Message-ID: <20200626070345.vuxic46l3agy3jay@mpHalley.localdomain>
Date:   Fri, 26 Jun 2020 09:03:45 +0200
From:   "javier.gonz@...sung.com" <javier@...igon.com>
To:     Damien Le Moal <Damien.LeMoal@....com>
Cc:     Kanchan Joshi <joshi.k@...sung.com>,
        "axboe@...nel.dk" <axboe@...nel.dk>,
        "viro@...iv.linux.org.uk" <viro@...iv.linux.org.uk>,
        "bcrl@...ck.org" <bcrl@...ck.org>,
        "asml.silence@...il.com" <asml.silence@...il.com>,
        "hch@...radead.org" <hch@...radead.org>,
        "linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
        "mb@...htnvm.io" <mb@...htnvm.io>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-aio@...ck.org" <linux-aio@...ck.org>,
        "io-uring@...r.kernel.org" <io-uring@...r.kernel.org>,
        "linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
        "selvakuma.s1@...sung.com" <selvakuma.s1@...sung.com>,
        "nj.shetty@...sung.com" <nj.shetty@...sung.com>
Subject: Re: [PATCH v2 0/2] zone-append support in io-uring and aio

On 26.06.2020 06:56, Damien Le Moal wrote:
>On 2020/06/26 15:37, javier.gonz@...sung.com wrote:
>> On 26.06.2020 03:11, Damien Le Moal wrote:
>>> On 2020/06/26 2:18, Kanchan Joshi wrote:
>>>> [Revised as per feedback from Damien, Pavel, Jens, Christoph, Matias, Wilcox]
>>>>
>>>> This patchset enables zone-append using io-uring/linux-aio, on block IO path.
>>>> Purpose is to provide zone-append consumption ability to applications which are
>>>> using zoned-block-device directly.
>>>>
>>>> The application may specify RWF_ZONE_APPEND flag with write when it wants to
>>>> send zone-append. RWF_* flags work with a certain subset of APIs e.g. uring,
>>>> aio, and pwritev2. An error is reported if zone-append is requested using
>>>> pwritev2. It is not in the scope of this patchset to support pwritev2 or any
>>>> other sync write API for reasons described later.
>>>>
>>>> Zone-append completion result --->
>>>> With zone-append, where write took place can only be known after completion.
>>>> So apart from usual return value of write, additional mean is needed to obtain
>>>> the actual written location.
>>>>
>>>> In aio, this is returned to application using res2 field of io_event -
>>>>
>>>> struct io_event {
>>>>         __u64           data;           /* the data field from the iocb */
>>>>         __u64           obj;            /* what iocb this event came from */
>>>>         __s64           res;            /* result code for this event */
>>>>         __s64           res2;           /* secondary result */
>>>> };
>>>>
>>>> In io-uring, cqe->flags is repurposed for zone-append result.
>>>>
>>>> struct io_uring_cqe {
>>>>         __u64   user_data;      /* sqe->data submission passed back */
>>>>         __s32   res;            /* result code for this event */
>>>>         __u32   flags;
>>>> };
>>>>
>>>> Since 32 bit flags is not sufficient, we choose to return zone-relative offset
>>>> in sector/512b units. This can cover zone-size represented by chunk_sectors.
>>>> Applications will have the trouble to combine this with zone start to know
>>>> disk-relative offset. But if more bits are obtained by pulling from res field
>>>> that too would compel application to interpret res field differently, and it
>>>> seems more painstaking than the former option.
>>>> To keep uniformity, even with aio, zone-relative offset is returned.
>>>
>>> I am really not a fan of this, to say the least. The input is byte offset, the
>>> output is 512B relative sector count... Arg... We really cannot do better than
>>> that ?
>>>
>>> At the very least, byte relative offset ? The main reason is that this is
>>> _somewhat_ acceptable for raw block device accesses since the "sector"
>>> abstraction has a clear meaning, but once we add iomap/zonefs async zone append
>>> support, we really will want to have byte unit as the interface is regular
>>> files, not block device file. We could argue that 512B sector unit is still
>>> around even for files (e.g. block counts in file stat). Bu the different unit
>>> for input and output of one operation is really ugly. This is not nice for the user.
>>>
>>
>> You can refer to the discussion with Jens, Pavel and Alex on the uring
>> interface. With the bits we have and considering the maximun zone size
>> supported, there is no space for a byte relative offset. We can take
>> some bits from cqe->res, but we were afraid this is not very
>> future-proof. Do you have a better idea?
>
>If you can take 8 bits, that gives you 40 bits, enough to support byte relative
>offsets for any zone size defined as a number of 512B sectors using an unsigned
>int. Max zone size is 2^31 sectors in that case, so 2^40 bytes. Unless I am
>already too tired and my math is failing me...

Yes, the match is correct. I was thinking more of the bits being needed
for other use-case that could collide with append. We considered this
and discard it for being messy - when Pavel brought up the 512B
alignment we saw it as a good alternative.

Note too that we would be able to translate to a byte offset in
iouring.h too so the user would not need to think of this.

I do not feel strongly on this, so the one that better fits the current
and near-future for uring, that is the one we will send on V3. Will give
it until next week for others to comment too.

>
>zone size is defined by chunk_sectors, which is used for raid and software raids
>too. This has been an unsigned int forever. I do not see the need for changing
>this to a 64bit anytime soon, if ever. A raid with a stripe size larger than 1TB
>does not really make any sense. Same for zone size...

Yes. I think already max zone sizes are pretty huge. But yes, this might
change, so we will take it when it happens.

[...]

Javier

Powered by blists - more mailing lists