Get rid of the need to do re-expand and revert on an iterator when we
encounter a short IO, or failure that warrants a retry. Use the new
state save/restore helpers instead.
We keep the iov_iter_state persistent across retries, if we need to
restart the read or write operation. If there's a pending retry, the
operation will always exit with the state correctly saved.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A common complaint is that using O_NONBLOCK files with io_uring can be a
bit of a pain. Be a bit nicer and allow normal retry IFF the file does
support async behavior. This makes it possible to use io_uring more
reliably with O_NONBLOCK files, for use cases where it either isn't
possible or feasible to modify the file flags.
Cc: stable@vger.kernel.org
Reported-and-tested-by: Dan Melnic <dmm@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
It might be inconvenient that direct open/accept deviates from the
update semantics and fails if the slot is taken instead of removing a
file sitting there. Implement this auto-removal.
Note that removal might need to allocate and so may fail. However, if an
empty slot is specified, it's guaraneed to not fail on the fd
installation side for valid userspace programs. It's needed for users
who can't tolerate such failures, e.g. accept where the other end
never retries.
Suggested-by: Franz-B. Tuneke <franz-bernhard.tuneke@tu-dortmund.de>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/c896f14ea46b0eaa6c09d93149e665c2c37979b4.1631632300.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move get_timespec() section in io_cqring_wait() before the sigmask
saving, otherwise we'll fail to restore sigmask once get_timespec()
returns error.
Fixes: c73ebb685f ("io_uring: add timeout support for io_uring_enter()")
Signed-off-by: Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
Link: https://lore.kernel.org/r/20210914143852.9663-1-xiaoguang.wang@linux.alibaba.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We need to re-check sqd->thread after we've dropped the lock. Pin
the sqd before doing the lockdep lock dance, and check if the thread
is alive after that. It's either NULL or alive, as the SQPOLL thread
cannot exit without holding the same sqd->lock.
Reported-and-tested-by: syzbot+337de45f13a4fd54d708@syzkaller.appspotmail.com
Fixes: fa84693b3c ("io_uring: ensure IORING_REGISTER_IOWQ_MAX_WORKERS works with SQPOLL")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When setting up the next segment, we check what type the iter is and
handle it accordingly. However, when incrementing and processed amount
we do not, and both iter advance and addr/len are adjusted, regardless
of type. Split the increment side just like we do on the setup side.
Fixes: 4017eb91a9 ("io_uring: make loop_rw_iter() use original user supplied pointers")
Cc: stable@vger.kernel.org
Reported-by: Valentina Palmiotti <vpalmiotti@gmail.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
-----BEGIN PGP SIGNATURE-----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=aNr5
-----END PGP SIGNATURE-----
Merge tag 'io_uring-5.15-2021-09-11' of git://git.kernel.dk/linux-block
Pull io_uring fixes from Jens Axboe:
- Fix an off-by-one in a BUILD_BUG_ON() check. Not a real issue right
now as we have plenty of flags left, but could become one. (Hao)
- Fix lockdep issue introduced in this merge window (me)
- Fix a few issues with the worker creation (me, Pavel, Qiang)
- Fix regression with wq_has_sleeper() for IOPOLL (Pavel)
- Timeout link error propagation fix (Pavel)
* tag 'io_uring-5.15-2021-09-11' of git://git.kernel.dk/linux-block:
io_uring: fix off-by-one in BUILD_BUG_ON check of __REQ_F_LAST_BIT
io_uring: fail links of cancelled timeouts
io-wq: fix memory leak in create_io_worker()
io-wq: fix silly logic error in io_task_work_match()
io_uring: drop ctx->uring_lock before acquiring sqd->lock
io_uring: fix missing mb() before waitqueue_active
io-wq: fix cancellation on create-worker failure
Build check of __REQ_F_LAST_BIT should be larger than, not equal or larger
than. It's perfectly valid to have __REQ_F_LAST_BIT be 32, as that means
that the last valid bit is 31 which does fit in the type.
Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
Link: https://lore.kernel.org/r/20210907032243.114190-1-haoxu@linux.alibaba.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull iov_iter fixes from Al Viro:
"Fixes for io-uring handling of iov_iter reexpands"
* 'work.iov_iter' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
io_uring: reexpand under-reexpanded iters
iov_iter: track truncated size
The SQPOLL thread dictates the lock order, and we hold the ctx->uring_lock
for all the registration opcodes. We also hold a ref to the ctx, and we
do drop the lock for other reasons to quiesce, so it's fine to drop the
ctx lock temporarily to grab the sqd->lock. This fixes the following
lockdep splat:
======================================================
WARNING: possible circular locking dependency detected
5.14.0-syzkaller #0 Not tainted
------------------------------------------------------
syz-executor.5/25433 is trying to acquire lock:
ffff888023426870 (&sqd->lock){+.+.}-{3:3}, at: io_register_iowq_max_workers fs/io_uring.c:10551 [inline]
ffff888023426870 (&sqd->lock){+.+.}-{3:3}, at: __io_uring_register fs/io_uring.c:10757 [inline]
ffff888023426870 (&sqd->lock){+.+.}-{3:3}, at: __do_sys_io_uring_register+0x10aa/0x2e70 fs/io_uring.c:10792
but task is already holding lock:
ffff8880885b40a8 (&ctx->uring_lock){+.+.}-{3:3}, at: __do_sys_io_uring_register+0x2e1/0x2e70 fs/io_uring.c:10791
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (&ctx->uring_lock){+.+.}-{3:3}:
__mutex_lock_common kernel/locking/mutex.c:596 [inline]
__mutex_lock+0x131/0x12f0 kernel/locking/mutex.c:729
__io_sq_thread fs/io_uring.c:7291 [inline]
io_sq_thread+0x65a/0x1370 fs/io_uring.c:7368
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
-> #0 (&sqd->lock){+.+.}-{3:3}:
check_prev_add kernel/locking/lockdep.c:3051 [inline]
check_prevs_add kernel/locking/lockdep.c:3174 [inline]
validate_chain kernel/locking/lockdep.c:3789 [inline]
__lock_acquire+0x2a07/0x54a0 kernel/locking/lockdep.c:5015
lock_acquire kernel/locking/lockdep.c:5625 [inline]
lock_acquire+0x1ab/0x510 kernel/locking/lockdep.c:5590
__mutex_lock_common kernel/locking/mutex.c:596 [inline]
__mutex_lock+0x131/0x12f0 kernel/locking/mutex.c:729
io_register_iowq_max_workers fs/io_uring.c:10551 [inline]
__io_uring_register fs/io_uring.c:10757 [inline]
__do_sys_io_uring_register+0x10aa/0x2e70 fs/io_uring.c:10792
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(&ctx->uring_lock);
lock(&sqd->lock);
lock(&ctx->uring_lock);
lock(&sqd->lock);
*** DEADLOCK ***
Fixes: 2e480058dd ("io-wq: provide a way to limit max number of workers")
Reported-by: syzbot+97fa56483f69d677969f@syzkaller.appspotmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In case of !SQPOLL, io_cqring_ev_posted_iopoll() doesn't provide a
memory barrier required by waitqueue_active(&ctx->poll_wait). There is
a wq_has_sleeper(), which does smb_mb() inside, but it's called only for
SQPOLL.
Fixes: 5fd4617840 ("io_uring: be smarter about waking multiple CQ ring waiters")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/2982e53bcea2274006ed435ee2a77197107d8a29.1631130542.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
-----BEGIN PGP SIGNATURE-----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=QoQg
-----END PGP SIGNATURE-----
Merge tag 'for-5.15/io_uring-2021-09-04' of git://git.kernel.dk/linux-block
Pull io_uring fixes from Jens Axboe:
"As sometimes happens, two reports came in around the merge window open
that led to some fixes. Hence this one is a bit bigger than usual
followup fixes, but most of it will be going towards stable, outside
of the fixes that are addressing regressions from this merge window.
In detail:
- postgres is a heavy user of signals between tasks, and if we're
unlucky this can interfere with io-wq worker creation. Make sure
we're resilient against unrelated signal handling. This set of
changes also includes hardening against allocation failures, which
could previously had led to stalls.
