From: Ming Lei <ming.lei@redhat.com>
To: Yoav Cohen <yoav@nvidia.com>
Cc: Jens Axboe <axboe@kernel.dk>,
"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
Uday Shankar <ushankar@purestorage.com>,
Caleb Sander Mateos <csander@purestorage.com>
Subject: Re: [PATCH 2/3] ublk: add feature UBLK_F_QUIESCE
Date: Thu, 12 Jun 2025 17:15:09 +0800 [thread overview]
Message-ID: <aEqanYVjHgVOjcwR@fedora> (raw)
In-Reply-To: <DM4PR12MB63280BC70C29A91E973E8080A974A@DM4PR12MB6328.namprd12.prod.outlook.com>
On Thu, Jun 12, 2025 at 08:17:49AM +0000, Yoav Cohen wrote:
> Hi Ming,
>
> Thank you very much, I managed to integrate the feature to our application and it seems to work perfectly fine during our update tests.
> Just a double check: when UBLK_F_USER_RECOVERY & UBLK_F_USER_RECOVERY_REISSUE
> and QUIECE_DEV was called - does any IO that will be completed using COMMIT_AND_FETCH with a failure (i.e result=-EIO) will be retry after the recovery stage?
>
UBLK_F_USER_RECOVERY_REISSUE supposes all inflight IOs are failed, so
these IOs will be re-delivered to ublk server after recovering to LIVE.
So you needn't to complete the IOs with -EIO for retrying them during
recovery, in short:
- if ublk server handles inflight IOs, complete them by sending COMMIT_AND_FETCH
with result before closing '/dev/ublkcN', and these IOs will not be re-issued
by driver after recovery
- otherwise, just ignore & discard inflight IOs, they all will be
re-issued by driver after recovery via UBLK_F_USER_RECOVERY_REISSUE.
Thanks,
Ming
next prev parent reply other threads:[~2025-06-12 9:15 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-22 16:35 [PATCH 0/3] ublk: add UBLK_F_QUIESCE Ming Lei
2025-05-22 16:35 ` [PATCH 1/3] selftests: ublk: add test case for UBLK_U_CMD_UPDATE_SIZE Ming Lei
2025-05-22 16:35 ` [PATCH 2/3] ublk: add feature UBLK_F_QUIESCE Ming Lei
2025-06-05 12:37 ` Yoav Cohen
2025-06-06 9:54 ` Ming Lei
2025-06-08 10:20 ` Yoav Cohen
2025-06-08 14:34 ` Ming Lei
2025-06-12 8:17 ` Yoav Cohen
2025-06-12 9:15 ` Ming Lei [this message]
2025-06-12 12:04 ` Yoav Cohen
2025-06-23 2:13 ` Ming Lei
2025-05-22 16:35 ` [PATCH 3/3] selftests: ublk: add test for UBLK_F_QUIESCE Ming Lei
2025-05-23 15:42 ` [PATCH 0/3] ublk: add UBLK_F_QUIESCE 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=aEqanYVjHgVOjcwR@fedora \
--to=ming.lei@redhat.com \
--cc=axboe@kernel.dk \
--cc=csander@purestorage.com \
--cc=linux-block@vger.kernel.org \
--cc=ushankar@purestorage.com \
--cc=yoav@nvidia.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.