All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@infradead.org>
To: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@infradead.org>,
	linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org,
	adilger@dilger.ca, martin.petersen@oracle.com
Subject: Re: [PATCH 11/11] nvme: add support for streams and directives
Date: Mon, 19 Jun 2017 11:53:14 -0700	[thread overview]
Message-ID: <20170619185314.GA32047@infradead.org> (raw)
In-Reply-To: <0e920111-7846-f7f6-2288-e375f6a43843@kernel.dk>

On Mon, Jun 19, 2017 at 08:53:08AM -0600, Jens Axboe wrote:
> Looking at it a bit more closely - there's a difference between
> assigning X number of streams (allocating) for use by the subsystem or
> per-ns, and having to manually open them. So I don't necessarily think
> there's a problem here, neither for us or on the device.

As far as I can tell the allocate resource is only supposed to be
for an individual namespace:

Section 9.3:

"Stream resources may be allocated for the exclusive use of a specified
 namespace associated with a particular Host Identifier using the Allocate
 Resources operation. Stream resources that are not allocated for the
 exclusive use of any namespace are available NVM subsystem stream
 resources as reported in NVM Subsystem Streams Available (NSSA) and may
 be used by any namespace that has the Streams Directive enabled and has
 not been allocated exclusive stream resources in response to an
 Allocate Resources operation"

Section 9.3.1.3:

"The Allocate Resources operation indicates the number of streams that the host requests for the exclusive use of the specified namespace."

I think this is pretty clear, but if not it would be good if you could
bring it into the working group so that we can clarify it further.

> Awaiting clarification from the vendor what their position/view of this
> is. Reading the spec, I do agree that it leans towards only needing
> allocation for a specific name space, but it doesn't explicitly say that
> you can use any of the available streams, without allocation, if they
> haven't been assigned to a specific name space. I would interpret it
> that way, though.

Let's get some text into an ECN either way to clarify it..

  reply	other threads:[~2017-06-19 18:53 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-16 17:24 [PATCHSET v6] Add support for write life time hints Jens Axboe
2017-06-16 17:24 ` [PATCH 01/11] fs: add support for an inode to carry write hint related data Jens Axboe
2017-06-16 17:24 ` [PATCH 02/11] block: add support for write hints in a bio Jens Axboe
2017-06-16 17:24 ` [PATCH 03/11] blk-mq: expose stream write hints through debugfs Jens Axboe
2017-06-16 17:24 ` [PATCH 04/11] fs: add support for allowing applications to pass in write life time hints Jens Axboe
2017-06-16 17:24 ` [PATCH 05/11] fs: add fcntl() interface for setting/getting " Jens Axboe
2017-06-16 17:24 ` [PATCH 06/11] fs: add O_DIRECT support for sending down " Jens Axboe
2017-06-16 17:24 ` [PATCH 07/11] fs: add support for buffered writeback to pass down write hints Jens Axboe
2017-06-16 17:24 ` [PATCH 08/11] ext4: add support for passing in write hints for buffered writes Jens Axboe
2017-06-16 17:24 ` [PATCH 09/11] xfs: " Jens Axboe
2017-06-16 17:24 ` [PATCH 10/11] btrfs: " Jens Axboe
2017-06-16 17:24 ` [PATCH 11/11] nvme: add support for streams and directives Jens Axboe
2017-06-16 18:09   ` Christoph Hellwig
2017-06-16 19:41     ` Jens Axboe
2017-06-16 19:56       ` Jens Axboe
2017-06-17 12:21       ` Christoph Hellwig
2017-06-17 14:20         ` Jens Axboe
2017-06-17 15:03           ` Christoph Hellwig
2017-06-17 15:11             ` Jens Axboe
2017-06-17 15:43               ` Jens Axboe
2017-06-19  6:25               ` Christoph Hellwig
2017-06-19 14:31                 ` Jens Axboe
2017-06-19 14:53                   ` Jens Axboe
2017-06-19 18:53                     ` Christoph Hellwig [this message]
2017-06-19 19:03                       ` Jens Axboe
  -- strict thread matches above, loose matches on Subject: below --
2017-06-17 19:59 [PATCHSET v7] Add support for write life time hints Jens Axboe
2017-06-17 19:59 ` [PATCH 11/11] nvme: add support for streams and directives Jens Axboe
2017-06-19  6:35   ` Christoph Hellwig
2017-06-19 15:04     ` Jens Axboe
2017-06-15  3:45 [PATCHSET v4] Add support for write life time hints Jens Axboe
2017-06-15  3:45 ` [PATCH 11/11] nvme: add support for streams and directives Jens Axboe
2017-06-14 19:05 [PATCHSET v3] Add support for write life time hints Jens Axboe
2017-06-14 19:05 ` [PATCH 11/11] nvme: add support for streams and directives Jens Axboe
2017-06-14 20:32   ` Christoph Hellwig
2017-06-14 20:43     ` Jens Axboe
2017-06-13 17:15 [PATCH 0/11] Add support for write life time hints Jens Axboe
2017-06-13 17:15 ` [PATCH 11/11] nvme: add support for streams and directives Jens Axboe
2017-06-13 19:47   ` Andreas Dilger
2017-06-13 20:25     ` Jens Axboe
2017-06-13 21:12   ` Andreas Dilger
2017-06-13 21:18     ` Jens Axboe

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=20170619185314.GA32047@infradead.org \
    --to=hch@infradead.org \
    --cc=adilger@dilger.ca \
    --cc=axboe@kernel.dk \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.