public inbox for [email protected]
 help / color / mirror / Atom feed
From: Kanchan Joshi <[email protected]>
To: [email protected], [email protected], [email protected],
	[email protected]
Cc: [email protected], [email protected],
	[email protected], [email protected],
	[email protected], [email protected],
	Kanchan Joshi <[email protected]>
Subject: [RFC PATCH v4 0/2] Async nvme passthrough via io_uring
Date: Thu, 25 Mar 2021 22:35:38 +0530	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: CGME20210325170659epcas5p314b54d01c60189f899f67aaeb9d87a13@epcas5p3.samsung.com

This series adds async passthrough capability for nvme block-dev over
io_uring.
The patches are on top of Jens uring-cmd series:
https://lore.kernel.org/linux-nvme/[email protected]/
https://git.kernel.dk/cgit/linux-block/log/?h=io_uring-fops.v4
The tree contains nvme-5.13 updates (pdu-enhancement patches of Keith)
as well.

Application is expected to allocate passthrough command structure, set
it up traditionally, and pass its address via
"block_uring_cmd->unused2[0]" field.
On completion, CQE is posted with completion-status after any ioctl
specific buffer/field update.

Tests are done by transforming fio io_uring read/write into
passthrough:
https://github.com/joshkan/fio/tree/uring_cmd_nvme_v4

Changes from v3:
1. Moved to v4 branch of Jens, adapted to interface changes
2. Extended support for NVME_IOCTL_IO_CMD64
3. Applied nvme feedback - hch, Keith
4. Appiled io_uring feedback - Jens, Stefan

Changes from v2:
1. Rebase against latest uring-cmd branch of Jens
2. Remove per-io nvme_command allocation
3. Disallow passthrough commands with non-zero command effects

Change from v1:
1. Rewire the work on top of Jens uring-cmd interface
2. Support only passthrough, and not other nvme ioctls

Kanchan Joshi (2):
  io_uring: add helpers for io_uring_cmd completion in submitter-task.
  nvme: wire up support for async passthrough

 drivers/nvme/host/core.c | 194 +++++++++++++++++++++++++++++++++------
 drivers/nvme/host/nvme.h |   3 +
 drivers/nvme/host/pci.c  |   1 +
 fs/io_uring.c            |  23 +++++
 include/linux/io_uring.h |  12 +++
 5 files changed, 206 insertions(+), 27 deletions(-)

-- 
2.25.1


       reply	other threads:[~2021-03-25 17:07 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20210325170659epcas5p314b54d01c60189f899f67aaeb9d87a13@epcas5p3.samsung.com>
2021-03-25 17:05 ` Kanchan Joshi [this message]
     [not found]   ` <CGME20210325170704epcas5p3aafad4845b9ea1a545d643121a0ee1e5@epcas5p3.samsung.com>
2021-03-25 17:05     ` [RFC PATCH v4 1/2] io_uring: add helpers for io_uring_cmd completion in submitter-task Kanchan Joshi
     [not found]   ` <CGME20210325170708epcas5p259755feb9f0a769e7390a3b6eebc0e01@epcas5p2.samsung.com>
2021-03-25 17:05     ` [RFC PATCH v4 2/2] nvme: wire up support for async passthrough Kanchan Joshi

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 \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /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