2005-01-07 01:45:51 +00:00
|
|
|
/*-
|
1997-01-03 19:50:26 +00:00
|
|
|
* Copyright (c) 1982, 1986, 1989, 1993
|
|
|
|
|
* The Regents of the University of California. 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.
|
|
|
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
|
* without specific prior written permission.
|
|
|
|
|
*
|
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS 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 THE REGENTS 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.
|
|
|
|
|
*
|
|
|
|
|
* From: @(#)if.h 8.1 (Berkeley) 6/10/93
|
1999-08-28 01:08:13 +00:00
|
|
|
* $FreeBSD$
|
1997-01-03 19:50:26 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#ifndef _NET_IF_VAR_H_
|
|
|
|
|
#define _NET_IF_VAR_H_
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Structures defining a network interface, providing a packet
|
|
|
|
|
* transport mechanism (ala level 0 of the PUP protocols).
|
|
|
|
|
*
|
|
|
|
|
* Each interface accepts output datagrams of a specified maximum
|
|
|
|
|
* length, and provides higher level routines with input datagrams
|
|
|
|
|
* received from its medium.
|
|
|
|
|
*
|
|
|
|
|
* Output occurs when the routine if_output is called, with three parameters:
|
|
|
|
|
* (*ifp->if_output)(ifp, m, dst, rt)
|
|
|
|
|
* Here m is the mbuf chain to be sent and dst is the destination address.
|
|
|
|
|
* The output routine encapsulates the supplied datagram if necessary,
|
|
|
|
|
* and then transmits it on its medium.
|
|
|
|
|
*
|
|
|
|
|
* On input, each interface unwraps the data received by it, and either
|
2003-01-01 18:49:04 +00:00
|
|
|
* places it on the input queue of an internetwork datagram routine
|
1997-01-03 19:50:26 +00:00
|
|
|
* and posts the associated software interrupt, or passes the datagram to a raw
|
|
|
|
|
* packet input routine.
|
|
|
|
|
*
|
|
|
|
|
* Routines exist for locating interfaces by their addresses
|
2003-01-01 18:49:04 +00:00
|
|
|
* or for locating an interface on a certain network, as well as more general
|
1997-01-03 19:50:26 +00:00
|
|
|
* routing and gateway routines maintaining information used to locate
|
|
|
|
|
* interfaces. These routines live in the files if.c and route.c
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#ifdef __STDC__
|
|
|
|
|
/*
|
|
|
|
|
* Forward structure declarations for function prototypes [sic].
|
|
|
|
|
*/
|
|
|
|
|
struct mbuf;
|
2001-09-12 08:38:13 +00:00
|
|
|
struct thread;
|
1997-01-03 19:50:26 +00:00
|
|
|
struct rtentry;
|
2001-10-17 18:07:05 +00:00
|
|
|
struct rt_addrinfo;
|
1997-01-03 19:50:26 +00:00
|
|
|
struct socket;
|
|
|
|
|
struct ether_header;
|
2005-02-22 13:04:05 +00:00
|
|
|
struct carp_if;
|
Merge the //depot/user/yar/vlan branch into CVS. It contains some collective
work by yar, thompsa and myself. The checksum offloading part also involves
work done by Mihail Balikov.
The most important changes:
o Instead of global linked list of all vlan softc use a per-trunk
hash. The size of hash is dynamically adjusted, depending on
number of entries. This changes struct ifnet, replacing counter
of vlans with a pointer to trunk structure. This change is an
improvement for setups with big number of VLANs, several interfaces
and several CPUs. It is a small regression for a setup with a single
VLAN interface.
An alternative to dynamic hash is a per-trunk static array with
4096 entries, which is a compile time option - VLAN_ARRAY. In my
experiments the array is not an improvement, probably because such
a big trunk structure doesn't fit into CPU cache.
o Introduce an UMA zone for VLAN tags. Since drivers depend on it,
the zone is declared in kern_mbuf.c, not in optional vlan(4) driver.
This change is a big improvement for any setup utilizing vlan(4).
o Use rwlock(9) instead of mutex(9) for locking. We are the first
ones to do this! :)
o Some drivers can do hardware VLAN tagging + hardware checksum
offloading. Add an infrastructure for this. Whenever vlan(4) is
attached to a parent or parent configuration is changed, the flags
on vlan(4) interface are updated.
In collaboration with: yar, thompsa
In collaboration with: Mihail Balikov <mihail.balikov interbgc.com>
2006-01-30 13:45:15 +00:00
|
|
|
struct ifvlantrunk;
|
1997-01-03 19:50:26 +00:00
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
#include <sys/queue.h> /* get TAILQ macros */
|
|
|
|
|
|
2000-11-26 21:47:01 +00:00
|
|
|
#ifdef _KERNEL
|
Lock down the network interface queues. The queue mutex must be obtained
before adding/removing packets from the queue. Also, the if_obytes and
if_omcasts fields should only be manipulated under protection of the mutex.
IF_ENQUEUE, IF_PREPEND, and IF_DEQUEUE perform all necessary locking on
the queue. An IF_LOCK macro is provided, as well as the old (mutex-less)
versions of the macros in the form _IF_ENQUEUE, _IF_QFULL, for code which
needs them, but their use is discouraged.
Two new macros are introduced: IF_DRAIN() to drain a queue, and IF_HANDOFF,
which takes care of locking/enqueue, and also statistics updating/start
if necessary.
2000-11-25 07:35:38 +00:00
|
|
|
#include <sys/mbuf.h>
|
2004-02-26 04:27:55 +00:00
|
|
|
#include <sys/eventhandler.h>
|
2000-11-26 21:47:01 +00:00
|
|
|
#endif /* _KERNEL */
|
2001-03-28 09:17:56 +00:00
|
|
|
#include <sys/lock.h> /* XXX */
|
|
|
|
|
#include <sys/mutex.h> /* XXX */
|
2001-09-06 02:40:43 +00:00
|
|
|
#include <sys/event.h> /* XXX */
|
2004-07-27 23:20:45 +00:00
|
|
|
#include <sys/_task.h>
|
Lock down the network interface queues. The queue mutex must be obtained
before adding/removing packets from the queue. Also, the if_obytes and
if_omcasts fields should only be manipulated under protection of the mutex.
IF_ENQUEUE, IF_PREPEND, and IF_DEQUEUE perform all necessary locking on
the queue. An IF_LOCK macro is provided, as well as the old (mutex-less)
versions of the macros in the form _IF_ENQUEUE, _IF_QFULL, for code which
needs them, but their use is discouraged.
Two new macros are introduced: IF_DRAIN() to drain a queue, and IF_HANDOFF,
which takes care of locking/enqueue, and also statistics updating/start
if necessary.
2000-11-25 07:35:38 +00:00
|
|
|
|
2003-10-31 18:32:15 +00:00
|
|
|
#define IF_DUNIT_NONE -1
|
|
|
|
|
|
2004-06-13 17:29:10 +00:00
|
|
|
#include <altq/if_altq.h>
|
|
|
|
|
|
2000-05-26 02:09:24 +00:00
|
|
|
TAILQ_HEAD(ifnethead, ifnet); /* we use TAILQs so that the order of */
|
|
|
|
|
TAILQ_HEAD(ifaddrhead, ifaddr); /* instantiation is preserved in the list */
|
|
|
|
|
TAILQ_HEAD(ifprefixhead, ifprefix);
|
2001-02-06 10:12:15 +00:00
|
|
|
TAILQ_HEAD(ifmultihead, ifmultiaddr);
|
2006-06-19 22:20:45 +00:00
|
|
|
TAILQ_HEAD(ifgrouphead, ifg_group);
|
1997-01-03 19:50:26 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Structure defining a queue for a network interface.
|
|
|
|
|
*/
|
|
|
|
|
struct ifqueue {
|
|
|
|
|
struct mbuf *ifq_head;
|
|
|
|
|
struct mbuf *ifq_tail;
|
|
|
|
|
int ifq_len;
|
|
|
|
|
int ifq_maxlen;
|
|
|
|
|
int ifq_drops;
|
Lock down the network interface queues. The queue mutex must be obtained
before adding/removing packets from the queue. Also, the if_obytes and
if_omcasts fields should only be manipulated under protection of the mutex.
IF_ENQUEUE, IF_PREPEND, and IF_DEQUEUE perform all necessary locking on
the queue. An IF_LOCK macro is provided, as well as the old (mutex-less)
versions of the macros in the form _IF_ENQUEUE, _IF_QFULL, for code which
needs them, but their use is discouraged.
Two new macros are introduced: IF_DRAIN() to drain a queue, and IF_HANDOFF,
which takes care of locking/enqueue, and also statistics updating/start
if necessary.
