xref: /dflybsd-src/share/man/man9/altq.9 (revision f09de992f9332956876cbc1cd9cd4cf4f2771904)

.\"	$OpenBSD: altq.9,v 1.4 2001/07/12 12:41:42 itojun Exp $
.\"	$NetBSD: altq.9,v 1.14 2007/06/24 19:26:58 rumble Exp $
.\"
.\" Copyright (C) 2001
.\" Sony Computer Science Laboratories Inc.  All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\"
.\" THIS SOFTWARE IS PROVIDED BY SONY CSL AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED.  IN NO EVENT SHALL SONY CSL OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.Dd November 18, 2007
.Dt ALTQ 9
.Os
.Sh NAME
.Nm ALTQ
.Nd kernel interfaces for manipulating output queues on network interfaces
.Sh SYNOPSIS
.In sys/types.h
.In sys/socket.h
.In net/if.h
.In net/ifq_var.h
.Ft static int
.Fn ifq_enqueue "struct ifaltq *_ifq" "struct mbuf *_m" "struct altq_pktattr *_pa"
.Ft static int
.Fn ifq_handoff "struct ifnet *_ifp" "struct mbuf *_m" "struct altq_pktattr *_pa"
.Ft struct mbuf *
.Fn ifq_dequeue "struct ifaltq *_ifq" "struct mbuf *_mpolled"
.Ft struct mbuf *
.Fn ifq_poll "struct ifaltq *_ifq"
.Ft void
.Fn ifq_purge "struct ifaltq *_ifq"
.Ft void
.Fn ifq_classify "struct ifaltq *_ifq" "struct mbuf *_m" "uint8_t _af" "struct altq_pktattr *_pa"
.Ft int
.Fn ifq_is_empty "struct ifaltq *_ifq"
.Ft void
.Fn ifq_set_maxlen "struct ifaltq *_ifq" "int _len"
.Ft void
.Fn ifq_set_ready "struct ifaltq *_ifq"
.Ft int
.Fn ifq_is_enabled "struct ifaltq *_ifq"
.Sh DESCRIPTION
The
.Nm
system is a framework to manage queueing disciplines on network
interfaces.
.Nm
introduces new functions to manipulate output queues.
The output queue functions are used to abstract queue operations and not to
touch the internal fields of the output queue structure.
.Pp
.Fn ifq_enqueue
and
.Fn ifq_handoff
enqueue a packet
.Fa _m
to the queue
.Fa _ifq .
The underlying queueing discipline may discard the packet.
They return 0 on success, or
.Er ENOBUFS
if the packet is discarded.
The packet pointed to by
.Fa _m
will be freed by the device driver on success or by the queueing discipline on
failure, so that the caller should not touch
.Fa _m
after calling
.Fn ifq_enqueue .
.Fn ifq_handoff
combines the enqueue operation with statistic generation and calls
.Fn if_start
upon successful enqueue to initiate the actual send.
.Pp
.Fn ifq_dequeue
dequeues a packet from the queue.
It returns the dequeued packet, or
.Dv NULL
if no packet is dequeued.
The caller must always check the return value
since a non-empty queue could return
.Dv NULL
under rate-limiting.
.Pp
.Fn ifq_poll
returns the next packet without removing it from the queue.
It is guaranteed by the underlying queueing discipline that
.Fn ifq_dequeue
immediately after
.Fn ifq_poll
returns the same packet.
.Pp
.Fn ifq_purge
discards all the packets in the queue.
The purge operation is needed since a non-work conserving queue cannot be
emptied by a dequeue loop.
.Pp
.Fn ifq_classify
classifies a packet to a scheduling class, and returns the result in
.Fa _pa .
.Pp
.Fn ifq_is_empty
can be used to check if the queue is empty.
Note that
.Fn ifq_dequeue
could still return
.Dv NULL
if the queueing discipline is non-work conserving.
.Pp
.Fn ifq_set_maxlen
sets the queue length limit to the default FIFO queue.
.Pp
.Fn ifq_set_ready
sets a flag to indicate that this driver is converted to use the new macros.
.Nm
can be enabled only on interfaces with this flag.
.Pp
.Fn ifq_is_enabled
returns 1 if
.Nm
is enabled for the queue, 0 if not.
.Sh QUEUEING DISCIPLINES
Queueing disciplines need to maintain
.Fa ifq_len
.Po
used by
.Fn ifq_is_empty
.Pc .
Queueing disciplines also need to guarantee that the same mbuf is returned if
.Fn ifq_dequeue
is called immediately after
.Fn ifq_poll .
.Sh SEE ALSO
.Xr altq 4 ,
.Xr pf 4 ,
.Xr pf.conf 5 ,
.Xr pfctl 8
.Sh HISTORY
The
.Nm
system first appeared in March 1997 and was imported into
.Dx 1.1 .