AIO_READ(2)

📦 version 15.1, in freebsd 14.5, freebsd 15.1

💖 AIO_READ(2) 💖 — asynchronous read from a file (REALTIME)

okay bestie, welcome to the System Calls Manual ✨ this one is about aio_read, aio_read2, and aio_readv — asynchronous reads, REALTIME energy, let’s get into it 💅

🎀 NAME

aio_read, aio_read2, aio_readv — asynchronous read from a file (REALTIME)

🌸 LIBRARY

Standard C Library (libc, -lc)

🦋 SYNOPSIS

#include <aio.h>

int
aio_read(struct aiocb *iocb);

int
aio_read2(struct aiocb *iocb, int flags);

#include <sys/uio.h>

int
aio_readv(struct aiocb *iocb);

✨ DESCRIPTION

The aio_read(), aio_read2(), and aio_readv() system calls let the calling process read from the descriptor iocb->aio_fildes 🫶 The syscalls return immediately after the read request has been enqueued to the descriptor; the read may or may not have completed at the time the call returns. It’s giving “I sent the request, don’t wait up” energy.

For the aio_read() and aio_readv() calls, the read begins at the offset iocb->aio_offset.

The aio_read() call will read iocb->aio_nbytes into the buffer pointed to by iocb->aio_buf, whereas aio_readv() reads the data into the iocb->aio_iovcnt buffers specified by the members of the iocb->aio_iov array. For aio_readv() the iovec structure is defined in readv(2) 🦋

The aio_read2() call takes the flags argument. If flags is passed as zero, the call behaves identically to aio_read(). The following flags can be specified by logical or:

  • AIO_OP2_FOFFSET — the read occurs at the file descriptor offset, which is advanced by the operation as done by the read(2) syscall. The iocb->aio_offset field is ignored 💅
  • AIO_OP2_VECTORED — similar to aio_readv(), the read buffers are specified by the aiocb->aio_iov array.

The iocb pointer may be subsequently used as an argument to aio_return() and aio_error() in order to determine return or error status for the enqueued operation while it is in progress.

If the request could not be enqueued (generally due to invalid arguments), then the call returns without having enqueued the request.

If the request is successfully enqueued, the value of iocb->aio_offset can be modified during the request as context, so this value must not be referenced after the request is enqueued. No cap, don’t touch it after enqueueing 🙈

The iocb->aio_sigevent structure can be used to request notification of the operation’s completion as described in aio(4).

💅 RESTRICTIONS

Bestie, these are the hard rules — the vibe check is strict:

  • The Asynchronous I/O Control Block structure pointed to by iocb and the buffer that the iocb->aio_buf member of that structure references must remain valid until the operation has completed.
  • The asynchronous I/O control buffer iocb should be zeroed before the system calls to avoid passing bogus context information to the kernel.
  • Modifications of the Asynchronous I/O Control Block structure or the buffer contents are not allowed while the request is queued.
  • If the file offset in iocb->aio_offset is past the offset maximum for iocb->aio_fildes, no I/O will occur.

🫶 RETURN VALUES

The aio_read(), aio_read2(), and aio_readv() functions return the value 0 if successful; otherwise the value -1 is returned and the global variable errno is set to indicate the error.

💀 DIAGNOSTICS

None.

That’s it. No diagnostics. The section is literally just None 😭

😭 ERRORS

The aio_read(), aio_read2(), and aio_readv() system calls will fail if:

  • [EAGAIN] — the request was not queued because of system resource limitations.
  • [EFAULT] — part of aio_iov points outside the process’s allocated address space.
  • [EINVAL] — the asynchronous notification method in iocb->aio_sigevent.sigev_notify is invalid or not supported.
  • [EOPNOTSUPP] — asynchronous read operations on the file descriptor iocb->aio_fildes are unsafe and unsafe asynchronous I/O operations are disabled.

The following conditions may be synchronously detected when the aio_read() or aio_readv() system call is made, or asynchronously, at any time thereafter. If they are detected at call time, the calls return -1 and set errno appropriately; otherwise the aio_return() system call must be called, and will return -1, and aio_error() must be called to determine the actual value that would have been returned in errno.

  • [EBADF] — the iocb->aio_fildes argument is invalid.
  • [EINVAL] — the offset iocb->aio_offset is not valid, the priority specified by iocb->aio_reqprio is not a valid priority, or the number of bytes specified by iocb->aio_nbytes is not valid.
  • [EOVERFLOW] — the file is a regular file, iocb->aio_nbytes is greater than zero, the starting offset in iocb->aio_offset is before the end of the file, but is at or beyond the iocb->aio_fildes offset maximum.

If the request is successfully enqueued, but subsequently cancelled or an error occurs, the value returned by the aio_return() system call is per the read(2) system call, and the value returned by the aio_error() system call is either one of the error returns from the read(2) system call, or one of:

  • [EBADF] — the iocb->aio_fildes argument is invalid for reading.
  • [ECANCELED] — the request was explicitly cancelled via a call to aio_cancel().
  • [EINVAL] — the offset iocb->aio_offset would be invalid.

🔥 SEE ALSO

👑 STANDARDS

The aio_read() system call is expected to conform to the IEEE Std 1003.1 (“POSIX.1”) standard. The aio_read2() and aio_readv() system calls are FreeBSD extensions, and should not be used in portable code. Periodt — keep those out of your portable stuff 💅

🌈 HISTORY

The aio_read() system call first appeared in FreeBSD 3.0. The aio_readv() system call first appeared in FreeBSD 13.0. The aio_read2() system call first appeared in FreeBSD 14.1. Iconic timeline, honestly ✨

🍓 AUTHORS

This manual page was written by Terry Lambert <terry@whistle.com>.

🧚 BUGS

Invalid information in iocb->_aiocb_private may confuse the kernel. Like, the kernel sees nonsense and goes ??? 💀

Sign-off vibe check: this page is from Debian, dated March 27, 2025, and it’s the AIO_READ(2) manual page. Byeee, go read asynchronously 💖✨

📦 where this page comes from