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: <20200630005625.2405062-1-daeho43@gmail.com> <961072bb-4c8f-b01e-666d-1f5e35a8b76d@huawei.com>
 <CACOAw_wQx5wjdWDX_WFebNS42t=wBuSh_k7oQ4v7abBv80SZXw@mail.gmail.com>
 <9d1afacc-6033-2bae-d55d-909d50f1904b@huawei.com> <CACOAw_zFLgeoomdHhRmzYMtCocTugW5AVxb2wZnoRAC9+hR9Gw@mail.gmail.com>
 <ee97fa6e-33df-04f9-ec37-6b7d7b495acb@huawei.com>
In-Reply-To: <ee97fa6e-33df-04f9-ec37-6b7d7b495acb@huawei.com>
From:   Daeho Jeong <daeho43@gmail.com>
Date:   Fri, 3 Jul 2020 11:46:42 +0900
Message-ID: <CACOAw_y23emnXdmbUJT_QDdF42EGJ9p4-ZuC9caJBGTpLWbR=g@mail.gmail.com>
Subject: Re: [f2fs-dev] [PATCH] f2fs: add symbolic link to kobject in sysfs
To:     Chao Yu <yuchao0@huawei.com>
Cc:     linux-kernel@vger.kernel.org,
        linux-f2fs-devel@lists.sourceforge.net, kernel-team@android.com,
        Daeho Jeong <daehojeong@google.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Got it.

Thanks~

2020=EB=85=84 7=EC=9B=94 3=EC=9D=BC (=EA=B8=88) =EC=98=A4=EC=A0=84 11:02, C=
hao Yu <yuchao0@huawei.com>=EB=8B=98=EC=9D=B4 =EC=9E=91=EC=84=B1:
>
> On 2020/7/1 20:12, Daeho Jeong wrote:
> >> On 2020/7/1 15:04, Daeho Jeong wrote:
> >>> Actually, I want to keep the mount number remaining to the same
> >>> number, even if it's re-mounted.
> >>
> >> Then once there is f2fs umounter, the order will be incorrect...
> >
> > Actually, we prepared this patch for a strictly controlled system like
> > Android to easily access the sysfs node for a specific partition like
> > userdata partition using a specific number.
>
> I'm not against Android defined interfaces, just be confused about the
> behavior that does not fully documented (at least, we should add this
> into f2fs doc, and specify this is android specified interface), somethin=
g
> like once one mount point was umounted, that sequential number @x in
> 'mount_@x" could be reused by later newly mounted point, it breaks the
> description: "in the order of mounting filesystem".
>
> > In this system, we don't worry about another unexpected f2fs umounter
> > interfering in between unmounting and mounting a partition.
> >
> > When we are under the condition that we can keep track of how many
> > times the userdata partition has been re-mounted, we might as well use
> > the original partition name like "/sys/fs/f2fs/dm-9".
> > This is for when we couldn't do that.
> > .
> >