- Some use cases that end up having a mix of bounded and unbounded
work would have starvation issues related to that. Split the
pending work lists to handle that better.
- Completion trace int -> unsigned -> long fix
- Fix issue with REGISTER_IOWQ_MAX_WORKERS and SQPOLL
- Fix regression with hash wait lock in this merge window
- Fix retry issued on block devices (Ming)
- Fix regression with links in this merge window (Pavel)
- Fix race with multi-shot poll and completions (Xiaoguang)
- Ensure regular file IO doesn't inadvertently skip completion
batching (Pavel)
- Ensure submissions are flushed after running task_work (Pavel)"
* tag 'for-5.15/io_uring-2021-09-04' of git://git.kernel.dk/linux-block:
io_uring: io_uring_complete() trace should take an integer
io_uring: fix possible poll event lost in multi shot mode
io_uring: prolong tctx_task_work() with flushing
io_uring: don't disable kiocb_done() CQE batching
io_uring: ensure IORING_REGISTER_IOWQ_MAX_WORKERS works with SQPOLL
io-wq: make worker creation resilient against signals
io-wq: get rid of FIXED worker flag
io-wq: only exit on fatal signals
io-wq: split bounded and unbounded work into separate lists
io-wq: fix queue stalling race
io_uring: don't submit half-prepared drain request
io_uring: fix queueing half-created requests
io-wq: ensure that hash wait lock is IRQ disabling
io_uring: retry in case of short read on block device
io_uring: IORING_OP_WRITE needs hash_reg_file set
io-wq: fix race between adding work and activating a free worker
[ 74.211232] BUG: KASAN: stack-out-of-bounds in iov_iter_revert+0x809/0x900
[ 74.212778] Read of size 8 at addr ffff888025dc78b8 by task
syz-executor.0/828
[ 74.214756] CPU: 0 PID: 828 Comm: syz-executor.0 Not tainted
5.14.0-rc3-next-20210730 #1
[ 74.216525] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996),
BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014
[ 74.219033] Call Trace:
[ 74.219683] dump_stack_lvl+0x8b/0xb3
[ 74.220706] print_address_description.constprop.0+0x1f/0x140
[ 74.224226] kasan_report.cold+0x7f/0x11b
[ 74.226085] iov_iter_revert+0x809/0x900
[ 74.227960] io_write+0x57d/0xe40
[ 74.232647] io_issue_sqe+0x4da/0x6a80
[ 74.242578] __io_queue_sqe+0x1ac/0xe60
[ 74.245358] io_submit_sqes+0x3f6e/0x76a0
[ 74.248207] __do_sys_io_uring_enter+0x90c/0x1a20
[ 74.257167] do_syscall_64+0x3b/0x90
[ 74.257984] entry_SYSCALL_64_after_hwframe+0x44/0xae
old_size = iov_iter_count();
...
iov_iter_revert(old_size - iov_iter_count());
If iov_iter_revert() is done base on the initial size as above, and the
iter is truncated and not reexpanded in the middle, it miscalculates
borders causing problems. This trace is due to no one reexpanding after
generic_write_checks().
Now iters store how many bytes has been truncated, so reexpand them to
the initial state right before reverting.
Cc: stable@vger.kernel.org
Reported-by: Palash Oswal <oswalpalash@gmail.com>
Reported-by: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Reported-and-tested-by: syzbot+9671693590ef5aad8953@syzkaller.appspotmail.com
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
IIUC, IORING_POLL_ADD_MULTI is similar to epoll's edge-triggered mode,
that means once one pure poll request returns one event(cqe), we'll
need to read or write continually until EAGAIN is returned, then I think
there is a possible poll event lost race in multi shot mode:
t1 poll request add | |
t2 | |
t3 event happens | |
t4 task work add | |
t5 | task work run |
t6 | commit one cqe |
t7 | | user app handles cqe
t8 | new event happen |
t9 | add back to waitqueue |
t10 |
After t6 but before t9, if new event happens, there'll be no wakeup
operation, and if user app has picked up this cqe in t7, read or write
until EAGAIN is returned. In t8, new event happens and will be lost,
though this race window maybe small.
To fix this possible race, add poll request back to waitqueue before
committing cqe.
Fixes: 88e41cf928 ("io_uring: add multishot mode for IORING_OP_POLL_ADD")
Signed-off-by: Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
Link: https://lore.kernel.org/r/20210903142436.5767-1-xiaoguang.wang@linux.alibaba.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
SQPOLL has a different thread doing submissions, we need to check for
that and use the right task context when updating the worker values.
Just hold the sqd->lock across the operation, this ensures that the
thread cannot go away while we poke at ->io_uring.
Link: https://github.com/axboe/liburing/issues/420
Fixes: 2e480058dd ("io-wq: provide a way to limit max number of workers")
Reported-by: Johannes Lundberg <johalun0@gmail.com>
Tested-by: Johannes Lundberg <johalun0@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In case of buffered reading from block device, when short read happens,
we should retry to read more, otherwise the IO will be completed
partially, for example, the following fio expects to read 2MB, but it
can only read 1M or less bytes:
fio --name=onessd --filename=/dev/nvme0n1 --filesize=2M \
--rw=randread --bs=2M --direct=0 --overwrite=0 --numjobs=1 \
--iodepth=1 --time_based=0 --runtime=2 --ioengine=io_uring \
--registerfiles --fixedbufs --gtod_reduce=1 --group_reporting
Fix the issue by allowing short read retry for block device, which sets
FMODE_BUF_RASYNC really.
Fixes: 9a173346bd ("io_uring: fix short read retries for non-reg files")
Cc: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/20210821150751.1290434-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
During some testing, it became evident that using IORING_OP_WRITE doesn't
hash buffered writes like the other writes commands do. That's simply
an oversight, and can cause performance regressions when doing buffered
writes with this command.
Correct that and add the flag, so that buffered writes are correctly
hashed when using the non-iovec based write command.
Cc: stable@vger.kernel.org
Fixes: 3a6820f2bb ("io_uring: add non-vectored read/write commands")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
-----BEGIN PGP SIGNATURE-----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=T7Zv
-----END PGP SIGNATURE-----
Merge tag 'for-5.15/io_uring-vfs-2021-08-30' of git://git.kernel.dk/linux-block
Pull io_uring mkdirat/symlinkat/linkat support from Jens Axboe:
"This adds io_uring support for mkdirat, symlinkat, and linkat"
* tag 'for-5.15/io_uring-vfs-2021-08-30' of git://git.kernel.dk/linux-block:
io_uring: add support for IORING_OP_LINKAT
io_uring: add support for IORING_OP_SYMLINKAT
io_uring: add support for IORING_OP_MKDIRAT
namei: update do_*() helpers to return ints
namei: make do_linkat() take struct filename
namei: add getname_uflags()
namei: make do_symlinkat() take struct filename
namei: make do_mknodat() take struct filename
namei: make do_mkdirat() take struct filename
namei: change filename_parentat() calling conventions
namei: ignore ERR/NULL names in putname()
-----BEGIN PGP SIGNATURE-----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=izfH
-----END PGP SIGNATURE-----
Merge tag 'io_uring-bio-cache.5-2021-08-30' of git://git.kernel.dk/linux-block
Pull support for struct bio recycling from Jens Axboe:
"This adds bio recycling support for polled IO, allowing quick reuse of
a bio for high IOPS scenarios via a percpu bio_set list.
It's good for almost a 10% improvement in performance, bumping our
per-core IO limit from ~3.2M IOPS to ~3.5M IOPS"
* tag 'io_uring-bio-cache.5-2021-08-30' of git://git.kernel.dk/linux-block:
bio: improve kerneldoc documentation for bio_alloc_kiocb()
block: provide bio_clear_hipri() helper
block: use the percpu bio cache in __blkdev_direct_IO
io_uring: enable use of bio alloc cache
block: clear BIO_PERCPU_CACHE flag if polling isn't supported
bio: add allocation cache abstraction
fs: add kiocb alloc cache flag
bio: optimize initialization of a bio
We allow updating normal timeouts, add support for adjusting timings of
linked timeouts as well.
