public inbox for [email protected]
 help / color / mirror / Atom feed
From: Jens Axboe <[email protected]>
To: Linus Torvalds <[email protected]>
Cc: io-uring <[email protected]>
Subject: [GIT PULL] io_uring fixes for 6.8-rc3
Date: Fri, 2 Feb 2024 08:13:38 -0700	[thread overview]
Message-ID: <[email protected]> (raw)

Hi Linus,

Fixes that should go into the 6.8 kernel release, and targeted for
stable as well. There are three parts here:

- Fix for missing retry for read multishot. If we trigger the execution
  of it and there's more than one buffer to be read, then we don't
  always read more than the first one. As it's edge triggered, this can
  lead to stalls.

- Limit inline receive multishot retries for fairness reasons. If we
  have a very bursty socket receiving data, we still need to ensure we
  process other requests as well. This is really two minor cleanups,
  then adding a way for poll reissue to trigger a requeue, and then
  finally having multishot receive utilize that.

- Fix for a weird corner case for non-multishot receive with MSG_WAITALL,
  using provided buffers, and setting the length to zero (to let the
  buffer dictate the receive size).

Please pull!


The following changes since commit 16bae3e1377846734ec6b87eee459c0f3551692c:

  io_uring: enable audit and restrict cred override for IORING_OP_FIXED_FD_INSTALL (2024-01-23 15:25:14 -0700)

are available in the Git repository at:

  git://git.kernel.dk/linux.git tags/io_uring-6.8-2024-02-01

for you to fetch changes up to 72bd80252feeb3bef8724230ee15d9f7ab541c6e:

  io_uring/net: fix sr->len for IORING_OP_RECV with MSG_WAITALL and buffers (2024-02-01 06:42:36 -0700)

----------------------------------------------------------------
io_uring-6.8-2024-02-01

----------------------------------------------------------------
Jens Axboe (6):
      io_uring/rw: ensure poll based multishot read retries appropriately
      io_uring/poll: move poll execution helpers higher up
      io_uring/net: un-indent mshot retry path in io_recv_finish()
      io_uring/poll: add requeue return code from poll multishot handling
      io_uring/net: limit inline multishot retries
      io_uring/net: fix sr->len for IORING_OP_RECV with MSG_WAITALL and buffers

 io_uring/io_uring.h |  8 +++++++-
 io_uring/net.c      | 50 ++++++++++++++++++++++++++++++++++++--------------
 io_uring/poll.c     | 49 ++++++++++++++++++++++++++++---------------------
 io_uring/poll.h     |  9 +++++++++
 io_uring/rw.c       | 10 +++++++++-
 5 files changed, 89 insertions(+), 37 deletions(-)

-- 
Jens Axboe


             reply	other threads:[~2024-02-02 15:13 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-02-02 15:13 Jens Axboe [this message]
2024-02-02 20:43 ` [GIT PULL] io_uring fixes for 6.8-rc3 pr-tracker-bot

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] \
    /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