2000-11-25 07:35:38 +00:00
|
|
|
struct mtx ifq_mtx;
|
1997-01-03 19:50:26 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Structure defining a network interface.
|
|
|
|
|
*
|
|
|
|
|
* (Would like to call this struct ``if'', but C isn't PL/1.)
|
|
|
|
|
*/
|
2001-10-02 18:08:34 +00:00
|
|
|
|
1997-01-03 19:50:26 +00:00
|
|
|
struct ifnet {
|
|
|
|
|
void *if_softc; /* pointer to driver state */
|
2005-06-10 16:49:24 +00:00
|
|
|
void *if_l2com; /* pointer to protocol bits */
|
2000-05-26 02:09:24 +00:00
|
|
|
TAILQ_ENTRY(ifnet) if_link; /* all struct ifnets are chained */
|
2003-10-31 18:32:15 +00:00
|
|
|
char if_xname[IFNAMSIZ]; /* external name (name + unit) */
|
|
|
|
|
const char *if_dname; /* driver name */
|
|
|
|
|
int if_dunit; /* unit or IF_DUNIT_NONE */
|
1997-01-03 19:50:26 +00:00
|
|
|
struct ifaddrhead if_addrhead; /* linked list of addresses per if */
|
2004-04-15 19:45:59 +00:00
|
|
|
/*
|
|
|
|
|
* if_addrhead is the list of all addresses associated to
|
2004-04-16 10:28:54 +00:00
|
|
|
* an interface.
|
|
|
|
|
* Some code in the kernel assumes that first element
|
|
|
|
|
* of the list has type AF_LINK, and contains sockaddr_dl
|
|
|
|
|
* addresses which store the link-level address and the name
|
2004-04-15 19:45:59 +00:00
|
|
|
* of the interface.
|
2004-04-16 10:28:54 +00:00
|
|
|
* However, access to the AF_LINK address through this
|
2005-11-11 16:04:59 +00:00
|
|
|
* field is deprecated. Use if_addr or ifaddr_byindex() instead.
|
2004-04-15 19:45:59 +00:00
|
|
|
*/
|
2004-08-15 06:24:42 +00:00
|
|
|
struct knlist if_klist; /* events attached to this if */
|
2001-09-06 02:40:43 +00:00
|
|
|
int if_pcount; /* number of promiscuous listeners */
|
2005-03-01 10:59:14 +00:00
|
|
|
struct carp_if *if_carp; /* carp interface structure */
|
1997-01-03 19:50:26 +00:00
|
|
|
struct bpf_if *if_bpf; /* packet filter structure */
|
|
|
|
|
u_short if_index; /* numeric abbreviation for this if */
|
|
|
|
|
short if_timer; /* time 'til if_watchdog called */
|
Merge the //depot/user/yar/vlan branch into CVS. It contains some collective
work by yar, thompsa and myself. The checksum offloading part also involves
work done by Mihail Balikov.
The most important changes:
o Instead of global linked list of all vlan softc use a per-trunk
hash. The size of hash is dynamically adjusted, depending on
number of entries. This changes struct ifnet, replacing counter
of vlans with a pointer to trunk structure. This change is an
improvement for setups with big number of VLANs, several interfaces
and several CPUs. It is a small regression for a setup with a single
VLAN interface.
An alternative to dynamic hash is a per-trunk static array with
4096 entries, which is a compile time option - VLAN_ARRAY. In my
experiments the array is not an improvement, probably because such
a big trunk structure doesn't fit into CPU cache.
o Introduce an UMA zone for VLAN tags. Since drivers depend on it,
the zone is declared in kern_mbuf.c, not in optional vlan(4) driver.
This change is a big improvement for any setup utilizing vlan(4).
o Use rwlock(9) instead of mutex(9) for locking. We are the first
ones to do this! :)
o Some drivers can do hardware VLAN tagging + hardware checksum
offloading. Add an infrastructure for this. Whenever vlan(4) is
attached to a parent or parent configuration is changed, the flags
on vlan(4) interface are updated.
In collaboration with: yar, thompsa
In collaboration with: Mihail Balikov <mihail.balikov interbgc.com>
2006-01-30 13:45:15 +00:00
|
|
|
struct ifvlantrunk *if_vlantrunk; /* pointer to 802.1q data */
|
2002-08-18 07:05:00 +00:00
|
|
|
int if_flags; /* up/down, broadcast, etc. */
|
2006-09-06 18:06:04 +00:00
|
|
|
int if_capabilities; /* interface features & capabilities */
|
|
|
|
|
int if_capenable; /* enabled features & capabilities */
|
1997-01-03 19:50:26 +00:00
|
|
|
void *if_linkmib; /* link-type-specific MIB data */
|
|
|
|
|
size_t if_linkmiblen; /* length of above data */
|
|
|
|
|
struct if_data if_data;
|
1997-01-07 19:15:32 +00:00
|
|
|
struct ifmultihead if_multiaddrs; /* multicast addresses configured */
|
|
|
|
|
int if_amcount; /* number of all-multicast requests */
|
1997-01-03 19:50:26 +00:00
|
|
|
/* procedure handles */
|
|
|
|
|
int (*if_output) /* output routine (enqueue) */
|
2002-03-19 21:54:18 +00:00
|
|
|
(struct ifnet *, struct mbuf *, struct sockaddr *,
|
|
|
|
|
struct rtentry *);
|
2002-11-14 23:36:28 +00:00
|
|
|
void (*if_input) /* input routine (from h/w driver) */
|
|
|
|
|
(struct ifnet *, struct mbuf *);
|
1997-01-03 19:50:26 +00:00
|
|
|
void (*if_start) /* initiate output routine */
|
2002-03-19 21:54:18 +00:00
|
|
|
(struct ifnet *);
|
1997-01-03 19:50:26 +00:00
|
|
|
int (*if_ioctl) /* ioctl routine */
|
2002-03-19 21:54:18 +00:00
|
|
|
(struct ifnet *, u_long, caddr_t);
|
1997-01-03 19:50:26 +00:00
|
|
|
void (*if_watchdog) /* timer routine */
|
2002-03-19 21:54:18 +00:00
|
|
|
(struct ifnet *);
|
1997-01-03 19:50:26 +00:00
|
|
|
void (*if_init) /* Init routine */
|
2002-03-19 21:54:18 +00:00
|
|
|
(void *);
|
1997-01-07 19:15:32 +00:00
|
|
|
int (*if_resolvemulti) /* validate/resolve multicast */
|
2002-03-19 21:54:18 +00:00
|
|
|
(struct ifnet *, struct sockaddr **, struct sockaddr *);
|
2005-11-11 16:04:59 +00:00
|
|
|
struct ifaddr *if_addr; /* pointer to link-level address */
|
2004-10-30 08:45:13 +00:00
|
|
|
void *if_spare2; /* spare pointer 2 */
|
|
|
|
|
void *if_spare3; /* spare pointer 3 */
|
2005-07-21 22:01:06 +00:00
|
|
|
int if_drv_flags; /* driver-managed status flags */
|
2004-10-30 08:45:13 +00:00
|
|
|
u_int if_spare_flags2; /* spare flags 2 */
|
2004-06-13 17:29:10 +00:00
|
|
|
struct ifaltq if_snd; /* output queue (includes altq) */
|
2003-12-07 05:49:21 +00:00
|
|
|
const u_int8_t *if_broadcastaddr; /* linklevel broadcast bytestring */
|
2004-04-04 06:14:55 +00:00
|
|
|
|
2005-06-05 03:13:13 +00:00
|
|
|
void *if_bridge; /* bridge glue */
|
|
|
|
|
|
2004-04-04 06:14:55 +00:00
|
|
|
struct lltable *lltables; /* list of L3-L2 resolution tables */
|
|
|
|
|
|
Modify the MAC Framework so that instead of embedding a (struct label)
in various kernel objects to represent security data, we embed a
(struct label *) pointer, which now references labels allocated using
a UMA zone (mac_label.c). This allows the size and shape of struct
label to be varied without changing the size and shape of these kernel
objects, which become part of the frozen ABI with 5-STABLE. This opens
the door for boot-time selection of the number of label slots, and hence
changes to the bound on the number of simultaneous labeled policies
at boot-time instead of compile-time. This also makes it easier to
embed label references in new objects as required for locking/caching
with fine-grained network stack locking, such as inpcb structures.
This change also moves us further in the direction of hiding the
structure of kernel objects from MAC policy modules, not to mention
dramatically reducing the number of '&' symbols appearing in both the
MAC Framework and MAC policy modules, and improving readability.
While this results in minimal performance change with MAC enabled, it
will observably shrink the size of a number of critical kernel data
structures for the !MAC case, and should have a small (but measurable)
performance benefit (i.e., struct vnode, struct socket) do to memory
conservation and reduced cost of zeroing memory.
NOTE: Users of MAC must recompile their kernel and all MAC modules as a
result of this change. Because this is an API change, third party
MAC modules will also need to be updated to make less use of the '&'
symbol.
