d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
173bc8ca15
freebsd-skq/sys/net/if_bridge.c

3457 lines
82 KiB
C
Raw Normal View History

Sync if_bridge to NetBSD r1.31 Rename conflicting variables when handling SNAP Ethernet frames. Obtained from: NetBSD Approved by: mlaier (mentor) Approved by: re (blanket)
2005-06-29 19:23:32 +00:00
/* $NetBSD: if_bridge.c,v 1.31 2005/06/01 19:45:34 jdc Exp $ */
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* Copyright 2001 Wasabi Systems, Inc.
* All rights reserved.
*
* Written by Jason R. Thorpe for Wasabi Systems, Inc.
*
* 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.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed for the NetBSD Project by
* Wasabi Systems, Inc.
* 4. The name of Wasabi Systems, Inc. may not be used to endorse
* or promote products derived from this software without specific prior
* written permission.
*
* THIS SOFTWARE IS PROVIDED BY WASABI SYSTEMS, INC. ``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 WASABI SYSTEMS, INC
* 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.
*/
/*
* Copyright (c) 1999, 2000 Jason L. Wright (jason@thought.net)
* 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.
*
* OpenBSD: if_bridge.c,v 1.60 2001/06/15 03:38:33 itojun Exp
*/
/*
* Network interface bridge support.
*
* TODO:
*
* - Currently only supports Ethernet-like interfaces (Ethernet,
* 802.11, VLANs on Ethernet, etc.) Figure out a nice way
* to bridge other types of interfaces (FDDI-FDDI, and maybe
* consider heterogenous bridges).
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_inet.h"
#include "opt_inet6.h"
#include <sys/param.h>
#include <sys/mbuf.h>
#include <sys/malloc.h>
#include <sys/protosw.h>
#include <sys/systm.h>
#include <sys/time.h>
#include <sys/socket.h> /* for net/if.h */
#include <sys/sockio.h>
#include <sys/ctype.h> /* string functions */
#include <sys/kernel.h>
#include <sys/random.h>
- Use the new bridgestp callback to once again flush our bridge routes when an interface is disabled. - Log port changes to syslog, defaulting to off
2006-08-02 03:54:28 +00:00
#include <sys/syslog.h>
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
#include <sys/sysctl.h>
#include <vm/uma.h>
#include <sys/module.h>
Sweep kernel replacing suser(9) calls with priv(9) calls, assigning specific privilege names to a broad range of privileges. These may require some future tweaking. Sponsored by: nCircle Network Security, Inc. Obtained from: TrustedBSD Project Discussed on: arch@ Reviewed (at least in part) by: mlaier, jmg, pjd, bde, ceri, Alex Lyashkov <umka at sevcity dot net>, Skip Ford <skip dot ford at verizon dot net>, Antoine Brodin <antoine dot brodin at laposte dot net>
2006-11-06 13:42:10 +00:00
#include <sys/priv.h>
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
#include <sys/proc.h>
#include <sys/lock.h>
#include <sys/mutex.h>
Conditionally compile out V_ globals while instantiating the appropriate container structures, depending on VIMAGE_GLOBALS compile time option. Make VIMAGE_GLOBALS a new compile-time option, which by default will not be defined, resulting in instatiations of global variables selected for V_irtualization (enclosed in #ifdef VIMAGE_GLOBALS blocks) to be effectively compiled out. Instantiate new global container structures to hold V_irtualized variables: vnet_net_0, vnet_inet_0, vnet_inet6_0, vnet_ipsec_0, vnet_netgraph_0, and vnet_gif_0. Update the VSYM() macro so that depending on VIMAGE_GLOBALS the V_ macros resolve either to the original globals, or to fields inside container structures, i.e. effectively #ifdef VIMAGE_GLOBALS #define V_rt_tables rt_tables #else #define V_rt_tables vnet_net_0._rt_tables #endif Update SYSCTL_V_*() macros to operate either on globals or on fields inside container structs. Extend the internal kldsym() lookups with the ability to resolve selected fields inside the virtualization container structs. This applies only to the fields which are explicitly registered for kldsym() visibility via VNET_MOD_DECLARE() and vnet_mod_register(), currently this is done only in sys/net/if.c. Fix a few broken instances of MODULE_GLOBAL() macro use in SCTP code, and modify the MODULE_GLOBAL() macro to resolve to V_ macros, which in turn result in proper code being generated depending on VIMAGE_GLOBALS. De-virtualize local static variables in sys/contrib/pf/net/pf_subr.c which were prematurely V_irtualized by automated V_ prepending scripts during earlier merging steps. PF virtualization will be done separately, most probably after next PF import. Convert a few variable initializations at instantiation to initialization in init functions, most notably in ipfw. Also convert TUNABLE_INT() initializers for V_ variables to TUNABLE_FETCH_INT() in initializer functions. Discussed at: devsummit Strassburg Reviewed by: bz, julian Approved by: julian (mentor) Obtained from: //depot/projects/vimage-commit2/... X-MFC after: never Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-12-10 23:12:39 +00:00
#include <sys/rwlock.h>
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
#include <net/bpf.h>
#include <net/if.h>
#include <net/if_clone.h>
#include <net/if_dl.h>
#include <net/if_types.h>
#include <net/if_var.h>
#include <net/pfil.h>
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
#include <net/vnet.h>
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
#include <netinet/in.h> /* for struct arpcom */
#include <netinet/in_systm.h>
#include <netinet/in_var.h>
#include <netinet/ip.h>
#include <netinet/ip_var.h>
#ifdef INET6
#include <netinet/ip6.h>
#include <netinet6/ip6_var.h>
#endif
carp(4) allows people to share a set of IP addresses and can only use IPv4/v6 for inter-node communication (according to my reading). Properly wrap the carp callouts in INET || INET6 and refelect this in sys/conf/files as well. While in theory this should be ok, it might be a bit optimistic to think that carp could build with inet6 only[1]. Discussed with: mlaier [1]
2009-06-11 10:26:38 +00:00
#if defined(INET) || defined(INET6)
Allow bridge and carp to play nicely together by returning the packet if its destined for a carp interface. Obtained from: OpenBSD MFC after: 2 weeks
2006-06-08 23:40:16 +00:00
#include <netinet/ip_carp.h>
#endif
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
#include <machine/in_cksum.h>
#include <netinet/if_ether.h> /* for struct arpcom */
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
#include <net/bridgestp.h>
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
#include <net/if_bridgevar.h>
#include <net/if_llc.h>
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
#include <net/if_vlan_var.h>
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
#include <net/route.h>
#include <netinet/ip_fw.h>
Start splitting ip_fw2.c and ip_fw.h into smaller components. At this time we pull out from ip_fw2.c the logging functions, and support for dynamic rules, and move kernel-only stuff into netinet/ipfw/ip_fw_private.h No ABI change involved in this commit, unless I made some mistake. ip_fw.h has changed, though not in the userland-visible part. Files touched by this commit: conf/files now references the two new source files netinet/ip_fw.h remove kernel-only definitions gone into netinet/ipfw/ip_fw_private.h. netinet/ipfw/ip_fw_private.h new file with kernel-specific ipfw definitions netinet/ipfw/ip_fw_log.c ipfw_log and related functions netinet/ipfw/ip_fw_dynamic.c code related to dynamic rules netinet/ipfw/ip_fw2.c removed the pieces that goes in the new files netinet/ipfw/ip_fw_nat.c minor rearrangement to remove LOOKUP_NAT from the main headers. This require a new function pointer. A bunch of other kernel files that included netinet/ip_fw.h now require netinet/ipfw/ip_fw_private.h as well. Not 100% sure i caught all of them. MFC after: 1 month
2009-12-15 16:15:14 +00:00
#include <netinet/ipfw/ip_fw_private.h>
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* Size of the route hash table. Must be a power of two.
*/
#ifndef BRIDGE_RTHASH_SIZE
#define BRIDGE_RTHASH_SIZE 1024
#endif
#define BRIDGE_RTHASH_MASK (BRIDGE_RTHASH_SIZE - 1)
/*
* Maximum number of addresses to cache.
*/
#ifndef BRIDGE_RTABLE_MAX
#define BRIDGE_RTABLE_MAX 100
#endif
/*
* Timeout (in seconds) for entries learned dynamically.
*/
#ifndef BRIDGE_RTABLE_TIMEOUT
#define BRIDGE_RTABLE_TIMEOUT (20 * 60) /* same as ARP */
#endif
/*
* Number of seconds between walks of the route list.
*/
#ifndef BRIDGE_RTABLE_PRUNE_PERIOD
#define BRIDGE_RTABLE_PRUNE_PERIOD (5 * 60)
#endif
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
/*
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
* List of capabilities to possibly mask on the member interface.
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
*/
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
#define BRIDGE_IFCAPS_MASK (IFCAP_TOE|IFCAP_TSO|IFCAP_TXCSUM)
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
When bridging LRO is causing a problem, the believe that it would work as long as all interfaces have TSO seems to be false, until the matter gets sorted out just disable LRO completely.
2009-08-24 21:04:51 +00:00
/*
* List of capabilities to strip
*/
#define BRIDGE_IFCAPS_STRIP IFCAP_LRO
Remove the dependency of bridgestp.h on if_bridgevar.h by moving a couple of private structures to if_bridge.c.
2006-07-27 21:01:48 +00:00
/*
* Bridge interface list entry.
*/
struct bridge_iflist {
LIST_ENTRY(bridge_iflist) bif_next;
struct ifnet *bif_ifp; /* member if */
struct bstp_port bif_stp; /* STP state */
uint32_t bif_flags; /* member if flags */
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
int bif_savedcaps; /* saved capabilities */
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
uint32_t bif_addrmax; /* max # of addresses */
uint32_t bif_addrcnt; /* cur. # of addresses */
uint32_t bif_addrexceeded;/* # of address violations */
Remove the dependency of bridgestp.h on if_bridgevar.h by moving a couple of private structures to if_bridge.c.