Reported-by: Victor Stewart <v@nametag.social>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A preparation patch. Keep all queued linked timeout in a list, so they
may be found and updated.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Certain use cases want to use CLOCK_BOOTTIME or CLOCK_REALTIME rather than
CLOCK_MONOTONIC, instead of the default CLOCK_MONOTONIC.
Add an IORING_TIMEOUT_BOOTTIME and IORING_TIMEOUT_REALTIME flag that
allows timeouts and linked timeouts to use the selected clock source.
Only one clock source may be selected, and we -EINVAL the request if more
than one is given. If neither BOOTIME nor REALTIME are selected, the
previous default of MONOTONIC is used.
Link: https://github.com/axboe/liburing/issues/369
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io-wq divides work into two categories:
1) Work that completes in a bounded time, like reading from a regular file
or a block device. This type of work is limited based on the size of
the SQ ring.
2) Work that may never complete, we call this unbounded work. The amount
of workers here is just limited by RLIMIT_NPROC.
For various uses cases, it's handy to have the kernel limit the maximum
amount of pending workers for both categories. Provide a way to do with
with a new IORING_REGISTER_IOWQ_MAX_WORKERS operation.
IORING_REGISTER_IOWQ_MAX_WORKERS takes an array of two integers and sets
the max worker count to what is being passed in for each category. The
old values are returned into that same array. If 0 is being passed in for
either category, it simply returns the current value.
The value is capped at RLIMIT_NPROC. This actually isn't that important
as it's more of a hint, if we're exceeding the value then our attempt
to fork a new worker will fail. This happens naturally already if more
than one node is in the system, as these values are per-node internally
for io-wq.
Reported-by: Johannes Lundberg <johalun0@gmail.com>
Link: https://github.com/axboe/liburing/issues/420
Signed-off-by: Jens Axboe <axboe@kernel.dk>
req->buf_index is u16 and so we rely on registered buffers indexes
fitting into it. Add a build check, so when the upper limit for the
number of buffers is lifted we get a compliation fail but not lurking
problems.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/787e8e1a17cea51ca6301426b1c4c4887b8bd676.1629920396.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Given a linkchain like this:
req0(link_flag)-->req1(link_flag)-->...-->reqn(no link_flag)
There is a problem:
- if some intermediate linked req like req1 's submittion fails, reqs
after it won't be cancelled.
- sqpoll disabled: maybe it's ok since users can get the error info
of req1 and stop submitting the following sqes.
- sqpoll enabled: definitely a problem, the following sqes will be
submitted in the next round.
The solution is to refactor the code logic to:
- if a linked req's submittion fails, just mark it and the head(if it
exists) as REQ_F_FAIL. Leverage req->result to indicate whether it
is failed or cancelled.
- submit or fail the whole chain when we come to the end of it.
Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/20210827094609.36052-3-haoxu@linux.alibaba.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
It's not necessary to free the request back to slab when we fail to
get sqe, just move it to state->free_list.
Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/20210825175856.194299-1-haoxu@linux.alibaba.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
As done with open opcodes, allow accept to skip installing fd into
processes' file tables and put it directly into io_uring's fixed file
table. Same restrictions and design as for open.
Suggested-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/6d16163f376fac7ac26a656de6b42199143e9721.1629888991.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Instead of opening a file into a process's file table as usual and then
registering the fd within io_uring, some users may want to skip the
first step and place it directly into io_uring's fixed file table.
This patch adds such a capability for IORING_OP_OPENAT and
IORING_OP_OPENAT2.
The behaviour is controlled by setting sqe->file_index, where 0 implies
the old behaviour using normal file tables. If non-zero value is
specified, then it will behave as described and place the file into a
fixed file slot sqe->file_index - 1. A file table should be already
created, the slot should be valid and empty, otherwise the operation
will fail.
Keep the error codes consistent with IORING_OP_FILES_UPDATE, ENXIO and
EINVAL on inappropriate fixed tables, and return EBADF on collision with
already registered file.
Note: IOSQE_FIXED_FILE can't be used to switch between modes, because
accept takes a file, and it already uses the flag with a different
meaning.
Suggested-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/e9b33d1163286f51ea707f87d95bd596dada1e65.1629888991.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
IORING_OP_LINKAT behaves like linkat(2) and takes the same flags and
arguments.
In some internal places 'hardlink' is used instead of 'link' to avoid
confusion with the SQE links. Name 'link' conflicts with the existing
'link' member of io_kiocb.
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Suggested-by: Christian Brauner <christian.brauner@ubuntu.com>
Link: https://lore.kernel.org/io-uring/20210514145259.wtl4xcsp52woi6ab@wittgenstein/
Signed-off-by: Dmitry Kadashev <dkadashev@gmail.com>
Acked-by: Christian Brauner <christian.brauner@ubuntu.com>
Link: https://lore.kernel.org/r/20210708063447.3556403-12-dkadashev@gmail.com
[axboe: add splice_fd_in check]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Mark polled IO as being safe for dipping into the bio allocation
cache, in case the targeted bio_set has it enabled.
This brings an IOPOLL gen2 Optane QD=128 workload from ~3.2M IOPS to
~3.5M IOPS.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
WARNING: CPU: 1 PID: 5870 at fs/io_uring.c:5975 io_try_cancel_userdata+0x30f/0x540 fs/io_uring.c:5975
CPU: 0 PID: 5870 Comm: iou-wrk-5860 Not tainted 5.14.0-rc6-next-20210820-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:io_try_cancel_userdata+0x30f/0x540 fs/io_uring.c:5975
Call Trace:
io_async_cancel fs/io_uring.c:6014 [inline]
io_issue_sqe+0x22d5/0x65a0 fs/io_uring.c:6407
io_wq_submit_work+0x1dc/0x300 fs/io_uring.c:6511
io_worker_handle_work+0xa45/0x1840 fs/io-wq.c:533
io_wqe_worker+0x2cc/0xbb0 fs/io-wq.c:582
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
io_try_cancel_userdata() can be called from io_async_cancel() executing
in the io-wq context, so the warning fires, which is there to alert
anyone accessing task->io_uring->io_wq in a racy way. However,
io_wq_put_and_exit() always first waits for all threads to complete,
so the only detail left is to zero tctx->io_wq after the context is
removed.
note: one little assumption is that when IO_WQ_WORK_CANCEL, the executor
won't touch ->io_wq, because io_wq_destroy() might cancel left pending
requests in such a way.
Cc: stable@vger.kernel.org
Reported-by: syzbot+b0c9d1588ae92866515f@syzkaller.appspotmail.com
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/dfdd37a80cfa9ffd3e59538929c99cdd55d8699e.1629721757.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Employ inline completion logic for read/write completions done via
io_req_task_complete(). If ->uring_lock is contended, just do normal
request completion, but if not, make tctx_task_work() to grab the lock
and do batched inline completions in io_req_task_complete().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/94589c3ce69eaed86a21bb1ec696407a54fab1aa.1629286357.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Many task_work handlers either grab ->uring_lock, or may benefit from
having it. Move locking logic out of individual handlers to a lazy
approach controlled by tctx_task_work(), so we don't keep doing
tons of mutex lock/unlock.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/d6a34e147f2507a2f3e2fa1e38a9c541dcad3929.1629286357.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We've previously had an issue where overflow flush unconditionally calls
io_cqring_ev_posted() even if it didn't flush any events to the ring,
causing wake and eventfd increment where no new events are available.
Some applications don't like that, see commit b18032bb0a for details.
This came up in discussion for another patch recently, hence add a
comment detailing what the relationship between calling the events
posted helper and CQ ring entries is.
Link: https://lore.kernel.org/io-uring/77a44fce-c831-16a6-8e80-9aee77f496a2@kernel.dk/
Signed-off-by: Jens Axboe <axboe@kernel.dk>
coml_nr in ctx_flush_and_put() is not protected by uring_lock, this
may cause problems when accessing in parallel:
say coml_nr > 0
ctx_flush_and put other context
if (compl_nr) get mutex
coml_nr > 0
do flush
coml_nr = 0
release mutex
get mutex
do flush (*)
release mutex
in (*) place, we call io_cqring_ev_posted() and users likely get
no events there. To avoid spurious events, re-check the value when
under the lock.
