From: "Javier González" <javier@javigon.com>
To: Damien Le Moal <Damien.LeMoal@wdc.com>
Cc: "Jens Axboe" <axboe@kernel.dk>,
"Niklas Cassel" <Niklas.Cassel@wdc.com>,
"Ajay Joshi" <Ajay.Joshi@wdc.com>,
"Sagi Grimberg" <sagi@grimberg.me>,
"Keith Busch" <Keith.Busch@wdc.com>,
"Dmitry Fomichev" <Dmitry.Fomichev@wdc.com>,
"Aravind Ramesh" <Aravind.Ramesh@wdc.com>,
"linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
"Hans Holmberg" <Hans.Holmberg@wdc.com>,
"Matias Bjørling" <mb@lightnvm.io>,
"Christoph Hellwig" <hch@lst.de>,
"Matias Bjorling" <Matias.Bjorling@wdc.com>
Subject: Re: [PATCH 5/5] nvme: support for zoned namespaces
Date: Tue, 16 Jun 2020 17:02:17 +0200 [thread overview]
Message-ID: <20200616150217.inezhntsehtcbjsw@MacBook-Pro.localdomain> (raw)
In-Reply-To: <CY4PR04MB37512BCDD74996057697F5CAE79D0@CY4PR04MB3751.namprd04.prod.outlook.com>
On 16.06.2020 14:42, Damien Le Moal wrote:
>On 2020/06/16 23:16, Javier González wrote:
>> On 16.06.2020 12:35, Damien Le Moal wrote:
>>> On 2020/06/16 21:24, Javier González wrote:
>>>> On 16.06.2020 14:06, Matias Bjørling wrote:
>>>>> On 16/06/2020 14.00, Javier González wrote:
>>>>>> On 16.06.2020 13:18, Matias Bjørling wrote:
>>>>>>> On 16/06/2020 12.41, Javier González wrote:
>>>>>>>> On 16.06.2020 08:34, Keith Busch wrote:
>>>>>>>>> Add support for NVM Express Zoned Namespaces (ZNS) Command Set defined
>>>>>>>>> in NVM Express TP4053. Zoned namespaces are discovered based on their
>>>>>>>>> Command Set Identifier reported in the namespaces Namespace
>>>>>>>>> Identification Descriptor list. A successfully discovered Zoned
>>>>>>>>> Namespace will be registered with the block layer as a host managed
>>>>>>>>> zoned block device with Zone Append command support. A namespace that
>>>>>>>>> does not support append is not supported by the driver.
>>>>>>>>
>>>>>>>> Why are we enforcing the append command? Append is optional on the
>>>>>>>> current ZNS specification, so we should not make this mandatory in the
>>>>>>>> implementation. See specifics below.
>>>>>>
>>>>>>>
>>>>>>> There is already general support in the kernel for the zone append
>>>>>>> command. Feel free to submit patches to emulate the support. It is
>>>>>>> outside the scope of this patchset.
>>>>>>>
>>>>>>
>>>>>> It is fine that the kernel supports append, but the ZNS specification
>>>>>> does not impose the implementation for append, so the driver should not
>>>>>> do that either.
>>>>>>
>>>>>> ZNS SSDs that choose to leave append as a non-implemented optional
>>>>>> command should not rely on emulated SW support, specially when
>>>>>> traditional writes work very fine for a large part of current ZNS use
>>>>>> cases.
>>>>>>
>>>>>> Please, remove this virtual constraint.
>>>>>
>>>>> The Zone Append command is mandatory for zoned block devices. Please
>>>>> see https://lwn.net/Articles/818709/ for the background.
>>>>
>>>> I do not see anywhere in the block layer that append is mandatory for
>>>> zoned devices. Append is emulated on ZBC, but beyond that there is no
>>>> mandatory bits. Please explain.
>>>
>>> This is to allow a single write IO path for all types of zoned block device for
>>> higher layers, e.g file systems. The on-going re-work of btrfs zone support for
>>> instance now relies 100% on zone append being supported. That significantly
>>> simplifies the file system support and more importantly remove the need for
>>> locking around block allocation and BIO issuing, allowing to preserve a fully
>>> asynchronous write path that can include workqueues for efficient CPU usage of
>>> things like encryption and compression. Without zone append, file system would
>>> either (1) have to reject these drives that do not support zone append, or (2)
>>> implement 2 different write IO path (slower regular write and zone append). None
>>> of these options are ideal, to say the least.
>>>
>>> So the approach is: mandate zone append support for ZNS devices. To allow other
>>> ZNS drives, an emulation similar to SCSI can be implemented, with that emulation
>>> ideally combined to work for both types of drives if possible.
>>
>> Enforcing QD=1 becomes a problem on devices with large zones. In
>> a ZNS device that has smaller zones this should not be a problem.
