With the previous fixes for number of files open checking, I added some
debug code to see if we had other spots where we're checking rlimit()
against the async io-wq workers. The only one I found was file size
checking, which we should also honor.
During write and fallocate prep, store the max file size and override
that for the current ask if we're in io-wq worker context.
Cc: stable@vger.kernel.org # 5.1+
Signed-off-by: Jens Axboe <axboe@kernel.dk>
It's a preparation patch removing io_wq_enqueue_hashed(), which
now should be done by io_wq_hash_work() + io_wq_enqueue().
Also, set hash value for dependant works, and do it as late as possible,
because req->file can be unavailable before. This hash will be ignored
by io-wq.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Ensure we keep the truncated value, if we did truncate it. If not, we
might read/write more than the registered buffer size.
Also for retry, ensure that we return the truncated mapped value for
the vectorized versions of the read/write commands.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When SETUP_IOPOLL and SETUP_SQPOLL are both enabled, applications don't need
to do io completion events polling again, they can rely on io_sq_thread to do
polling work, which can reduce cpu usage and uring_lock contention.
I modify fio io_uring engine codes a bit to evaluate the performance:
static int fio_ioring_getevents(struct thread_data *td, unsigned int min,
continue;
}
- if (!o->sqpoll_thread) {
+ if (o->sqpoll_thread && o->hipri) {
r = io_uring_enter(ld, 0, actual_min,
IORING_ENTER_GETEVENTS);
if (r < 0) {
and use "fio -name=fiotest -filename=/dev/nvme0n1 -iodepth=$depth -thread
-rw=read -ioengine=io_uring -hipri=1 -sqthread_poll=1 -direct=1 -bs=4k
-size=10G -numjobs=1 -time_based -runtime=120"
original codes
--------------------------------------------------------------------
iodepth | 4 | 8 | 16 | 32 | 64
bw | 1133MB/s | 1519MB/s | 2090MB/s | 2710MB/s | 3012MB/s
fio cpu usage | 100% | 100% | 100% | 100% | 100%
--------------------------------------------------------------------
with patch
--------------------------------------------------------------------
iodepth | 4 | 8 | 16 | 32 | 64
bw | 1196MB/s | 1721MB/s | 2351MB/s | 2977MB/s | 3357MB/s
fio cpu usage | 63.8% | 74.4%% | 81.1% | 83.7% | 82.4%
--------------------------------------------------------------------
bw improve | 5.5% | 13.2% | 12.3% | 9.8% | 11.5%
--------------------------------------------------------------------
From above test results, we can see that bw has above 5.5%~13%
improvement, and fio process's cpu usage also drops much. Note this
won't improve io_sq_thread's cpu usage when SETUP_IOPOLL|SETUP_SQPOLL
are both enabled, in this case, io_sq_thread always has 100% cpu usage.
I think this patch will be friendly to applications which will often use
io_uring_wait_cqe() or similar from liburing.
Signed-off-by: Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If CONFIG_NET is not set, gcc warns:
fs/io_uring.c:3110:12: warning: io_setup_async_msg defined but not used [-Wunused-function]
static int io_setup_async_msg(struct io_kiocb *req,
^~~~~~~~~~~~~~~~~~
There are many funcions wraped by CONFIG_NET, move them
together to simplify code, also fix this warning.
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Minor tweaks.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Not easy to tell if we're going over the size of bits we can shove
in req->flags, so add an end-of-bits marker and a BUILD_BUG_ON()
check for it.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We have IORING_OP_PROVIDE_BUFFERS, but the only way to remove buffers
is to trigger IO on them. The usual case of shrinking a buffer pool
would be to just not replenish the buffers when IO completes, and
instead just free it. But it may be nice to have a way to manually
remove a number of buffers from a given group, and
IORING_OP_REMOVE_BUFFERS provides that functionality.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This adds support for the vectored read. This is limited to supporting
just 1 segment in the iov, and is provided just for convenience for
applications that use IORING_OP_READV already.
The iov helpers will be used for IORING_OP_RECVMSG as well.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If a server process has tons of pending socket connections, generally
it uses epoll to wait for activity. When the socket is ready for reading
(or writing), the task can select a buffer and issue a recv/send on the
given fd.
Now that we have fast (non-async thread) support, a task can have tons
of pending reads or writes pending. But that means they need buffers to
back that data, and if the number of connections is high enough, having
them preallocated for all possible connections is unfeasible.
With IORING_OP_PROVIDE_BUFFERS, an application can register buffers to
use for any request. The request then sets IOSQE_BUFFER_SELECT in the
sqe, and a given group ID in sqe->buf_group. When the fd becomes ready,
a free buffer from the specified group is selected. If none are
available, the request is terminated with -ENOBUFS. If successful, the
CQE on completion will contain the buffer ID chosen in the cqe->flags
member, encoded as:
(buffer_id << IORING_CQE_BUFFER_SHIFT) | IORING_CQE_F_BUFFER;
Once a buffer has been consumed by a request, it is no longer available
and must be registered again with IORING_OP_PROVIDE_BUFFERS.
