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;
IronPort-SDR: uBJpJVK5qupXcnq2qec+No96Bn1Su7vZUKrObZeMz9TxmZEoxQs3MrVcZADYxMrwrzropmqvR4
 0siwJAJiFCU24eDELkpYEZdKiFC3kj+Yn4eDabXVjNBQPQGiCq+p/WyOaAaNujCVquG/h+Sfb6
 QUaulFBO0mIe3SbodRF7p/JDr72uz3u7jguWcPyfOq7gGmeBa9oWaRRurllOdneThZBjR2GiLA
 O3cWiyGR+C2P3GxiC3qtx8RqnXVd1Htl/Du6g7vNYZv4WZjwhfOiXbgiWGSVQk84Z5wwfaEV1I
 zB0=
From:   Damien Le Moal <Damien.LeMoal@wdc.com>
To:     Kanchan Joshi <joshiiitr@gmail.com>
CC:     "hch@infradead.org" <hch@infradead.org>,
        Jens Axboe <axboe@kernel.dk>,
        Pavel Begunkov <asml.silence@gmail.com>,
        Kanchan Joshi <joshi.k@samsung.com>,
        "viro@zeniv.linux.org.uk" <viro@zeniv.linux.org.uk>,
        "bcrl@kvack.org" <bcrl@kvack.org>,
        Matthew Wilcox <willy@infradead.org>,
        "linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linux-aio@kvack.org" <linux-aio@kvack.org>,
        "io-uring@vger.kernel.org" <io-uring@vger.kernel.org>,
        "linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
        "linux-api@vger.kernel.org" <linux-api@vger.kernel.org>,
        SelvaKumar S <selvakuma.s1@samsung.com>,
        Nitesh Shetty <nj.shetty@samsung.com>,
        Javier Gonzalez <javier.gonz@samsung.com>,
        Johannes Thumshirn <Johannes.Thumshirn@wdc.com>
Subject: Re: [PATCH v4 6/6] io_uring: add support for zone-append
Thread-Topic: [PATCH v4 6/6] io_uring: add support for zone-append
Thread-Index: AQHWYdbcc0q15qREKECGO7brHi6zEg==
Date:   Fri, 31 Jul 2020 08:14:22 +0000
Message-ID: <MWHPR04MB375863C20C1EF2CB27E62703E74E0@MWHPR04MB3758.namprd04.prod.outlook.com>
References: <b0b7159d-ed10-08ad-b6c7-b85d45f60d16@kernel.dk>
 <e871eef2-8a93-fdbc-b762-2923526a2db4@gmail.com>
 <80d27717-080a-1ced-50d5-a3a06cf06cd3@kernel.dk>
 <da4baa8c-76b0-7255-365c-d8b58e322fd0@gmail.com>
 <65a7e9a6-aede-31ce-705c-b7f94f079112@kernel.dk>
 <d4f9a5d3-1df2-1060-94fa-f77441a89299@gmail.com>
 <CA+1E3rJ3SoLU9aYcugAQgJnSPnJtcCwjZdMREXS3FTmXgy3yow@mail.gmail.com>
 <f030a338-cd52-2e83-e1da-bdbca910d49e@kernel.dk>
 <CA+1E3rKxZk2CatTuPcQq5d14vXL9_9LVb2_+AfR2m9xn2WTZdg@mail.gmail.com>
 <MWHPR04MB3758DC08EA17780E498E9EC0E74E0@MWHPR04MB3758.namprd04.prod.outlook.com>
 <20200731064526.GA25674@infradead.org>
 <MWHPR04MB37581344328A42EA7F5ED13EE74E0@MWHPR04MB3758.namprd04.prod.outlook.com>
 <CA+1E3rLM4G4SwzD6RWsK6Ssp7NmhiPedZDjrqN3kORQr9fxCtw@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
wdcipoutbound: EOP-TRUE
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: f78441f6-822e-4509-3638-08d83529bb26
X-MS-Exchange-CrossTenant-originalarrivaltime: 31 Jul 2020 08:14:22.1030
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: b61c8803-16f3-4c35-9b17-6f65f441df86
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-CrossTenant-userprincipalname: dqyMjdRWZCRxLPLB3QQc8kWxOH+glnqDW2ky0shuMsW2RyaKrFx77R6rPAdOI9ICqO/a9hb+oY0kLpDPfomo3w==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: MWHPR04MB0879
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 2020/07/31 16:59, Kanchan Joshi wrote:=0A=
> On Fri, Jul 31, 2020 at 12:29 PM Damien Le Moal <Damien.LeMoal@wdc.com> w=
rote:=0A=
>>=0A=
>> On 2020/07/31 15:45, hch@infradead.org wrote:=0A=
>>> On Fri, Jul 31, 2020 at 06:42:10AM +0000, Damien Le Moal wrote:=0A=
>>>>> - We may not be able to use RWF_APPEND, and need exposing a new=0A=
>>>>> type/flag (RWF_INDIRECT_OFFSET etc.) user-space. Not sure if this=0A=
>>>>> sounds outrageous, but is it OK to have uring-only flag which can be=
=0A=
>>>>> combined with RWF_APPEND?=0A=
>>>>=0A=
>>>> Why ? Where is the problem ? O_APPEND/RWF_APPEND is currently meaningl=
ess for=0A=
>>>> raw block device accesses. We could certainly define a meaning for the=
se in the=0A=
>>>> context of zoned block devices.=0A=
>>>=0A=
>>> We can't just add a meaning for O_APPEND on block devices now,=0A=
>>> as it was previously silently ignored.  I also really don't think any=
=0A=
>>> of these semantics even fit the block device to start with.  If you=0A=
>>> want to work on raw zones use zonefs, that's what is exists for.=0A=
>>=0A=
>> Which is fine with me. Just trying to say that I think this is exactly t=
he=0A=
>> discussion we need to start with. What interface do we implement...=0A=
>>=0A=
>> Allowing zone append only through zonefs as the raw block device equival=
ent, all=0A=
>> the O_APPEND/RWF_APPEND semantic is defined and the "return written offs=
et"=0A=
>> implementation in VFS would be common for all file systems, including re=
gular=0A=
>> ones. Beside that, there is I think the question of short writes... Not =
sure if=0A=
>> short writes can currently happen with async RWF_APPEND writes to regula=
r files.=0A=
>> I think not but that may depend on the FS.=0A=
> =0A=
> generic_write_check_limits (called by generic_write_checks, used by=0A=
> most FS) may make it short, and AFAIK it does not depend on=0A=
> async/sync.=0A=
=0A=
Johannes has a patch (not posted yet) fixing all this for zonefs,=0A=
differentiating sync and async cases, allow short writes or not, etc. This =
was=0A=
done by not using generic_write_check_limits() and instead writing a=0A=
zonefs_check_write() function that is zone append friendly.=0A=
=0A=
We can post that as a base for the discussion on semantic if you want...=0A=
=0A=
> This was one of the reason why we chose to isolate the operation by a=0A=
> different IOCB flag and not by IOCB_APPEND alone.=0A=
=0A=
For zonefs, the plan is:=0A=
* For the sync write case, zone append is always used.=0A=
* For the async write case, if we see IOCB_APPEND, then zone append BIOs ar=
e=0A=
used. If not, regular write BIOs are used.=0A=
=0A=
Simple enough I think. No need for a new flag.=0A=
=0A=
-- =0A=
Damien Le Moal=0A=
Western Digital Research=0A=