Suggestions from: bmilekic
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
2003-11-12 03:14:31 +00:00
|
|
|
struct label *if_label; /* interface MAC label */
|
2003-10-17 15:46:31 +00:00
|
|
|
|
2004-04-04 06:14:55 +00:00
|
|
|
/* these are only used by IPv6 */
|
|
|
|
|
struct ifprefixhead if_prefixhead; /* list of prefixes per if */
|
2003-10-17 15:46:31 +00:00
|
|
|
void *if_afdata[AF_MAX];
|
2003-10-24 16:57:59 +00:00
|
|
|
int if_afdata_initialized;
|
|
|
|
|
struct mtx if_afdata_mtx;
|
2004-07-27 23:20:45 +00:00
|
|
|
struct task if_starttask; /* task for IFF_NEEDSGIANT */
|
2005-04-20 09:30:54 +00:00
|
|
|
struct task if_linktask; /* task for link change events */
|
2005-08-02 17:43:35 +00:00
|
|
|
struct mtx if_addr_mtx; /* mutex to protect address lists */
|
2005-11-08 20:08:34 +00:00
|
|
|
LIST_ENTRY(ifnet) if_clones; /* interfaces of a cloner */
|
2006-06-19 22:20:45 +00:00
|
|
|
TAILQ_HEAD(, ifg_list) if_groups; /* linked list of groups per if */
|
|
|
|
|
/* protected by if_addr_mtx */
|
|
|
|
|
void *if_pf_kif;
|
1997-01-03 19:50:26 +00:00
|
|
|
};
|
Lock down the network interface queues. The queue mutex must be obtained
before adding/removing packets from the queue. Also, the if_obytes and
if_omcasts fields should only be manipulated under protection of the mutex.
IF_ENQUEUE, IF_PREPEND, and IF_DEQUEUE perform all necessary locking on
the queue. An IF_LOCK macro is provided, as well as the old (mutex-less)
versions of the macros in the form _IF_ENQUEUE, _IF_QFULL, for code which
needs them, but their use is discouraged.
Two new macros are introduced: IF_DRAIN() to drain a queue, and IF_HANDOFF,
which takes care of locking/enqueue, and also statistics updating/start
if necessary.
2000-11-25 07:35:38 +00:00
|
|
|
|
2002-03-19 21:54:18 +00:00
|
|
|
typedef void if_init_f_t(void *);
|
1997-01-03 19:50:26 +00:00
|
|
|
|
2004-04-18 01:15:32 +00:00
|
|
|
/*
|
|
|
|
|
* XXX These aliases are terribly dangerous because they could apply
|
|
|
|
|
* to anything.
|
|
|
|
|
*/
|
1997-01-03 19:50:26 +00:00
|
|
|
#define if_mtu if_data.ifi_mtu
|
|
|
|
|
#define if_type if_data.ifi_type
|
|
|
|
|
#define if_physical if_data.ifi_physical
|
|
|
|
|
#define if_addrlen if_data.ifi_addrlen
|
|
|
|
|
#define if_hdrlen if_data.ifi_hdrlen
|
|
|
|
|
#define if_metric if_data.ifi_metric
|
2004-05-03 13:48:35 +00:00
|
|
|
#define if_link_state if_data.ifi_link_state
|
1997-01-03 19:50:26 +00:00
|
|
|
#define if_baudrate if_data.ifi_baudrate
|
2000-03-27 19:14:27 +00:00
|
|
|
#define if_hwassist if_data.ifi_hwassist
|
1997-01-03 19:50:26 +00:00
|
|
|
#define if_ipackets if_data.ifi_ipackets
|
|
|
|
|
#define if_ierrors if_data.ifi_ierrors
|
|
|
|
|
#define if_opackets if_data.ifi_opackets
|
|
|
|
|
#define if_oerrors if_data.ifi_oerrors
|
|
|
|
|
#define if_collisions if_data.ifi_collisions
|
|
|
|
|
#define if_ibytes if_data.ifi_ibytes
|
|
|
|
|
#define if_obytes if_data.ifi_obytes
|
|
|
|
|
#define if_imcasts if_data.ifi_imcasts
|
|
|
|
|
#define if_omcasts if_data.ifi_omcasts
|
|
|
|
|
#define if_iqdrops if_data.ifi_iqdrops
|
|
|
|
|
#define if_noproto if_data.ifi_noproto
|
|
|
|
|
#define if_lastchange if_data.ifi_lastchange
|
|
|
|
|
#define if_recvquota if_data.ifi_recvquota
|
|
|
|
|
#define if_xmitquota if_data.ifi_xmitquota
|
2004-10-25 17:04:40 +00:00
|
|
|
#define if_rawoutput(if, m, sa) if_output(if, m, sa, (struct rtentry *)NULL)
|
1997-01-03 19:50:26 +00:00
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
/* for compatibility with other BSDs */
|
|
|
|
|
#define if_addrlist if_addrhead
|
|
|
|
|
#define if_list if_link
|
2006-08-04 21:27:40 +00:00
|
|
|
#define if_name(ifp) ((ifp)->if_xname)
|
1999-11-22 02:45:11 +00:00
|
|
|
|
2005-08-02 17:43:35 +00:00
|
|
|
/*
|
|
|
|
|
* Locks for address lists on the network interface.
|
|
|
|
|
*/
|
|
|
|
|
#define IF_ADDR_LOCK_INIT(if) mtx_init(&(if)->if_addr_mtx, \
|
|
|
|
|
"if_addr_mtx", NULL, MTX_DEF)
|
|
|
|
|
#define IF_ADDR_LOCK_DESTROY(if) mtx_destroy(&(if)->if_addr_mtx)
|
|
|
|
|
#define IF_ADDR_LOCK(if) mtx_lock(&(if)->if_addr_mtx)
|
|
|
|
|
#define IF_ADDR_UNLOCK(if) mtx_unlock(&(if)->if_addr_mtx)
|
|
|
|
|
#define IF_ADDR_LOCK_ASSERT(if) mtx_assert(&(if)->if_addr_mtx, MA_OWNED)
|
|
|
|
|
|
1997-01-03 19:50:26 +00:00
|
|
|
/*
|
|
|
|
|
* Output queues (ifp->if_snd) and slow device input queues (*ifp->if_slowq)
|
|
|
|
|
* are queues of messages stored on ifqueue structures
|
|
|
|
|
* (defined above). Entries are added to and deleted from these structures
|
|
|
|
|
* by these macros, which should be called with ipl raised to splimp().
|
|
|
|
|
*/
|
Change and clean the mutex lock interface.
mtx_enter(lock, type) becomes:
mtx_lock(lock) for sleep locks (MTX_DEF-initialized locks)
mtx_lock_spin(lock) for spin locks (MTX_SPIN-initialized)
similarily, for releasing a lock, we now have:
mtx_unlock(lock) for MTX_DEF and mtx_unlock_spin(lock) for MTX_SPIN.
We change the caller interface for the two different types of locks
because the semantics are entirely different for each case, and this
makes it explicitly clear and, at the same time, it rids us of the
extra `type' argument.
The enter->lock and exit->unlock change has been made with the idea
that we're "locking data" and not "entering locked code" in mind.
Further, remove all additional "flags" previously passed to the
lock acquire/release routines with the exception of two:
MTX_QUIET and MTX_NOSWITCH
The functionality of these flags is preserved and they can be passed
to the lock/unlock routines by calling the corresponding wrappers:
mtx_{lock, unlock}_flags(lock, flag(s)) and
mtx_{lock, unlock}_spin_flags(lock, flag(s)) for MTX_DEF and MTX_SPIN
locks, respectively.
Re-inline some lock acq/rel code; in the sleep lock case, we only
inline the _obtain_lock()s in order to ensure that the inlined code
fits into a cache line. In the spin lock case, we inline recursion and
actually only perform a function call if we need to spin. This change
has been made with the idea that we generally tend to avoid spin locks
and that also the spin locks that we do have and are heavily used
(i.e. sched_lock) do recurse, and therefore in an effort to reduce
function call overhead for some architectures (such as alpha), we
inline recursion for this case.
Create a new malloc type for the witness code and retire from using
the M_DEV type. The new type is called M_WITNESS and is only declared
if WITNESS is enabled.
Begin cleaning up some machdep/mutex.h code - specifically updated the
"optimized" inlined code in alpha/mutex.h and wrote MTX_LOCK_SPIN
and MTX_UNLOCK_SPIN asm macros for the i386/mutex.h as we presently
need those.
Finally, caught up to the interface changes in all sys code.
Contributors: jake, jhb, jasone (in no particular order)
2001-02-09 06:11:45 +00:00
|
|
|
#define IF_LOCK(ifq) mtx_lock(&(ifq)->ifq_mtx)
|
|
|
|
|
#define IF_UNLOCK(ifq) mtx_unlock(&(ifq)->ifq_mtx)
|
2004-06-13 17:29:10 +00:00
|
|
|
#define IF_LOCK_ASSERT(ifq) mtx_assert(&(ifq)->ifq_mtx, MA_OWNED)
|
Lock down the network interface queues. The queue mutex must be obtained
before adding/removing packets from the queue. Also, the if_obytes and
if_omcasts fields should only be manipulated under protection of the mutex.
IF_ENQUEUE, IF_PREPEND, and IF_DEQUEUE perform all necessary locking on
the queue. An IF_LOCK macro is provided, as well as the old (mutex-less)
versions of the macros in the form _IF_ENQUEUE, _IF_QFULL, for code which
needs them, but their use is discouraged.
Two new macros are introduced: IF_DRAIN() to drain a queue, and IF_HANDOFF,
which takes care of locking/enqueue, and also statistics updating/start
if necessary.