2006-07-27 21:01:48 +00:00
};
/*
* Bridge route node.
*/
struct bridge_rtnode {
LIST_ENTRY(bridge_rtnode) brt_hash; /* hash table linkage */
LIST_ENTRY(bridge_rtnode) brt_list; /* list linkage */
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
struct bridge_iflist *brt_dst; /* destination if */
Remove the dependency of bridgestp.h on if_bridgevar.h by moving a couple of private structures to if_bridge.c.
2006-07-27 21:01:48 +00:00
unsigned long brt_expire; /* expiration time */
uint8_t brt_flags; /* address flags */
uint8_t brt_addr[ETHER_ADDR_LEN];
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
uint16_t brt_vlan; /* vlan id */
Remove the dependency of bridgestp.h on if_bridgevar.h by moving a couple of private structures to if_bridge.c.
2006-07-27 21:01:48 +00:00
};
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
#define brt_ifp brt_dst->bif_ifp
Remove the dependency of bridgestp.h on if_bridgevar.h by moving a couple of private structures to if_bridge.c.
2006-07-27 21:01:48 +00:00
/*
* Software state for each bridge.
*/
struct bridge_softc {
struct ifnet *sc_ifp; /* make this an interface */
LIST_ENTRY(bridge_softc) sc_list;
struct mtx sc_mtx;
struct cv sc_cv;
uint32_t sc_brtmax; /* max # of addresses */
uint32_t sc_brtcnt; /* cur. # of addresses */
uint32_t sc_brttimeout; /* rt timeout in seconds */
struct callout sc_brcallout; /* bridge callout */
uint32_t sc_iflist_ref; /* refcount for sc_iflist */
uint32_t sc_iflist_xcnt; /* refcount for sc_iflist */
LIST_HEAD(, bridge_iflist) sc_iflist; /* member interface list */
LIST_HEAD(, bridge_rtnode) *sc_rthash; /* our forwarding table */
LIST_HEAD(, bridge_rtnode) sc_rtlist; /* list version of above */
uint32_t sc_rthash_key; /* key for hash */
LIST_HEAD(, bridge_iflist) sc_spanlist; /* span ports list */
struct bstp_state sc_stp; /* STP state */
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
uint32_t sc_brtexceeded; /* # of cache drops */
bridge_delete_member is called via the event handler from if_detach after the LLADDR is reclaimed which causes a null pointer deref with inherit_mac enabled. Record the ifnet pointer of the interface and then compare that to find when to re-assign the bridge address. Submitted by: sam
2009-02-13 19:20:25 +00:00
struct ifnet *sc_ifaddr; /* member mac copied from */
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
u_char sc_defaddr[6]; /* Default MAC address */
Remove the dependency of bridgestp.h on if_bridgevar.h by moving a couple of private structures to if_bridge.c.
2006-07-27 21:01:48 +00:00
};
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
static struct mtx bridge_list_mtx;
eventhandler_tag bridge_detach_cookie = NULL;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
int bridge_rtable_prune_period = BRIDGE_RTABLE_PRUNE_PERIOD;
uma_zone_t bridge_rtnode_zone;
Revise network interface cloning to take an optional opaque parameter that can specify configuration parameters: o rev cloner api's to add optional parameter block o add SIOCCREATE2 that accepts parameter data o rev vlan support to use new api (maintain old code) Reviewed by: arch@
2006-07-09 06:04:01 +00:00
static int bridge_clone_create(struct if_clone *, int, caddr_t);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void bridge_clone_destroy(struct ifnet *);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int bridge_ioctl(struct ifnet *, u_long, caddr_t);
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
static void bridge_mutecaps(struct bridge_softc *);
static void bridge_set_ifcap(struct bridge_softc *, struct bridge_iflist *,
int);
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
static void bridge_ifdetach(void *arg __unused, struct ifnet *);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
static void bridge_init(void *);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void bridge_dummynet(struct mbuf *, struct ifnet *);
static void bridge_stop(struct ifnet *, int);
static void bridge_start(struct ifnet *);
static struct mbuf *bridge_input(struct ifnet *, struct mbuf *);
static int bridge_output(struct ifnet *, struct mbuf *, struct sockaddr *,
struct rtentry *);
Remove the dependency of bridgestp.h on if_bridgevar.h by moving a couple of private structures to if_bridge.c.
2006-07-27 21:01:48 +00:00
static void bridge_enqueue(struct bridge_softc *, struct ifnet *,
struct mbuf *);
static void bridge_rtdelete(struct bridge_softc *, struct ifnet *ifp, int);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
static void bridge_forward(struct bridge_softc *, struct bridge_iflist *,
struct mbuf *m);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void bridge_timer(void *);
static void bridge_broadcast(struct bridge_softc *, struct ifnet *,
struct mbuf *, int);
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
static void bridge_span(struct bridge_softc *, struct mbuf *);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int bridge_rtupdate(struct bridge_softc *, const uint8_t *,
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
uint16_t, struct bridge_iflist *, int, uint8_t);
static struct ifnet *bridge_rtlookup(struct bridge_softc *, const uint8_t *,
uint16_t);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void bridge_rttrim(struct bridge_softc *);
static void bridge_rtage(struct bridge_softc *);
static void bridge_rtflush(struct bridge_softc *, int);
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
static int bridge_rtdaddr(struct bridge_softc *, const uint8_t *,
uint16_t);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int bridge_rtable_init(struct bridge_softc *);
static void bridge_rtable_fini(struct bridge_softc *);
Fix two bugs with the bridge - code expects memcmp() to return a signed value, our memcmp() returns 0 if args are equal and > 0 if not. - It's possible to hijack interface for static entry. If bridge recieves packet from interface marked as learning it will replace the bridge_rtnode entry for the source address even if such entry marked as static. Submitted by: Gleb Kurtsov <k-gleb yandex.ru> MFC after: 3 days
2006-01-31 21:21:28 +00:00
static int bridge_rtnode_addr_cmp(const uint8_t *, const uint8_t *);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static struct bridge_rtnode *bridge_rtnode_lookup(struct bridge_softc *,
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
const uint8_t *, uint16_t);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int bridge_rtnode_insert(struct bridge_softc *,
struct bridge_rtnode *);
static void bridge_rtnode_destroy(struct bridge_softc *,
struct bridge_rtnode *);
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
static void bridge_rtable_expire(struct ifnet *, int);
- Use the new bridgestp callback to once again flush our bridge routes when an interface is disabled. - Log port changes to syslog, defaulting to off
2006-08-02 03:54:28 +00:00
static void bridge_state_change(struct ifnet *, int);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static struct bridge_iflist *bridge_lookup_member(struct bridge_softc *,
const char *name);
static struct bridge_iflist *bridge_lookup_member_if(struct bridge_softc *,
struct ifnet *ifp);
static void bridge_delete_member(struct bridge_softc *,
If we have been called from ether_ifdetach() then do not try and clear the promisc flag from the member interface, this is a no-op anyway since the interface is disappearing. The driver may have already released its resources such as miibus and this is likely to panic the kernel. Submitted and tested by: Wojciech A. Koszek MFC after: 2 weeks
2005-10-23 22:30:07 +00:00
struct bridge_iflist *, int);
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
static void bridge_delete_span(struct bridge_softc *,
struct bridge_iflist *);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int bridge_ioctl_add(struct bridge_softc *, void *);
static int bridge_ioctl_del(struct bridge_softc *, void *);
static int bridge_ioctl_gifflags(struct bridge_softc *, void *);
static int bridge_ioctl_sifflags(struct bridge_softc *, void *);
static int bridge_ioctl_scache(struct bridge_softc *, void *);
static int bridge_ioctl_gcache(struct bridge_softc *, void *);
static int bridge_ioctl_gifs(struct bridge_softc *, void *);
static int bridge_ioctl_rts(struct bridge_softc *, void *);
static int bridge_ioctl_saddr(struct bridge_softc *, void *);
static int bridge_ioctl_sto(struct bridge_softc *, void *);
static int bridge_ioctl_gto(struct bridge_softc *, void *);
static int bridge_ioctl_daddr(struct bridge_softc *, void *);
static int bridge_ioctl_flush(struct bridge_softc *, void *);
static int bridge_ioctl_gpri(struct bridge_softc *, void *);
static int bridge_ioctl_spri(struct bridge_softc *, void *);
static int bridge_ioctl_ght(struct bridge_softc *, void *);
static int bridge_ioctl_sht(struct bridge_softc *, void *);
static int bridge_ioctl_gfd(struct bridge_softc *, void *);
static int bridge_ioctl_sfd(struct bridge_softc *, void *);
static int bridge_ioctl_gma(struct bridge_softc *, void *);
static int bridge_ioctl_sma(struct bridge_softc *, void *);
static int bridge_ioctl_sifprio(struct bridge_softc *, void *);
static int bridge_ioctl_sifcost(struct bridge_softc *, void *);
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
static int bridge_ioctl_sifmaxaddr(struct bridge_softc *, void *);
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
static int bridge_ioctl_addspan(struct bridge_softc *, void *);
static int bridge_ioctl_delspan(struct bridge_softc *, void *);
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
static int bridge_ioctl_gbparam(struct bridge_softc *, void *);
static int bridge_ioctl_grte(struct bridge_softc *, void *);
static int bridge_ioctl_gifsstp(struct bridge_softc *, void *);
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
static int bridge_ioctl_sproto(struct bridge_softc *, void *);
static int bridge_ioctl_stxhc(struct bridge_softc *, void *);
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int bridge_pfil(struct mbuf **, struct ifnet *, struct ifnet *,
int);
static int bridge_ip_checkbasic(struct mbuf **mp);
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
#ifdef INET6
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int bridge_ip6_checkbasic(struct mbuf **mp);
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
#endif /* INET6 */
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
static int bridge_fragment(struct ifnet *, struct mbuf *,
struct ether_header *, int, struct llc *);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
/* The default bridge vlan is 1 (IEEE 802.1Q-2003 Table 9-2) */
#define VLANTAGOF(_m) \
(_m->m_flags & M_VLANTAG) ? EVL_VLANOFTAG(_m->m_pkthdr.ether_vtag) : 1
Change the passing of callbacks to a struct in case this needs to be extended in the future.
