Linux-NVME Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: Christoph Hellwig <hch@lst.de>, Keith Busch <kbusch@kernel.org>
Cc: Kanchan Joshi <joshi.k@samsung.com>,
	martin.petersen@oracle.com, sagi@grimberg.me,
	linux-nvme@lists.infradead.org, bvanassche@acm.org,
	javier.gonz@samsung.com, gost.dev@samsung.com,
	Hui Qi <hui81.qi@samsung.com>,
	Nitesh Shetty <nj.shetty@samsung.com>
Subject: Re: [PATCH v3] nvme: enable FDP support
Date: Tue, 2 Jul 2024 09:49:35 -0600	[thread overview]
Message-ID: <d7a308f4-7b1f-4136-b655-c72474992151@kernel.dk> (raw)
In-Reply-To: <20240702153631.GA534@lst.de>

On 7/2/24 9:36 AM, Christoph Hellwig wrote:
> On Tue, Jul 02, 2024 at 09:15:06AM -0600, Keith Busch wrote:
>> My first concern for applying something like this is what kind of
>> maintenance burdens does this create, or any potential harm to users who
>> don't subscribe to the feature? Nothing here looks alarming to me, and
>> there is a clear demand to be able to access these features like this.
>> Maybe it works out for some workloads, maybe it doesn't, but I don't see
>> a need to block this at this point.
> 
> We're just abusing the interface, and giving how badly designed and
> intentionally Linux-hostile it was I see no point.
> 
> If Samsung and Meta care enough about good Linux I/O temperature hint
> support we'll get what we want in the technical working group, please
> help working on that!

I'm with Keith on this one - there's no real maintenance burden to bear
for this feature, it's pretty well contained. There's no point gate
keeping it based on changing the spec, as that will take a long time to
get done. And there's no reason to hold this back until this happens
based on mostly ideological reasons, when there are actual users that
could use it now. What do we have to lose?

-- 
Jens Axboe



  reply	other threads:[~2024-07-02 15:49 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20240702103348epcas5p1015eafbddf4795558843cd74b0453b12@epcas5p1.samsung.com>
2024-07-02 10:26 ` [PATCH v3] nvme: enable FDP support Kanchan Joshi
2024-07-02 11:39   ` Christoph Hellwig
2024-07-02 15:15     ` Keith Busch
2024-07-02 15:36       ` Christoph Hellwig
2024-07-02 15:49         ` Jens Axboe [this message]
2024-07-02 15:51           ` Christoph Hellwig
2024-07-03  8:40     ` Kanchan Joshi
2024-07-03 15:08       ` Christoph Hellwig

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=d7a308f4-7b1f-4136-b655-c72474992151@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=gost.dev@samsung.com \
    --cc=hch@lst.de \
    --cc=hui81.qi@samsung.com \
    --cc=javier.gonz@samsung.com \
    --cc=joshi.k@samsung.com \
    --cc=kbusch@kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=martin.petersen@oracle.com \
    --cc=nj.shetty@samsung.com \
    --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