2000-11-25 07:35:38 +00:00
|
|
|
#define _IF_QFULL(ifq) ((ifq)->ifq_len >= (ifq)->ifq_maxlen)
|
|
|
|
|
#define _IF_DROP(ifq) ((ifq)->ifq_drops++)
|
|
|
|
|
#define _IF_QLEN(ifq) ((ifq)->ifq_len)
|
|
|
|
|
|
|
|
|
|
#define _IF_ENQUEUE(ifq, m) do { \
|
|
|
|
|
(m)->m_nextpkt = NULL; \
|
|
|
|
|
if ((ifq)->ifq_tail == NULL) \
|
|
|
|
|
(ifq)->ifq_head = m; \
|
|
|
|
|
else \
|
|
|
|
|
(ifq)->ifq_tail->m_nextpkt = m; \
|
|
|
|
|
(ifq)->ifq_tail = m; \
|
|
|
|
|
(ifq)->ifq_len++; \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IF_ENQUEUE(ifq, m) do { \
|
|
|
|
|
IF_LOCK(ifq); \
|
|
|
|
|
_IF_ENQUEUE(ifq, m); \
|
|
|
|
|
IF_UNLOCK(ifq); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define _IF_PREPEND(ifq, m) do { \
|
|
|
|
|
(m)->m_nextpkt = (ifq)->ifq_head; \
|
|
|
|
|
if ((ifq)->ifq_tail == NULL) \
|
|
|
|
|
(ifq)->ifq_tail = (m); \
|
|
|
|
|
(ifq)->ifq_head = (m); \
|
|
|
|
|
(ifq)->ifq_len++; \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IF_PREPEND(ifq, m) do { \
|
|
|
|
|
IF_LOCK(ifq); \
|
|
|
|
|
_IF_PREPEND(ifq, m); \
|
|
|
|
|
IF_UNLOCK(ifq); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define _IF_DEQUEUE(ifq, m) do { \
|
|
|
|
|
(m) = (ifq)->ifq_head; \
|
|
|
|
|
if (m) { \
|
2004-10-25 17:04:40 +00:00
|
|
|
if (((ifq)->ifq_head = (m)->m_nextpkt) == NULL) \
|
Lock down the network interface queues. The queue mutex must be obtained
before adding/removing packets from the queue. Also, the if_obytes and
if_omcasts fields should only be manipulated under protection of the mutex.
IF_ENQUEUE, IF_PREPEND, and IF_DEQUEUE perform all necessary locking on
the queue. An IF_LOCK macro is provided, as well as the old (mutex-less)
versions of the macros in the form _IF_ENQUEUE, _IF_QFULL, for code which
needs them, but their use is discouraged.
Two new macros are introduced: IF_DRAIN() to drain a queue, and IF_HANDOFF,
which takes care of locking/enqueue, and also statistics updating/start
if necessary.
2000-11-25 07:35:38 +00:00
|
|
|
(ifq)->ifq_tail = NULL; \
|
|
|
|
|
(m)->m_nextpkt = NULL; \
|
|
|
|
|
(ifq)->ifq_len--; \
|
|
|
|
|
} \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IF_DEQUEUE(ifq, m) do { \
|
|
|
|
|
IF_LOCK(ifq); \
|
|
|
|
|
_IF_DEQUEUE(ifq, m); \
|
|
|
|
|
IF_UNLOCK(ifq); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
2004-06-13 17:29:10 +00:00
|
|
|
#define _IF_POLL(ifq, m) ((m) = (ifq)->ifq_head)
|
|
|
|
|
#define IF_POLL(ifq, m) _IF_POLL(ifq, m)
|
|
|
|
|
|
|
|
|
|
#define _IF_DRAIN(ifq) do { \
|
Lock down the network interface queues. The queue mutex must be obtained
before adding/removing packets from the queue. Also, the if_obytes and
if_omcasts fields should only be manipulated under protection of the mutex.
IF_ENQUEUE, IF_PREPEND, and IF_DEQUEUE perform all necessary locking on
the queue. An IF_LOCK macro is provided, as well as the old (mutex-less)
versions of the macros in the form _IF_ENQUEUE, _IF_QFULL, for code which
needs them, but their use is discouraged.
Two new macros are introduced: IF_DRAIN() to drain a queue, and IF_HANDOFF,
which takes care of locking/enqueue, and also statistics updating/start
if necessary.
2000-11-25 07:35:38 +00:00
|
|
|
struct mbuf *m; \
|
|
|
|
|
for (;;) { \
|
|
|
|
|
_IF_DEQUEUE(ifq, m); \
|
|
|
|
|
if (m == NULL) \
|
|
|
|
|
break; \
|
|
|
|
|
m_freem(m); \
|
|
|
|
|
} \
|
|
|
|
|
} while (0)
|
1997-01-03 19:50:26 +00:00
|
|
|
|
2004-06-13 17:29:10 +00:00
|
|
|
#define IF_DRAIN(ifq) do { \
|
|
|
|
|
IF_LOCK(ifq); \
|
|
|
|
|
_IF_DRAIN(ifq); \
|
|
|
|
|
IF_UNLOCK(ifq); \
|
|
|
|
|
} while(0)
|
|
|
|
|
|
1999-12-29 04:46:21 +00:00
|
|
|
#ifdef _KERNEL
|
2004-02-26 04:27:55 +00:00
|
|
|
/* interface address change event */
|
|
|
|
|
typedef void (*ifaddr_event_handler_t)(void *, struct ifnet *);
|
|
|
|
|
EVENTHANDLER_DECLARE(ifaddr_event, ifaddr_event_handler_t);
|
|
|
|
|
/* new interface arrival event */
|
|
|
|
|
typedef void (*ifnet_arrival_event_handler_t)(void *, struct ifnet *);
|
|
|
|
|
EVENTHANDLER_DECLARE(ifnet_arrival_event, ifnet_arrival_event_handler_t);
|
|
|
|
|
/* interface departure event */
|
|
|
|
|
typedef void (*ifnet_departure_event_handler_t)(void *, struct ifnet *);
|
|
|
|
|
EVENTHANDLER_DECLARE(ifnet_departure_event, ifnet_departure_event_handler_t);
|
|
|
|
|
|
2006-06-19 22:20:45 +00:00
|
|
|
/*
|
|
|
|
|
* interface groups
|
|
|
|
|
*/
|
|
|
|
|
struct ifg_group {
|
|
|
|
|
char ifg_group[IFNAMSIZ];
|
|
|
|
|
u_int ifg_refcnt;
|
|
|
|
|
void *ifg_pf_kif;
|
|
|
|
|
TAILQ_HEAD(, ifg_member) ifg_members;
|
|
|
|
|
TAILQ_ENTRY(ifg_group) ifg_next;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
struct ifg_member {
|
|
|
|
|
TAILQ_ENTRY(ifg_member) ifgm_next;
|
|
|
|
|
struct ifnet *ifgm_ifp;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
struct ifg_list {
|
|
|
|
|
struct ifg_group *ifgl_group;
|
|
|
|
|
TAILQ_ENTRY(ifg_list) ifgl_next;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/* group attach event */
|
|
|
|
|
typedef void (*group_attach_event_handler_t)(void *, struct ifg_group *);
|
|
|
|
|
EVENTHANDLER_DECLARE(group_attach_event, group_attach_event_handler_t);
|
|
|
|
|
/* group detach event */
|
|
|
|
|
typedef void (*group_detach_event_handler_t)(void *, struct ifg_group *);
|
|
|
|
|
EVENTHANDLER_DECLARE(group_detach_event, group_detach_event_handler_t);
|
|
|
|
|
/* group change event */
|
|
|
|
|
typedef void (*group_change_event_handler_t)(void *, const char *);
|
|
|
|
|
EVENTHANDLER_DECLARE(group_change_event, group_change_event_handler_t);
|
|
|
|
|
|
2003-10-24 16:57:59 +00:00
|
|
|
#define IF_AFDATA_LOCK_INIT(ifp) \
|
|
|
|
|
mtx_init(&(ifp)->if_afdata_mtx, "if_afdata", NULL, MTX_DEF)
|
|
|
|
|
#define IF_AFDATA_LOCK(ifp) mtx_lock(&(ifp)->if_afdata_mtx)
|
|
|
|
|
#define IF_AFDATA_TRYLOCK(ifp) mtx_trylock(&(ifp)->if_afdata_mtx)
|
|
|
|
|
#define IF_AFDATA_UNLOCK(ifp) mtx_unlock(&(ifp)->if_afdata_mtx)
|
|
|
|
|
#define IF_AFDATA_DESTROY(ifp) mtx_destroy(&(ifp)->if_afdata_mtx)
|
|
|
|
|
|
2004-10-19 18:11:55 +00:00
|
|
|
#define IFF_LOCKGIANT(ifp) do { \
|
|
|
|
|
if ((ifp)->if_flags & IFF_NEEDSGIANT) \
|
|
|
|
|
mtx_lock(&Giant); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFF_UNLOCKGIANT(ifp) do { \
|
|
|
|
|
if ((ifp)->if_flags & IFF_NEEDSGIANT) \
|
|
|
|
|
mtx_unlock(&Giant); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
