From: Jens Axboe <axboe@kernel.dk>
To: io-uring@vger.kernel.org
Cc: juanlu@fastmail.com
Subject: [PATCHSET v2] Cancel requests at ring close time
Date: Fri, 11 Sep 2026 09:45:24 -0600 [thread overview]
Message-ID: <20260911154811.646705-1-axboe@kernel.dk> (raw)
Hi,
For v1, see here:
https://lore.kernel.org/io-uring/20260909141010.21064-1-axboe@kernel.dk/
And for the current tree, go here:
https://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux.git/log/?h=io_uring-exit-cancel.7
Changes since v1:
- Reorder the series so the fixes for the current tree come first,
then the exit/release cancelation changes on top
- Add fix for io-wq completions that finish asynchronously (eg a
direct read completing via ->ki_complete), ensuring CQE posting and
file putting are sane.
- Ditto for IOPOLL completions reaped from io_do_iopoll()
- Put request files synchronously before posting the CQEs in the batch
flush, fixing an issue with SQPOLL.
- Don't clear REQ_F_REFCOUNT from the completion path, a linked timeout
can still take a reference on the request until it's disarmed
--
Jens Axboe
next reply other threads:[~2026-09-11 15:48 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-11 15:45 Jens Axboe [this message]
2026-09-11 15:45 ` [PATCH 01/10] io_uring/io-wq: put the request file before posting a completion Jens Axboe
2026-09-11 15:45 ` [PATCH 02/10] io_uring: post io-wq completions from the last request reference Jens Axboe
2026-09-11 15:45 ` [PATCH 03/10] io_uring/rw: don't reap io-wq IOPOLL completions while io-wq has a reference Jens Axboe
2026-09-11 15:45 ` [PATCH 04/10] io_uring: put request files before posting the completions Jens Axboe
2026-09-11 15:45 ` [PATCH 05/10] io_uring/uring_cmd: only cancel requests of the given task Jens Axboe
2026-09-11 15:45 ` [PATCH 06/10] io_uring/notif: count pending zerocopy notifications per ring Jens Axboe
2026-09-11 15:45 ` [PATCH 07/10] io_uring/cancel: cancel and wait for all requests on process exit Jens Axboe
2026-09-11 15:45 ` [PATCH 08/10] io_uring: run cancelations synchronously on ring release Jens Axboe
2026-09-11 15:45 ` [PATCH 09/10] io_uring: drop registered files and buffers at release time Jens Axboe
2026-09-11 15:45 ` [PATCH 10/10] io_uring: wait for in-flight requests on ring release 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=20260911154811.646705-1-axboe@kernel.dk \
--to=axboe@kernel.dk \
--cc=io-uring@vger.kernel.org \
--cc=juanlu@fastmail.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.