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], [email protected],
Kanchan Joshi <[email protected]>
Subject: [PATCH for-next 0/4] nvme-multipathing for uring-passthrough
Date: Mon, 11 Jul 2022 16:31:51 +0530 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: CGME20220711110753epcas5p4169b9e288d15ca35740dbb66a6f6983a@epcas5p4.samsung.com
nvme passthrough lacks multipathing capability and some of us have
already expressed interest to see this plumbed. Most recently during LSFMM,
around 2 months back.
This series wires up multipathing for uring-passthrough commands.
Attempt is not to affect the common path (i.e. when
requeue/failover does not trigger) with allocation or deferral. The
most important design bit is to treat "struct io_uring_cmd" in the same
way as "struct bio" is treated by the block-based nvme multipath.
Uring-commands are queued when path is not available, and resubmitted on
discovery of new path. Also if passthrough command on multipath-node is
failed, it is resubmitted on a different path.
Testing:
Using the upstream fio that support uring-passthrough:
fio -iodepth=16 -rw=randread -ioengine=io_uring_cmd -bs=4k -numjobs=4
-size=1G -iodepth_batch_submit=16 -group_reporting -cmd_type=nvme
-filename=/dev/ng0n1 -name=uring-pt
1. Multiple failover - every command is retried 1-5 times before completion
2. Fail nvme_find_path() - this tests completion post requeue
3. Combine above two
4. Repeat above but for passthrough commands which do not generate bio
(e.g. flush command)
Anuj Gupta (2):
nvme: compact nvme_uring_cmd_pdu struct
nvme-multipath: add multipathing for uring-passthrough commands
Kanchan Joshi (2):
io_uring, nvme: rename a function
io_uring: grow a field in struct io_uring_cmd
drivers/nvme/host/ioctl.c | 157 +++++++++++++++++++++++++++++-----
drivers/nvme/host/multipath.c | 36 +++++++-
drivers/nvme/host/nvme.h | 26 ++++++
include/linux/io_uring.h | 44 +++++++++-
io_uring/uring_cmd.c | 4 +-
5 files changed, 237 insertions(+), 30 deletions(-)
--
2.25.1
next parent reply other threads:[~2022-07-11 11:28 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CGME20220711110753epcas5p4169b9e288d15ca35740dbb66a6f6983a@epcas5p4.samsung.com>
2022-07-11 11:01 ` Kanchan Joshi [this message]
[not found] ` <CGME20220711110800epcas5p3d338dd486fd778c5ba5bfe93a91ec8bd@epcas5p3.samsung.com>
2022-07-11 11:01 ` [PATCH for-next 1/4] io_uring, nvme: rename a function Kanchan Joshi
[not found] ` <CGME20220711110812epcas5p33aa90b23aa62fb11722aa8195754becf@epcas5p3.samsung.com>
2022-07-11 11:01 ` [PATCH for-next 2/4] nvme: compact nvme_uring_cmd_pdu struct Kanchan Joshi
[not found] ` <CGME20220711110824epcas5p22c8e945cb8c3c3ac46c8c2b5ab55db9b@epcas5p2.samsung.com>
2022-07-11 11:01 ` [PATCH for-next 3/4] io_uring: grow a field in struct io_uring_cmd Kanchan Joshi
[not found] ` <CGME20220711110827epcas5p3fd81f142f55ca3048abc38a9ef0d0089@epcas5p3.samsung.com>
2022-07-11 11:01 ` [PATCH for-next 4/4] nvme-multipath: add multipathing for uring-passthrough commands Kanchan Joshi
2022-07-11 13:51 ` Sagi Grimberg
2022-07-11 15:12 ` Stefan Metzmacher
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] \
[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