2004-10-30 09:39:13 +00:00
|
|
|
int if_handoff(struct ifqueue *ifq, struct mbuf *m, struct ifnet *ifp,
|
|
|
|
|
int adjust);
|
2004-06-13 17:29:10 +00:00
|
|
|
#define IF_HANDOFF(ifq, m, ifp) \
|
|
|
|
|
if_handoff((struct ifqueue *)ifq, m, ifp, 0)
|
|
|
|
|
#define IF_HANDOFF_ADJ(ifq, m, ifp, adj) \
|
|
|
|
|
if_handoff((struct ifqueue *)ifq, m, ifp, adj)
|
1997-01-03 19:50:26 +00:00
|
|
|
|
2004-07-27 23:20:45 +00:00
|
|
|
void if_start(struct ifnet *);
|
|
|
|
|
|
2004-06-13 17:29:10 +00:00
|
|
|
#define IFQ_ENQUEUE(ifq, m, err) \
|
|
|
|
|
do { \
|
|
|
|
|
IF_LOCK(ifq); \
|
|
|
|
|
if (ALTQ_IS_ENABLED(ifq)) \
|
|
|
|
|
ALTQ_ENQUEUE(ifq, m, NULL, err); \
|
|
|
|
|
else { \
|
|
|
|
|
if (_IF_QFULL(ifq)) { \
|
|
|
|
|
m_freem(m); \
|
|
|
|
|
(err) = ENOBUFS; \
|
|
|
|
|
} else { \
|
|
|
|
|
_IF_ENQUEUE(ifq, m); \
|
|
|
|
|
(err) = 0; \
|
|
|
|
|
} \
|
|
|
|
|
} \
|
|
|
|
|
if (err) \
|
|
|
|
|
(ifq)->ifq_drops++; \
|
|
|
|
|
IF_UNLOCK(ifq); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_DEQUEUE_NOLOCK(ifq, m) \
|
|
|
|
|
do { \
|
|
|
|
|
if (TBR_IS_ENABLED(ifq)) \
|
2004-06-15 01:45:19 +00:00
|
|
|
(m) = tbr_dequeue_ptr(ifq, ALTDQ_REMOVE); \
|
2004-06-13 17:29:10 +00:00
|
|
|
else if (ALTQ_IS_ENABLED(ifq)) \
|
|
|
|
|
ALTQ_DEQUEUE(ifq, m); \
|
|
|
|
|
else \
|
|
|
|
|
_IF_DEQUEUE(ifq, m); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_DEQUEUE(ifq, m) \
|
|
|
|
|
do { \
|
|
|
|
|
IF_LOCK(ifq); \
|
|
|
|
|
IFQ_DEQUEUE_NOLOCK(ifq, m); \
|
|
|
|
|
IF_UNLOCK(ifq); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_POLL_NOLOCK(ifq, m) \
|
|
|
|
|
do { \
|
|
|
|
|
if (TBR_IS_ENABLED(ifq)) \
|
2004-06-15 01:45:19 +00:00
|
|
|
(m) = tbr_dequeue_ptr(ifq, ALTDQ_POLL); \
|
2004-06-13 17:29:10 +00:00
|
|
|
else if (ALTQ_IS_ENABLED(ifq)) \
|
|
|
|
|
ALTQ_POLL(ifq, m); \
|
|
|
|
|
else \
|
|
|
|
|
_IF_POLL(ifq, m); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_POLL(ifq, m) \
|
|
|
|
|
do { \
|
|
|
|
|
IF_LOCK(ifq); \
|
|
|
|
|
IFQ_POLL_NOLOCK(ifq, m); \
|
|
|
|
|
IF_UNLOCK(ifq); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_PURGE_NOLOCK(ifq) \
|
|
|
|
|
do { \
|
|
|
|
|
if (ALTQ_IS_ENABLED(ifq)) { \
|
|
|
|
|
ALTQ_PURGE(ifq); \
|
|
|
|
|
} else \
|
|
|
|
|
_IF_DRAIN(ifq); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_PURGE(ifq) \
|
|
|
|
|
do { \
|
|
|
|
|
IF_LOCK(ifq); \
|
|
|
|
|
IFQ_PURGE_NOLOCK(ifq); \
|
|
|
|
|
IF_UNLOCK(ifq); \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_SET_READY(ifq) \
|
|
|
|
|
do { ((ifq)->altq_flags |= ALTQF_READY); } while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_LOCK(ifq) IF_LOCK(ifq)
|
|
|
|
|
#define IFQ_UNLOCK(ifq) IF_UNLOCK(ifq)
|
|
|
|
|
#define IFQ_LOCK_ASSERT(ifq) IF_LOCK_ASSERT(ifq)
|
|
|
|
|
#define IFQ_IS_EMPTY(ifq) ((ifq)->ifq_len == 0)
|
|
|
|
|
#define IFQ_INC_LEN(ifq) ((ifq)->ifq_len++)
|
|
|
|
|
#define IFQ_DEC_LEN(ifq) (--(ifq)->ifq_len)
|
|
|
|
|
#define IFQ_INC_DROPS(ifq) ((ifq)->ifq_drops++)
|
|
|
|
|
#define IFQ_SET_MAXLEN(ifq, len) ((ifq)->ifq_maxlen = (len))
|
|
|
|
|
|
Rename IFF_RUNNING to IFF_DRV_RUNNING, IFF_OACTIVE to IFF_DRV_OACTIVE,
and move both flags from ifnet.if_flags to ifnet.if_drv_flags, making
and documenting the locking of these flags the responsibility of the
device driver, not the network stack. The flags for these two fields
will be mutually exclusive so that they can be exposed to user space as
though they were stored in the same variable.
Provide #defines to provide the old names #ifndef _KERNEL, so that user
applications (such as ifconfig) can use the old flag names. Using the
old names in a device driver will result in a compile error in order to
help device driver writers adopt the new model.
When exposing the interface flags to user space, via interface ioctls
or routing sockets, or the two fields together. Since the driver flags
cannot currently be set for user space, no new logic is currently
required to handle this case.
Add some assertions that general purpose network stack routines, such
as if_setflags(), are not improperly used on driver-owned flags.
With this change, a large number of very minor network stack races are
closed, subject to correct device driver locking. Most were likely
never triggered.
Driver sweep to follow; many thanks to pjd and bz for the line-by-line
review they gave this patch.
Reviewed by: pjd, bz
MFC after: 7 days
2005-08-09 10:16:17 +00:00
|
|
|
/*
|
|
|
|
|
* The IFF_DRV_OACTIVE test should really occur in the device driver, not in
|
|
|
|
|
* the handoff logic, as that flag is locked by the device driver.
|
|
|
|
|
*/
|
2004-06-13 17:29:10 +00:00
|
|
|
#define IFQ_HANDOFF_ADJ(ifp, m, adj, err) \
|
|
|
|
|
do { \
|
|
|
|
|
int len; \
|
|
|
|
|
short mflags; \
|
|
|
|
|
\
|
|
|
|
|
len = (m)->m_pkthdr.len; \
|
|
|
|
|
mflags = (m)->m_flags; \
|
|
|
|
|
IFQ_ENQUEUE(&(ifp)->if_snd, m, err); \
|
|
|
|
|
if ((err) == 0) { \
|
|
|
|
|
(ifp)->if_obytes += len + (adj); \
|
|
|
|
|
if (mflags & M_MCAST) \
|
|
|
|
|
(ifp)->if_omcasts++; \
|
Rename IFF_RUNNING to IFF_DRV_RUNNING, IFF_OACTIVE to IFF_DRV_OACTIVE,
and move both flags from ifnet.if_flags to ifnet.if_drv_flags, making
and documenting the locking of these flags the responsibility of the
device driver, not the network stack. The flags for these two fields
will be mutually exclusive so that they can be exposed to user space as
though they were stored in the same variable.
Provide #defines to provide the old names #ifndef _KERNEL, so that user
applications (such as ifconfig) can use the old flag names. Using the
old names in a device driver will result in a compile error in order to
help device driver writers adopt the new model.
When exposing the interface flags to user space, via interface ioctls
or routing sockets, or the two fields together. Since the driver flags
cannot currently be set for user space, no new logic is currently
required to handle this case.
Add some assertions that general purpose network stack routines, such
as if_setflags(), are not improperly used on driver-owned flags.
With this change, a large number of very minor network stack races are
closed, subject to correct device driver locking. Most were likely
never triggered.
Driver sweep to follow; many thanks to pjd and bz for the line-by-line
review they gave this patch.
