man/io_uring_enter.2

.\" Copyright (C) 2019 Jens Axboe <axboe@kernel.dk>
.\" Copyright (C) 2019 Red Hat, Inc.
.\"
.\" SPDX-License-Identifier: LGPL-2.0-or-later
.\"
.TH io_uring_enter 2 2019-01-22 "Linux" "Linux Programmer's Manual"
.SH NAME
io_uring_enter \- initiate and/or complete asynchronous I/O
.SH SYNOPSIS
.nf
.BR "#include <liburing.h>"
.PP
.BI "int io_uring_enter(unsigned int " fd ", unsigned int " to_submit ,
.BI "                   unsigned int " min_complete ", unsigned int " flags ,
.BI "                   sigset_t *" sig );
.PP
.BI "int io_uring_enter2(unsigned int " fd ", unsigned int " to_submit ,
.BI "                    unsigned int " min_complete ", unsigned int " flags ,
.BI "                    sigset_t *" sig ", size_t " sz );
.fi
.PP
.SH DESCRIPTION
.PP
.BR io_uring_enter (2)
is used to initiate and complete I/O using the shared submission and
completion queues setup by a call to
.BR io_uring_setup (2).
A single call can both submit new I/O and wait for completions of I/O
initiated by this call or previous calls to
.BR io_uring_enter (2).

.I fd
is the file descriptor returned by
.BR io_uring_setup (2).
.I to_submit
specifies the number of I/Os to submit from the submission queue.
.I flags
is a bitmask of the following values:
.TP
.B IORING_ENTER_GETEVENTS
If this flag is set, then the system call will wait for the specified
number of events in
.I min_complete
before returning. This flag can be set along with
.I to_submit
to both submit and complete events in a single system call.
.TP
.B IORING_ENTER_SQ_WAKEUP
If the ring has been created with
.B IORING_SETUP_SQPOLL,
then this flag asks the kernel to wakeup the SQ kernel thread to submit IO.
.TP
.B IORING_ENTER_SQ_WAIT
If the ring has been created with
.B IORING_SETUP_SQPOLL,
then the application has no real insight into when the SQ kernel thread has
consumed entries from the SQ ring. This can lead to a situation where the
application can no longer get a free SQE entry to submit, without knowing
when it one becomes available as the SQ kernel thread consumes them. If
the system call is used with this flag set, then it will wait until at least
one entry is free in the SQ ring.
.TP
.B IORING_ENTER_EXT_ARG
Since kernel 5.11, the system calls arguments have been modified to look like
the following:

.nf
.BI "int io_uring_enter2(unsigned int " fd ", unsigned int " to_submit ,
.BI "                    unsigned int " min_complete ", unsigned int " flags ,
.BI "                    const void *" arg ", size_t " argsz );
.fi

which behaves just like the original definition by default. However, if
.B IORING_ENTER_EXT_ARG
is set, then instead of a
.I sigset_t
being passed in, a pointer to a
.I struct io_uring_getevents_arg
is used instead and
.I argsz
must be set to the size of this structure. The definition is as follows:

.nf
.BI "struct io_uring_getevents_arg {
.BI "        __u64   sigmask;
.BI "        __u32   sigmask_sz;
.BI "        __u32   pad;
.BI "        __u64   ts;
.BI "};
.fi

which allows passing in both a signal mask as well as pointer to a
.I struct __kernel_timespec
timeout value. If
.I ts
is set to a valid pointer, then this time value indicates the timeout for
waiting on events. If an application is waiting on events and wishes to
stop waiting after a specified amount of time, then this can be accomplished
directly in version 5.11 and newer by using this feature.
.TP
.B IORING_ENTER_REGISTERED_RING
If the ring file descriptor has been registered through use of
.B IORING_REGISTER_RING_FDS,
then setting this flag will tell the kernel that the
.I ring_fd
passed in is the registered ring offset rather than a normal file descriptor.

.PP
.PP
If the io_uring instance was configured for polling, by specifying
.B IORING_SETUP_IOPOLL
in the call to
.BR io_uring_setup (2),
then min_complete has a slightly different meaning.  Passing a value
of 0 instructs the kernel to return any events which are already complete,
without blocking.  If
.I min_complete
is a non-zero value, the kernel will still return immediately if any
completion events are available.  If no event completions are
available, then the call will poll either until one or more
completions become available, or until the process has exceeded its
scheduler time slice.

Note that, for interrupt driven I/O (where
.B IORING_SETUP_IOPOLL
was not specified in the call to
.BR io_uring_setup (2)),
an application may check the completion queue for event completions
without entering the kernel at all.
.PP
When the system call returns that a certain amount of SQEs have been
consumed and submitted, it's safe to reuse SQE entries in the ring. This is
true even if the actual IO submission had to be punted to async context,
which means that the SQE may in fact not have been submitted yet. If the
kernel requires later use of a particular SQE entry, it will have made a
private copy of it.

.I sig
is a pointer to a signal mask (see
.BR sigprocmask (2));
if
.I sig
is not NULL,
.BR io_uring_enter (2)
first replaces the current signal mask by the one pointed to by
.IR sig ,
then waits for events to become available in the completion queue, and
then restores the original signal mask.  The following
.BR io_uring_enter (2)
call:
.PP
.in +4n
.EX
ret = io_uring_enter(fd, 0, 1, IORING_ENTER_GETEVENTS, &sig);
.EE
.in
.PP
is equivalent to
.I atomically
executing the following calls:
.PP
.in +4n
.EX
pthread_sigmask(SIG_SETMASK, &sig, &orig);
ret = io_uring_enter(fd, 0, 1, IORING_ENTER_GETEVENTS, NULL);
pthread_sigmask(SIG_SETMASK, &orig, NULL);
.EE
.in
.PP
See the description of
.BR pselect (2)
for an explanation of why the
.I sig
parameter is necessary.

Submission queue entries are represented using the following data
structure:
.PP
.in +4n
.EX
/*
 * IO submission data structure (Submission Queue Entry)
 */
struct io_uring_sqe {
    __u8    opcode;         /* type of operation for this sqe */
    __u8    flags;          /* IOSQE_ flags */
    __u16   ioprio;         /* ioprio for the request */
    __s32   fd;             /* file descriptor to do IO on */
    union {
        __u64   off;            /* offset into file */
        __u64   addr2;
    };
    union {
        __u64   addr;       /* pointer to buffer or iovecs */
        __u64   splice_off_in;
    }
    __u32   len;            /* buffer size or number of iovecs */
    union {
        __kernel_rwf_t  rw_flags;
        __u32    fsync_flags;
        __u16    poll_events;   /* compatibility */
        __u32    poll32_events; /* word-reversed for BE */
        __u32    sync_range_flags;
        __u32    msg_flags;
        __u32    timeout_flags;
        __u32    accept_flags;
        __u32    cancel_flags;
        __u32    open_flags;
        __u32    statx_flags;
        __u32    fadvise_advice;
        __u32    splice_flags;
        __u32    rename_flags;
        __u32    unlink_flags;
        __u32    hardlink_flags;
    };
    __u64    user_data;     /* data to be passed back at completion time */
    union {
    struct {
        /* index into fixed buffers, if used */
            union {
                /* index into fixed buffers, if used */
                __u16    buf_index;
                /* for grouped buffer selection */
                __u16    buf_group;
            }
        /* personality to use, if used */
        __u16    personality;
        union {
            __s32    splice_fd_in;
            __u32    file_index;
	};
    };
    __u64    __pad2[3];
    };
};
.EE
.in
.PP
The
.I opcode
describes the operation to be performed.  It can be one of:
.TP
.B IORING_OP_NOP
Do not perform any I/O.  This is useful for testing the performance of
the io_uring implementation itself.
.TP
.B IORING_OP_READV
.TP
.B IORING_OP_WRITEV
Vectored read and write operations, similar to
.BR preadv2 (2)
and
.BR pwritev2 (2).
If the file is not seekable,
.I off
must be set to zero or -1.

.TP
.B IORING_OP_READ_FIXED
.TP
.B IORING_OP_WRITE_FIXED
Read from or write to pre-mapped buffers.  See
.BR io_uring_register (2)
for details on how to setup a context for fixed reads and writes.

.TP
.B IORING_OP_FSYNC
File sync.  See also
.BR fsync (2).
Note that, while I/O is initiated in the order in which it appears in
the submission queue, completions are unordered.  For example, an
application which places a write I/O followed by an fsync in the
submission queue cannot expect the fsync to apply to the write.  The
two operations execute in parallel, so the fsync may complete before
the write is issued to the storage.  The same is also true for
previously issued writes that have not completed prior to the fsync.

.TP
.B IORING_OP_POLL_ADD
Poll the
.I fd
specified in the submission queue entry for the events
specified in the
.I poll_events
field.  Unlike poll or epoll without
.BR EPOLLONESHOT ,
by default this interface always works in one shot mode.  That is, once the poll
operation is completed, it will have to be resubmitted.

If
.B IORING_POLL_ADD_MULTI
is set in the SQE
.I len
field, then the poll will work in multi shot mode instead. That means it'll
repatedly trigger when the requested event becomes true, and hence multiple
CQEs can be generated from this single SQE. The CQE
.I flags
field will have
.B IORING_CQE_F_MORE
set on completion if the application should expect further CQE entries from
the original request. If this flag isn't set on completion, then the poll
request has been terminated and no further events will be generated. This mode
is available since 5.13.

If
.B IORING_POLL_UPDATE_EVENTS
is set in the SQE
.I len
field, then the request will update an existing poll request with the mask of
events passed in with this request. The lookup is based on the
.I user_data
field of the original SQE submitted, and this values is passed in the
.I addr
field of the SQE. This mode is available since 5.13.

If
.B IORING_POLL_UPDATE_USER_DATA
is set in the SQE
.I len
field, then the request will update the
.I user_data
of an existing poll request based on the value passed in the
.I off
field. This mode is available since 5.13.

This command works like
an async
.BR poll(2)
and the completion event result is the returned mask of events. For the
variants that update
.I user_data
or
.I events
, the completion result will be similar to
.B IORING_OP_POLL_REMOVE.

.TP
.B IORING_OP_POLL_REMOVE
Remove an existing poll request.  If found, the
.I res
field of the
.I "struct io_uring_cqe"
will contain 0.  If not found,
.I res
will contain
.B -ENOENT,
or
.B -EALREADY
if the poll request was in the process of completing already.

.TP
.B IORING_OP_EPOLL_CTL
Add, remove or modify entries in the interest list of
.BR epoll (7).
See
.BR epoll_ctl (2)
for details of the system call.
.I fd
holds the file descriptor that represents the epoll instance,
.I addr
holds the file descriptor to add, remove or modify,
.I len
holds the operation (EPOLL_CTL_ADD, EPOLL_CTL_DEL, EPOLL_CTL_MOD) to perform and,
.I off
holds a pointer to the
.I epoll_events
structure. Available since 5.6.

.TP
.B IORING_OP_SYNC_FILE_RANGE
Issue the equivalent of a \fBsync_file_range\fR (2) on the file descriptor. The
.I fd
field is the file descriptor to sync, the
.I off
field holds the offset in bytes, the
.I len
field holds the length in bytes, and the
.I sync_range_flags
field holds the flags for the command. See also
.BR sync_file_range (2)
for the general description of the related system call. Available since 5.2.

.TP
.B IORING_OP_SENDMSG
Issue the equivalent of a
.BR sendmsg(2)
system call.
.I fd
must be set to the socket file descriptor,
.I addr
must contain a pointer to the msghdr structure, and
.I msg_flags
holds the flags associated with the system call. See also
.BR sendmsg (2)
for the general description of the related system call. Available since 5.3.

This command also supports the following modifiers in
.I ioprio:

.PP
.in +12
.B IORING_RECVSEND_POLL_FIRST
If set, io_uring will assume the socket is currently full and attempting to
send data will be unsuccessful. For this case, io_uring will arm internal
poll and trigger a send of the data when there is enough space available.
This initial send attempt can be wasteful for the case where the socket
is expected to be full, setting this flag will bypass the initial send
attempt and go straight to arming poll. If poll does indicate that data can
be sent, the operation will proceed.
.EE
.in
.PP

.TP
.B IORING_OP_RECVMSG
Works just like IORING_OP_SENDMSG, except for
.BR recvmsg(2)
instead. See the description of IORING_OP_SENDMSG. Available since 5.3.

This command also supports the following modifiers in
.I ioprio:

.PP
.in +12
.B IORING_RECVSEND_POLL_FIRST
If set, io_uring will assume the socket is currently empty and attempting to
receive data will be unsuccessful. For this case, io_uring will arm internal
poll and trigger a receive of the data when the socket has data to be read.
This initial receive attempt can be wasteful for the case where the socket
is expected to be empty, setting this flag will bypass the initial receive
attempt and go straight to arming poll. If poll does indicate that data is
ready to be received, the operation will proceed.
.EE
.in
.PP

.TP
.B IORING_OP_SEND
Issue the equivalent of a
.BR send(2)
system call.
.I fd
must be set to the socket file descriptor,
.I addr
must contain a pointer to the buffer,
.I len
denotes the length of the buffer to send, and
.I msg_flags
holds the flags associated with the system call. See also
.BR send(2)
for the general description of the related system call. Available since 5.6.

This command also supports the following modifiers in
.I ioprio:

.PP
.in +12
.B IORING_RECVSEND_POLL_FIRST
If set, io_uring will assume the socket is currently full and attempting to
send data will be unsuccessful. For this case, io_uring will arm internal
poll and trigger a send of the data when there is enough space available.
This initial send attempt can be wasteful for the case where the socket
is expected to be full, setting this flag will bypass the initial send
attempt and go straight to arming poll. If poll does indicate that data can
be sent, the operation will proceed.
.EE
.in
.PP

.TP
.B IORING_OP_RECV
Works just like IORING_OP_SEND, except for
.BR recv(2)
instead. See the description of IORING_OP_SEND. Available since 5.6.

This command also supports the following modifiers in
.I ioprio:

.PP
.in +12
.B IORING_RECVSEND_POLL_FIRST
If set, io_uring will assume the socket is currently empty and attempting to
receive data will be unsuccessful. For this case, io_uring will arm internal
poll and trigger a receive of the data when the socket has data to be read.
This initial receive attempt can be wasteful for the case where the socket
is expected to be empty, setting this flag will bypass the initial receive
attempt and go straight to arming poll. If poll does indicate that data is
ready to be received, the operation will proceed.
.EE
.in
.PP

.TP
.B IORING_OP_TIMEOUT
This command will register a timeout operation. The
.I addr
field must contain a pointer to a struct timespec64 structure,
.I len
must contain 1 to signify one timespec64 structure,
.I timeout_flags
may contain IORING_TIMEOUT_ABS
for an absolute timeout value, or 0 for a relative timeout.
.I off
may contain a completion event count. A timeout
will trigger a wakeup event on the completion ring for anyone waiting for
events. A timeout condition is met when either the specified timeout expires,
or the specified number of events have completed. Either condition will
trigger the event. If set to 0, completed events are not counted, which
effectively acts like a timer. io_uring timeouts use the
.B CLOCK_MONOTONIC
clock source. The request will complete with
.I -ETIME
if the timeout got completed through expiration of the timer, or
.I 0
if the timeout got completed through requests completing on their own. If
the timeout was canceled before it expired, the request will complete with
.I -ECANCELED.
Available since 5.4.

Since 5.15, this command also supports the following modifiers in
.I timeout_flags:

.PP
.in +12
.B IORING_TIMEOUT_BOOTTIME
If set, then the clocksource used is
.I CLOCK_BOOTTIME
instead of
.I CLOCK_MONOTONIC.
This clocksource differs in that it includes time elapsed if the system was
suspend while having a timeout request in-flight.

.B IORING_TIMEOUT_REALTIME
If set, then the clocksource used is
.I CLOCK_REALTIME
instead of
.I CLOCK_MONOTONIC.
.EE
.in
.PP

.TP
.B IORING_OP_TIMEOUT_REMOVE
If
.I timeout_flags are zero, then it attempts to remove an existing timeout
operation.
.I addr
must contain the
.I user_data
field of the previously issued timeout operation. If the specified timeout
request is found and canceled successfully, this request will terminate
with a result value of
.I 0
If the timeout request was found but expiration was already in progress,
this request will terminate with a result value of
.I -EBUSY
If the timeout request wasn't found, the request will terminate with a result
value of
.I -ENOENT
Available since 5.5.

If
.I timeout_flags
contain
.I IORING_TIMEOUT_UPDATE,
instead of removing an existing operation, it updates it.
.I addr
and return values are same as before.
.I addr2
field must contain a pointer to a struct timespec64 structure.
.I timeout_flags
may also contain IORING_TIMEOUT_ABS, in which case the value given is an
absolute one, not a relative one.
Available since 5.11.

.TP
.B IORING_OP_ACCEPT
Issue the equivalent of an
.BR accept4(2)
system call.
.I fd
must be set to the socket file descriptor,
.I addr
must contain the pointer to the sockaddr structure, and
.I addr2
must contain a pointer to the socklen_t addrlen field. Flags can be passed using
the
.I accept_flags
field. See also
.BR accept4(2)
for the general description of the related system call. Available since 5.5.

If the
.I file_index
field is set to a positive number, the file won't be installed into the
normal file table as usual but will be placed into the fixed file table at index
.I file_index - 1.
In this case, instead of returning a file descriptor, the result will contain
either 0 on success or an error. If the index points to a valid empty slot, the
installation is guaranteed to not fail. If there is already a file in the slot,
it will be replaced, similar to
.B IORING_OP_FILES_UPDATE.
Please note that only io_uring has access to such files and no other syscall
can use them. See
.B IOSQE_FIXED_FILE
and
.B IORING_REGISTER_FILES.