Requests need to support this feature. For now, IORING_OP_READ and
IORING_OP_RECV support it. This is checked on SQE submission, a CQE with
res == -EOPNOTSUPP will be posted if attempted on unsupported requests.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
IORING_OP_PROVIDE_BUFFERS uses the buffer registration infrastructure to
support passing in an addr/len that is associated with a buffer ID and
buffer group ID. The group ID is used to index and lookup the buffers,
while the buffer ID can be used to notify the application which buffer
in the group was used. The addr passed in is the starting buffer address,
and length is each buffer length. A number of buffers to add with can be
specified, in which case addr is incremented by length for each addition,
and each buffer increments the buffer ID specified.
No validation is done of the buffer ID. If the application provides
buffers within the same group with identical buffer IDs, then it'll have
a hard time telling which buffer ID was used. The only restriction is
that the buffer ID can be a max of 16-bits in size, so USHRT_MAX is the
maximum ID that can be used.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This just prepares the ring for having lists of buffers associated with
it, that the application can provide for SQEs to consume instead of
providing their own.
The buffers are organized by group ID.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
First it changes io-wq interfaces. It replaces {get,put}_work() with
free_work(), which guaranteed to be called exactly once. It also enforces
free_work() callback to be non-NULL.
io_uring follows the changes and instead of putting a submission reference
in io_put_req_async_completion(), it will be done in io_free_work(). As
removes io_get_work() with corresponding refcount_inc(), the ref balance
is maintained.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If after dropping the submission reference req->refs == 1, the request
is done, because this one is for io_put_work() and will be dropped
synchronously shortly after. In this case it's safe to steal a next
work from the request.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There will be no use for @nxt in the handlers, and it's doesn't work
anyway, so purge it
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The rule is simple, any async handler gets a submission ref and should
put it at the end. Make them all follow it, and so more consistent.
This is a preparation patch, and as io_wq_assign_next() currently won't
ever work, this doesn't care to use io_put_req_find_next() instead of
io_put_req().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
refcount_inc_not_zero() -> refcount_inc() fix.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Clang warns:
fs/io_uring.c:4178:6: warning: variable 'mask' is used uninitialized
whenever 'if' condition is false [-Wsometimes-uninitialized]
if (def->pollin)
^~~~~~~~~~~
fs/io_uring.c:4182:2: note: uninitialized use occurs here
mask |= POLLERR | POLLPRI;
^~~~
fs/io_uring.c:4178:2: note: remove the 'if' if its condition is always
true
if (def->pollin)
^~~~~~~~~~~~~~~~
fs/io_uring.c:4154:15: note: initialize the variable 'mask' to silence
this warning
__poll_t mask, ret;
^
= 0
1 warning generated.
io_op_defs has many definitions where pollin is not set so mask indeed
might be uninitialized. Initialize it to zero and change the next
assignment to |=, in case further masks are added in the future to avoid
missing changing the assignment then.
Fixes: d7718a9d25 ("io_uring: use poll driven retry for files that support it")
Link: https://github.com/ClangBuiltLinux/linux/issues/916
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io-wq cares about IO_WQ_WORK_UNBOUND flag only while enqueueing, so
it's useless setting it for a next req of a link. Thus, removed it
from io_prep_linked_timeout(), and inline the function.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
After __io_queue_sqe() ended up in io_queue_async_work(), it's already
known that there is no @nxt req, so skip the check and return from the
function.
Also, @nxt initialisation now can be done just before
io_put_req_find_next(), as there is no jumping until it's checked.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently io_uring tries any request in a non-blocking manner, if it can,
and then retries from a worker thread if we get -EAGAIN. Now that we have
a new and fancy poll based retry backend, use that to retry requests if
the file supports it.
This means that, for example, an IORING_OP_RECVMSG on a socket no longer
requires an async thread to complete the IO. If we get -EAGAIN reading
from the socket in a non-blocking manner, we arm a poll handler for
notification on when the socket becomes readable. When it does, the
pending read is executed directly by the task again, through the io_uring
task work handlers. Not only is this faster and more efficient, it also
means we're not generating potentially tons of async threads that just
sit and block, waiting for the IO to complete.
The feature is marked with IORING_FEAT_FAST_POLL, meaning that async
pollable IO is fast, and that poll<link>other_op is fast as well.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add a pollin/pollout field to the request table, and have commands that
we can safely poll for properly marked.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For poll requests, it's not uncommon to link a read (or write) after
the poll to execute immediately after the file is marked as ready.