Reviewed by: pjd, bz
MFC after: 7 days
2005-08-09 10:16:17 +00:00
|
|
|
if (((ifp)->if_drv_flags & IFF_DRV_OACTIVE) == 0) \
|
2004-07-27 23:20:45 +00:00
|
|
|
if_start(ifp); \
|
2004-06-13 17:29:10 +00:00
|
|
|
} \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_HANDOFF(ifp, m, err) \
|
2004-06-15 03:40:39 +00:00
|
|
|
IFQ_HANDOFF_ADJ(ifp, m, 0, err)
|
2004-06-13 17:29:10 +00:00
|
|
|
|
|
|
|
|
#define IFQ_DRV_DEQUEUE(ifq, m) \
|
|
|
|
|
do { \
|
|
|
|
|
(m) = (ifq)->ifq_drv_head; \
|
|
|
|
|
if (m) { \
|
|
|
|
|
if (((ifq)->ifq_drv_head = (m)->m_nextpkt) == NULL) \
|
|
|
|
|
(ifq)->ifq_drv_tail = NULL; \
|
|
|
|
|
(m)->m_nextpkt = NULL; \
|
|
|
|
|
(ifq)->ifq_drv_len--; \
|
|
|
|
|
} else { \
|
|
|
|
|
IFQ_LOCK(ifq); \
|
|
|
|
|
IFQ_DEQUEUE_NOLOCK(ifq, m); \
|
|
|
|
|
while ((ifq)->ifq_drv_len < (ifq)->ifq_drv_maxlen) { \
|
|
|
|
|
struct mbuf *m0; \
|
|
|
|
|
IFQ_DEQUEUE_NOLOCK(ifq, m0); \
|
|
|
|
|
if (m0 == NULL) \
|
|
|
|
|
break; \
|
|
|
|
|
m0->m_nextpkt = NULL; \
|
|
|
|
|
if ((ifq)->ifq_drv_tail == NULL) \
|
|
|
|
|
(ifq)->ifq_drv_head = m0; \
|
|
|
|
|
else \
|
|
|
|
|
(ifq)->ifq_drv_tail->m_nextpkt = m0; \
|
|
|
|
|
(ifq)->ifq_drv_tail = m0; \
|
|
|
|
|
(ifq)->ifq_drv_len++; \
|
|
|
|
|
} \
|
|
|
|
|
IFQ_UNLOCK(ifq); \
|
|
|
|
|
} \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_DRV_PREPEND(ifq, m) \
|
|
|
|
|
do { \
|
|
|
|
|
(m)->m_nextpkt = (ifq)->ifq_drv_head; \
|
2004-07-14 13:31:41 +00:00
|
|
|
if ((ifq)->ifq_drv_tail == NULL) \
|
|
|
|
|
(ifq)->ifq_drv_tail = (m); \
|
2004-06-13 17:29:10 +00:00
|
|
|
(ifq)->ifq_drv_head = (m); \
|
|
|
|
|
(ifq)->ifq_drv_len++; \
|
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
|
|
#define IFQ_DRV_IS_EMPTY(ifq) \
|
|
|
|
|
(((ifq)->ifq_drv_len == 0) && ((ifq)->ifq_len == 0))
|
|
|
|
|
|
|
|
|
|
#define IFQ_DRV_PURGE(ifq) \
|
|
|
|
|
do { \
|
2004-07-14 13:31:41 +00:00
|
|
|
struct mbuf *m, *n = (ifq)->ifq_drv_head; \
|
|
|
|
|
while((m = n) != NULL) { \
|
|
|
|
|
n = m->m_nextpkt; \
|
2004-06-13 17:29:10 +00:00
|
|
|
m_freem(m); \
|
|
|
|
|
} \
|
|
|
|
|
(ifq)->ifq_drv_head = (ifq)->ifq_drv_tail = NULL; \
|
|
|
|
|
(ifq)->ifq_drv_len = 0; \
|
|
|
|
|
IFQ_PURGE(ifq); \
|
|
|
|
|
} while (0)
|
1999-08-06 13:53:03 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* 72 was chosen below because it is the size of a TCP/IP
|
|
|
|
|
* header (40) + the minimum mss (32).
|
|
|
|
|
*/
|
|
|
|
|
#define IF_MINMTU 72
|
|
|
|
|
#define IF_MAXMTU 65535
|
|
|
|
|
|
1999-12-29 04:46:21 +00:00
|
|
|
#endif /* _KERNEL */
|
1997-01-03 19:50:26 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* The ifaddr structure contains information about one address
|
|
|
|
|
* of an interface. They are maintained by the different address families,
|
|
|
|
|
* are allocated and attached when an address is set, and are linked
|
|
|
|
|
* together so all addresses for an interface can be located.
|
2004-04-15 19:45:59 +00:00
|
|
|
*
|
|
|
|
|
* NOTE: a 'struct ifaddr' is always at the beginning of a larger
|
|
|
|
|
* chunk of malloc'ed memory, where we store the three addresses
|
|
|
|
|
* (ifa_addr, ifa_dstaddr and ifa_netmask) referenced here.
|
1997-01-03 19:50:26 +00:00
|
|
|
*/
|
|
|
|
|
struct ifaddr {
|
|
|
|
|
struct sockaddr *ifa_addr; /* address of interface */
|
|
|
|
|
struct sockaddr *ifa_dstaddr; /* other end of p-to-p link */
|
|
|
|
|
#define ifa_broadaddr ifa_dstaddr /* broadcast address interface */
|
|
|
|
|
struct sockaddr *ifa_netmask; /* used to determine subnet */
|
2000-10-19 23:15:54 +00:00
|
|
|
struct if_data if_data; /* not all members are meaningful */
|
1997-01-03 19:50:26 +00:00
|
|
|
struct ifnet *ifa_ifp; /* back-pointer to interface */
|
2000-05-26 02:09:24 +00:00
|
|
|
TAILQ_ENTRY(ifaddr) ifa_link; /* queue macro glue */
|
1997-01-03 19:50:26 +00:00
|
|
|
void (*ifa_rtrequest) /* check or clean routes (+ or -)'d */
|
2002-03-19 21:54:18 +00:00
|
|
|
(int, struct rtentry *, struct rt_addrinfo *);
|
1997-01-03 19:50:26 +00:00
|
|
|
u_short ifa_flags; /* mostly rt_flags for cloning */
|
1999-05-16 17:09:20 +00:00
|
|
|
u_int ifa_refcnt; /* references to this structure */
|
1997-01-03 19:50:26 +00:00
|
|
|
int ifa_metric; /* cost of going out this interface */
|
1997-08-28 01:17:12 +00:00
|
|
|
int (*ifa_claim_addr) /* check if an addr goes to this if */
|
2002-03-19 21:54:18 +00:00
|
|
|
(struct ifaddr *, struct sockaddr *);
|
2002-12-18 11:46:59 +00:00
|
|
|
struct mtx ifa_mtx;
|
1997-01-03 19:50:26 +00:00
|
|
|
};
|
|
|
|
|
#define IFA_ROUTE RTF_UP /* route installed */
|
|
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
/* for compatibility with other BSDs */
|
|
|
|
|
#define ifa_list ifa_link
|
|
|
|
|
|
2002-12-18 11:46:59 +00:00
|
|
|
#define IFA_LOCK_INIT(ifa) \
|
|
|
|
|
mtx_init(&(ifa)->ifa_mtx, "ifaddr", NULL, MTX_DEF)
|
|
|
|
|
#define IFA_LOCK(ifa) mtx_lock(&(ifa)->ifa_mtx)
|
|
|
|
|
#define IFA_UNLOCK(ifa) mtx_unlock(&(ifa)->ifa_mtx)
|
|
|
|
|
#define IFA_DESTROY(ifa) mtx_destroy(&(ifa)->ifa_mtx)
|
|
|
|
|
|
1999-11-05 14:41:39 +00:00
|
|
|
/*
|
|
|
|
|
* The prefix structure contains information about one prefix
|
|
|
|
|
* of an interface. They are maintained by the different address families,
|
2002-12-30 21:18:15 +00:00
|
|
|
* are allocated and attached when a prefix or an address is set,
|
1999-11-22 02:45:11 +00:00
|
|
|
* and are linked together so all prefixes for an interface can be located.
|
1999-11-05 14:41:39 +00:00
|
|
|
*/
|
|
|
|
|
struct ifprefix {
|
|
|
|
|
struct sockaddr *ifpr_prefix; /* prefix of interface */
|
|
|
|
|
struct ifnet *ifpr_ifp; /* back-pointer to interface */
|
2000-05-26 02:09:24 +00:00
|
|
|
TAILQ_ENTRY(ifprefix) ifpr_list; /* queue macro glue */
|
1999-11-05 14:41:39 +00:00
|
|
|
u_char ifpr_plen; /* prefix length in bits */
|
|
|
|
|
u_char ifpr_type; /* protocol dependent prefix type */
|
|
|
|
|
};
|
|
|
|
|
|
1997-01-07 19:15:32 +00:00
|
|
|
/*
|
|
|
|
|
* Multicast address structure. This is analogous to the ifaddr
|
|
|
|
|
* structure except that it keeps track of multicast addresses.
|
|
|
|
|
*/
|
|
|
|
|
struct ifmultiaddr {
|
2001-02-06 10:12:15 +00:00
|
|
|
TAILQ_ENTRY(ifmultiaddr) ifma_link; /* queue macro glue */
|
1997-01-08 13:20:25 +00:00
|
|
|
struct sockaddr *ifma_addr; /* address this membership is for */
|
|
|
|
|
struct sockaddr *ifma_lladdr; /* link-layer translation, if any */
|
|
|
|
|
struct ifnet *ifma_ifp; /* back-pointer to interface */
|
|
|
|
|
u_int ifma_refcount; /* reference count */
|
|
|
|
|
void *ifma_protospec; /* protocol-specific state, if any */
|
2007-03-20 00:36:10 +00:00
|
|
|
struct ifmultiaddr *ifma_llifma; /* pointer to ifma for ifma_lladdr */
|
1997-01-07 19:15:32 +00:00
|
|
|
};
|
|
|
|
|
|
1999-12-29 04:46:21 +00:00
|
|
|
#ifdef _KERNEL
|
2002-12-18 12:41:03 +00:00
|
|
|
#define IFAFREE(ifa) \
|
|
|
|
|
do { \
|
|
|
|
|
IFA_LOCK(ifa); \
|
|
|
|
|
KASSERT((ifa)->ifa_refcnt > 0, \
|
|
|
|
|
("ifa %p !(ifa_refcnt > 0)", ifa)); \
|
|
|
|
|
if (--(ifa)->ifa_refcnt == 0) { \
|
|
|
|
|
IFA_DESTROY(ifa); \
|
|
|
|
|
free(ifa, M_IFADDR); \
|
|
|
|
|
} else \
|
|
|
|
|
IFA_UNLOCK(ifa); \
|
2002-12-18 11:46:59 +00:00
|
|
|
} while (0)
|
|
|
|
|
|
2002-12-18 12:41:03 +00:00
|
|
|
#define IFAREF(ifa) \
|
|
|
|
|
do { \
|
|
|
|
|
IFA_LOCK(ifa); \
|
|
|
|
|
++(ifa)->ifa_refcnt; \
|
|
|
|
|
IFA_UNLOCK(ifa); \
|
1999-05-06 18:13:11 +00:00
|
|
|
} while (0)
|
1997-01-03 19:50:26 +00:00
|
|
|
|
2002-12-22 05:35:03 +00:00
|
|
|
extern struct mtx ifnet_lock;
|
2002-12-27 00:24:35 +00:00
|
|
|
#define IFNET_LOCK_INIT() \
|
|
|
|
|
mtx_init(&ifnet_lock, "ifnet", NULL, MTX_DEF | MTX_RECURSE)
|
2002-12-22 05:35:03 +00:00
|
|
|
#define IFNET_WLOCK() mtx_lock(&ifnet_lock)
|
|
|
|
|
#define IFNET_WUNLOCK() mtx_unlock(&ifnet_lock)
|
|
|
|
|
#define IFNET_RLOCK() IFNET_WLOCK()
|
|
|
|
|
#define IFNET_RUNLOCK() IFNET_WUNLOCK()
|
|
|
|
|
|
2001-09-06 02:40:43 +00:00
|
|
|
struct ifindex_entry {
|
2001-12-14 19:32:47 +00:00
|
|
|
struct ifnet *ife_ifnet;
|
2004-06-16 09:47:26 +00:00
|
|
|
struct cdev *ife_dev;
|
2001-09-06 02:40:43 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
#define ifnet_byindex(idx) ifindex_table[(idx)].ife_ifnet
|
2004-04-16 10:28:54 +00:00
|
|
|
/*
|
|
|
|
|
* Given the index, ifaddr_byindex() returns the one and only
|
|
|
|
|
* link-level ifaddr for the interface. You are not supposed to use
|
|
|
|
|
* it to traverse the list of addresses associated to the interface.
|
|
|
|
|
*/
|
2005-11-11 16:04:59 +00:00
|
|
|
#define ifaddr_byindex(idx) ifnet_byindex(idx)->if_addr
|
2001-09-06 02:40:43 +00:00
|
|
|
#define ifdev_byindex(idx) ifindex_table[(idx)].ife_dev
|
|
|
|
|
|
1997-01-03 19:50:26 +00:00
|
|
|
extern struct ifnethead ifnet;
|
2001-09-06 02:40:43 +00:00
|
|
|
extern struct ifindex_entry *ifindex_table;
|
1997-01-03 19:50:26 +00:00
|
|
|
extern int ifqmaxlen;
|
2001-01-29 11:06:26 +00:00
|
|
|
extern struct ifnet *loif; /* first loopback interface */
|
1997-01-03 19:50:26 +00:00
|
|
|
extern int if_index;
|
|
|
|
|
|
2006-06-19 22:20:45 +00:00
|
|
|
int if_addgroup(struct ifnet *, const char *);
|
|
|
|
|
int if_delgroup(struct ifnet *, const char *);
|
2002-03-19 21:54:18 +00:00
|
|
|
int if_addmulti(struct ifnet *, struct sockaddr *, struct ifmultiaddr **);
|
|
|
|
|
int if_allmulti(struct ifnet *, int);
|
2005-06-10 16:49:24 +00:00
|
|
|
struct ifnet* if_alloc(u_char);
|
2002-03-19 21:54:18 +00:00
|
|
|
void if_attach(struct ifnet *);
|
|
|
|
|
int if_delmulti(struct ifnet *, struct sockaddr *);
|
2007-03-20 00:36:10 +00:00
|
|
|
void if_delmulti_ifma(struct ifmultiaddr *);
|
2002-03-19 21:54:18 +00:00
|
|
|
void if_detach(struct ifnet *);
|
2005-05-25 13:52:03 +00:00
|
|
|
void if_purgeaddrs(struct ifnet *);
|
2002-03-19 21:54:18 +00:00
|
|
|
void if_down(struct ifnet *);
|
2005-06-10 16:49:24 +00:00
|
|
|
void if_free(struct ifnet *);
|
|
|
|
|
void if_free_type(struct ifnet *, u_char);
|
2003-10-31 18:32:15 +00:00
|
|
|
void if_initname(struct ifnet *, const char *, int);
|
2004-12-08 05:45:59 +00:00
|
|
|
void if_link_state_change(struct ifnet *, int);
|
2002-09-24 17:35:08 +00:00
|
|
|
int if_printf(struct ifnet *, const char *, ...) __printflike(2, 3);
|
2002-03-19 21:54:18 +00:00
|
|
|
int if_setlladdr(struct ifnet *, const u_char *, int);
|
|
|
|
|
void if_up(struct ifnet *);
|
|
|
|
|
/*void ifinit(void);*/ /* declared in systm.h for main() */
|
|
|
|
|
int ifioctl(struct socket *, u_long, caddr_t, struct thread *);
|
|
|
|
|
int ifpromisc(struct ifnet *, int);
|
|
|
|
|
struct ifnet *ifunit(const char *);
|
|
|
|
|
|
|
|
|
|
struct ifaddr *ifa_ifwithaddr(struct sockaddr *);
|
2006-09-06 17:12:10 +00:00
|
|
|
struct ifaddr *ifa_ifwithbroadaddr(struct sockaddr *);
|
2002-03-19 21:54:18 +00:00
|
|
|
struct ifaddr *ifa_ifwithdstaddr(struct sockaddr *);
|
|
|
|
|
struct ifaddr *ifa_ifwithnet(struct sockaddr *);
|
|
|
|
|
struct ifaddr *ifa_ifwithroute(int, struct sockaddr *, struct sockaddr *);
|
|
|
|
|
struct ifaddr *ifaof_ifpforaddr(struct sockaddr *, struct ifnet *);
|
|
|
|
|
|
|
|
|
|
int if_simloop(struct ifnet *ifp, struct mbuf *m, int af, int hlen);
|
|
|
|
|
|
2005-06-10 16:49:24 +00:00
|
|
|
typedef void *if_com_alloc_t(u_char type, struct ifnet *ifp);
|
|
|
|
|
typedef void if_com_free_t(void *com, u_char type);
|
|
|
|
|
void if_register_com_alloc(u_char type, if_com_alloc_t *a, if_com_free_t *f);
|
|
|
|
|
void if_deregister_com_alloc(u_char type);
|
|
|
|
|
|
2001-10-14 20:17:53 +00:00
|
|
|
#define IF_LLADDR(ifp) \
|
2005-11-11 16:04:59 +00:00
|
|
|
LLADDR((struct sockaddr_dl *)((ifp)->if_addr->ifa_addr))
|
2001-10-14 20:17:53 +00:00
|
|
|
|
Device Polling code for -current.
Non-SMP, i386-only, no polling in the idle loop at the moment.
To use this code you must compile a kernel with
options DEVICE_POLLING
and at runtime enable polling with
sysctl kern.polling.enable=1
The percentage of CPU reserved to userland can be set with
sysctl kern.polling.user_frac=NN (default is 50)
while the remainder is used by polling device drivers and netisr's.
These are the only two variables that you should need to touch. There
are a few more parameters in kern.polling but the default values
are adequate for all purposes. See the code in kern_poll.c for
more details on them.
Polling in the idle loop will be implemented shortly by introducing
a kernel thread which does the job. Until then, the amount of CPU
dedicated to polling will never exceed (100-user_frac).
The equivalent (actually, better) code for -stable is at
http://info.iet.unipi.it/~luigi/polling/
and also supports polling in the idle loop.