2007-03-09 19:34:55 +00:00
static struct bstp_cb_ops bridge_ops = {
.bcb_state = bridge_state_change,
.bcb_rtage = bridge_rtable_expire
};
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
SYSCTL_DECL(_net_link);
SYSCTL_NODE(_net_link, IFT_BRIDGE, bridge, CTLFLAG_RW, 0, "Bridge");
When pfil(9) is enabled the bridge only considers ETHERTYPE_ARP, ETHERTYPE_IP and ETHERTYPE_IPV6 frames. Change this to be a sysctl knob so that is able to still bridge non-IP packets if desired. Also return early if all pfil_* sysctls are turned off, the user obviously does not want to filter on the bridge.
2005-12-29 09:39:15 +00:00
static int pfil_onlyip = 1; /* only pass IP[46] packets when pfil is enabled */
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
static int pfil_bridge = 1; /* run pfil hooks on the bridge interface */
static int pfil_member = 1; /* run pfil hooks on the member interface */
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
static int pfil_ipfw = 0; /* layer2 filter with ipfw */
Revert r1.80 as the ethernet header was inadvertently stripped from ARP packets. Reimplement this correctly and use a sysctl that defaults to off so the user doesnt get any suprises if ipfw blocks the ARP packet. MFC after: 3 days
2006-09-22 21:57:52 +00:00
static int pfil_ipfw_arp = 0; /* layer2 filter with ipfw */
Allow additional packet filtering on the physical interface for locally destined packets, disabled by default. PR: kern/116051 Submitted by: Eygene Ryabinkin Approved by: re (bmah) MFC after: 2 weeks
2007-09-16 21:09:15 +00:00
static int pfil_local_phys = 0; /* run pfil hooks on the physical interface for
locally destined packets */
- Use the new bridgestp callback to once again flush our bridge routes when an interface is disabled. - Log port changes to syslog, defaulting to off
2006-08-02 03:54:28 +00:00
static int log_stp = 0; /* log STP state changes */
Put the bridge mac inheritance behind a sysctl with the default off as this still needs all the edge cases fixed. Submitted by: Eygene Ryabinkin
2008-09-08 03:28:26 +00:00
static int bridge_inherit_mac = 0; /* share MAC with first bridge member */
When pfil(9) is enabled the bridge only considers ETHERTYPE_ARP, ETHERTYPE_IP and ETHERTYPE_IPV6 frames. Change this to be a sysctl knob so that is able to still bridge non-IP packets if desired. Also return early if all pfil_* sysctls are turned off, the user obviously does not want to filter on the bridge.
2005-12-29 09:39:15 +00:00
SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_onlyip, CTLFLAG_RW,
&pfil_onlyip, 0, "Only pass IP packets when pfil is enabled");
Revert r1.80 as the ethernet header was inadvertently stripped from ARP packets. Reimplement this correctly and use a sysctl that defaults to off so the user doesnt get any suprises if ipfw blocks the ARP packet. MFC after: 3 days
2006-09-22 21:57:52 +00:00
SYSCTL_INT(_net_link_bridge, OID_AUTO, ipfw_arp, CTLFLAG_RW,
&pfil_ipfw_arp, 0, "Filter ARP packets through IPFW layer2");
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_bridge, CTLFLAG_RW,
&pfil_bridge, 0, "Packet filter on the bridge interface");
SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_member, CTLFLAG_RW,
&pfil_member, 0, "Packet filter on the member interface");
Allow additional packet filtering on the physical interface for locally destined packets, disabled by default. PR: kern/116051 Submitted by: Eygene Ryabinkin Approved by: re (bmah) MFC after: 2 weeks
2007-09-16 21:09:15 +00:00
SYSCTL_INT(_net_link_bridge, OID_AUTO, pfil_local_phys, CTLFLAG_RW,
&pfil_local_phys, 0,
"Packet filter on the physical interface for locally destined packets");
- Use the new bridgestp callback to once again flush our bridge routes when an interface is disabled. - Log port changes to syslog, defaulting to off
2006-08-02 03:54:28 +00:00
SYSCTL_INT(_net_link_bridge, OID_AUTO, log_stp, CTLFLAG_RW,
&log_stp, 0, "Log STP state changes");
Put the bridge mac inheritance behind a sysctl with the default off as this still needs all the edge cases fixed. Submitted by: Eygene Ryabinkin
2008-09-08 03:28:26 +00:00
SYSCTL_INT(_net_link_bridge, OID_AUTO, inherit_mac, CTLFLAG_RW,
&bridge_inherit_mac, 0,
"Inherit MAC address from the first bridge member");
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
struct bridge_control {
int (*bc_func)(struct bridge_softc *, void *);
int bc_argsize;
int bc_flags;
};
#define BC_F_COPYIN 0x01 /* copy arguments in */
#define BC_F_COPYOUT 0x02 /* copy arguments out */
#define BC_F_SUSER 0x04 /* do super-user check */
const struct bridge_control bridge_control_table[] = {
{ bridge_ioctl_add, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_del, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_gifflags, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_COPYOUT },
{ bridge_ioctl_sifflags, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_scache, sizeof(struct ifbrparam),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_gcache, sizeof(struct ifbrparam),
BC_F_COPYOUT },
{ bridge_ioctl_gifs, sizeof(struct ifbifconf),
BC_F_COPYIN|BC_F_COPYOUT },
{ bridge_ioctl_rts, sizeof(struct ifbaconf),
BC_F_COPYIN|BC_F_COPYOUT },
{ bridge_ioctl_saddr, sizeof(struct ifbareq),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_sto, sizeof(struct ifbrparam),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_gto, sizeof(struct ifbrparam),
BC_F_COPYOUT },
{ bridge_ioctl_daddr, sizeof(struct ifbareq),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_flush, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_gpri, sizeof(struct ifbrparam),
BC_F_COPYOUT },
{ bridge_ioctl_spri, sizeof(struct ifbrparam),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_ght, sizeof(struct ifbrparam),
BC_F_COPYOUT },
{ bridge_ioctl_sht, sizeof(struct ifbrparam),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_gfd, sizeof(struct ifbrparam),
BC_F_COPYOUT },
{ bridge_ioctl_sfd, sizeof(struct ifbrparam),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_gma, sizeof(struct ifbrparam),
BC_F_COPYOUT },
{ bridge_ioctl_sma, sizeof(struct ifbrparam),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_sifprio, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_sifcost, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_SUSER },
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
{ bridge_ioctl_addspan, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_delspan, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_SUSER },
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
{ bridge_ioctl_gbparam, sizeof(struct ifbropreq),
BC_F_COPYOUT },
{ bridge_ioctl_grte, sizeof(struct ifbrparam),
BC_F_COPYOUT },
{ bridge_ioctl_gifsstp, sizeof(struct ifbpstpconf),
Fix SIOCGDRVSPEC/BRDGGIFSSTP ioctl: make it copyin() the user provided buffer length before trying to use it. Reviewed by: thompsa Approved by: bz (mentor) MFC after: 3 days
2006-12-03 21:50:57 +00:00
BC_F_COPYIN|BC_F_COPYOUT },
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
{ bridge_ioctl_sproto, sizeof(struct ifbrparam),
BC_F_COPYIN|BC_F_SUSER },
{ bridge_ioctl_stxhc, sizeof(struct ifbrparam),
BC_F_COPYIN|BC_F_SUSER },
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
{ bridge_ioctl_sifmaxaddr, sizeof(struct ifbreq),
BC_F_COPYIN|BC_F_SUSER },
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
};
const int bridge_control_table_size =
sizeof(bridge_control_table) / sizeof(bridge_control_table[0]);
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
LIST_HEAD(, bridge_softc) bridge_list;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
IFC_SIMPLE_DECLARE(bridge, 0);
static int
bridge_modevent(module_t mod, int type, void *data)
{
switch (type) {
case MOD_LOAD:
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
mtx_init(&bridge_list_mtx, "if_bridge list", NULL, MTX_DEF);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if_clone_attach(&bridge_cloner);
bridge_rtnode_zone = uma_zcreate("bridge_rtnode",
sizeof(struct bridge_rtnode), NULL, NULL, NULL, NULL,
UMA_ALIGN_PTR, 0);
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
LIST_INIT(&bridge_list);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_input_p = bridge_input;
bridge_output_p = bridge_output;
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
bridge_dn_p = bridge_dummynet;
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
bridge_detach_cookie = EVENTHANDLER_REGISTER(
ifnet_departure_event, bridge_ifdetach, NULL,
EVENTHANDLER_PRI_ANY);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
break;
case MOD_UNLOAD:
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
EVENTHANDLER_DEREGISTER(ifnet_departure_event,
bridge_detach_cookie);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if_clone_detach(&bridge_cloner);
uma_zdestroy(bridge_rtnode_zone);
bridge_input_p = NULL;
bridge_output_p = NULL;
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
bridge_dn_p = NULL;
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
mtx_destroy(&bridge_list_mtx);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
break;
default:
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (EOPNOTSUPP);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (0);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
static moduledata_t bridge_mod = {
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
"if_bridge",
bridge_modevent,
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
0
};
DECLARE_MODULE(if_bridge, bridge_mod, SI_SUB_PSEUDO, SI_ORDER_ANY);
bridgestp is now a seperate module.