Fixes: 2c32395d81 ("io_uring: fix __tctx_task_work() ctx race")
Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
Link: https://lore.kernel.org/r/20210820221954.61815-1-haoxu@linux.alibaba.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Now allocated rsrc table uses PAGE_SIZE as the size of 2nd-level, and
accessing this table relies on each level index from fixed TABLE_SHIFT
(12 - 3) in 4k page case. In order to correctly work in non-4k page,
define TABLE_SHIFT as non-fixed (PAGE_SHIFT - shift of data) for
2nd-level table entry number.
Signed-off-by: wangyangbo <wangyangbo@uniontech.com>
Link: https://lore.kernel.org/r/20210819055657.27327-1-wangyangbo@uniontech.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Now with IRQ completions done via IRQ, almost all requests freeing
are done from the context of submitter task, so it makes sense to
extend task_put optimisation from io_req_free_batch_finish() to cover
all the cases including task_work by moving it into io_put_task().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/824a7cbd745ddeee4a0f3ff85c558a24fd005872.1629302453.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We have two checks of task->flags & PF_EXITING left:
1) In io_req_task_submit(), which is called in task_work and hence always
in the context of the original task. That means that
req->task == current, and hence checking ->flags is totally fine.
2) In io_poll_rewait(), where we need to stop re-arming poll to prevent
it interfering with cancelation. This is only run from task_work as
well, and hence for this case too req->task == current.
Add a comment to both spots detailing that.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Move earlier the check for whether __io_queue_proc() tries to poll
already polled waitqueue, and do the same for the second poll entry, if
any. Shouldn't really matter, but at least it would have a more
predictable behaviour.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/8cb428cfe8ade0fd055859fabb878db8777d4c2f.1629228203.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We have io_req_complete_post() to post a CQE and put the request. It
takes care of all synchronisation and is more concise and efficent, so
replace all hancoded occurrences of
"lock; post CQE; unlock; + put_req()" with io_req_complete_post().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/2c83463458a613f9d870e5147eb134da2aa70779.1629228203.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Linked timeout handling during issuing is heavy, it adds extra
instructions and forces to save the next linked timeout before
io_issue_sqe().
Follwing the same reasoning as in refcounting patches, a request can't
be freed by the time it returns from io_issue_sqe(), so now we don't
need to do io_prep_linked_timeout() in advance, and it can be delayed to
colder paths optimising the generic path.
Also, it should also save quite a lot for requests with linked timeouts
and completed inline on timeout spinlocking + hrtimer_start() +
hrtimer_try_to_cancel() and so on.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/19bfc9a0d26c5c5f1e359f7650afe807ca8ef879.1628981736.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Tracking linked timeouts as infligh was needed to make sure that io-wq
is not destroyed by io_uring_cancel_generic() racing with
io_async_cancel_one() accessing it. Now, cancellations issued by linked
timeouts are done in the task context, so it's already synchronised.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e1b05cf47cb69df2305efdbee8cf7ba36f46c1a3.1628981736.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_req_free_batch() has a __must_hold annotation referencing a
request being passed in, but we're passing in the context.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We can merge two spin_unlock() operations to one since we removed some
code not long ago.
Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When doing cancellation, we use a parameter to indicate where it's from
do_exit or exec. So a boolean value is good enough for this, remove the
struct files* as it is not necessary.
Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
[axboe: fixup io_uring_files_cancel for !CONFIG_IO_URING]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
As submission references are gone, there is only one initial reference
left. Instead of actually doing atomic refcounting, add a flag
indicating whether we're going to take more refs or doing any other sync
magic. The flag should be set before the request may get used in
parallel.
Together with the previous patch it saves 2 refcount atomics per request
for IOPOLL and IRQ completions, and 1 atomic per req for inline
completions, with some exceptions. In particular, currently, there are
three cases, when the refcounting have to be enabled:
- Polling, including apoll. Because double poll entries takes a ref.
Might get relaxed in the near future.
- Link timeouts, enabled for both, the timeout and the request it's
bound to, because they work in-parallel and we need to synchronise
to cancel one of them on completion.
- When a request gets in io-wq, because it doesn't hold uring_lock and
we need guarantees of submission references.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/8b204b6c5f6643062270a1913d6d3a7f8f795fd9.1628705069.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Requests are by default given with two references, submission and
completion. Completion references are straightforward, they represent
request ownership and are put when a request is completed or so.
Submission references are a bit more trickier. They're needed when
io_issue_sqe() followed deep into the submission stack (e.g. in fs,
block, drivers, etc.), request may have given away for concurrent
execution or already completed, and the code unwinding back to
io_issue_sqe() may be accessing some pieces of our requests, e.g.
file or iov.
Now, we prevent such async/in-depth completions by pushing requests
through task_work. Punting to io-wq is also done through task_works,
apart from a couple of cases with a pretty well known context. So,
there're two cases:
1) io_issue_sqe() from the task context and protected by ->uring_lock.
Either requests return back to io_uring or handed to task_work, which
won't be executed because we're currently controlling that task. So,
we can be sure that requests are staying alive all the time and we don't
need submission references to pin them.
2) io_issue_sqe() from io-wq, which doesn't hold the mutex. The role of
submission reference is played by io-wq reference, which is put by
io_wq_submit_work(). Hence, it should be fine.
Considering that, we can carefully kill the submission reference.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/6b68f1c763229a590f2a27148aee77767a8d7750.1628705069.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We have no hard/soft IRQ users of this lock left, remove any IRQ
disabling/saving and restoring when grabbing this lock.
This is straight forward with no users entering with IRQs disabled
anymore, the only thing to look out for is the waitqueue poll head
lock which nests inside the completion lock. That needs IRQs disabled,
and hence we have to do that now instead of relying on the outer lock
doing so.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This is in preparation to making the completion lock work outside of
hard/soft IRQ context.
Add a timeout_lock to handle the ordering of timeout completions or
cancelations with the timeouts actually triggering.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For requests with non-fixed files, instead of grabbing just one
reference, we get by the number of left requests, so the following
requests using the same file can take it without atomics.
However, it's not all win. If there is one request in the middle
not using files or having a fixed file, we'll need to put back the left
references. Even worse if an application submits requests dealing with
different files, it will do a put for each new request, so doubling the
number of atomics needed. Also, even if not used, it's still takes some
cycles in the submission path.
If a file used many times, it rather makes sense to pre-register it, if
not, we may fall in the described pitfall. So, this optimisation is a
matter of use case. Go with the simpliest code-wise way, remove it.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
After recent fixes, tctx_task_work() always does proper spinlocking
before looking into ->task_list, so now we don't need atomics for
->task_state, replace it with non-atomic task_running using the critical
section.
Tide it up, combine two separate block with spinlocking, and always try
to splice in there, so we do less locking when new requests are arriving
during the function execution.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
[axboe: fix missing ->task_running reset on task_work_add() failure]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We cache all the reference to task + tctx, so if io_put_task() is
called by the corresponding task itself, we can save on atomics and
return the refs right back into the cache.
It's beneficial for all inline completions, and also iopolling, when
polling and submissions are done by the same task, including
SQPOLL|IOPOLL.
Note: io_uring_cancel_generic() can return refs to the cache as well,
so those should be flushed in the loop for tctx_inflight() to work
right.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/6fe9646b3cb70e46aca1f58426776e368c8926b3.1628471125.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If io_ring_exit_work() can't get it done in 5 minutes, something is
going very wrong, don't keep spinning at HZ / 20 rate, it doesn't help
and it may take much of CPU time if there is a lot of workers stuck as
such.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/9e2d1ca81d569f6bc628af1a42ff6663bff7ce9c.1628471125.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Turns out we always init struct io_wait_queue in io_cqring_wait(), even
if it's not used after, i.e. there are already enough of CQEs. And often
it's exactly what happens, for instance, requests may have been
completed inline, or in case of io_uring_enter(submit=N, wait=1).
It shows up in my profiler, so optimise it by delaying the struct init.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/6f1b81c60b947d165583dc333947869c3d85d037.1628471125.git.asml.silence@gmail.com
[axboe: fixed up for new cqring wait]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
IOPOLL users should care more about getting completions for requests
they submitted, but not in "device did/completed something". Currently,
io_do_iopoll() may return a positive number, which will instruct
io_iopoll_check() to break the loop and end the syscall, even if there
is not enough CQEs or none at all.