NOTE to Alpha developers:
There is really nothing in this code that is i386-specific.
If you move the 2 lines supporting the new option from
sys/conf/{files,options}.i386 to sys/conf/{files,options} I am
pretty sure that this should work on the Alpha as well, just that
I do not have a suitable test box to try it. If someone feels like
trying it, I would appreciate it.
NOTE to other developers:
sure some things could be done better, and as always I am open to
constructive criticism, which a few of you have already given and
I greatly appreciated.
However, before proposing radical architectural changes, please
take some time to possibly try out this code, or at the very least
read the comments in kern_poll.c, especially re. the reason why I
am using a soft netisr and cannot (I believe) replace it with a
simple timeout.
Quick description of files touched by this commit:
sys/conf/files.i386
new file kern/kern_poll.c
sys/conf/options.i386
new option
sys/i386/i386/trap.c
poll in trap (disabled by default)
sys/kern/kern_clock.c
initialization and hardclock hooks.
sys/kern/kern_intr.c
minor swi_net changes
sys/kern/kern_poll.c
the bulk of the code.
sys/net/if.h
new flag
sys/net/if_var.h
declaration for functions used in device drivers.
sys/net/netisr.h
NETISR_POLL
sys/dev/fxp/if_fxp.c
sys/dev/fxp/if_fxpvar.h
sys/pci/if_dc.c
sys/pci/if_dcreg.h
sys/pci/if_sis.c
sys/pci/if_sisreg.h
device driver modifications
2001-12-14 17:56:12 +00:00
|
|
|
#ifdef DEVICE_POLLING
|
Big polling(4) cleanup.
o Axe poll in trap.
o Axe IFF_POLLING flag from if_flags.
o Rework revision 1.21 (Giant removal), in such a way that
poll_mtx is not dropped during call to polling handler.
This fixes problem with idle polling.
o Make registration and deregistration from polling in a
functional way, insted of next tick/interrupt.
o Obsolete kern.polling.enable. Polling is turned on/off
with ifconfig.
Detailed kern_poll.c changes:
- Remove polling handler flags, introduced in 1.21. The are not
needed now.
- Forget and do not check if_flags, if_capenable and if_drv_flags.
- Call all registered polling handlers unconditionally.
- Do not drop poll_mtx, when entering polling handlers.
- In ether_poll() NET_LOCK_GIANT prior to locking poll_mtx.
- In netisr_poll() axe the block, where polling code asks drivers
to unregister.
- In netisr_poll() and ether_poll() do polling always, if any
handlers are present.
- In ether_poll_[de]register() remove a lot of error hiding code. Assert
that arguments are correct, instead.
- In ether_poll_[de]register() use standard return values in case of
error or success.
- Introduce poll_switch() that is a sysctl handler for kern.polling.enable.
poll_switch() goes through interface list and enabled/disables polling.
A message that kern.polling.enable is deprecated is printed.
Detailed driver changes:
- On attach driver announces IFCAP_POLLING in if_capabilities, but
not in if_capenable.
- On detach driver calls ether_poll_deregister() if polling is enabled.
- In polling handler driver obtains its lock and checks IFF_DRV_RUNNING
flag. If there is no, then unlocks and returns.
- In ioctl handler driver checks for IFCAP_POLLING flag requested to
be set or cleared. Driver first calls ether_poll_[de]register(), then
obtains driver lock and [dis/en]ables interrupts.
- In interrupt handler driver checks IFCAP_POLLING flag in if_capenable.
If present, then returns.This is important to protect from spurious
interrupts.
Reviewed by: ru, sam, jhb
2005-10-01 18:56:19 +00:00
|
|
|
enum poll_cmd { POLL_ONLY, POLL_AND_CHECK_STATUS };
|
Device Polling code for -current.
Non-SMP, i386-only, no polling in the idle loop at the moment.
To use this code you must compile a kernel with
options DEVICE_POLLING
and at runtime enable polling with
sysctl kern.polling.enable=1
The percentage of CPU reserved to userland can be set with
sysctl kern.polling.user_frac=NN (default is 50)
while the remainder is used by polling device drivers and netisr's.
These are the only two variables that you should need to touch. There
are a few more parameters in kern.polling but the default values
are adequate for all purposes. See the code in kern_poll.c for
more details on them.
Polling in the idle loop will be implemented shortly by introducing
a kernel thread which does the job. Until then, the amount of CPU
dedicated to polling will never exceed (100-user_frac).
The equivalent (actually, better) code for -stable is at
http://info.iet.unipi.it/~luigi/polling/
and also supports polling in the idle loop.
NOTE to Alpha developers:
There is really nothing in this code that is i386-specific.
If you move the 2 lines supporting the new option from
sys/conf/{files,options}.i386 to sys/conf/{files,options} I am
pretty sure that this should work on the Alpha as well, just that
I do not have a suitable test box to try it. If someone feels like
trying it, I would appreciate it.
NOTE to other developers:
sure some things could be done better, and as always I am open to
constructive criticism, which a few of you have already given and
I greatly appreciated.
However, before proposing radical architectural changes, please
take some time to possibly try out this code, or at the very least
read the comments in kern_poll.c, especially re. the reason why I
am using a soft netisr and cannot (I believe) replace it with a
simple timeout.
Quick description of files touched by this commit:
sys/conf/files.i386
new file kern/kern_poll.c
sys/conf/options.i386
new option
sys/i386/i386/trap.c
poll in trap (disabled by default)
sys/kern/kern_clock.c
initialization and hardclock hooks.
sys/kern/kern_intr.c
minor swi_net changes
sys/kern/kern_poll.c
the bulk of the code.
sys/net/if.h
new flag
sys/net/if_var.h
declaration for functions used in device drivers.
sys/net/netisr.h
NETISR_POLL
sys/dev/fxp/if_fxp.c
sys/dev/fxp/if_fxpvar.h
sys/pci/if_dc.c
sys/pci/if_dcreg.h
sys/pci/if_sis.c
sys/pci/if_sisreg.h
device driver modifications
2001-12-14 17:56:12 +00:00
|
|
|
|
2002-03-19 21:54:18 +00:00
|
|
|
typedef void poll_handler_t(struct ifnet *ifp, enum poll_cmd cmd, int count);
|
|
|
|
|
int ether_poll_register(poll_handler_t *h, struct ifnet *ifp);
|
|
|
|
|
int ether_poll_deregister(struct ifnet *ifp);
|
Device Polling code for -current.
Non-SMP, i386-only, no polling in the idle loop at the moment.
To use this code you must compile a kernel with
options DEVICE_POLLING
and at runtime enable polling with
sysctl kern.polling.enable=1
The percentage of CPU reserved to userland can be set with
sysctl kern.polling.user_frac=NN (default is 50)
while the remainder is used by polling device drivers and netisr's.
These are the only two variables that you should need to touch. There
are a few more parameters in kern.polling but the default values
are adequate for all purposes. See the code in kern_poll.c for
more details on them.
Polling in the idle loop will be implemented shortly by introducing
a kernel thread which does the job. Until then, the amount of CPU
dedicated to polling will never exceed (100-user_frac).
The equivalent (actually, better) code for -stable is at
http://info.iet.unipi.it/~luigi/polling/
and also supports polling in the idle loop.
NOTE to Alpha developers:
There is really nothing in this code that is i386-specific.
If you move the 2 lines supporting the new option from
sys/conf/{files,options}.i386 to sys/conf/{files,options} I am
pretty sure that this should work on the Alpha as well, just that
I do not have a suitable test box to try it. If someone feels like
trying it, I would appreciate it.
NOTE to other developers:
sure some things could be done better, and as always I am open to
constructive criticism, which a few of you have already given and
I greatly appreciated.
However, before proposing radical architectural changes, please
take some time to possibly try out this code, or at the very least
read the comments in kern_poll.c, especially re. the reason why I
am using a soft netisr and cannot (I believe) replace it with a
simple timeout.
Quick description of files touched by this commit:
sys/conf/files.i386
new file kern/kern_poll.c
sys/conf/options.i386
new option
sys/i386/i386/trap.c
poll in trap (disabled by default)
sys/kern/kern_clock.c
initialization and hardclock hooks.
sys/kern/kern_intr.c
minor swi_net changes
sys/kern/kern_poll.c
the bulk of the code.
sys/net/if.h
new flag
sys/net/if_var.h
declaration for functions used in device drivers.
sys/net/netisr.h
NETISR_POLL
sys/dev/fxp/if_fxp.c
sys/dev/fxp/if_fxpvar.h
sys/pci/if_dc.c
sys/pci/if_dcreg.h
sys/pci/if_sis.c
sys/pci/if_sisreg.h
device driver modifications
2001-12-14 17:56:12 +00:00
|
|
|
#endif /* DEVICE_POLLING */
|
|
|
|
|
|
1999-12-29 04:46:21 +00:00
|
|
|
#endif /* _KERNEL */
|
1997-01-03 19:50:26 +00:00
|
|
|
|
|
|
|
|
#endif /* !_NET_IF_VAR_H_ */
|