Available since 5.5.

.TP
.B IORING_OP_ASYNC_CANCEL
Attempt to cancel an already issued request.
.I addr
must contain the
.I user_data
field of the request that should be canceled. The cancelation request will
complete with one of the following results codes. If found, the
.I res
field of the cqe will contain 0. If not found,
.I res
will contain -ENOENT. If found and attempted canceled, the
.I res
field will contain -EALREADY. In this case, the request may or may not
terminate. In general, requests that are interruptible (like socket IO) will
get canceled, while disk IO requests cannot be canceled if already started.
Available since 5.5.

.TP
.B IORING_OP_LINK_TIMEOUT
This request must be linked with another request through
.I IOSQE_IO_LINK
which is described below. Unlike
.I IORING_OP_TIMEOUT,
.I IORING_OP_LINK_TIMEOUT
acts on the linked request, not the completion queue. The format of the command
is otherwise like
.I IORING_OP_TIMEOUT,
except there's no completion event count as it's tied to a specific request.
If used, the timeout specified in the command will cancel the linked command,
unless the linked command completes before the timeout. The timeout will
complete with
.I -ETIME
if the timer expired and the linked request was attempted canceled, or
.I -ECANCELED
if the timer got canceled because of completion of the linked request. Like
.B IORING_OP_TIMEOUT
the clock source used is
.B CLOCK_MONOTONIC
Available since 5.5.


.TP
.B IORING_OP_CONNECT
Issue the equivalent of a
.BR connect(2)
system call.
.I fd
must be set to the socket file descriptor,
.I addr
must contain the const pointer to the sockaddr structure, and
.I off
must contain the socklen_t addrlen field. See also
.BR connect(2)
for the general description of the related system call. Available since 5.5.

.TP
.B IORING_OP_FALLOCATE
Issue the equivalent of a
.BR fallocate(2)
system call.
.I fd
must be set to the file descriptor,
.I len
must contain the mode associated with the operation,
.I off
must contain the offset on which to operate, and
.I addr
must contain the length. See also
.BR fallocate(2)
for the general description of the related system call. Available since 5.6.

.TP
.B IORING_OP_FADVISE
Issue the equivalent of a
.BR posix_fadvise(2)
system call.
.I fd
must be set to the file descriptor,
.I off
must contain the offset on which to operate,
.I len
must contain the length, and
.I fadvise_advice
must contain the advice associated with the operation. See also
.BR posix_fadvise(2)
for the general description of the related system call. Available since 5.6.

.TP
.B IORING_OP_MADVISE
Issue the equivalent of a
.BR madvise(2)
system call.
.I addr
must contain the address to operate on,
.I len
must contain the length on which to operate,
and
.I fadvise_advice
must contain the advice associated with the operation. See also
.BR madvise(2)
for the general description of the related system call. Available since 5.6.

.TP
.B IORING_OP_OPENAT
Issue the equivalent of a
.BR openat(2)
system call.
.I fd
is the
.I dirfd
argument,
.I addr
must contain a pointer to the
.I *pathname
argument,
.I open_flags
should contain any flags passed in, and
.I len
is access mode of the file. See also
.BR openat(2)
for the general description of the related system call. Available since 5.6.

If the
.I file_index
field is set to a positive number, the file won't be installed into the
normal file table as usual but will be placed into the fixed file table at index
.I file_index - 1.
In this case, instead of returning a file descriptor, the result will contain
either 0 on success or an error. If the index points to a valid empty slot, the
installation is guaranteed to not fail. If there is already a file in the slot,
it will be replaced, similar to
.B IORING_OP_FILES_UPDATE.
Please note that only io_uring has access to such files and no other syscall
can use them. See
.B IOSQE_FIXED_FILE
and
.B IORING_REGISTER_FILES.

Available since 5.15.

.TP
.B IORING_OP_OPENAT2
Issue the equivalent of a
.BR openat2(2)
system call.
.I fd
is the
.I dirfd
argument,
.I addr
must contain a pointer to the
.I *pathname
argument,
.I len
should contain the size of the open_how structure, and
.I off
should be set to the address of the open_how structure. See also
.BR openat2(2)
for the general description of the related system call. Available since 5.6.

If the
.I file_index
field is set to a positive number, the file won't be installed into the
normal file table as usual but will be placed into the fixed file table at index
.I file_index - 1.
In this case, instead of returning a file descriptor, the result will contain
either 0 on success or an error. If the index points to a valid empty slot, the
installation is guaranteed to not fail. If there is already a file in the slot,
it will be replaced, similar to
.B IORING_OP_FILES_UPDATE.
Please note that only io_uring has access to such files and no other syscall
can use them. See
.B IOSQE_FIXED_FILE
and
.B IORING_REGISTER_FILES.

Available since 5.15.

.TP
.B IORING_OP_CLOSE
Issue the equivalent of a
.BR close(2)
system call.
.I fd
is the file descriptor to be closed. See also
.BR close(2)
for the general description of the related system call. Available since 5.6.
If the
.I file_index
field is set to a positive number, this command can be used to close files
that were direct opened through
.B IORING_OP_OPENAT
,
.B IORING_OP_OPENAT2
, or
.B IORING_OP_ACCEPT
using the io_uring specific direct descriptors. Note that only one of the
descriptor fields may be set. The direct close feature is available since
the 5.15 kernel, where direct descriptors were introduced.

.TP
.B IORING_OP_STATX
Issue the equivalent of a
.BR statx(2)
system call.
.I fd
is the
.I dirfd
argument,
.I addr
must contain a pointer to the
.I *pathname
string,
.I statx_flags
is the
.I flags
argument,
.I len
should be the
.I mask
argument, and
.I off
must contain a pointer to the
.I statxbuf
to be filled in. See also
.BR statx(2)
for the general description of the related system call. Available since 5.6.