Don't return positive numbers, so io_iopoll_check() exits only when it
gets an actual error, need reschedule or got enough CQEs.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/641a88f751623b6758303b3171f0a4141f06726e.1628471125.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We prefer nornal task_works even if it would fail requests inside. Kill
a PF_EXITING check in io_req_task_work_add(), task_work_add() handles
well dying tasks, i.e. return error when can't enqueue due to late
stages of do_exit().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/fc14297e8441cd8f5d1743a2488cf0df09bf48ac.1628471125.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If we use fixed files, we can be sure (almost) that REQ_F_ISREG is set.
However, for non-reg files io_prep_rw() still will look into inode to
double check, and that's expensive and can be avoided.
The only caveat is that it only currently works with 64+ bit
architectures, see FFS_ISREG, so we should consider that.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/0a62780c491ca2522cd52db4ae3f16e03aafed0f.1628471125.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Optimise io_file_get() with registered files, which is in a hot path,
by inlining parts of the function. Saves a function call, and
inefficiencies of passing arguments, e.g. evaluating
(sqe_flags & IOSQE_FIXED_FILE).
It couldn't have been done before as compilers were refusing to inline
it because of the function size.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/52115cd6ce28f33bd0923149c0e6cb611084a0b1.1628471125.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Instead of hand-coded two-level tables for registered files, allocate
them with kvmalloc(). In many cases small enough tables are enough, and
so can be kmalloc()'ed removing an extra memory load and a bunch of bit
logic instructions from the hot path. If the table is larger, we trade
off all the pros with a TLB-assisted memory lookup.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/280421d3b48775dabab773006bb5588c7b2dabc0.1628471125.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently we only wake the first waiter, even if we have enough entries
posted to satisfy multiple waiters. Improve that situation so that
every waiter knows how much the CQ tail has to advance before they can
be safely woken up.
With this change, if we have N waiters each asking for 1 event and we get
4 completions, then we wake up 4 waiters. If we have N waiters asking
for 2 completions and we get 4 completions, then we wake up the first
two. Previously, only the first waiter would've been woken up.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We currently check for ret != 0 to indicate error, but '1' is a valid
return and just indicates that the allocation succeeded with a wrap.
Correct the check to be for < 0, like it was before the xarray
conversion.
Cc: stable@vger.kernel.org
Fixes: 61cf93700f ("io_uring: Convert personality_idr to XArray")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If an SQPOLL based ring is newly created and an application issues an
io_uring_enter(2) system call on it, then we can return a spurious
-EOWNERDEAD error. This happens because there's nothing to submit, and
if the caller doesn't specify any other action, the initial error
assignment of -EOWNERDEAD never gets overwritten. This causes us to
return it directly, even if it isn't valid.
Move the error assignment into the actual failure case instead.
Cc: stable@vger.kernel.org
Fixes: d9d05217cb ("io_uring: stop SQPOLL submit on creator's death")
Reported-by: Sherlock Holo sherlockya@gmail.com
Link: https://github.com/axboe/liburing/issues/413
Signed-off-by: Jens Axboe <axboe@kernel.dk>
__io_rsrc_put_work() might need ->uring_lock, so nobody should wait for
rsrc nodes holding the mutex. However, that's exactly what
io_ring_ctx_free() does with io_wait_rsrc_data().
Split it into rsrc wait + dealloc, and move the first one out of the
lock.
Cc: stable@vger.kernel.org
Fixes: b60c8dce33 ("io_uring: preparation for rsrc tagging")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/0130c5c2693468173ec1afab714e0885d2c9c363.1628559783.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ammar reports that he's seeing a lockdep splat on running test/rsrc_tags
from the regression suite:
======================================================
WARNING: possible circular locking dependency detected
5.14.0-rc3-bluetea-test-00249-gc7d102232649 #5 Tainted: G OE
------------------------------------------------------
kworker/2:4/2684 is trying to acquire lock:
ffff88814bb1c0a8 (&ctx->uring_lock){+.+.}-{3:3}, at: io_rsrc_put_work+0x13d/0x1a0
but task is already holding lock:
ffffc90001c6be70 ((work_completion)(&(&ctx->rsrc_put_work)->work)){+.+.}-{0:0}, at: process_one_work+0x1bc/0x530
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 ((work_completion)(&(&ctx->rsrc_put_work)->work)){+.+.}-{0:0}:
__flush_work+0x31b/0x490
io_rsrc_ref_quiesce.part.0.constprop.0+0x35/0xb0
__do_sys_io_uring_register+0x45b/0x1060
do_syscall_64+0x35/0xb0
entry_SYSCALL_64_after_hwframe+0x44/0xae
-> #0 (&ctx->uring_lock){+.+.}-{3:3}:
__lock_acquire+0x119a/0x1e10
lock_acquire+0xc8/0x2f0
__mutex_lock+0x86/0x740
io_rsrc_put_work+0x13d/0x1a0
process_one_work+0x236/0x530
worker_thread+0x52/0x3b0
kthread+0x135/0x160
ret_from_fork+0x1f/0x30
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock((work_completion)(&(&ctx->rsrc_put_work)->work));
lock(&ctx->uring_lock);
lock((work_completion)(&(&ctx->rsrc_put_work)->work));
lock(&ctx->uring_lock);
*** DEADLOCK ***
2 locks held by kworker/2:4/2684:
#0: ffff88810004d938 ((wq_completion)events){+.+.}-{0:0}, at: process_one_work+0x1bc/0x530
#1: ffffc90001c6be70 ((work_completion)(&(&ctx->rsrc_put_work)->work)){+.+.}-{0:0}, at: process_one_work+0x1bc/0x530
stack backtrace:
CPU: 2 PID: 2684 Comm: kworker/2:4 Tainted: G OE 5.14.0-rc3-bluetea-test-00249-gc7d102232649 #5
Hardware name: Acer Aspire ES1-421/OLVIA_BE, BIOS V1.05 07/02/2015
Workqueue: events io_rsrc_put_work
Call Trace:
dump_stack_lvl+0x6a/0x9a
check_noncircular+0xfe/0x110
__lock_acquire+0x119a/0x1e10
lock_acquire+0xc8/0x2f0
? io_rsrc_put_work+0x13d/0x1a0
__mutex_lock+0x86/0x740
? io_rsrc_put_work+0x13d/0x1a0
? io_rsrc_put_work+0x13d/0x1a0
? io_rsrc_put_work+0x13d/0x1a0
? process_one_work+0x1ce/0x530
io_rsrc_put_work+0x13d/0x1a0
process_one_work+0x236/0x530
worker_thread+0x52/0x3b0
? process_one_work+0x530/0x530
kthread+0x135/0x160
? set_kthread_struct+0x40/0x40
ret_from_fork+0x1f/0x30
which is due to holding the ctx->uring_lock when flushing existing
pending work, while the pending work flushing may need to grab the uring
lock if we're using IOPOLL.
Fix this by dropping the uring_lock a bit earlier as part of the flush.
Cc: stable@vger.kernel.org
Link: https://github.com/axboe/liburing/issues/404
Tested-by: Ammar Faizi <ammarfaizi2@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The compiler should be forbidden from any strange optimization for async
writes to user visible data-structures. Without proper protection, the
compiler can cause write-tearing or invent writes that would confuse the
userspace.
However, there are writes to sq_flags which are not protected by
WRITE_ONCE(). Use WRITE_ONCE() for these writes.
This is purely a theoretical issue. Presumably, any compiler is very
unlikely to do such optimizations.
Fixes: 75b28affdd ("io_uring: allocate the two rings together")
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Nadav Amit <namit@vmware.com>
Link: https://lore.kernel.org/r/20210808001342.964634-3-namit@vmware.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When using SQPOLL, the submission queue polling thread calls
task_work_run() to run queued work. However, when work is added with
TWA_SIGNAL - as done by io_uring itself - the TIF_NOTIFY_SIGNAL remains
set afterwards and is never cleared.