2006-07-26 22:15:15 +00:00
MODULE_DEPEND(if_bridge, bridgestp, 1, 1, 1);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
/*
* handler for net.link.bridge.pfil_ipfw
*/
static int
sysctl_pfil_ipfw(SYSCTL_HANDLER_ARGS)
{
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
int enable = pfil_ipfw;
int error;
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
error = sysctl_handle_int(oidp, &enable, 0, req);
enable = (enable) ? 1 : 0;
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
if (enable != pfil_ipfw) {
pfil_ipfw = enable;
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
/*
* Disable pfil so that ipfw doesnt run twice, if the user
* really wants both then they can re-enable pfil_bridge and/or
When pfil(9) is enabled the bridge only considers ETHERTYPE_ARP, ETHERTYPE_IP and ETHERTYPE_IPV6 frames. Change this to be a sysctl knob so that is able to still bridge non-IP packets if desired. Also return early if all pfil_* sysctls are turned off, the user obviously does not want to filter on the bridge.
2005-12-29 09:39:15 +00:00
* pfil_member. Also allow non-ip packets as ipfw can filter by
* layer2 type.
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
*/
if (pfil_ipfw) {
When pfil(9) is enabled the bridge only considers ETHERTYPE_ARP, ETHERTYPE_IP and ETHERTYPE_IPV6 frames. Change this to be a sysctl knob so that is able to still bridge non-IP packets if desired. Also return early if all pfil_* sysctls are turned off, the user obviously does not want to filter on the bridge.
2005-12-29 09:39:15 +00:00
pfil_onlyip = 0;
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
pfil_bridge = 0;
pfil_member = 0;
}
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
}
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (error);
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
}
SYSCTL_PROC(_net_link_bridge, OID_AUTO, ipfw, CTLTYPE_INT|CTLFLAG_RW,
&pfil_ipfw, 0, &sysctl_pfil_ipfw, "I", "Layer2 filter with IPFW");
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* bridge_clone_create:
*
* Create a new bridge instance.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Revise network interface cloning to take an optional opaque parameter that can specify configuration parameters: o rev cloner api's to add optional parameter block o add SIOCCREATE2 that accepts parameter data o rev vlan support to use new api (maintain old code) Reviewed by: arch@
2006-07-09 06:04:01 +00:00
bridge_clone_create(struct if_clone *ifc, int unit, caddr_t params)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
Since we are using random ethernet addresses for the bridge, it is possible that we might have address collisions, so make sure that this hardware address isn't already in use on another bridge. Submitted by: csjp MFC after: 1 month
2006-03-03 09:12:21 +00:00
struct bridge_softc *sc, *sc2;
struct ifnet *bifp, *ifp;
int retry;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
sc = malloc(sizeof(*sc), M_DEVBUF, M_WAITOK|M_ZERO);
Catch up with the struct ifnet changes and use if_alloc(). Reviewed by: brooks Approved by: mlaier (mentor)
2005-06-10 23:52:01 +00:00
ifp = sc->sc_ifp = if_alloc(IFT_ETHER);
if (ifp == NULL) {
free(sc, M_DEVBUF);
return (ENOSPC);
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Move the lock init until after if_alloc in case the allocation fails and we free the softc and return. MFC after: 3 days
2007-02-23 19:37:55 +00:00
BRIDGE_LOCK_INIT(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
sc->sc_brtmax = BRIDGE_RTABLE_MAX;
sc->sc_brttimeout = BRIDGE_RTABLE_TIMEOUT;
/* Initialize our routing table. */
bridge_rtable_init(sc);
Mark the callouts as MPSAFE as if_bridge has been giant-free since day 1. Use the SMP friendly callout_init_mtx() while we are here. Approved by: mlaier (mentor) MFC after: 3 days
2005-08-18 20:17:00 +00:00
callout_init_mtx(&sc->sc_brcallout, &sc->sc_mtx, 0);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
LIST_INIT(&sc->sc_iflist);
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
LIST_INIT(&sc->sc_spanlist);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
ifp->if_softc = sc;
if_initname(ifp, ifc->ifc_name, unit);
The bridge cant hear its own transmissions so set IFF_SIMPLEX. PR: kern/102361 Tested by: Radim Kolar <hsn@netmag.cz> MFC after: 3 days
2006-08-25 20:11:56 +00:00
ifp->if_flags = IFF_BROADCAST | IFF_SIMPLEX | IFF_MULTICAST;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
ifp->if_ioctl = bridge_ioctl;
ifp->if_start = bridge_start;
ifp->if_init = bridge_init;
ifp->if_type = IFT_BRIDGE;
IFQ_SET_MAXLEN(&ifp->if_snd, ifqmaxlen);
ifp->if_snd.ifq_drv_maxlen = ifqmaxlen;
IFQ_SET_READY(&ifp->if_snd);
/*
Use bit operations to get a locally administered address rather than using a hardcoded OUI code.
2006-06-12 22:43:37 +00:00
* Generate a random ethernet address with a locally administered
* address.
Since we are using random ethernet addresses for the bridge, it is possible that we might have address collisions, so make sure that this hardware address isn't already in use on another bridge. Submitted by: csjp MFC after: 1 month
2006-03-03 09:12:21 +00:00
*
* Since we are using random ethernet addresses for the bridge, it is
* possible that we might have address collisions, so make sure that
* this hardware address isn't already in use on another bridge.
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
*/
Since we are using random ethernet addresses for the bridge, it is possible that we might have address collisions, so make sure that this hardware address isn't already in use on another bridge. Submitted by: csjp MFC after: 1 month
2006-03-03 09:12:21 +00:00
for (retry = 1; retry != 0;) {
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
arc4rand(sc->sc_defaddr, ETHER_ADDR_LEN, 1);
sc->sc_defaddr[0] &= ~1; /* clear multicast bit */
sc->sc_defaddr[0] |= 2; /* set the LAA bit */
Since we are using random ethernet addresses for the bridge, it is possible that we might have address collisions, so make sure that this hardware address isn't already in use on another bridge. Submitted by: csjp MFC after: 1 month
2006-03-03 09:12:21 +00:00
retry = 0;
mtx_lock(&bridge_list_mtx);
LIST_FOREACH(sc2, &bridge_list, sc_list) {
bifp = sc2->sc_ifp;
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
if (memcmp(sc->sc_defaddr,
IF_LLADDR(bifp), ETHER_ADDR_LEN) == 0)
Since we are using random ethernet addresses for the bridge, it is possible that we might have address collisions, so make sure that this hardware address isn't already in use on another bridge. Submitted by: csjp MFC after: 1 month
2006-03-03 09:12:21 +00:00
retry = 1;
}
mtx_unlock(&bridge_list_mtx);
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Change the passing of callbacks to a struct in case this needs to be extended in the future.
2007-03-09 19:34:55 +00:00
bstp_attach(&sc->sc_stp, &bridge_ops);
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
ether_ifattach(ifp, sc->sc_defaddr);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/* Now undo some of the damage... */
ifp->if_baudrate = 0;
ifp->if_type = IFT_BRIDGE;
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
mtx_lock(&bridge_list_mtx);
LIST_INSERT_HEAD(&bridge_list, sc, sc_list);
mtx_unlock(&bridge_list_mtx);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
/*
* bridge_clone_destroy:
*
* Destroy a bridge instance.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_clone_destroy(struct ifnet *ifp)
{
struct bridge_softc *sc = ifp->if_softc;
struct bridge_iflist *bif;
BRIDGE_LOCK(sc);
bridge_stop(ifp, 1);
Fix a panic in softclock() if the interface is destroyed with a bpf consumer attached. This is caused by bpf_detachd clearing IFF_PROMISC on the interface which does a SIOCSIFFLAGS ioctl. The problem here is that while the interface has been stopped, IFF_UP has not been cleared so IFF_UP != IFF_DRV_RUNNING, this causes the ioctl function to init() the interface which resets the callouts. The destroy then completes and frees the softc but softclock will panic on a dead callout pointer. Ensure ifp->if_flags matches reality by clearing IFF_UP when we destroy. Silence from: rwatson Approved by: mlaier (mentor) MFC after: 3 days
2005-08-27 01:17:42 +00:00
ifp->if_flags &= ~IFF_UP;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
while ((bif = LIST_FIRST(&sc->sc_iflist)) != NULL)
If we have been called from ether_ifdetach() then do not try and clear the promisc flag from the member interface, this is a no-op anyway since the interface is disappearing. The driver may have already released its resources such as miibus and this is likely to panic the kernel. Submitted and tested by: Wojciech A. Koszek MFC after: 2 weeks
2005-10-23 22:30:07 +00:00
bridge_delete_member(sc, bif, 0);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
while ((bif = LIST_FIRST(&sc->sc_spanlist)) != NULL) {
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
bridge_delete_span(sc, bif);
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_UNLOCK(sc);
Mark the callouts as MPSAFE as if_bridge has been giant-free since day 1. Use the SMP friendly callout_init_mtx() while we are here. Approved by: mlaier (mentor) MFC after: 3 days
2005-08-18 20:17:00 +00:00
callout_drain(&sc->sc_brcallout);
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
mtx_lock(&bridge_list_mtx);
LIST_REMOVE(sc, sc_list);
mtx_unlock(&bridge_list_mtx);
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
bstp_detach(&sc->sc_stp);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
ether_ifdetach(ifp);
Catch up with the struct ifnet changes and use if_alloc(). Reviewed by: brooks Approved by: mlaier (mentor)
2005-06-10 23:52:01 +00:00
if_free_type(ifp, IFT_ETHER);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/* Tear down the routing table. */
bridge_rtable_fini(sc);
BRIDGE_LOCK_DESTROY(sc);
free(sc, M_DEVBUF);
}
/*
* bridge_ioctl:
*
* Handle a control request from the operator.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl(struct ifnet *ifp, u_long cmd, caddr_t data)
{
struct bridge_softc *sc = ifp->if_softc;
- Check if_type of "addm <interface>" before setting the interface's MTU to the if_bridge(4) interface. This fixes a bug that MTU value of "addm <interface>" is used even when it is invalid for the if_bridge(4) member: # ifconfig bridge0 create # ifconfig bridge0 bridge0: flags=8802<BROADCAST,SIMPLEX,MULTICAST> metric 0 mtu 1500 ... # ifconfig bridge0 addm lo0 ifconfig: BRDGADD lo0: Invalid argument # ifconfig bridge0 bridge0: flags=8802<BROADCAST,SIMPLEX,MULTICAST> metric 0 mtu 16384 ... - Do not ignore MTU value of an interface even when if_type == IFT_GIF. This fixes MTU mismatch when an if_bridge(4) interface has a gif(4) interface and no other interface as the member, and it is directly used for L2 communication with EtherIP tunneling enabled. - Implement SIOCSIFMTU ioctl. Changing the MTU is allowed only when all members have the same MTU value.
