[GIT PULL] io_uring fixes for 5.12-rc6

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Linus,

Nothing really major in here, and finally nothing really related to
signals. A few minor fixups related to the threading changes, and some
general fixes, that's it.

There's the pending gdb-get-confused-about-arch, but that's more of a
cosmetic issue, nothing that hinder use of it. And given that other
archs will likely be affected by that oddity too, better to postpone any
changes there until 5.13 imho.

This pull request contains:

- Combine signal checking for SQPOLL with parking, as we need to have
  the sqd lock dropped for both cases.

- S_ISBLK read/writes are bounded work, they were mistakenly bundled as
  unbounded.

- Move IO reissue for reg/blk into the read/write part itself, making it
  just be like -EAGAIN retries.

- Cancelation tightening

- Fix for a silly issue introduced with a prior fix this merge window,
  where a failed ring setup could lead to a crash.

- Ensure that we don't overflow ->comm with 10 billion PIDs

- Fix for an iov_iter revert issue earlier in this release.

Please pull!


The following changes since commit 2b8ed1c94182dbbd0163d0eb443a934cbf6b0d85:

  io_uring: remove unsued assignment to pointer io (2021-03-27 14:09:11 -0600)

are available in the Git repository at:

  git://git.kernel.dk/linux-block.git tags/io_uring-5.12-2021-04-02

for you to fetch changes up to 230d50d448acb6639991440913299e50cacf1daf:

  io_uring: move reissue into regular IO path (2021-04-02 09:24:20 -0600)

----------------------------------------------------------------
io_uring-5.12-2021-04-02

----------------------------------------------------------------
Jens Axboe (3):
      io_uring: drop sqd lock before handling signals for SQPOLL
      io_uring: don't mark S_ISBLK async work as unbounded
      io_uring: move reissue into regular IO path

Pavel Begunkov (4):
      io_uring: always go for cancellation spin on exec
      io_uring: handle setup-failed ctx in kill_timeouts
      io_uring/io-wq: protect against sprintf overflow
      io_uring: fix EIOCBQUEUED iter revert

 fs/io-wq.c    |  4 ++--
 fs/io_uring.c | 50 ++++++++++++++++++++++++++++++--------------------
 2 files changed, 32 insertions(+), 22 deletions(-)

-- 
Jens Axboe




[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux