man/io_uring_setup_buf_ring.3

.\" Copyright (C) 2022 Jens Axboe <axboe@kernel.dk>
.\"
.\" SPDX-License-Identifier: LGPL-2.0-or-later
.\"
.TH io_uring_setup_buf_ring 3 "Mar 07, 2023" "liburing-2.4" "liburing Manual"
.SH NAME
io_uring_setup_buf_ring \- setup and register buffer ring for provided buffers
.SH SYNOPSIS
.nf
.B #include <liburing.h>
.PP
.BI "struct io_uring_buf_ring *io_uring_setup_buf_ring(struct io_uring *" ring ",
.BI "                            unsigned int " nentries ",
.BI "                            int " bgid ",
.BI "                            unsigned int " flags ",
.BI "                            int *" ret ");"
.BI "
.fi
.SH DESCRIPTION
.PP
The
.BR io_uring_setup_buf_ring (3)
function registers a shared buffer ring to be used with provided buffers. For
the request types that support it, provided buffers are given to the ring and
one is selected by a request if it has
.B IOSQE_BUFFER_SELECT
set in the SQE
.IR flags ,
when the request is ready to receive data. This allows both clear ownership
of the buffer lifetime, and a way to have more read/receive type of operations
in flight than buffers available.

The
.I ring
argument must pointer to the ring for which the provided buffer ring is being
registered,
.I nentries
is the number of entries requested in the buffer ring. This argument must be
a power-of 2 in size.
.I bgid
is the chosen buffer group ID,
.I flags
are modifier flags for the operation, and
.I *ret
is is a pointer to an integer for the error value if any part of the ring
allocation and registration fails.

The
.I flags
argument can be set to one of the following values:
.TP
.B IORING_REGISTER_PBUF_RING
The buffers in this ring can be incrementally consumed. With partial
consumption, each completion of a given buffer ID will continue where the
previous one left off, or from the start if no completions have been seen yet.
When more completions should be expected for a given buffer ID, the CQE will
have
.B IORING_CQE_F_BUF_MORE
set in the
.I flags
member.
.PP

Under the covers, this function uses
.BR io_uring_register_buf_ring (3)
to register the ring, and handles the allocation of the ring rather than
letting the application open code it.

To unregister and free a buffer group ID setup with this function, the
application must call
.BR io_uring_free_buf_ring (3) .

Available since 5.19.

.SH RETURN VALUE
On success
.BR io_uring_setup_buf_ring (3)
returns a pointer to the buffer ring. On failure it returns
.BR NULL
and sets
.I *ret
to -errno.
.SH NOTES
Note that even if the kernel supports this feature, registering a provided
buffer ring may still fail with
.B -EINVAL
if the host is a 32-bit architecture and the memory being passed in resides in
high memory.
.SH SEE ALSO
.BR io_uring_register_buf_ring (3),
.BR io_uring_buf_ring_init (3),
.BR io_uring_buf_ring_add (3),
.BR io_uring_buf_ring_advance (3),
.BR io_uring_buf_ring_cq_advance (3)