Since the poll completion is called inside the waitqueue wake up handler,
we have to punt that linked request to async context. This slows down
the processing, and actually means it's faster to not use a link for this
use case.
We also run into problems if the completion_lock is contended, as we're
doing a different lock ordering than the issue side is. Hence we have
to do trylock for completion, and if that fails, go async. Poll removal
needs to go async as well, for the same reason.
eventfd notification needs special case as well, to avoid stack blowing
recursion or deadlocks.
These are all deficiencies that were inherited from the aio poll
implementation, but I think we can do better. When a poll completes,
simply queue it up in the task poll list. When the task completes the
list, we can run dependent links inline as well. This means we never
have to go async, and we can remove a bunch of code associated with
that, and optimizations to try and make that run faster. The diffstat
speaks for itself.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Store the io_kiocb in the private field instead of the poll entry, this
is in preparation for allowing multiple waitqueues.
No functional changes in this patch.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
IO_WQ_WORK_CB is used only for linked timeouts, which will be armed
before the work setup (i.e. mm, override creds, etc). The setup
shouldn't take long, so it's ok to arm it a bit later and get rid
of IO_WQ_WORK_CB.
Make io-wq call work->func() only once, callbacks will handle the rest.
i.e. the linked timeout handler will do the actual issue. And as a
bonus, it removes an extra indirect call.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Deduplicate call to io_cqring_fill_event(), plain and easy
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add support for splice(2).
- output file is specified as sqe->fd, so it's handled by generic code
- hash_reg_file handled by generic code as well
- len is 32bit, but should be fine
- the fd_in is registered file, when SPLICE_F_FD_IN_FIXED is set, which
is a splice flag (i.e. sqe->splice_flags).
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Preparation without functional changes. Adds io_get_file(), that allows
to grab files not only into req->file.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
req->in_async is not really needed, it only prevents propagation of
@nxt for fast not-blocked submissions. Remove it.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_prep_async_worker() called io_wq_assign_next() do many useless checks:
io_req_work_grab_env() was already called during prep, and @do_hashed
is not ever used. Add io_prep_next_work() -- simplified version, that
can be called io-wq.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Many operations define custom work.func before getting into an io-wq.
There are several points against:
- it calls io_wq_assign_next() from outside io-wq, that may be confusing
- sync context would go unnecessary through io_req_cancelled()
- prototypes are quite different, so work!=old_work looks strange
- makes async/sync responsibilities fuzzy
- adds extra overhead
Don't call generic path and io-wq handlers from each other, but use
helpers instead
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Don't drop an early reference, hang on to it and let the caller drop
it. This makes it behave more like "regular" requests.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If the -EAGAIN happens because of a static condition, then a poll
or later retry won't fix it. We must call it again from blocking
condition. Play it safe and ensure that any -EAGAIN condition from read
or write must retry from async context.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We must set MSG_CMSG_COMPAT if we're in compatability mode, otherwise
the iovec import for these commands will not do the right thing and fail
the command with -EINVAL.
Found by running the test suite compiled as 32-bit.
Cc: stable@vger.kernel.org
Fixes: aa1fa28fc7 ("io_uring: add support for recvmsg()")
Fixes: 0fa03c624d ("io_uring: add support for sendmsg()")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Follow the pattern used with other *_show_fdinfo functions and only
define and use io_uring_show_fdinfo and its helper functions if
CONFIG_PROC_FS is set.
Fixes: 87ce955b24 ("io_uring: add ->show_fdinfo() for the io_uring file descriptor")
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Unlike the other core import helpers, import_single_range() returns 0 on
success, not the length imported. This means that links that depend on
the result of non-vec based IORING_OP_{READ,WRITE} that were added for
5.5 get errored when they should not be.
Fixes: 3a6820f2bb ("io_uring: add non-vectored read/write commands")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If work completes inline, then we should pick up a dependent link item
in __io_queue_sqe() as well. If we don't do so, we're forced to go async
with that item, which is suboptimal.
This also fixes an issue with io_put_req_find_next(), which always looks
up the next work item. That should only be done if we're dropping the
last reference to the request, to prevent multiple lookups of the same
work item.
Outside of being a fix, this also enables a good cleanup series for 5.7,
where we never have to pass 'nxt' around or into the work handlers.
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
After making ext4 support iopoll method:
let ext4_file_operations's iopoll method be iomap_dio_iopoll(),
we found fio can easily hang in fio_ioring_getevents() with below fio
job:
rm -f testfile; sync;
sudo fio -name=fiotest -filename=testfile -iodepth=128 -thread
-rw=write -ioengine=io_uring -hipri=1 -sqthread_poll=1 -direct=1
-bs=4k -size=10G -numjobs=8 -runtime=2000 -group_reporting
with IORING_SETUP_SQPOLL and IORING_SETUP_IOPOLL enabled.