2010-01-31 08:16:37 +00:00
struct ifreq *ifr = (struct ifreq *)data;
struct bridge_iflist *bif;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
struct thread *td = curthread;
union {
struct ifbreq ifbreq;
struct ifbifconf ifbifconf;
struct ifbareq ifbareq;
struct ifbaconf ifbaconf;
struct ifbrparam ifbrparam;
Sync with the OpenBSD port of RSTP - use flags rather than sperate ioctls for edge, p2p - implement p2p and autop2p flags - define large pathcost constant as ULL - show bridgeid and rootid in ifconfig Obtained from: Reyk Floeter <reyk@openbsd.org>
2006-11-27 00:35:40 +00:00
struct ifbropreq ifbropreq;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
} args;
struct ifdrv *ifd = (struct ifdrv *) data;
const struct bridge_control *bc;
int error = 0;
switch (cmd) {
Add support for multicast to the bridge and allow inet6 addresses to be assigned to the interface. IPv6 auto-configuration is disabled. An IPv6 link-local address has a link-local scope within one link, the spec is unclear for the bridge case and it may cause scope violation. An address can be assigned in the usual way; ifconfig bridge0 inet6 xxxx:... Tested by: bmah Reviewed by: ume (netinet6) Approved by: mlaier (mentor) MFC after: 1 week
2005-09-06 21:11:59 +00:00
case SIOCADDMULTI:
case SIOCDELMULTI:
break;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
case SIOCGDRVSPEC:
case SIOCSDRVSPEC:
if (ifd->ifd_cmd >= bridge_control_table_size) {
error = EINVAL;
break;
}
bc = &bridge_control_table[ifd->ifd_cmd];
if (cmd == SIOCGDRVSPEC &&
(bc->bc_flags & BC_F_COPYOUT) == 0) {
error = EINVAL;
break;
}
else if (cmd == SIOCSDRVSPEC &&
(bc->bc_flags & BC_F_COPYOUT) != 0) {
error = EINVAL;
break;
}
if (bc->bc_flags & BC_F_SUSER) {
Sweep kernel replacing suser(9) calls with priv(9) calls, assigning specific privilege names to a broad range of privileges. These may require some future tweaking. Sponsored by: nCircle Network Security, Inc. Obtained from: TrustedBSD Project Discussed on: arch@ Reviewed (at least in part) by: mlaier, jmg, pjd, bde, ceri, Alex Lyashkov <umka at sevcity dot net>, Skip Ford <skip dot ford at verizon dot net>, Antoine Brodin <antoine dot brodin at laposte dot net>
2006-11-06 13:42:10 +00:00
error = priv_check(td, PRIV_NET_BRIDGE);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (error)
break;
}
if (ifd->ifd_len != bc->bc_argsize ||
ifd->ifd_len > sizeof(args)) {
error = EINVAL;
break;
}
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
bzero(&args, sizeof(args));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (bc->bc_flags & BC_F_COPYIN) {
error = copyin(ifd->ifd_data, &args, ifd->ifd_len);
if (error)
break;
}
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
BRIDGE_LOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
error = (*bc->bc_func)(sc, &args);
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
BRIDGE_UNLOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (error)
break;
if (bc->bc_flags & BC_F_COPYOUT)
error = copyout(&args, ifd->ifd_data, ifd->ifd_len);
break;
case SIOCSIFFLAGS:
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
if (!(ifp->if_flags & IFF_UP) &&
(ifp->if_drv_flags & IFF_DRV_RUNNING)) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* If interface is marked down and it is running,
* then stop and disable it.
*/
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
BRIDGE_LOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_stop(ifp, 1);
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
BRIDGE_UNLOCK(sc);
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
} else if ((ifp->if_flags & IFF_UP) &&
!(ifp->if_drv_flags & IFF_DRV_RUNNING)) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* If interface is marked up and it is stopped, then
* start it.
*/
Fix a panic when bringing up the bridge interface. We were casting a ifnet pointer to a softc which is no longer valid since the ifnet struct was split out from the softc. Approved by: mlaier (mentor) Approved by: re (blanket)
2005-06-27 21:58:12 +00:00
(*ifp->if_init)(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
break;
case SIOCSIFMTU:
- Check if_type of "addm <interface>" before setting the interface's MTU to the if_bridge(4) interface. This fixes a bug that MTU value of "addm <interface>" is used even when it is invalid for the if_bridge(4) member: # ifconfig bridge0 create # ifconfig bridge0 bridge0: flags=8802<BROADCAST,SIMPLEX,MULTICAST> metric 0 mtu 1500 ... # ifconfig bridge0 addm lo0 ifconfig: BRDGADD lo0: Invalid argument # ifconfig bridge0 bridge0: flags=8802<BROADCAST,SIMPLEX,MULTICAST> metric 0 mtu 16384 ... - Do not ignore MTU value of an interface even when if_type == IFT_GIF. This fixes MTU mismatch when an if_bridge(4) interface has a gif(4) interface and no other interface as the member, and it is directly used for L2 communication with EtherIP tunneling enabled. - Implement SIOCSIFMTU ioctl. Changing the MTU is allowed only when all members have the same MTU value.
2010-01-31 08:16:37 +00:00
if (ifr->ifr_mtu < 576) {
error = EINVAL;
break;
}
if (LIST_EMPTY(&sc->sc_iflist)) {
sc->sc_ifp->if_mtu = ifr->ifr_mtu;
break;
}
BRIDGE_LOCK(sc);
LIST_FOREACH(bif, &sc->sc_iflist, bif_next) {
if (bif->bif_ifp->if_mtu != ifr->ifr_mtu) {
log(LOG_NOTICE, "%s: invalid MTU: %lu(%s)"
" != %d\n", sc->sc_ifp->if_xname,
bif->bif_ifp->if_mtu,
bif->bif_ifp->if_xname, ifr->ifr_mtu);
error = EINVAL;
break;
}
}
if (!error)
sc->sc_ifp->if_mtu = ifr->ifr_mtu;
BRIDGE_UNLOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
break;
default:
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
/*
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
* drop the lock as ether_ioctl() will call bridge_start() and
* cause the lock to be recursed.
*/
error = ether_ioctl(ifp, cmd, data);
break;
}
return (error);
}
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
/*
* bridge_mutecaps:
*
* Clear or restore unwanted capabilities on the member interface
*/
static void
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
bridge_mutecaps(struct bridge_softc *sc)
{
struct bridge_iflist *bif;
int enabled, mask;
/* Initial bitmask of capabilities to test */
mask = BRIDGE_IFCAPS_MASK;
LIST_FOREACH(bif, &sc->sc_iflist, bif_next) {
/* Every member must support it or its disabled */
mask &= bif->bif_savedcaps;
}
LIST_FOREACH(bif, &sc->sc_iflist, bif_next) {
enabled = bif->bif_ifp->if_capenable;
When bridging LRO is causing a problem, the believe that it would work as long as all interfaces have TSO seems to be false, until the matter gets sorted out just disable LRO completely.
2009-08-24 21:04:51 +00:00
enabled &= ~BRIDGE_IFCAPS_STRIP;
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
/* strip off mask bits and enable them again if allowed */
enabled &= ~BRIDGE_IFCAPS_MASK;
enabled |= mask;
bridge_set_ifcap(sc, bif, enabled);
}
}
static void
bridge_set_ifcap(struct bridge_softc *sc, struct bridge_iflist *bif, int set)
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
{
struct ifnet *ifp = bif->bif_ifp;
struct ifreq ifr;
int error;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
bzero(&ifr, sizeof(ifr));
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
ifr.ifr_reqcap = set;
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
if (ifp->if_capenable != set) {
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
error = (*ifp->if_ioctl)(ifp, SIOCSIFCAP, (caddr_t)&ifr);
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
if (error)
if_printf(sc->sc_ifp,
"error setting interface capabilities on %s\n",
ifp->if_xname);
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
}
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* bridge_lookup_member:
*
* Lookup a bridge member interface.
*/
Make four more functions static that were missed in the last commit.
2005-10-14 20:57:02 +00:00
static struct bridge_iflist *
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_lookup_member(struct bridge_softc *sc, const char *name)
{
struct bridge_iflist *bif;
struct ifnet *ifp;
BRIDGE_LOCK_ASSERT(sc);
LIST_FOREACH(bif, &sc->sc_iflist, bif_next) {
ifp = bif->bif_ifp;
if (strcmp(ifp->if_xname, name) == 0)
return (bif);
}
return (NULL);
}
/*
* bridge_lookup_member_if:
*
* Lookup a bridge member interface by ifnet*.
*/
Make four more functions static that were missed in the last commit.