.TP
.B IORING_OP_READ
.TP
.B IORING_OP_WRITE
Issue the equivalent of a
.BR pread(2)
or
.BR pwrite(2)
system call.
.I fd
is the file descriptor to be operated on,
.I addr
contains the buffer in question,
.I len
contains the length of the IO operation, and
.I offs
contains the read or write offset. If
.I fd
does not refer to a seekable file,
.I off
must be set to zero or -1. If
.I offs
is set to
.B -1
, the offset will use (and advance) the file position, like the
.BR read(2)
and
.BR write(2)
system calls. These are non-vectored versions of the
.B IORING_OP_READV
and
.B IORING_OP_WRITEV
opcodes. See also
.BR read(2)
and
.BR write(2)
for the general description of the related system call. Available since 5.6.

.TP
.B IORING_OP_SPLICE
Issue the equivalent of a
.BR splice(2)
system call.
.I splice_fd_in
is the file descriptor to read from,
.I splice_off_in
is an offset to read from,
.I fd
is the file descriptor to write to,
.I off
is an offset from which to start writing to. A sentinel value of
.B -1
is used to pass the equivalent of a NULL for the offsets to
.BR splice(2).
.I len
contains the number of bytes to copy.
.I splice_flags
contains a bit mask for the flag field associated with the system call.
Please note that one of the file descriptors must refer to a pipe.
See also
.BR splice(2)
for the general description of the related system call. Available since 5.7.

.TP
.B IORING_OP_TEE
Issue the equivalent of a
.BR tee(2)
system call.
.I splice_fd_in
is the file descriptor to read from,
.I fd
is the file descriptor to write to,
.I len
contains the number of bytes to copy, and
.I splice_flags
contains a bit mask for the flag field associated with the system call.
Please note that both of the file descriptors must refer to a pipe.
See also
.BR tee(2)
for the general description of the related system call. Available since 5.8.

.TP
.B IORING_OP_FILES_UPDATE
This command is an alternative to using
.B IORING_REGISTER_FILES_UPDATE
which then works in an async fashion, like the rest of the io_uring commands.
The arguments passed in are the same.
.I addr
must contain a pointer to the array of file descriptors,
.I len
must contain the length of the array, and
.I off
must contain the offset at which to operate. Note that the array of file
descriptors pointed to in
.I addr
must remain valid until this operation has completed. Available since 5.6.

.TP
.B IORING_OP_PROVIDE_BUFFERS
This command allows an application to register a group of buffers to be used
by commands that read/receive data. Using buffers in this manner can eliminate
the need to separate the poll + read, which provides a convenient point in
time to allocate a buffer for a given request. It's often infeasible to have
as many buffers available as pending reads or receive. With this feature, the
application can have its pool of buffers ready in the kernel, and when the
file or socket is ready to read/receive data, a buffer can be selected for the
operation.
.I fd
must contain the number of buffers to provide,
.I addr
must contain the starting address to add buffers from,
.I len
must contain the length of each buffer to add from the range,
.I buf_group
must contain the group ID of this range of buffers, and
.I off
must contain the starting buffer ID of this range of buffers. With that set,
the kernel adds buffers starting with the memory address in
.I addr,
each with a length of
.I len.
Hence the application should provide
.I len * fd
worth of memory in
.I addr.
Buffers are grouped by the group ID, and each buffer within this group will be
identical in size according to the above arguments. This allows the application
to provide different groups of buffers, and this is often used to have
differently sized buffers available depending on what the expectations are of
the individual request. When submitting a request that should use a provided
buffer, the
.B IOSQE_BUFFER_SELECT
flag must be set, and
.I buf_group
must be set to the desired buffer group ID where the buffer should be selected
from. Available since 5.7.

.TP
.B IORING_OP_REMOVE_BUFFERS
Remove buffers previously registered with
.B IORING_OP_PROVIDE_BUFFERS.
.I fd
must contain the number of buffers to remove, and
.I buf_group
must contain the buffer group ID from which to remove the buffers. Available
since 5.7.

.TP
.B IORING_OP_SHUTDOWN
Issue the equivalent of a
.BR shutdown(2)
system call.
.I fd
is the file descriptor to the socket being shutdown, and
.I len
must be set to the
.I how
argument. No no other fields should be set. Available since 5.11.

.TP
.B IORING_OP_RENAMEAT
Issue the equivalent of a
.BR renameat2(2)
system call.
.I fd
should be set to the
.I olddirfd,
.I addr
should be set to the
.I oldpath,
.I len
should be set to the
.I newdirfd,
.I addr
should be set to the
.I oldpath,
.I addr2
should be set to the
.I newpath,
and finally
.I rename_flags
should be set to the
.I flags
passed in to
.BR renameat2(2).
Available since 5.11.

.TP
.B IORING_OP_UNLINKAT
Issue the equivalent of a
.BR unlinkat2(2)
system call.
.I fd
should be set to the
.I dirfd,
.I addr
should be set to the
.I pathname,
and
.I unlink_flags
should be set to the
.I flags
being passed in to
.BR unlinkat(2).
Available since 5.11.

.TP
.B IORING_OP_MKDIRAT
Issue the equivalent of a
.BR mkdirat2(2)
system call.
.I fd
should be set to the
.I dirfd,
.I addr
should be set to the
.I pathname,
and
.I len
should be set to the
.I mode
being passed in to
.BR mkdirat(2).
Available since 5.15.

.TP
.B IORING_OP_SYMLINKAT
Issue the equivalent of a
.BR symlinkat2(2)
system call.
.I fd
should be set to the
.I newdirfd,
.I addr
should be set to the
.I target
and
.I addr2
should be set to the
.I linkpath
being passed in to
.BR symlinkat(2).
Available since 5.15.

.TP
.B IORING_OP_LINKAT
Issue the equivalent of a
.BR linkat2(2)
system call.
.I fd
should be set to the
.I olddirfd,
.I addr
should be set to the
.I oldpath,
.I len
should be set to the
.I newdirfd,
.I addr2
should be set to the
.I newpath,
and
.I hardlink_flags
should be set to the
.I flags
being passed in to
.BR linkat(2).
Available since 5.15.

