linux-nvme.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Sagi Grimberg <sagi@grimberg.me>
To: Chaitanya Kulkarni <Chaitanya.Kulkarni@wdc.com>,
	"hch@lst.de" <hch@lst.de>
Cc: "linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>
Subject: Re: [RFC PATCH 0/2] nvmet: add polling support
Date: Thu, 12 Dec 2019 12:32:00 -0800	[thread overview]
Message-ID: <ed3638c6-7506-4ac6-a2ab-df432b2111b6@grimberg.me> (raw)
In-Reply-To: <BYAPR04MB57495A09DE5E7652E2B38AAF86550@BYAPR04MB5749.namprd04.prod.outlook.com>


>> percpu threads per namespace? Sounds like the wrong approach. These
>> threads will compete for cpu time with the main nvmet contexts.
>>
> That make sense, how about a global threadpool for target which can be
> shared between all the subsystem and their name-spaces just like we
> have buffered_io_wq?
>> Have you considered having the main nvmet contexts incorporate polling
>> activity between I/Os? Don't have a great dea on how to do it from first
>> thought...
>>
> 
> I am not able to understand nvmet context, can you please elaborate ?
> Are you referring to the pattern we one have in the
> nvme_execute_rq_polled() ?

No, we would want non-selective polling. Right now we have nvmet context
starting from the transport going to submit I/O to the backend, or
starting from the backend going to submit to the transport.

Ideally, we'd have these contexts to do the polling instead of a 
different thread that will poll for as much as it can taking away
cpu time?

One way to do it is to place a intermediate thread that will sit between
the transport and the backend but that would yield an additional context
switch in the I/O path (not ideal).

_______________________________________________
linux-nvme mailing list
linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme

  reply	other threads:[~2019-12-12 20:32 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-10  6:25 Chaitanya Kulkarni
2019-12-10  6:25 ` [RFC PATCH 1/2] nvmet: add bdev-ns " Chaitanya Kulkarni
2020-01-20 12:52   ` Max Gurtovoy
2020-01-21 19:22     ` Chaitanya Kulkarni
2020-01-23 14:23       ` Max Gurtovoy
2020-01-30 18:19         ` Chaitanya Kulkarni
2019-12-10  6:25 ` [RFC PATCH 2/2] nvmet: add file-ns " Chaitanya Kulkarni
2019-12-12  1:01 ` [RFC PATCH 0/2] nvmet: add " Sagi Grimberg
2019-12-12  5:44   ` Chaitanya Kulkarni
2019-12-12 20:32     ` Sagi Grimberg [this message]
2020-01-20  5:13       ` Chaitanya Kulkarni
2020-01-20  4:48   ` Chaitanya Kulkarni

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=ed3638c6-7506-4ac6-a2ab-df432b2111b6@grimberg.me \
    --to=sagi@grimberg.me \
    --cc=Chaitanya.Kulkarni@wdc.com \
    --cc=hch@lst.de \
    --cc=linux-nvme@lists.infradead.org \
    --subject='Re: [RFC PATCH 0/2] nvmet: add polling support' \
    /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

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).