Consequently, when the submission queue polling thread checks whether
signal_pending(), it may always find a pending signal, if
task_work_add() was ever called before.
The impact of this bug might be different on different kernel versions.
It appears that on 5.14 it would only cause unnecessary calculation and
prevent the polling thread from sleeping. On 5.13, where the bug was
found, it stops the polling thread from finding newly submitted work.
Instead of task_work_run(), use tracehook_notify_signal() that clears
TIF_NOTIFY_SIGNAL. Test for TIF_NOTIFY_SIGNAL in addition to
current->task_works to avoid a race in which task_works is cleared but
the TIF_NOTIFY_SIGNAL is set.
Fixes: 685fe7feed ("io-wq: eliminate the need for a manager thread")
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Nadav Amit <namit@vmware.com>
Link: https://lore.kernel.org/r/20210808001342.964634-2-namit@vmware.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For pure poll requests, it doesn't remove the second poll wait entry
when it's done, neither after vfs_poll() or in the poll completion
handler. We should remove the second poll wait entry.
And we use io_poll_remove_double() rather than io_poll_remove_waitqs()
since the latter has some redundant logic.
Fixes: 88e41cf928 ("io_uring: add multishot mode for IORING_OP_POLL_ADD")
Cc: stable@vger.kernel.org # 5.13+
Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
Link: https://lore.kernel.org/r/20210728030322.12307-1-haoxu@linux.alibaba.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Some setups, like SCSI, can throw spurious -EAGAIN off the softirq
completion path. Normally we expect this to happen inline as part
of submission, but apparently SCSI has a weird corner case where it
can happen as part of normal completions.
This should be solved by having the -EAGAIN bubble back up the stack
as part of submission, but previous attempts at this failed and we're
not just quite there yet. Instead we currently use REQ_F_REISSUE to
handle this case.
For now, catch it in io_rw_should_reissue() and prevent a reissue
from a bogus path.
Cc: stable@vger.kernel.org
Reported-by: Fabian Ebner <f.ebner@proxmox.com>
Tested-by: Fabian Ebner <f.ebner@proxmox.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
As a safeguard, if we're going to queue async work, do it from task_work
from the original task. This ensures that we can always sanely create
threads, regards of what the reissue context may be.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We use a bit to manage if we need to add the shared task_work, but
a list + lock for the pending work. Before aborting a current run
of the task_work we check if the list is empty, but we do so without
grabbing the lock that protects it. This can lead to races where
we think we have nothing left to run, where in practice we could be
racing with a task adding new work to the list. If we do hit that
race condition, we could be left with work items that need processing,
but the shared task_work is not active.
Ensure that we grab the lock before checking if the list is empty,
so we know if it's safe to exit the run or not.
Link: https://lore.kernel.org/io-uring/c6bd5987-e9ae-cd02-49d0-1b3ac1ef65b1@tnonline.net/
Cc: stable@vger.kernel.org # 5.11+
Reported-by: Forza <forza@tnonline.net>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Catch an illegal case to queue async from an unrelated task that got
the ring fd passed to it. This should not be possible to hit, but
better be proactive and catch it explicitly. io-wq is extended to
check for early IO_WQ_WORK_CANCEL being set on a work item as well,
so it can run the request through the normal cancelation path.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There are two reasons why this shouldn't be done:
1) Ring is exiting, and we're canceling requests anyway. Any request
should be canceled anyway. In theory, this could iterate for a
number of times if someone else is also driving the target block
queue into request starvation, however the likelihood of this
happening is miniscule.
2) If the original task decided to pass the ring to another task, then
we don't want to be reissuing from this context as it may be an
unrelated task or context. No assumptions should be made about
the context in which ->release() is run. This can only happen for pure
read/write, and we'll get -EFAULT on them anyway.
Link: https://lore.kernel.org/io-uring/YPr4OaHv0iv0KTOc@zeniv-ca.linux.org.uk/
Reported-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
A previous commit shuffled some code around, and inadvertently used
struct file after fdput() had been called on it. As we can't touch
the file post fdput() dropping our reference, move the fdput() to
after that has been done.
Cc: Pavel Begunkov <asml.silence@gmail.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/io-uring/YPnqM0fY3nM5RdRI@zeniv-ca.linux.org.uk/
Fixes: f2a48dd09b ("io_uring: refactor io_sq_offload_create()")
Reported-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
__io_queue_proc() can enqueue both poll entries and still fail
afterwards, so the callers trying to cancel it should also try to remove
the second poll entry (if any).
For example, it may leave the request alive referencing a io_uring
context but not accessible for cancellation:
[ 282.599913][ T1620] task:iou-sqp-23145 state:D stack:28720 pid:23155 ppid: 8844 flags:0x00004004
[ 282.609927][ T1620] Call Trace:
[ 282.613711][ T1620] __schedule+0x93a/0x26f0
[ 282.634647][ T1620] schedule+0xd3/0x270
[ 282.638874][ T1620] io_uring_cancel_generic+0x54d/0x890
[ 282.660346][ T1620] io_sq_thread+0xaac/0x1250
[ 282.696394][ T1620] ret_from_fork+0x1f/0x30
Cc: stable@vger.kernel.org
Fixes: 18bceab101 ("io_uring: allow POLL_ADD with double poll_wait() users")
Reported-and-tested-by: syzbot+ac957324022b7132accf@syzkaller.appspotmail.com
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/0ec1228fc5eda4cb524eeda857da8efdc43c331c.1626774457.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If __io_queue_proc() fails to add a second poll entry, e.g. kmalloc()
failed, but it goes on with a third waitqueue, it may succeed and
overwrite the error status. Count the number of poll entries we added,
so we can set pt->error to zero at the beginning and find out when the
mentioned scenario happens.
Cc: stable@vger.kernel.org
Fixes: 18bceab101 ("io_uring: allow POLL_ADD with double poll_wait() users")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/9d6b9e561f88bcc0163623b74a76c39f712151c3.1626774457.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When we use delayed_work for fallback execution of requests, current
will be not of the submitter task, and so checks in io_req_task_submit()
may not behave as expected. Currently, it leaves inline completions not
flushed, so making io_ring_exit_work() to hang. Use the submitter task
for all those checks.
Cc: stable@vger.kernel.org
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/cb413c715bed0bc9c98b169059ea9c8a2c770715.1625881431.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Colin reports that Coverity complains about checking for poll being
non-zero after having dereferenced it multiple times. This is a valid
complaint, and actually a leftover from back when this code was based
on the aio poll code.
Kill the redundant check.
Link: https://lore.kernel.org/io-uring/fe70c532-e2a7-3722-58a1-0fa4e5c5ff2c@canonical.com/
Reported-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We have requests like IORING_OP_FILES_UPDATE that don't go through
->iopoll_list but get completed in place under ->uring_lock, and so
after dropping the lock io_iopoll_check() should expect that some CQEs
might have get completed in a meanwhile.
Currently such events won't be accounted in @nr_events, and the loop
will continue to poll even if there is enough of CQEs. It shouldn't be a
problem as it's not likely to happen and so, but not nice either. Just
return earlier in this case, it should be enough.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/66ef932cc66a34e3771bbae04b2953a8058e9d05.1625747741.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If one entered io_req_task_work_add() not seeing PF_EXITING, it will set
a ->task_state bit and try task_work_add(), which may fail by that
moment. If that happens the function would try to cancel the request.
However, in a meanwhile there might come other io_req_task_work_add()
callers, which will see the bit set and leave their requests in the
list, which will never be executed.
Don't propagate an error, but clear the bit first and then fallback
all requests that we can splice from the list. The callback functions
have to be able to deal with PF_EXITING, so poll and apoll was modified
via changing io_poll_rewait().
Fixes: 7cbf1722d5 ("io_uring: provide FIFO ordering for task_work")
Reported-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/060002f19f1fdbd130ba24aef818ea4d3080819b.1625142209.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Since we don't really use req->task_work anymore, get rid of it together
with the nasty ->func aliasing between ->io_task_work and ->task_work,
and hide ->fallback_node inside of io_task_work.
Also, as task_work is gone now, replace the callback type from
task_work_func_t to a function taking io_kiocb to avoid casting and
simplify code.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When task_work_add() fails, we use ->exit_task_work to queue the work.