2005-10-14 20:57:02 +00:00
static struct bridge_iflist *
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_lookup_member_if(struct bridge_softc *sc, struct ifnet *member_ifp)
{
struct bridge_iflist *bif;
BRIDGE_LOCK_ASSERT(sc);
LIST_FOREACH(bif, &sc->sc_iflist, bif_next) {
if (bif->bif_ifp == member_ifp)
return (bif);
}
return (NULL);
}
/*
* bridge_delete_member:
*
* Delete the specified member interface.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
If we have been called from ether_ifdetach() then do not try and clear the promisc flag from the member interface, this is a no-op anyway since the interface is disappearing. The driver may have already released its resources such as miibus and this is likely to panic the kernel. Submitted and tested by: Wojciech A. Koszek MFC after: 2 weeks
2005-10-23 22:30:07 +00:00
bridge_delete_member(struct bridge_softc *sc, struct bridge_iflist *bif,
int gone)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
struct ifnet *ifs = bif->bif_ifp;
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
struct ifnet *fif = NULL;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_LOCK_ASSERT(sc);
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
if (bif->bif_flags & IFBIF_STP)
use two stage creation of stp ports, this means that the stp variables can be set before the port is marked STP and they will no longer be overwrittten
2006-11-26 18:43:48 +00:00
bstp_disable(&bif->bif_stp);
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
ifs->if_bridge = NULL;
BRIDGE_XLOCK(sc);
LIST_REMOVE(bif, bif_next);
BRIDGE_XDROP(sc);
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
/*
* If removing the interface that gave the bridge its mac address, set
* the mac address of the bridge to the address of the next member, or
* to its default address if no members are left.
*/
bridge_delete_member is called via the event handler from if_detach after the LLADDR is reclaimed which causes a null pointer deref with inherit_mac enabled. Record the ifnet pointer of the interface and then compare that to find when to re-assign the bridge address. Submitted by: sam
2009-02-13 19:20:25 +00:00
if (bridge_inherit_mac && sc->sc_ifaddr == ifs) {
if (LIST_EMPTY(&sc->sc_iflist)) {
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
bcopy(sc->sc_defaddr,
IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN);
bridge_delete_member is called via the event handler from if_detach after the LLADDR is reclaimed which causes a null pointer deref with inherit_mac enabled. Record the ifnet pointer of the interface and then compare that to find when to re-assign the bridge address. Submitted by: sam
2009-02-13 19:20:25 +00:00
sc->sc_ifaddr = NULL;
} else {
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
fif = LIST_FIRST(&sc->sc_iflist)->bif_ifp;
bcopy(IF_LLADDR(fif),
IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN);
bridge_delete_member is called via the event handler from if_detach after the LLADDR is reclaimed which causes a null pointer deref with inherit_mac enabled. Record the ifnet pointer of the interface and then compare that to find when to re-assign the bridge address. Submitted by: sam
2009-02-13 19:20:25 +00:00
sc->sc_ifaddr = fif;
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
}
Declare a new EVENTHANDLER called iflladdr_event which signals that the L2 address on an interface has changed. This lets stacked interfaces such as vlan(4) detect that their lower interface has changed and adjust things in order to keep working. Previously this situation broke at least vlan(4) and lagg(4) configurations. The EVENTHANDLER_INVOKE call was not placed within if_setlladdr() due to the risk of a loop. PR: kern/142927 Submitted by: Nikolay Denev
2010-01-18 20:34:00 +00:00
EVENTHANDLER_INVOKE(iflladdr_event, sc->sc_ifp);
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
}
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
bridge_mutecaps(sc); /* recalcuate now this interface is removed */
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtdelete(sc, ifs, IFBF_FLUSHALL);
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
KASSERT(bif->bif_addrcnt == 0,
("%s: %d bridge routes referenced", __func__, bif->bif_addrcnt));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Tell bridgestp that we are about to free the memory so it can cleanup.
2006-08-02 02:59:24 +00:00
BRIDGE_UNLOCK(sc);
Reorder the bridge add and delete routines to avoid calling ifpromisc() with the bridge lock held.
2009-05-01 19:46:42 +00:00
if (!gone) {
switch (ifs->if_type) {
case IFT_ETHER:
case IFT_L2VLAN:
/*
* Take the interface out of promiscuous mode.
*/
(void) ifpromisc(ifs, 0);
break;
case IFT_GIF:
break;
default:
#ifdef DIAGNOSTIC
panic("bridge_delete_member: impossible");
#endif
break;
}
/* reneable any interface capabilities */
bridge_set_ifcap(sc, bif, bif->bif_savedcaps);
}
use two stage creation of stp ports, this means that the stp variables can be set before the port is marked STP and they will no longer be overwrittten
2006-11-26 18:43:48 +00:00
bstp_destroy(&bif->bif_stp); /* prepare to free */
Tell bridgestp that we are about to free the memory so it can cleanup.
2006-08-02 02:59:24 +00:00
BRIDGE_LOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
free(bif, M_DEVBUF);
}
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
/*
* bridge_delete_span:
*
* Delete the specified span interface.
*/
static void
bridge_delete_span(struct bridge_softc *sc, struct bridge_iflist *bif)
{
BRIDGE_LOCK_ASSERT(sc);
KASSERT(bif->bif_ifp->if_bridge == NULL,
("%s: not a span interface", __func__));
LIST_REMOVE(bif, bif_next);
free(bif, M_DEVBUF);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_add(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
struct bridge_iflist *bif = NULL;
struct ifnet *ifs;
int error = 0;
ifs = ifunit(req->ifbr_ifsname);
if (ifs == NULL)
return (ENOENT);
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
if (ifs->if_ioctl == NULL) /* must be supported */
return (EINVAL);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
/* If it's in the span list, it can't be a member. */
LIST_FOREACH(bif, &sc->sc_spanlist, bif_next)
if (ifs == bif->bif_ifp)
return (EBUSY);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (ifs->if_bridge == sc)
return (EEXIST);
if (ifs->if_bridge != NULL)
return (EBUSY);
Use M_ZERO for the bridge_iflist to ensure there are no unexpected suprises.
2005-12-17 10:12:20 +00:00
bif = malloc(sizeof(*bif), M_DEVBUF, M_NOWAIT|M_ZERO);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (bif == NULL)
return (ENOMEM);
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
bif->bif_ifp = ifs;
bif->bif_flags = IFBIF_LEARNING | IFBIF_DISCOVER;
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
bif->bif_savedcaps = ifs->if_capenable;
Add code that clears certain capabilities from the member interface, these are restored when its removed from the bridge. At the moment we only clear IFCAP_TXCSUM. Since a locally generated packet on the bridge may be sent out any one or more interfaces it cant be assumed that every card does hardware csums. Most bridges don't generate a lot of traffic themselves so turning off offloading won't hurt, bridged packets are unaffected. Tested by: Bruce Walker (bmw borderware.com) MFC after: 5 days
2006-01-14 03:51:31 +00:00
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
switch (ifs->if_type) {
case IFT_ETHER:
case IFT_L2VLAN:
Add RFC 3378 EtherIP support. This change makes it possible to add gif interfaces to bridges, which will then send and receive IP protocol 97 packets. Packets are Ethernet frames with an EtherIP header prepended. Obtained from: NetBSD MFC after: 2 weeks
2005-12-21 21:29:45 +00:00
case IFT_GIF:
Reorder the bridge add and delete routines to avoid calling ifpromisc() with the bridge lock held.
2009-05-01 19:46:42 +00:00
/* permitted interface types */
Add RFC 3378 EtherIP support. This change makes it possible to add gif interfaces to bridges, which will then send and receive IP protocol 97 packets. Packets are Ethernet frames with an EtherIP header prepended. Obtained from: NetBSD MFC after: 2 weeks
2005-12-21 21:29:45 +00:00
break;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
default:
error = EINVAL;
goto out;
}
- Check if_type of "addm <interface>" before setting the interface's MTU to the if_bridge(4) interface. This fixes a bug that MTU value of "addm <interface>" is used even when it is invalid for the if_bridge(4) member: # ifconfig bridge0 create # ifconfig bridge0 bridge0: flags=8802<BROADCAST,SIMPLEX,MULTICAST> metric 0 mtu 1500 ... # ifconfig bridge0 addm lo0 ifconfig: BRDGADD lo0: Invalid argument # ifconfig bridge0 bridge0: flags=8802<BROADCAST,SIMPLEX,MULTICAST> metric 0 mtu 16384 ... - Do not ignore MTU value of an interface even when if_type == IFT_GIF. This fixes MTU mismatch when an if_bridge(4) interface has a gif(4) interface and no other interface as the member, and it is directly used for L2 communication with EtherIP tunneling enabled. - Implement SIOCSIFMTU ioctl. Changing the MTU is allowed only when all members have the same MTU value.
2010-01-31 08:16:37 +00:00
/* Allow the first Ethernet member to define the MTU */
if (LIST_EMPTY(&sc->sc_iflist))
sc->sc_ifp->if_mtu = ifs->if_mtu;
else if (sc->sc_ifp->if_mtu != ifs->if_mtu) {
if_printf(sc->sc_ifp, "invalid MTU: %lu(%s) != %lu\n",
ifs->if_mtu, ifs->if_xname, sc->sc_ifp->if_mtu);
error = EINVAL;
goto out;
}
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
/*
* Assign the interface's MAC address to the bridge if it's the first
* member and the MAC address of the bridge has not been changed from
* the default randomly generated one.