.TP
.B IORING_OP_MSG_RING
Send a message to an io_uring.
.I fd
must be set to a file descriptor of a ring that the application has access to,
.I len
can be set to any 32-bit value that the application wishes to pass on, and
.I off
should be set any 64-bit value that the application wishes to send. On the
target ring, a CQE will be posted with the
.I res
field matching the
.I len
set, and a
.I user_data
field matching the
.I off
value being passed in. This request type can be used to either just wake or
interrupt anyone waiting for completions on the target ring, or it can be used
to pass messages via the two fields. Available since 5.18.

.TP
.B IORING_OP_SOCKET
Issue the equivalent of a
.BR socket(2)
system call.
.I fd
must contain the communication domain,
.I off
must contain the communication type,
.I len
must contain the protocol, and
.I rw_flags
is currently unused and must be set to zero. See also
.BR socket(2)
for the general description of the related system call. Available since 5.19.

If the
.I file_index
field is set to a positive number, the file won't be installed into the
normal file table as usual but will be placed into the fixed file table at index
.I file_index - 1.
In this case, instead of returning a file descriptor, the result will contain
either 0 on success or an error. If the index points to a valid empty slot, the
installation is guaranteed to not fail. If there is already a file in the slot,
it will be replaced, similar to
.B IORING_OP_FILES_UPDATE.
Please note that only io_uring has access to such files and no other syscall
can use them. See
.B IOSQE_FIXED_FILE
and
.B IORING_REGISTER_FILES.

Available since 5.19.

.TP
.B IORING_OP_SEND_ZC
Issue the zerocopy equivalent of a
.BR send(2)
system call. Similar to IORING_OP_SEND, but tries to avoid making intermediate
copies of data. Zerocopy execution is not guaranteed and may fall back to
copying. The request may also fail with
.B -EOPNOTSUPP ,
when a protocol doesn't support zerocopy, in which case users are recommended
to use copying sends instead.

The
.I flags
field of the first
.I "struct io_uring_cqe"
may likely contain
.B IORING_CQE_F_MORE ,
which means that there will be a second completion event / notification for
the request, with the
.I user_data
field set to the same value. The user must not modify the data buffer until the
notification is posted. The first cqe follows the usual rules and so its
.I res
field will contain the number of bytes sent or a negative error code. The
notification's
.I res
field will be set to zero and the
.I flags
field will contain
.B IORING_CQE_F_NOTIF .
The two step model is needed because the kernel may hold on to buffers for a
long time, e.g. waiting for a TCP ACK, and having a separate cqe for request
completions allows userspace to push more data without extra delays. Note,
notifications are only responsible for controlling the lifetime of the buffers,
and as such don't mean anything about whether the data has atually been sent
out or received by the other end. Even errored requests may generate a
notification, and the user must check for
.B IORING_CQE_F_MORE
rather than relying on the result.

.I fd
must be set to the socket file descriptor,
.I addr
must contain a pointer to the buffer,
.I len
denotes the length of the buffer to send, and
.I msg_flags
holds the flags associated with the system call. When
.I addr2
is non-zero it points to the address of the target with
.I addr_len
specifying its size, turning the request into a
.BR sendto(2)
system call equivalent.

Available since 6.0.

This command also supports the following modifiers in
.I ioprio:

.PP
.in +12
.B IORING_RECVSEND_POLL_FIRST
If set, io_uring will assume the socket is currently full and attempting to
send data will be unsuccessful. For this case, io_uring will arm internal
poll and trigger a send of the data when there is enough space available.
This initial send attempt can be wasteful for the case where the socket
is expected to be full, setting this flag will bypass the initial send
attempt and go straight to arming poll. If poll does indicate that data can
be sent, the operation will proceed.

.B IORING_RECVSEND_FIXED_BUF
If set, instructs io_uring to use a pre-mapped buffer. The
.I buf_index
field should contain an index into an array of fixed buffers. See
.BR io_uring_register (2)
for details on how to setup a context for fixed buffer I/O.
.EE
.in
.PP

.TP
.B IORING_OP_WAITID
Issue the equivalent of a
.BR waitid(2)
system call.
.I len
must contain the idtype being queried/waited for and
.I fd
must contain the 'pid' (or id) being waited for.
.I file_index
is the 'options' being set (the child state changes to wait for).
.I addr2
is a pointer to siginfo_t, if any, being filled in. See also
.BR waitid(2)
for the general description of the related system call. Available since 6.5.