That will be run only in the cancellation path, which happens either
when the ctx is dying or one of tasks with inflight requests is exiting
or executing. There is a good chance that such a request would just get
stuck in the list potentially hodling a file, all io_uring rsrc
recycling or some other resources. Nothing terrible, it'll go away at
some point, but we don't want to lock them up for longer than needed.
Replace that hand made ->exit_task_work with delayed_work + llist
inspired by fput_many().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We currently spin in iopoll() when requests to be iopolled are for
same file(device), while one device may have multiple hardware queues.
given an example:
hw_queue_0 | hw_queue_1
req(30us) req(10us)
If we first spin on iopolling for the hw_queue_0. the avg latency would
be (30us + 30us) / 2 = 30us. While if we do round robin, the avg
latency would be (30us + 10us) / 2 = 20us since we reap the request in
hw_queue_1 in time. So it's better to do spinning only when requests
are in same hardware queue.
Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Most of requests are allocated from an internal cache, so it's waste of
time fully initialising them every time. Instead, let's pre-init some of
the fields we can during initial allocation (e.g. kmalloc(), see
io_alloc_req()) and keep them valid on request recycling. There are four
of them in this patch:
->ctx is always stays the same
->link is NULL on free, it's an invariant
->result is not even needed to init, just a precaution
->async_data we now clean in io_dismantle_req() as it's likely to
never be allocated.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/892ba0e71309bba9fe9e0142472330bbf9d8f05d.1624739600.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Since cancellation got moved before exit_signals(), there is no one left
who can call io_run_task_work() with PF_EXIING set, so remove the check.
Note that __io_req_task_submit() still needs a similar check.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/f7f305ececb1e6044ea649fb983ca754805bb884.1624739600.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
gcc 11 goes a weird path and duplicates most of io_arm_poll_handler()
for READ and WRITE cases. Help it and move all pollin vs pollout
specific bits under a single if-else, so there is no temptation for this
kind of unfolding.
before vs after:
text data bss dec hex filename
85362 12650 8 98020 17ee4 ./fs/io_uring.o
85186 12650 8 97844 17e34 ./fs/io_uring.o
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/1deea0037293a922a0358e2958384b2e42437885.1624739600.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
It is quite frequent that when an operation fails and returns EAGAIN,
the data becomes available between that failure and the call to
vfs_poll() done by io_arm_poll_handler().
Detecting the situation and reissuing the operation is much faster
than going ahead and push the operation to the io-wq.
Performance improvement testing has been performed with:
Single thread, 1 TCP connection receiving a 5 Mbps stream, no sqpoll.
4 measurements have been taken:
1. The time it takes to process a read request when data is already available
2. The time it takes to process by calling twice io_issue_sqe() after vfs_poll() indicated that data was available
3. The time it takes to execute io_queue_async_work()
4. The time it takes to complete a read request asynchronously
2.25% of all the read operations did use the new path.
ready data (baseline)
avg 3657.94182918628
min 580
max 20098
stddev 1213.15975908162
reissue completion
average 7882.67567567568
min 2316
max 28811
stddev 1982.79172973284
insert io-wq time
average 8983.82276995305
min 3324
max 87816
stddev 2551.60056552038
async time completion
average 24670.4758861127
min 10758
max 102612
stddev 3483.92416873804
Conclusion:
On average reissuing the sqe with the patch code is 1.1uSec faster and
in the worse case scenario 59uSec faster than placing the request on
io-wq
On average completion time by reissuing the sqe with the patch code is
16.79uSec faster and in the worse case scenario 73.8uSec faster than
async completion.
Signed-off-by: Olivier Langlois <olivier@trillion01.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/9e8441419bb1b8f3c3fcc607b2713efecdef2136.1624364038.git.olivier@trillion01.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We can't support IOPOLL with non-pollable request types, and we should
check for unused/reserved fields like we do for other request types.
Fixes: 14a1143b68 ("io_uring: add support for IORING_OP_UNLINKAT")
Cc: stable@vger.kernel.org
Reported-by: Dmitry Kadashev <dkadashev@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We can't support IOPOLL with non-pollable request types, and we should
check for unused/reserved fields like we do for other request types.
Fixes: 80a261fd00 ("io_uring: add support for IORING_OP_RENAMEAT")
Cc: stable@vger.kernel.org
Reported-by: Dmitry Kadashev <dkadashev@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The magic number used to cap the number of entries extracted from an
io_uring instance SQ before moving to the other instances is an
interesting parameter to experiment with.
A define has been created to make it easy to change its value from a
single location.
Signed-off-by: Olivier Langlois <olivier@trillion01.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/b401640063e77ad3e9f921e09c9b3ac10a8bb923.1624473200.git.olivier@trillion01.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If task_state is cleared, io_req_task_work_add() will go the slow path
adding a task_work, setting the task_state, waking up the task and so
on. Not to mention it's expensive. tctx_task_work() first clears the
state and then executes all the work items queued, so if any of them
resubmits or adds new task_work items, it would unnecessarily go through
the slow path of io_req_task_work_add().
Let's clear the ->task_state at the end. We still have to check
->task_list for emptiness afterward to synchronise with
io_req_task_work_add(), do that, and set the state back if we're going
to retry, because clearing not-ours task_state on the next iteration
would be buggy.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/1ef72cdac7022adf0cd7ce4bfe3bb5c82a62eb93.1623949695.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Entering tctx_task_work() with empty task_list is a strange scenario,
that can happen only on rare occasion during task exit, so let's not
check for task_list emptiness in advance and do it do-while style. The
code still correct for the empty case, just would do extra work about
which we don't care.
Do extra step and do the check before cond_resched(), so we don't
resched if have nothing to execute.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/c4173e288e69793d03c7d7ce826f9d28afba718a.1623949695.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We don't need a full copy of tctx->task_list in tctx_task_work(), but
only a first one, so just assign node directly.
Taking into account that task_works are run in a context of a task,
it's very unlikely to first see non-empty tctx->task_list and then
splice it empty, can only happen with task_work cancellations that is
not-normal slow path anyway. Hence, get rid of the check in the end,
it's there not for validity but "performance" purposes.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/d076c83fedb8253baf43acb23b8fafd7c5da1714.1623949695.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
tctx_task_work() tries to fetch a next batch of requests, but before it
would flush completions from the previous batch that may be sub-optimal.
E.g. io_req_task_queue() executes a head of the link where all the
linked may be enqueued through the same io_req_task_queue(). And there
are more cases for that.
Do the flushing at the end, so it can cache completions of several waves
of a single tctx_task_work(), and do the flush at the very end.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/3cac83934e4fbce520ff8025c3524398b3ae0270.1623949695.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently, if req->creds is not NULL, then there are creds assigned.
Track the invariant with a new flag in req->flags. No need to clear the
field at init, and also cleanup can be efficiently moved into
io_clean_op().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/5f8baeb8d3b909487f555542350e2eac97005556.1623949695.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io-wq defaults to per-node masks for IO workers. This works fine by
default, but isn't particularly handy for workloads that prefer more
specific affinities, for either performance or isolation reasons.
This adds IORING_REGISTER_IOWQ_AFF that allows the user to pass in a CPU
mask that is then applied to IO thread workers, and an
IORING_UNREGISTER_IOWQ_AFF that simply resets the masks back to the
default of per-node.
Note that no care is given to existing IO threads, they will need to go
through a reschedule before the affinity is correct if they are already
running or sleeping.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The req pointer uniquely identify a specific request.
Having it in traces can provide valuable insights that is not possible
to have if the calling process is reusing the same user_data value.
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Olivier Langlois <olivier@trillion01.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In most cases io_commit_cqring() is just an smp_store_release(), and
it's hot enough, especially for IRQ rw, to want it to save on a function
call. Mark it inline and extract a non-inlined slow path doing drain
and timeout flushing. The inlined part is pretty slim to not cause
binary bloating.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/7350f8b6b92caa50a48a80be39909f0d83eddd93.1623772051.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Place all drain_next logic into io_drain_req(), so it's never executed
if there was no drained requests before. The only thing we need is to
set ->drain_active if we see a request with IOSQE_IO_DRAIN, do that in
io_init_req() where flags are definitely in registers.