*/
Put the bridge mac inheritance behind a sysctl with the default off as this still needs all the edge cases fixed. Submitted by: Eygene Ryabinkin
2008-09-08 03:28:26 +00:00
if (bridge_inherit_mac && LIST_EMPTY(&sc->sc_iflist) &&
bridge_delete_member is called via the event handler from if_detach after the LLADDR is reclaimed which causes a null pointer deref with inherit_mac enabled. Record the ifnet pointer of the interface and then compare that to find when to re-assign the bridge address. Submitted by: sam
2009-02-13 19:20:25 +00:00
!memcmp(IF_LLADDR(sc->sc_ifp), sc->sc_defaddr, ETHER_ADDR_LEN)) {
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
bcopy(IF_LLADDR(ifs), IF_LLADDR(sc->sc_ifp), ETHER_ADDR_LEN);
bridge_delete_member is called via the event handler from if_detach after the LLADDR is reclaimed which causes a null pointer deref with inherit_mac enabled. Record the ifnet pointer of the interface and then compare that to find when to re-assign the bridge address. Submitted by: sam
2009-02-13 19:20:25 +00:00
sc->sc_ifaddr = ifs;
Declare a new EVENTHANDLER called iflladdr_event which signals that the L2 address on an interface has changed. This lets stacked interfaces such as vlan(4) detect that their lower interface has changed and adjust things in order to keep working. Previously this situation broke at least vlan(4) and lagg(4) configurations. The EVENTHANDLER_INVOKE call was not placed within if_setlladdr() due to the risk of a loop. PR: kern/142927 Submitted by: Nikolay Denev
2010-01-18 20:34:00 +00:00
EVENTHANDLER_INVOKE(iflladdr_event, sc->sc_ifp);
bridge_delete_member is called via the event handler from if_detach after the LLADDR is reclaimed which causes a null pointer deref with inherit_mac enabled. Record the ifnet pointer of the interface and then compare that to find when to re-assign the bridge address. Submitted by: sam
2009-02-13 19:20:25 +00:00
}
Set bridge MAC addresses to the MAC address of their first interface unless locally configured. This is more in line with the behaviour of other popular bridging implementations and makes bridges more predictable after reboots for example. Reviewed by: thompsa MFC after: 1 week
2008-07-01 08:14:58 +00:00
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
ifs->if_bridge = sc;
use two stage creation of stp ports, this means that the stp variables can be set before the port is marked STP and they will no longer be overwrittten
2006-11-26 18:43:48 +00:00
bstp_create(&sc->sc_stp, &bif->bif_stp, bif->bif_ifp);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* XXX: XLOCK HERE!?!
*
* NOTE: insert_***HEAD*** should be safe for the traversals.
*/
LIST_INSERT_HEAD(&sc->sc_iflist, bif, bif_next);
Be smarter about disabling interface capabilities. TOE/TSO/TXCSUM will only be disabled if one (or more) of the member interfaces does not support it. Always turn off LRO since we can not bridge a combined frame. Tested by: Stefan Lambrev
2008-07-03 15:58:30 +00:00
/* Set interface capabilities to the intersection set of all members */
bridge_mutecaps(sc);
Reorder the bridge add and delete routines to avoid calling ifpromisc() with the bridge lock held.
2009-05-01 19:46:42 +00:00
switch (ifs->if_type) {
case IFT_ETHER:
case IFT_L2VLAN:
/*
* Place the interface into promiscuous mode.
*/
BRIDGE_UNLOCK(sc);
error = ifpromisc(ifs, 1);
BRIDGE_LOCK(sc);
break;
}
if (error)
bridge_delete_member(sc, bif, 0);
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
out:
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (error) {
if (bif != NULL)
free(bif, M_DEVBUF);
}
return (error);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_del(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
struct bridge_iflist *bif;
bif = bridge_lookup_member(sc, req->ifbr_ifsname);
if (bif == NULL)
return (ENOENT);
If we have been called from ether_ifdetach() then do not try and clear the promisc flag from the member interface, this is a no-op anyway since the interface is disappearing. The driver may have already released its resources such as miibus and this is likely to panic the kernel. Submitted and tested by: Wojciech A. Koszek MFC after: 2 weeks
2005-10-23 22:30:07 +00:00
bridge_delete_member(sc, bif, 0);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_gifflags(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
struct bridge_iflist *bif;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
struct bstp_port *bp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bif = bridge_lookup_member(sc, req->ifbr_ifsname);
if (bif == NULL)
return (ENOENT);
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
bp = &bif->bif_stp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
req->ifbr_ifsflags = bif->bif_flags;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
req->ifbr_state = bp->bp_state;
req->ifbr_priority = bp->bp_priority;
req->ifbr_path_cost = bp->bp_path_cost;
req->ifbr_portno = bif->bif_ifp->if_index & 0xfff;
req->ifbr_proto = bp->bp_protover;
req->ifbr_role = bp->bp_role;
req->ifbr_stpflags = bp->bp_flags;
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
req->ifbr_addrcnt = bif->bif_addrcnt;
req->ifbr_addrmax = bif->bif_addrmax;
req->ifbr_addrexceeded = bif->bif_addrexceeded;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Sync with the OpenBSD port of RSTP - use flags rather than sperate ioctls for edge, p2p - implement p2p and autop2p flags - define large pathcost constant as ULL - show bridgeid and rootid in ifconfig Obtained from: Reyk Floeter <reyk@openbsd.org>
2006-11-27 00:35:40 +00:00
/* Copy STP state options as flags */
if (bp->bp_operedge)
req->ifbr_ifsflags |= IFBIF_BSTP_EDGE;
if (bp->bp_flags & BSTP_PORT_AUTOEDGE)
req->ifbr_ifsflags |= IFBIF_BSTP_AUTOEDGE;
These days P2P means peer-2-peer (also well known from serveral filesharing protocols) while PointToPoint has been PtP links. Change the variables accordingly while the code is still fresh and undocumented. Requested by: bz
2006-12-11 23:46:40 +00:00
if (bp->bp_ptp_link)
req->ifbr_ifsflags |= IFBIF_BSTP_PTP;
if (bp->bp_flags & BSTP_PORT_AUTOPTP)
req->ifbr_ifsflags |= IFBIF_BSTP_AUTOPTP;
Add two new flags to if_bridge(4) indicating whether the edge flag of the bridge port and path cost have been administratively set or calculated automatically by RSTP. Make sure to transition from non-edge to edge when the port goes down and the edge flag was manually set before. This is needed to comply with the condition ((!portEnabled && AdminEdge) || ....) in the Bridge Detection State Machine (IEE802.1D-2004, p. 171). Reviewed by: thompsa Approved by: bz (mentor)
2006-12-04 14:45:02 +00:00
if (bp->bp_flags & BSTP_PORT_ADMEDGE)
req->ifbr_ifsflags |= IFBIF_BSTP_ADMEDGE;
if (bp->bp_flags & BSTP_PORT_ADMCOST)
req->ifbr_ifsflags |= IFBIF_BSTP_ADMCOST;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_sifflags(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
struct bridge_iflist *bif;
Sync with the OpenBSD port of RSTP - use flags rather than sperate ioctls for edge, p2p - implement p2p and autop2p flags - define large pathcost constant as ULL - show bridgeid and rootid in ifconfig Obtained from: Reyk Floeter <reyk@openbsd.org>
2006-11-27 00:35:40 +00:00
struct bstp_port *bp;
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
int error;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bif = bridge_lookup_member(sc, req->ifbr_ifsname);
if (bif == NULL)
return (ENOENT);
Sync with the OpenBSD port of RSTP - use flags rather than sperate ioctls for edge, p2p - implement p2p and autop2p flags - define large pathcost constant as ULL - show bridgeid and rootid in ifconfig Obtained from: Reyk Floeter <reyk@openbsd.org>
2006-11-27 00:35:40 +00:00
bp = &bif->bif_stp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
if (req->ifbr_ifsflags & IFBIF_SPAN)
/* SPAN is readonly */
return (EINVAL);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (req->ifbr_ifsflags & IFBIF_STP) {
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
if ((bif->bif_flags & IFBIF_STP) == 0) {
use two stage creation of stp ports, this means that the stp variables can be set before the port is marked STP and they will no longer be overwrittten
2006-11-26 18:43:48 +00:00
error = bstp_enable(&bif->bif_stp);
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
if (error)
return (error);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
} else {
if ((bif->bif_flags & IFBIF_STP) != 0)
use two stage creation of stp ports, this means that the stp variables can be set before the port is marked STP and they will no longer be overwrittten
2006-11-26 18:43:48 +00:00
bstp_disable(&bif->bif_stp);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Sync with the OpenBSD port of RSTP - use flags rather than sperate ioctls for edge, p2p - implement p2p and autop2p flags - define large pathcost constant as ULL - show bridgeid and rootid in ifconfig Obtained from: Reyk Floeter <reyk@openbsd.org>
2006-11-27 00:35:40 +00:00
/* Pass on STP flags */
bstp_set_edge(bp, req->ifbr_ifsflags & IFBIF_BSTP_EDGE ? 1 : 0);
bstp_set_autoedge(bp, req->ifbr_ifsflags & IFBIF_BSTP_AUTOEDGE ? 1 : 0);
These days P2P means peer-2-peer (also well known from serveral filesharing protocols) while PointToPoint has been PtP links. Change the variables accordingly while the code is still fresh and undocumented. Requested by: bz
2006-12-11 23:46:40 +00:00
bstp_set_ptp(bp, req->ifbr_ifsflags & IFBIF_BSTP_PTP ? 1 : 0);
bstp_set_autoptp(bp, req->ifbr_ifsflags & IFBIF_BSTP_AUTOPTP ? 1 : 0);
Sync with the OpenBSD port of RSTP - use flags rather than sperate ioctls for edge, p2p - implement p2p and autop2p flags - define large pathcost constant as ULL - show bridgeid and rootid in ifconfig Obtained from: Reyk Floeter <reyk@openbsd.org>
2006-11-27 00:35:40 +00:00
/* Save the bits relating to the bridge */
bif->bif_flags = req->ifbr_ifsflags & IFBIFMASK;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_scache(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
sc->sc_brtmax = param->ifbrp_csize;
bridge_rttrim(sc);
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_gcache(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
param->ifbrp_csize = sc->sc_brtmax;
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_gifs(struct bridge_softc *sc, void *arg)
{
struct ifbifconf *bifc = arg;
struct bridge_iflist *bif;
struct ifbreq breq;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
char *buf, *outbuf;
int count, buflen, len, error = 0;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
count = 0;
LIST_FOREACH(bif, &sc->sc_iflist, bif_next)
count++;
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
LIST_FOREACH(bif, &sc->sc_spanlist, bif_next)
count++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
buflen = sizeof(breq) * count;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (bifc->ifbic_len == 0) {
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
bifc->ifbic_len = buflen;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
BRIDGE_UNLOCK(sc);