.PP
The
.I flags
field is a bit mask. The supported flags are:
.TP
.B IOSQE_FIXED_FILE
When this flag is specified,
.I fd
is an index into the files array registered with the io_uring instance (see the
.B IORING_REGISTER_FILES
section of the
.BR io_uring_register (2)
man page). Note that this isn't always available for all commands. If used on
a command that doesn't support fixed files, the SQE will error with
.B -EBADF.
Available since 5.1.
.TP
.B IOSQE_IO_DRAIN
When this flag is specified, the SQE will not be started before previously
submitted SQEs have completed, and new SQEs will not be started before this
one completes. Available since 5.2.
.TP
.B IOSQE_IO_LINK
When this flag is specified, the SQE forms a link with the next SQE in the
submission ring. That next SQE will not be started before the previous request
completes. This, in effect, forms a chain of SQEs, which can be arbitrarily
long. The tail of the chain is denoted by the first SQE that does not have this
flag set. Chains are not supported across submission boundaries. Even if the
last SQE in a submission has this flag set, it will still terminate the current
chain. This flag has no effect on previous SQE submissions, nor does it impact
SQEs that are outside of the chain tail. This means that multiple chains can be
executing in parallel, or chains and individual SQEs. Only members inside the
chain are serialized. A chain of SQEs will be broken, if any request in that
chain ends in error. io_uring considers any unexpected result an error. This
means that, eg, a short read will also terminate the remainder of the chain.
If a chain of SQE links is broken, the remaining unstarted part of the chain
will be terminated and completed with
.B -ECANCELED
as the error code. Available since 5.3.
.TP
.B IOSQE_IO_HARDLINK
Like IOSQE_IO_LINK, but it doesn't sever regardless of the completion result.
Note that the link will still sever if we fail submitting the parent request,
hard links are only resilient in the presence of completion results for
requests that did submit correctly. IOSQE_IO_HARDLINK implies IOSQE_IO_LINK.
Available since 5.5.
.TP
.B IOSQE_ASYNC
Normal operation for io_uring is to try and issue an sqe as non-blocking first,
and if that fails, execute it in an async manner. To support more efficient
overlapped operation of requests that the application knows/assumes will
always (or most of the time) block, the application can ask for an sqe to be
issued async from the start. Available since 5.6.
.TP
.B IOSQE_BUFFER_SELECT
Used in conjunction with the
.B IORING_OP_PROVIDE_BUFFERS
command, which registers a pool of buffers to be used by commands that read
or receive data. When buffers are registered for this use case, and this
flag is set in the command, io_uring will grab a buffer from this pool when
the request is ready to receive or read data. If successful, the resulting CQE
will have
.B IORING_CQE_F_BUFFER
set in the flags part of the struct, and the upper
.B IORING_CQE_BUFFER_SHIFT
bits will contain the ID of the selected buffers. This allows the application
to know exactly which buffer was selected for the operation. If no buffers
are available and this flag is set, then the request will fail with
.B -ENOBUFS
as the error code. Once a buffer has been used, it is no longer available in
the kernel pool. The application must re-register the given buffer again when
it is ready to recycle it (eg has completed using it). Available since 5.7.
.TP
.B IOSQE_CQE_SKIP_SUCCESS
Don't generate a CQE if the request completes successfully. If the request
fails, an appropriate CQE will be posted as usual and if there is no
.B IOSQE_IO_HARDLINK,
CQEs for all linked requests will be omitted. The notion of failure/success is
opcode specific and is the same as with breaking chains of
.B IOSQE_IO_LINK.
One special case is when the request has a linked timeout, then the CQE
generation for the linked timeout is decided solely by whether it has
.B IOSQE_CQE_SKIP_SUCCESS
set, regardless whether it timed out or was canceled. In other words, if a
linked timeout has the flag set, it's guaranteed to not post a CQE.

The semantics are chosen to accommodate several use cases. First, when all but
the last request of a normal link without linked timeouts are marked with the
flag, only one CQE per link is posted. Additionally, it enables suppression of
CQEs in cases where the side effects of a successfully executed operation is
enough for userspace to know the state of the system. One such example would
be writing to a synchronisation file.

This flag is incompatible with
.B IOSQE_IO_DRAIN.
Using both of them in a single ring is undefined behavior, even when they are
not used together in a single request. Currently, after the first request with
.B IOSQE_CQE_SKIP_SUCCESS,
all subsequent requests marked with drain will be failed at submission time.
Note that the error reporting is best effort only, and restrictions may change
in the future.

Available since 5.17.

.PP
.I ioprio
specifies the I/O priority.  See
.BR ioprio_get (2)
for a description of Linux I/O priorities.

.I fd
specifies the file descriptor against which the operation will be
performed, with the exception noted above.

If the operation is one of
.B IORING_OP_READ_FIXED
or
.BR IORING_OP_WRITE_FIXED ,
.I addr
and
.I len
must fall within the buffer located at
.I buf_index
in the fixed buffer array.  If the operation is either
.B IORING_OP_READV
or
.BR IORING_OP_WRITEV ,
then
.I addr
points to an iovec array of
.I len
entries.

.IR rw_flags ,
specified for read and write operations, contains a bitwise OR of
per-I/O flags, as described in the
.BR preadv2 (2)
man page.

The
.I fsync_flags
bit mask may contain either 0, for a normal file integrity sync, or
.B IORING_FSYNC_DATASYNC
to provide data sync only semantics.  See the descriptions of
.B O_SYNC
and
.B O_DSYNC
in the
.BR open (2)
manual page for more information.

The bits that may be set in
.I poll_events
are defined in \fI<poll.h>\fP, and documented in
.BR poll (2).

.I user_data
is an application-supplied value that will be copied into
the completion queue entry (see below).
.I buf_index
is an index into an array of fixed buffers, and is only valid if fixed
buffers were registered.
.I personality
is the credentials id to use for this operation. See
.BR io_uring_register(2)
for how to register personalities with io_uring. If set to 0, the current
personality of the submitting task is used.
.PP
Once the submission queue entry is initialized, I/O is submitted by
placing the index of the submission queue entry into the tail of the
submission queue.  After one or more indexes are added to the queue,
and the queue tail is advanced, the
.BR io_uring_enter (2)
system call can be invoked to initiate the I/O.

Completions use the following data structure:
.PP
.in +4n
.EX
/*
 * IO completion data structure (Completion Queue Entry)
 */
struct io_uring_cqe {
    __u64    user_data; /* sqe->data submission passed back */
    __s32    res;       /* result code for this event */
    __u32    flags;
};
.EE
.in
.PP
.I user_data
is copied from the field of the same name in the submission queue
entry.  The primary use case is to store data that the application
will need to access upon completion of this particular I/O.  The
.I flags
is used for certain commands, like
.B IORING_OP_POLL_ADD
or in conjunction with
.B IOSQE_BUFFER_SELECT
or
.B IORING_OP_MSG_RING,
, see those entries for details.
.I res
is the operation-specific result, but io_uring-specific errors
(e.g. flags or opcode invalid) are returned through this field.
They are described in section
.B CQE ERRORS.
.PP
For read and write opcodes, the
return values match
.I errno
values documented in the
.BR preadv2 (2)
and
.BR pwritev2 (2)
man pages, with
.I
res
holding the equivalent of
.I -errno
for error cases, or the transferred number of bytes in case the operation
is successful. Hence both error and success return can be found in that
field in the CQE. For other request types, the return values are documented
in the matching man page for that type, or in the opcodes section above for
io_uring-specific opcodes.
.PP
.SH RETURN VALUE
.BR io_uring_enter (2)
returns the number of I/Os successfully consumed.  This can be zero
if
.I to_submit
was zero or if the submission queue was empty. Note that if the ring was
created with
.B IORING_SETUP_SQPOLL
specified, then the return value will generally be the same as
.I to_submit
as submission happens outside the context of the system call.

