trigger 1000 checkpoint operations
> >    "for i in `seq 1 1000`; do touch test_dir1/file; fsync test_dir1;
> >     done"
> > - thread B =3D> gererating async. I/O
> >    "fio --rw=3Dwrite --numjobs=3D1 --bs=3D128k --runtime=3D3600 --time_=
based=3D1
> >         --filename=3Dtest_img --name=3Dtest"
> >
> > In "bg" cgroup,
> > - thread C =3D> trigger repeated checkpoint operations
> >    "echo $$ > /dev/blkio/bg/tasks; while true; do touch test_dir2/file;
> >     fsync test_dir2; done"
> >
> > We've measured thread A's execution time.
> >
> > [ w/o patch ]
> > Elapsed Time: Avg. 68 seconds
> > [ w/  patch ]
> > Elapsed Time: Avg. 48 seconds
> >
> > Signed-off-by: Daeho Jeong<daehojeong@google.com>
> > Signed-off-by: Sungjong Seo<sj1557.seo@samsung.com>