>
>Let's be precise: this is not running the drive at QD=1, it is "at most one
>write *request* per zone". If the FS is simultaneously using multiple block
>groups mapped to different zones, you will get a total write QD > 1, and as many
>reads as you want.
>
>> Would you agree that it is possible to have a write path that relies on
>> QD=1, where the FS / application has the responsibility for enforcing
>> this? Down the road this QD can be increased if the device is able to
>> buffer the writes.
>
>Doing QD=1 per zone for writes at the FS layer, that is, at the BIO layer does
>not work. This is because BIOs can be as large as the FS wants them to be. Such
>large BIO will be split into multiple requests in the block layer, resulting in
>more than one write per zone. That is why the zone write locking is at the
>scheduler level, between BIO split and request dispatch. That avoids the
>multiple requests fragments of a large BIO to be reordered and fail. That is
>mandatory as the block layer itself can occasionally reorder requests and lower
>levels such as AHCI HW is also notoriously good at reversing sequential
>requests. For NVMe with multi-queue, the IO issuing process getting rescheduled
>on a different CPU can result in sequential IOs being in different queues, with
>the likely result of an out-of-order execution. All cases are avoided with zone
>write locking and at most one write request dispatch per zone as recommended by
>the ZNS specifications (ZBC and ZAC standards for SMR HDDs are silent on this).
>
I understand. I agree that the current FSs supporting ZNS follow this
approach and it makes sense that there is a common interface that
simplifies the FS implementation. See the comment below on the part I
believe we see things differently.
>> I would be OK with some FS implementations to rely on append and impose
>> the constraint that append has to be supported (and it would be our job
>> to change that), but I would like to avoid the driver rejecting
>> initializing the device because current FS implementations have
>> implemented this logic.
>
>What is the difference between the driver rejecting drives and the FS rejecting
>the same drives ? That has the same end result to me: an entire class of devices
>cannot be used as desired by the user. Implementing zone append emulation avoids
>the rejection entirely while still allowing the FS to have a single write IO
>path, thus simplifying the code.
The difference is that users that use a raw ZNS device submitting I/O
through the kernel would still be able to use these devices. The result
would be that the ZNS SSD is recognized and initialized, but the FS
format fails.
>
>> We can agree that a number of initial customers will use these devices
>> raw, using the in-kernel I/O path, but without a FS on top.
>>
>> Thoughts?
>>
>>> and note that
>>> this emulation would require the drive to be operated with mq-deadline to enable
>>> zone write locking for preserving write command order. While on a HDD the
>>> performance penalty is minimal, it will likely be significant on a SSD.
>>
>> Exactly my concern. I do not want ZNS SSDs to be impacted by this type
>> of design decision at the driver level.
>
>But your proposed FS level approach would end up doing the exact same thing with
>the same limitation and so the same potential performance impact. The block
>layer generic approach has the advantage that we do not bother the higher levels
>with the implementation of in-order request dispatch guarantees. File systems
>are complex enough. The less complexity is required for zone support, the better.
This depends very much on how the FS / application is managing
stripping. At the moment our main use case is enabling user-space
applications submitting I/Os to raw ZNS devices through the kernel.
Can we enable this use case to start with?
Thanks,
Javier
_______________________________________________
linux-nvme mailing list
linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
next prev parent reply other threads:[~2020-06-16 15:02 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-15 23:34 [PATCH 0/5] nvme support for zoned namespace command set Keith Busch
2020-06-15 23:34 ` [PATCH 1/5] block: add capacity field to zone descriptors Keith Busch