There are two issues that results in this hang, one reason is that
when IORING_SETUP_SQPOLL and IORING_SETUP_IOPOLL are enabled, fio
does not use io_uring_enter to get completed events, it relies on
kernel io_sq_thread to poll for completed events.
Another reason is that there is a race: when io_submit_sqes() in
io_sq_thread() submits a batch of sqes, variable 'inflight' will
record the number of submitted reqs, then io_sq_thread will poll for
reqs which have been added to poll_list. But note, if some previous
reqs have been punted to io worker, these reqs will won't be in
poll_list timely. io_sq_thread() will only poll for a part of previous
submitted reqs, and then find poll_list is empty, reset variable
'inflight' to be zero. If app just waits these deferred reqs and does
not wake up io_sq_thread again, then hang happens.
For app that entirely relies on io_sq_thread to poll completed requests,
let io_iopoll_req_issued() wake up io_sq_thread properly when adding new
element to poll_list, and when io_sq_thread prepares to sleep, check
whether poll_list is empty again, if not empty, continue to poll.
Signed-off-by: Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We somehow never free the idr, even though we init it for every ctx.
Free it when the rest of the ring data is freed.
Fixes: 071698e13a ("io_uring: allow registering credentials")
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If we have a chain of requests and they don't all use the same
credentials, then the head of the chain will be issued with the
credentails of the tail of the chain.
Ensure __io_queue_sqe() overrides the credentials, if they are different.
Once we do that, we can clean up the creds handling as well, by only
having io_submit_sqe() do the lookup of a personality. It doesn't need
to assign it, since __io_queue_sqe() now always does the right thing.
Fixes: 75c6a03904 ("io_uring: support using a registered personality for commands")
Reported-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Since commit a3a0e43fd7 ("io_uring: don't enter poll loop if we have
CQEs pending"), if we already events pending, we won't enter poll loop.
In case SETUP_IOPOLL and SETUP_SQPOLL are both enabled, if app has
been terminated and don't reap pending events which are already in cq
ring, and there are some reqs in poll_list, io_sq_thread will enter
__io_iopoll_check(), and find pending events, then return, this loop
will never have a chance to exit.
I have seen this issue in fio stress tests, to fix this issue, let
io_sq_thread call io_iopoll_getevents() with argument 'min' being zero,
and remove __io_iopoll_check().
Fixes: a3a0e43fd7 ("io_uring: don't enter poll loop if we have CQEs pending")
Signed-off-by: Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This patch drops 'cur_mm' before calling cond_resched(), to prevent
the sq_thread from spinning even when the user process is finished.
Before this patch, if the user process ended without closing the
io_uring fd, the sq_thread continues to spin until the
'sq_thread_idle' timeout ends.
In the worst case where the 'sq_thread_idle' parameter is bigger than
INT_MAX, the sq_thread will spin forever.
Fixes: 6c271ce2f1 ("io_uring: add submission polling")
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_cleanup_req() should be called before req->io is freed, and so
shouldn't be after __io_free_req() -> __io_req_aux_free(). Also,
it will be ignored for in io_free_req_many(), which use
__io_req_aux_free().
Place cleanup_req() into __io_req_aux_free().
Fixes: 99bc4c3853 ("io_uring: fix iovec leaks")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The "kmsg" pointer can't be NULL and we have already dereferenced it so
a check here would be useless.
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
fallocate_finish() is missing cancellation check. Add it.
It's safe to do that, as only flags setup and sqe fields copy are done
before it gets into __io_fallocate().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Carter reported an issue where he could produce a stall on ring exit,
when we're cleaning up requests that match the given file table. For
this particular test case, a combination of a few things caused the
issue:
- The cq ring was overflown
- The request being canceled was in the overflow list
The combination of the above means that the cq overflow list holds a
reference to the request. The request is canceled correctly, but since
the overflow list holds a reference to it, the final put won't happen.
Since the final put doesn't happen, the request remains in the inflight.
Hence we never finish the cancelation flush.
Fix this by removing requests from the overflow list if we're canceling
them.
Cc: stable@vger.kernel.org # 5.5
Reported-by: Carter Li 李通洲 <carter.li@eoitek.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Jonas reports that he sometimes sees -97/-22 error returns from
sendmsg, if it gets punted async. This is due to not retaining the
sockaddr_storage between calls. Include that in the state we copy when
going async.
Cc: stable@vger.kernel.org # 5.3+
Reported-by: Jonas Bonn <jonas@norrbonn.se>
Tested-by: Jonas Bonn <jonas@norrbonn.se>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Normally we cancel all work we track, but for untracked work we could
leave the async worker behind until that work completes. This is totally
fine, but does leave resources pending after the task is gone until that
work completes.
Cancel work that this task queued up when it goes away.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
