From: Pavel Begunkov <asml.silence@gmail.com>
To: Jens Axboe <axboe@kernel.dk>, io-uring@vger.kernel.org
Subject: [PATCH v3 5.12 0/6] remove task file notes
Date: Fri, 5 Mar 2021 12:58:35 +0000 [thread overview]
Message-ID: <cover.1614942979.git.asml.silence@gmail.com> (raw)
Introduce a mapping from ctx to all tctx, and using that removes
file notes, i.e. taking a io_uring file note previously stored in
task->io_uring->xa. It's needed because we don't free io_uring ctx
until all submitters die/exec, and it became worse after killing
->flush(). There are rough corner in a form of not behaving nicely,
I'll address in follow-up patches.
The torture is as simple as below. It will get OOM in no time. Also,
I plan to use it to fix recently broken cancellations.
while (1) {
assert(!io_uring_queue_init(8, &ring, 0));
io_uring_queue_exit(&ring);
}
v2: rebase (resolve conflicts)
drop taken 2 patches
v3: use jiffies in 6/6 (Jens)
Pavel Begunkov (6):
io_uring: make del_task_file more forgiving
io_uring: introduce ctx to tctx back map
io_uring: do ctx initiated file note removal
io_uring: don't take task ring-file notes
io_uring: index io_uring->xa by ctx not file
io_uring: warn when ring exit takes too long
fs/io_uring.c | 131 +++++++++++++++++++++++++++++++--------
include/linux/io_uring.h | 2 +-
2 files changed, 106 insertions(+), 27 deletions(-)
--
2.24.0
next reply other threads:[~2021-03-05 13:03 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-05 12:58 Pavel Begunkov [this message]
2021-03-05 12:58 ` [PATCH v3 1/6] io_uring: make del_task_file more forgiving Pavel Begunkov
2021-03-05 12:58 ` [PATCH v3 2/6] io_uring: introduce ctx to tctx back map Pavel Begunkov
2021-03-05 12:58 ` [PATCH v3 3/6] io_uring: do ctx initiated file note removal Pavel Begunkov
2021-03-05 12:58 ` [PATCH v3 4/6] io_uring: don't take task ring-file notes Pavel Begunkov
2021-03-05 12:58 ` [PATCH v3 5/6] io_uring: index io_uring->xa by ctx not file Pavel Begunkov
2021-03-05 12:58 ` [PATCH v3 6/6] io_uring: warn when ring exit takes too long Pavel Begunkov
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=cover.1614942979.git.asml.silence@gmail.com \
--to=asml.silence@gmail.com \
--cc=axboe@kernel.dk \
--cc=io-uring@vger.kernel.org \
/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.