2020-06-15 23:49 ` Chaitanya Kulkarni
2020-06-16 10:28 ` Javier González
2020-06-16 13:47 ` Daniel Wagner
2020-06-16 13:54 ` Johannes Thumshirn
2020-06-16 15:41 ` Martin K. Petersen
2020-06-15 23:34 ` [PATCH 2/5] null_blk: introduce zone capacity for zoned device Keith Busch
2020-06-15 23:46 ` Chaitanya Kulkarni
2020-06-16 14:18 ` Daniel Wagner
2020-06-16 15:48 ` Martin K. Petersen
2020-06-15 23:34 ` [PATCH 3/5] nvme: implement I/O Command Sets Command Set support Keith Busch
2020-06-16 10:33 ` Javier González
2020-06-16 17:14 ` Niklas Cassel
2020-06-16 15:58 ` Martin K. Petersen
2020-06-16 17:01 ` Keith Busch
2020-06-17 9:50 ` Niklas Cassel
2020-06-16 17:06 ` Niklas Cassel
2020-06-17 2:01 ` Martin K. Petersen
2020-06-15 23:34 ` [PATCH 4/5] nvme: support for multi-command set effects Keith Busch
2020-06-16 10:34 ` Javier González
2020-06-16 16:03 ` Martin K. Petersen
2020-06-15 23:34 ` [PATCH 5/5] nvme: support for zoned namespaces Keith Busch
2020-06-16 10:41 ` Javier González
2020-06-16 11:18 ` Matias Bjørling
2020-06-16 12:00 ` Javier González
2020-06-16 12:06 ` Matias Bjørling
2020-06-16 12:24 ` Javier González
2020-06-16 12:27 ` Matias Bjørling
2020-06-16 12:35 ` Damien Le Moal
[not found] ` <CGME20200616130815uscas1p1be34e5fceaa548eac31fb30790a689d4@uscas1p1.samsung.com>
2020-06-16 13:08 ` Judy Brock
2020-06-16 13:32 ` Matias Bjørling
2020-06-16 13:34 ` Damien Le Moal
2020-06-16 14:16 ` Javier González
2020-06-16 14:42 ` Damien Le Moal
2020-06-16 15:02 ` Javier González [this message]
2020-06-16 15:20 ` Matias Bjørling
2020-06-16 16:03 ` Javier González
2020-06-16 16:07 ` Matias Bjorling
2020-06-16 16:21 ` Javier González
2020-06-16 16:25 ` Matias Bjørling
2020-06-16 15:48 ` Keith Busch
2020-06-16 15:55 ` Javier González
2020-06-16 16:04 ` Matias Bjorling
2020-06-16 16:07 ` Keith Busch
2020-06-16 16:13 ` Javier González
2020-06-17 0:38 ` Damien Le Moal
2020-06-17 6:18 ` Javier González
2020-06-17 6:54 ` Damien Le Moal
2020-06-17 7:11 ` Javier González
2020-06-17 7:29 ` Damien Le Moal
2020-06-17 7:34 ` Javier González
2020-06-17 0:14 ` Damien Le Moal
2020-06-17 6:09 ` Javier González
2020-06-17 6:47 ` Damien Le Moal
2020-06-17 7:02 ` Javier González
2020-06-17 7:24 ` Damien Le Moal
2020-06-17 7:29 ` Javier González
[not found] ` <CGME20200616123503uscas1p22ce22054a1b4152a20437b5abdd55119@uscas1p2.samsung.com>
2020-06-16 12:35 ` Judy Brock
2020-06-16 12:37 ` Damien Le Moal
2020-06-16 12:37 ` Matias Bjørling
2020-06-16 13:12 ` Judy Brock
2020-06-16 13:18 ` Judy Brock
2020-06-16 13:32 ` Judy Brock
2020-06-16 13:39 ` Damien Le Moal
2020-06-17 7:43 ` Christoph Hellwig
2020-06-17 12:01 ` Martin K. Petersen
2020-06-17 15:00 ` Javier González
2020-06-17 14:42 ` Javier González
2020-06-17 17:57 ` Matias Bjørling
2020-06-17 18:28 ` Javier González
2020-06-17 18:55 ` Matias Bjorling
2020-06-17 19:09 ` Javier González
2020-06-17 19:23 ` Matias Bjørling
2020-06-17 19:40 ` Javier González
2020-06-17 23:44 ` Heiner Litz
2020-06-18 1:55 ` Keith Busch
2020-06-18 4:24 ` Heiner Litz
2020-06-18 5:15 ` Damien Le Moal
2020-06-18 20:47 ` Heiner Litz
2020-06-18 21:04 ` Matias Bjorling
2020-06-18 21:19 ` Keith Busch
2020-06-18 22:05 ` Heiner Litz
2020-06-19 0:57 ` Damien Le Moal
2020-06-19 10:29 ` Matias Bjorling
2020-06-19 18:08 ` Heiner Litz
2020-06-19 18:10 ` Keith Busch
2020-06-19 18:17 ` Heiner Litz
2020-06-19 18:22 ` Keith Busch
2020-06-19 18:25 ` Matias Bjørling
2020-06-19 18:40 ` Heiner Litz
2020-06-19 18:18 ` Matias Bjørling
2020-06-20 6:33 ` Christoph Hellwig
2020-06-20 17:52 ` Heiner Litz
2022-03-02 21:11 ` Luis Chamberlain
2020-06-17 2:08 ` Martin K. Petersen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200616150217.inezhntsehtcbjsw@MacBook-Pro.localdomain \
--to=javier@javigon.com \
--cc=Ajay.Joshi@wdc.com \
--cc=Aravind.Ramesh@wdc.com \
--cc=Damien.LeMoal@wdc.com \
--cc=Dmitry.Fomichev@wdc.com \
--cc=Hans.Holmberg@wdc.com \
--cc=Keith.Busch@wdc.com \
--cc=Matias.Bjorling@wdc.com \
--cc=Niklas.Cassel@wdc.com \
--cc=axboe@kernel.dk \
--cc=hch@lst.de \
--cc=linux-block@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=mb@lightnvm.io \
--cc=sagi@grimberg.me \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).