Home | History | Annotate | Line # | Download | only in man9 
 $NetBSD: pfil.9,v 1.40 2022/01/15 17:54:01 wiz Exp $

 Copyright (c) 1996 Matthew R. Green
 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 THE AUTHOR ``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 THE AUTHOR 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 January 15, 2022
.Dt PFIL 9
.Os
.Sh NAME
.Nm pfil ,
.Nm pfil_head_create ,
.Nm pfil_head_destroy ,
.Nm pfil_head_get ,
.Nm pfil_hook_get ,
.Nm pfil_add_hook ,
.Nm pfil_remove_hook ,
.Nm pfil_run_hooks ,
.Nm pfil_add_ihook ,
.Nm pfil_remove_ihook ,
.Nm pfil_run_addrhooks ,
.Nm pfil_run_ifhooks
.Nd packet filter interface
.Sh SYNOPSIS
n sys/param.h n sys/mbuf.h n net/if.h n net/pfil.h .Ft pfil_head_t *
.Fn pfil_head_create "int type" "void *key"
.Ft int
.Fn pfil_head_destroy "pfil_head_t *ph"
.Ft pfil_head_t *
.Fn pfil_head_get "int type" "void *key"
.Ft struct packet_filter_hook *
.Fn pfil_hook_get "int dir" "pfil_head_t *ph"
.Ft int
.Fn pfil_add_hook "pfil_func_t func" "void *arg" "int flags" "pfil_head_t *ph"
.Ft int
.Fn pfil_remove_hook "pfil_func_t func" "void *arg" "int flags" "pfil_head_t *ph"
.Ft int
.Fn (*func) "void *arg" "struct mbuf **mp" "struct ifnet *" "int dir"
.Ft int
.Fn pfil_run_hooks "pfil_head_t *ph" "struct mbuf **mp" "struct ifnet *ifp" "int dir"
.Ft int
.Fn pfil_add_ihook "pfil_ifunc_t ifunc" "void *arg" "int flags" "pfil_head_t *ph"
.Ft int
.Fn pfil_remove_ihook "pfil_ifunc_t ifunc" "void *arg" "int flags" "pfil_head_t *ph"
.Ft void
.Fn (*ifunc) "void *arg" "unsigned long cmd" "void *ptr"
.Ft void
.Fn pfil_run_addrhooks "pfil_head_t *ph" "unsigned long" "struct ifaddr *ifa"
.Ft void
.Fn pfil_run_ifhooks "pfil_head_t *ph" "unsigned long" "struct ifnet *ifp"
.Sh DESCRIPTION
The
.Nm
framework allows for a specified function to be invoked for every
incoming or outgoing packet for a particular network I/O stream.
These hooks may be used to implement a firewall or perform packet
transformations.

p
Packet filtering points are created with
.Fn pfil_head_create .
Filtering points are identified by a
data link
.Vt ( int )
.Fa type
and a
.Vt ( void * )
.Fa key .
If a packet filtering point already exists for that data link
.Fa type
and
.Fa key
then the
.Fn pfil_head_create
function returns
.Dv NULL .
Packet filters use the
.Fn pfil_head_get
function specifying the data link
.Fa type
and the
.Fa key
to look up the filtering point with which they register themselves.
The
.Fa key
is unique to the filtering point.
The data link
.Fa type
is a
.Xr bpf 4
.Dv DLT_ Ns Ar type
constant indicating what kind of header is present on the packet
at the filtering point.
Filtering points may be destroyed with the
.Fn pfil_head_destroy
function.

p
Packet filters register/unregister themselves with a filtering point
with the
.Fn pfil_add_hook
and
.Fn pfil_remove_hook
functions, respectively.
The head is looked up using the
.Fn pfil_head_get
function, which takes the data link
.Fa type
and the
.Fa key
that the packet filter expects.
Filters may provide an argument to be passed to the filter when
invoked on a packet.

p
When a filter is invoked, the packet appears just as if it
.Dq came off the wire .
That is, all protocol fields are in network byte order.
The filter is called with its specified argument, the pointer to the
pointer to the mbuf containing the packet, the pointer to the network
interface that the packet is traversing, and the direction (either
.Dv PFIL_IN
or
.Dv PFIL_OUT ,
see also below) that the packet is traveling.
The filter may change which mbuf the
.Vt "mbuf **"
argument references.
The filter returns an errno if the packet processing is to stop, or 0
if the processing is to continue.
If the packet processing is to stop, it is the responsibility of the
filter to free the packet.

p
The
.Fa flags
parameter, used in the
.Fn pfil_add_hook
and
.Fn pfil_remove_hook
functions, indicates when the filter should be called.
The flags are:

p
l -tag -offset indent -width ".Dv PFIL_ALL" -compact t Dv PFIL_IN call me on incoming packets
t Dv PFIL_OUT call me on outgoing packets
t Dv PFIL_ALL call me on all of the above
.El

p
By the same token, event handlers register/unregister themselves
with the
.Fn pfil_add_ihook
and
.Fn pfil_remove_ihook
functions, respectively.
The event handler is called with its specified argument, the event id
(either
.Dv PFIL_IFNET_ATTACH
or
.Dv PFIL_IFNET_DETACH ,
see also below) or ioctl number, and the pointer
to the network interface or the pointer to the ifaddr.

p
The
.Fa flags
parameter, used in the
.Fn pfil_add_ihook
and
.Fn pfil_remove_ihook
functions, indicates when the filter should be called.
The flags are:

p
l -tag -offset indent -width ".Dv PFIL_IFADDR" -compact t Dv PFIL_IFADDR call me on interface reconfig
.Fa ( cmd
is ioctl #)
t Dv PFIL_IFNET call me on interface attach/detach
.Fa ( cmd
is either
.Dv PFIL_IFNET_ATTACH
or
.Dv PFIL_IFNET_DETACH )
.El
.Sh SEE ALSO
.Xr bpf 4
.Sh HISTORY
The
.Nm
interface first appeared in
.Nx 1.3 .
The
.Nm
input and output lists were originally implemented as
n sys/queue.h .Dv LIST
structures;
however this was changed in
.Nx 1.4
to
.Dv TAILQ
structures.
This change was to allow the input and output filters to be processed in
reverse order, to allow the same path to be taken, in or out of the kernel.

p
The
.Nm
interface was changed in 1.4T to accept a 3rd parameter to both
.Fn pfil_add_hook
and
.Fn pfil_remove_hook ,
introducing the capability of per-protocol filtering.
This was done primarily in order to support filtering of IPv6.

p
In 1.5K, the
.Nm
framework was changed to work with an arbitrary number of filtering points,
as well as be less IP-centric.

p
.Fn pfil_add_ihook
and
.Fn pfil_remove_ihook
were added in
.Nx 8.0 .
.Sh AUTHORS
.An -nosplit
The
.Nm
interface was designed and implemented by
.An Matthew R. Green ,
with help from
.An Darren Reed ,
.An Jason R. Thorpe ,
and
.An Charles M. Hannum .
.An Darren Reed
added support for IPv6 in addition to IPv4.
.An Jason R. Thorpe
added support for multiple hooks and other clean up.
.Sh BUGS
The current
.Nm
implementation will need changes to suit a threaded kernel model.