The errors related to a submission queue entry will be returned through a
completion queue entry (see section
.B CQE ERRORS),
rather than through the system call itself.

Errors that occur not on behalf of a submission queue entry are returned via the
system call directly. On such an error, a negative error code is returned. The
caller should not rely on
.I errno
variable.
.PP
.SH ERRORS
These are the errors returned by
.BR io_uring_enter (2)
system call.
.TP
.B EAGAIN
The kernel was unable to allocate memory for the request, or otherwise ran out
of resources to handle it. The application should wait for some completions and
try again.
.TP
.B EBADF
.I fd
is not a valid file descriptor.
.TP
.B EBADFD
.I fd
is a valid file descriptor, but the io_uring ring is not in the right state
(enabled). See
.BR io_uring_register (2)
for details on how to enable the ring.
.TP
.B EBADR
At least one CQE was dropped even with the
.B IORING_FEAT_NODROP
feature, and there are no otherwise available CQEs. This clears the error state
and so with no other changes the next call to
.BR io_uring_setup (2)
will not have this error. This error should be extremely rare and indicates the
machine is running critically low on memory and. It may be reasonable for the
application to terminate running unless it is able to safely handle any CQE
being lost.
.TP
.B EBUSY
If the
.B IORING_FEAT_NODROP
feature flag is set, then
.B EBUSY
will be returned if there were overflow entries,
.B IORING_ENTER_GETEVENTS
flag is set and not all of the overflow entries were able to be flushed to
the CQ ring.

Without
.B IORING_FEAT_NODROP
the application is attempting to overcommit the number of requests it can have
pending. The application should wait for some completions and try again. May
occur if the application tries to queue more requests than we have room for in
the CQ ring, or if the application attempts to wait for more events without
having reaped the ones already present in the CQ ring.
.TP
.B EEXIST
The thread submitting the work is invalid.
.TP
.B EINVAL
Some bits in the
.I flags
argument are invalid.
.TP
.B EFAULT
An invalid user space address was specified for the
.I sig
argument.
.TP
.B ENXIO
The io_uring instance is in the process of being torn down.
.TP
.B EOPNOTSUPP
.I fd
does not refer to an io_uring instance.
.TP
.B EINTR
The operation was interrupted by a delivery of a signal before it could
complete; see
.BR signal(7).
Can happen while waiting for events with
.B IORING_ENTER_GETEVENTS.

.SH CQE ERRORS
These io_uring-specific errors are returned as a negative value in the
.I res
field of the completion queue entry.
.TP
.B EACCES
The
.I flags
field or
.I opcode
in a submission queue entry is not allowed due to registered restrictions.
See
.BR io_uring_register (2)
for details on how restrictions work.
.TP
.B EBADF
The
.I fd
field in the submission queue entry is invalid, or the
.B IOSQE_FIXED_FILE
flag was set in the submission queue entry, but no files were registered
with the io_uring instance.
.TP
.B EFAULT
buffer is outside of the process' accessible address space
.TP
.B EFAULT
.B IORING_OP_READ_FIXED
or
.B IORING_OP_WRITE_FIXED
was specified in the
.I opcode
field of the submission queue entry, but either buffers were not
registered for this io_uring instance, or the address range described
by
.I addr
and
.I len
does not fit within the buffer registered at
.IR buf_index .
.TP
.B EINVAL
The
.I flags
field or
.I opcode
in a submission queue entry is invalid.
.TP
.B EINVAL
The
.I buf_index
member of the submission queue entry is invalid.
.TP
.B EINVAL
The
.I personality
field in a submission queue entry is invalid.
.TP
.B EINVAL
.B IORING_OP_NOP
was specified in the submission queue entry, but the io_uring context
was setup for polling
.RB ( IORING_SETUP_IOPOLL
was specified in the call to io_uring_setup).
.TP
.B EINVAL
.B IORING_OP_READV
or
.B IORING_OP_WRITEV
was specified in the submission queue entry, but the io_uring instance
has fixed buffers registered.
.TP
.B EINVAL
.B IORING_OP_READ_FIXED
or
.B IORING_OP_WRITE_FIXED
was specified in the submission queue entry, and the
.I buf_index
is invalid.
.TP
.B EINVAL
.BR IORING_OP_READV ,
.BR IORING_OP_WRITEV ,
.BR IORING_OP_READ_FIXED ,
.B IORING_OP_WRITE_FIXED
or
.B IORING_OP_FSYNC
was specified in the submission queue entry, but the io_uring instance
was configured for IOPOLLing, or any of
.IR addr ,
.IR ioprio ,
.IR off ,
.IR len ,
or
.I buf_index
was set in the submission queue entry.
.TP
.B EINVAL
.B IORING_OP_POLL_ADD
or
.B IORING_OP_POLL_REMOVE
was specified in the
.I opcode
field of the submission queue entry, but the io_uring instance was
configured for busy-wait polling
.RB ( IORING_SETUP_IOPOLL ),
or any of
.IR ioprio ,
.IR off ,
.IR len ,
or
.I buf_index
was non-zero in the submission queue entry.
.TP
.B EINVAL
.B IORING_OP_POLL_ADD
was specified in the
.I opcode
field of the submission queue entry, and the
.I addr
field was non-zero.
.TP
.B EOPNOTSUPP
.I opcode
is valid, but not supported by this kernel.
.TP
.B EOPNOTSUPP
.B IOSQE_BUFFER_SELECT
was set in the
.I flags
field of the submission queue entry, but the
.I opcode
doesn't support buffer selection.