Also, all drain-related code is encapsulated in io_drain_req(), makes it
cleaner.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/68bf4f7395ddaafbf1a26bd97b57d57d45a9f900.1623772051.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
->drain_used is one way, which is not optimal if users use DRAIN but
very rarely. However, we can just clear it in io_drain_req() when all
drained before requests are gone. Also rename the flag to reflect the
change and be more clear about it.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/7f37a240857546a94df6348507edddacab150460.1623772051.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
fs/io_uring.c: In function 'io_alloc_page_table':
include/linux/minmax.h:20:28: warning: comparison of distinct pointer
types lacks a cast
Cast everything to size_t using min_t.
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Fixes: 9123c8ffce ("io_uring: add helpers for 2 level table alloc")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/50f420a956bca070a43810d4a805293ed54f39d8.1623759527.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The sqe_ptr argument has been gone since 709b302fad (io_uring:
simplify io_get_sqring, 2020-04-08), made the return value of the
function. Update the comment accordingly.
Signed-off-by: Fam Zheng <fam.zheng@bytedance.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/20210604164256.12242-1-fam.zheng@bytedance.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Replace drain checks with one-way flag set upon seeing the first
IOSQE_IO_DRAIN request. There are several places where it cuts cycles
well:
1) It's much faster than the fast check with two
conditions in io_drain_req() including pretty complex
list_empty_careful().
2) We can mark io_queue_sqe() inline now, that's a huge win.
3) It replaces timeout and drain checks in io_commit_cqring() with a
single flags test. Also great not touching ->defer_list there without a
reason so limiting cache bouncing.
It adds a small amount of overhead to drain path, but it's negligible.
The main nuisance is that once it meets any DRAIN request in io_uring
instance lifetime it will _always_ go through a slower path, so
drain-less and offset-mode timeout less applications are preferable.
The overhead in that case would be not big, but it's worth to bear in
mind.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/98d2fff8c4da5144bb0d08499f591d4768128ea3.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
->uring_lock is prevalently used for submission, even though it protects
many other things like iopoll, registeration, selected bufs, and more.
And it's placed together with ->cq_wait poked on completion and CQ
waiting sides. Move them apart, ->uring_lock goes to the submission
data, and cq_wait to completion related chunk. The last one requires
some reshuffling so everything needed by io_cqring_ev_posted*() is in
one cacheline.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/dea5e845caee4c98aa0922b46d713154d81f7bd8.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We use several wait_queue_head's for different purposes, but namings are
confusing. First rename ctx->cq_wait into ctx->poll_wait, because this
one is used for polling an io_uring instance. Then rename ctx->wait into
ctx->cq_wait, which is responsible for CQE waiting.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/47b97a097780c86c67b20b6ccc4e077523dce682.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There are no users of ->sq_check_overflow, only ->cq_check_overflow is
used. Combine it and move out of completion related part of struct
io_ring_ctx.
A not so obvious benefit of it is fitting all completion side fields
into a single cacheline. It was taking 2 lines before with 56B padding,
and io_cqring_ev_posted*() were still touching both of them.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/25927394964df31d113e3c729416af573afff5f5.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
submit_state.link is used only to assemble a link and not used for
actual submission, so clear it before io_queue_sqe() in io_submit_sqe(),
awhile it's hot and in caches and queueing doesn't spoil it. May also
potentially help compiler with spilling or to do other optimisations.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/1579939426f3ad6b55af3005b1389bbbed7d780d.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_commit_cqring() might be very hot and we definitely don't want to
touch ->timeout_list there, because 1) it's shared with the submission
side so might lead to cache bouncing and 2) may need to load an extra
cache line, especially for IRQ completions.
We're interested in it at the completion side only when there are
offset-mode timeouts, which are not so popular. Replace
list_empty(->timeout_list) hot path check with a new one-way flag, which
is set when we prepare the first offset-mode timeout.
note: the flag sits in the same line as briefly used after ->rings
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e4892ec68b71a69f92ffbea4a1499be3ec0d463b.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Kill ->cached_sq_dropped and wire DRAIN sequence number correction via
->cq_extra, which is there exactly for that purpose. User visible
dropped counter will be populated by incrementing it instead of keeping
a copy, similarly as it was done not so long ago with cq_overflow.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/088aceb2707a534d531e2770267c4498e0507cc1.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Since moving locked_free_* out of struct io_submit_state
ctx->submit_state is accessed on submission side only, so move it into
the submission section. Same goes for rsrc table pointers/nodes/etc.,
they must be taken and checked during submission because sync'ed by
uring_lock, so move them there as well.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/8a5899a50afc6ccca63249e716f580b246f3dec6.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
ctx->flags are heavily used by both, completion and submission sides, so
move it out from the ctx fields related to submissions. Instead, place
it together with ctx->refs, because it's already cacheline-aligned and
so pads lots of space, and both almost never change. Also, in most
occasions they are accessed together as refs are taken at submission
time and put back during completion.
Do same with ctx->rings, where the pointer itself is never modified
apart from ring init/free.
Note: in percpu mode, struct percpu_ref doesn't modify the struct itself
but takes indirection with ref->percpu_count_ptr.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/4c48c173e63d35591383ba2b87e8b8e8dfdbd23d.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
sq_array and sq_sqes are always used together, however they are in
different cachelines, where the borderline is right before
cq_overflow_list is rather rarely touched. Move the fields together so
it loads only one cacheline.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/3ef2411a94874da06492506a8897eff679244f49.1623709150.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Static analysis is warning that the sizeof being used is should be
of *data->tags[i] and not data->tags[i]. Although these are the same
size on 64 bit systems it is not a portable assumption to assume
this is true for all cases. Fix this by using a temporary pointer
tag_slot to make the code a clearer.
Addresses-Coverity: ("Sizeof not portable")
Fixes: d878c81610 ("io_uring: hide rsrc tag copy into generic helpers")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/20210615130011.57387-1-colin.king@canonical.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
tctx in submission part is always synchronised because is executed from
the task's context, so we can batch allocate tctx/task references and
store them across syscall boundaries. It avoids enough of operations,
including an atomic for getting task ref and a percpu_counter_add()
function call, which still fallback to spinlock for large batching
cases (around >=32). Should be good for SQPOLL submitting in small
portions and coming at some moment bpf submissions.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/14b327b973410a3eec1f702ecf650e100513aca9.1623634181.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There is a complaint against sys_io_uring_enter() blocking if it submits
stdin reads. The problem is in __io_file_supports_async(), which
sees that it's a cdev and allows it to be processed inline.
Punt char devices using generic rules of io_file_supports_async(),
including checking for presence of *_iter() versions of rw callbacks.
Apparently, it will affect most of cdevs with some exceptions like
null and zero devices.
Cc: stable@vger.kernel.org
Reported-by: Birk Hirdman <lonjil@gmail.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/d60270856b8a4560a639ef5f76e55eb563633599.1623236455.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Relax buffer registration restictions, which filters out file backed
memory, and allow shmem/memfd as they have normal anonymous pages
underneath.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
struct io_submit_state contains struct io_comp_state and so
locked_free_*, that renders cachelines around ->locked_free* being
invalidated on most non-inline completions, that may terrorise caches if
submissions and completions are done by different tasks.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/290cb5412b76892e8631978ee8ab9db0c6290dd5.1621201931.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There are two copies of cq_overflow, shared with userspace and internal
cached one. It was needed for DRAIN accounting, but now we have yet
another knob to tune the accounting, i.e. cq_extra, and we can throw
away the internal counter and just increment the one in the shared ring.
If user modifies it as so never gets the right overflow value ever
again, it's its problem, even though before we would have restored it
back by next overflow.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/8427965f5175dd051febc63804909861109ce859.1621201931.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
ring has two types of resource-related fields: used for request
submission, and field needed for update/registration. Reshuffle them
into these two groups for better locality and readability. The second
group is not in the hot path, so it's natural to place them somewhere in
the end. Also update an outdated comment.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/05b34795bb4440f4ec4510f08abd5a31830f8ca0.1621201931.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
