Followup io_uring fixes for 5.12-rc
Jens Axboe
maintainerHi Linus,
A collection of later fixes that we should get into this release:
- Series of submission cleanups (Pavel)
- A few fixes for issues from earlier this merge window (Pavel, me)
- IOPOLL resubmission fix
- task_work locking fix (Hao)
Please pull!
The following changes since commit 0b81e80c813f92520667c872d499a2dba8377be6:
io_uring: tctx->task_lock should be IRQ safe (2021-02-16 11:11:20 -0700)
are available in the Git repository at:
git://git.kernel.dk/linux-block.git tags/for-5.12/io_uring-2021-02-25
for you to fetch changes up to cb5e1b81304e089ee3ca948db4d29f71902eb575:
Revert "io_uring: wait potential ->release() on resurrect" (2021-02-25 07:37:35 -0700)
----------------------------------------------------------------
for-5.12/io_uring-2021-02-25
----------------------------------------------------------------
Hao Xu (1):
io_uring: don't hold uring_lock when calling io_run_task_work*
Jens Axboe (3):
io_uring: make the !CONFIG_NET helpers a bit more robust
io_uring: don't attempt IO reissue from the ring exit path Revert "io_uring: wait potential ->release() on resurrect"
Pavel Begunkov (21):
io_uring: fix read memory leak
io_uring: kill fictitious submit iteration index
io_uring: keep io_*_prep() naming consistent
io_uring: don't duplicate ->file check in sfr
io_uring: move io_init_req()'s definition
io_uring: move io_init_req() into io_submit_sqe()
io_uring: move req link into submit_state
io_uring: don't submit link on error
io_uring: split sqe-prep and async setup
io_uring: do io_*_prep() early in io_submit_sqe()
io_uring: don't do async setup for links' heads
io_uring: fail links more in io_submit_sqe()
io_uring: don't take uring_lock during iowq cancel
io_uring: fail io-wq submission from a task_work
io_uring: zero ref_node after killing it
io_uring: keep generic rsrc infra generic
io_uring: wait potential ->release() on resurrect
io_uring: fix leaving invalid req->flags
io_uring: run task_work on io_uring_register()
io_uring: clear request count when freeing caches
io_uring: fix locked_free_list caches_free()
fs/io_uring.c | 686 +++++++++++++++++++++++++++++----------------------------- 1 file changed, 346 insertions(+), 340 deletions(-)
Signature · 1 line
Jens Axboe
Quoted 2 lines · Jens Axboe
A collection of later fixes that we should get into this release:
So you decided to not try to push the "user threads" rewrite for 5.12?
Probably just as well.
Linus
Jens Axboe
maintainerQuoted 7 lines · Linus Torvalds
Quoted 2 lines · Jens Axboe
A collection of later fixes that we should get into this release:
So you decided to not try to push the "user threads" rewrite for 5.12?
Probably just as well.
Heh, I actually did decide to push it, just wanted to flush this out first... I have it all ready to go, it's based on (a few patches back) of this branch. I'll still send this out later today, it's totally up to you of course, and the email has a lengthier explanation.
There's a few minor issues around referencing since we get auto-exit of threads, where we managed it fully before with kthreads. But nothing really major, and we have it mostly flushed out already. So those followup fixes should be ready to go in a day or so. It's what syzbot is tripping up on, and they are known and mostly edge cases.
Signature · 1 line
Jens Axboe
The pull request you sent on Thu, 25 Feb 2021 15:27:00 -0700:
The pull request you sent on Thu, 25 Feb 2021 15:27:00 -0700:
Quoted 1 line
git://git.kernel.dk/linux-block.gittags/for-5.12/io_uring-2021-02-25
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/efba6d3a7c4bb59f0750609fae0f9644d82304b6
Thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/prtracker.html