outbuf = malloc(buflen, M_TEMP, M_WAITOK | M_ZERO);
BRIDGE_LOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
count = 0;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
buf = outbuf;
len = min(bifc->ifbic_len, buflen);
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
bzero(&breq, sizeof(breq));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
LIST_FOREACH(bif, &sc->sc_iflist, bif_next) {
if (len < sizeof(breq))
break;
strlcpy(breq.ifbr_ifsname, bif->bif_ifp->if_xname,
sizeof(breq.ifbr_ifsname));
Sync with the OpenBSD port of RSTP - use flags rather than sperate ioctls for edge, p2p - implement p2p and autop2p flags - define large pathcost constant as ULL - show bridgeid and rootid in ifconfig Obtained from: Reyk Floeter <reyk@openbsd.org>
2006-11-27 00:35:40 +00:00
/* Fill in the ifbreq structure */
error = bridge_ioctl_gifflags(sc, &breq);
if (error)
break;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
memcpy(buf, &breq, sizeof(breq));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
count++;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
buf += sizeof(breq);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
len -= sizeof(breq);
}
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) {
if (len < sizeof(breq))
break;
strlcpy(breq.ifbr_ifsname, bif->bif_ifp->if_xname,
sizeof(breq.ifbr_ifsname));
breq.ifbr_ifsflags = bif->bif_flags;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
breq.ifbr_portno = bif->bif_ifp->if_index & 0xfff;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
memcpy(buf, &breq, sizeof(breq));
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
count++;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
buf += sizeof(breq);
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
len -= sizeof(breq);
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
BRIDGE_UNLOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bifc->ifbic_len = sizeof(breq) * count;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
error = copyout(outbuf, bifc->ifbic_req, bifc->ifbic_len);
BRIDGE_LOCK(sc);
free(outbuf, M_TEMP);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (error);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_rts(struct bridge_softc *sc, void *arg)
{
struct ifbaconf *bac = arg;
struct bridge_rtnode *brt;
struct ifbareq bareq;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
char *buf, *outbuf;
int count, buflen, len, error = 0;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (bac->ifbac_len == 0)
return (0);
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
count = 0;
LIST_FOREACH(brt, &sc->sc_rtlist, brt_list)
count++;
buflen = sizeof(bareq) * count;
BRIDGE_UNLOCK(sc);
outbuf = malloc(buflen, M_TEMP, M_WAITOK | M_ZERO);
BRIDGE_LOCK(sc);
count = 0;
buf = outbuf;
len = min(bac->ifbac_len, buflen);
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
bzero(&bareq, sizeof(bareq));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
LIST_FOREACH(brt, &sc->sc_rtlist, brt_list) {
if (len < sizeof(bareq))
goto out;
strlcpy(bareq.ifba_ifsname, brt->brt_ifp->if_xname,
sizeof(bareq.ifba_ifsname));
memcpy(bareq.ifba_dst, brt->brt_addr, sizeof(brt->brt_addr));
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
bareq.ifba_vlan = brt->brt_vlan;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC &&
It is better to use time_uptime here since it is monotonic. Pointed out by: glebius
2006-01-02 22:23:03 +00:00
time_uptime < brt->brt_expire)
bareq.ifba_expire = brt->brt_expire - time_uptime;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
else
bareq.ifba_expire = 0;
bareq.ifba_flags = brt->brt_flags;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
memcpy(buf, &bareq, sizeof(bareq));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
count++;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
buf += sizeof(bareq);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
len -= sizeof(bareq);
}
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
out:
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
BRIDGE_UNLOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bac->ifbac_len = sizeof(bareq) * count;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
error = copyout(outbuf, bac->ifbac_req, bac->ifbac_len);
BRIDGE_LOCK(sc);
free(outbuf, M_TEMP);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (error);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_saddr(struct bridge_softc *sc, void *arg)
{
struct ifbareq *req = arg;
struct bridge_iflist *bif;
int error;
bif = bridge_lookup_member(sc, req->ifba_ifsname);
if (bif == NULL)
return (ENOENT);
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
error = bridge_rtupdate(sc, req->ifba_dst, req->ifba_vlan, bif, 1,
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
req->ifba_flags);
return (error);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_sto(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
sc->sc_brttimeout = param->ifbrp_ctime;
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_gto(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
param->ifbrp_ctime = sc->sc_brttimeout;
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_daddr(struct bridge_softc *sc, void *arg)
{
struct ifbareq *req = arg;
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
return (bridge_rtdaddr(sc, req->ifba_dst, req->ifba_vlan));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_flush(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
bridge_rtflush(sc, req->ifbr_ifsflags);
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_gpri(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
struct bstp_state *bs = &sc->sc_stp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
param->ifbrp_prio = bs->bs_bridge_priority;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_spri(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
return (bstp_set_priority(&sc->sc_stp, param->ifbrp_prio));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_ght(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
struct bstp_state *bs = &sc->sc_stp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
param->ifbrp_hellotime = bs->bs_bridge_htime >> 8;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_sht(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
return (bstp_set_htime(&sc->sc_stp, param->ifbrp_hellotime));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_gfd(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
struct bstp_state *bs = &sc->sc_stp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
param->ifbrp_fwddelay = bs->bs_bridge_fdelay >> 8;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_sfd(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
return (bstp_set_fdelay(&sc->sc_stp, param->ifbrp_fwddelay));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_gma(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
struct bstp_state *bs = &sc->sc_stp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
param->ifbrp_maxage = bs->bs_bridge_max_age >> 8;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_sma(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
return (bstp_set_maxage(&sc->sc_stp, param->ifbrp_maxage));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_sifprio(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
struct bridge_iflist *bif;
bif = bridge_lookup_member(sc, req->ifbr_ifsname);
if (bif == NULL)
return (ENOENT);
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
return (bstp_set_port_priority(&bif->bif_stp, req->ifbr_priority));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_ioctl_sifcost(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
struct bridge_iflist *bif;
bif = bridge_lookup_member(sc, req->ifbr_ifsname);
if (bif == NULL)
return (ENOENT);
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
return (bstp_set_path_cost(&bif->bif_stp, req->ifbr_path_cost));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
static int
bridge_ioctl_sifmaxaddr(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
struct bridge_iflist *bif;
bif = bridge_lookup_member(sc, req->ifbr_ifsname);
if (bif == NULL)
return (ENOENT);
bif->bif_addrmax = req->ifbr_addrmax;
return (0);
}
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
static int
bridge_ioctl_addspan(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
struct bridge_iflist *bif = NULL;
struct ifnet *ifs;
ifs = ifunit(req->ifbr_ifsname);
if (ifs == NULL)
return (ENOENT);
LIST_FOREACH(bif, &sc->sc_spanlist, bif_next)
if (ifs == bif->bif_ifp)
return (EBUSY);
if (ifs->if_bridge != NULL)
return (EBUSY);
switch (ifs->if_type) {
case IFT_ETHER:
Allow gif interfaces to be added as span ports, the user may want to send a copy of all packets to the other side of the world.
2006-06-20 21:28:18 +00:00
case IFT_GIF:
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
case IFT_L2VLAN:
break;
default:
return (EINVAL);
}
bif = malloc(sizeof(*bif), M_DEVBUF, M_NOWAIT|M_ZERO);
if (bif == NULL)
return (ENOMEM);
bif->bif_ifp = ifs;
bif->bif_flags = IFBIF_SPAN;
LIST_INSERT_HEAD(&sc->sc_spanlist, bif, bif_next);
return (0);
}
static int
bridge_ioctl_delspan(struct bridge_softc *sc, void *arg)
{
struct ifbreq *req = arg;
struct bridge_iflist *bif;
struct ifnet *ifs;
ifs = ifunit(req->ifbr_ifsname);
if (ifs == NULL)
return (ENOENT);
LIST_FOREACH(bif, &sc->sc_spanlist, bif_next)
if (ifs == bif->bif_ifp)
break;
if (bif == NULL)
return (ENOENT);
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
bridge_delete_span(sc, bif);
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
return (0);
}
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
static int
bridge_ioctl_gbparam(struct bridge_softc *sc, void *arg)
{
struct ifbropreq *req = arg;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
struct bstp_state *bs = &sc->sc_stp;
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
struct bstp_port *root_port;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
req->ifbop_maxage = bs->bs_bridge_max_age >> 8;
req->ifbop_hellotime = bs->bs_bridge_htime >> 8;
req->ifbop_fwddelay = bs->bs_bridge_fdelay >> 8;
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
root_port = bs->bs_root_port;
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
if (root_port == NULL)
req->ifbop_root_port = 0;
else
req->ifbop_root_port = root_port->bp_ifp->if_index;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
req->ifbop_holdcount = bs->bs_txholdcount;
req->ifbop_priority = bs->bs_bridge_priority;
req->ifbop_protocol = bs->bs_protover;
req->ifbop_root_path_cost = bs->bs_root_pv.pv_cost;
Sync with the OpenBSD port of RSTP - use flags rather than sperate ioctls for edge, p2p - implement p2p and autop2p flags - define large pathcost constant as ULL - show bridgeid and rootid in ifconfig Obtained from: Reyk Floeter <reyk@openbsd.org>
2006-11-27 00:35:40 +00:00
req->ifbop_bridgeid = bs->bs_bridge_pv.pv_dbridge_id;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
req->ifbop_designated_root = bs->bs_root_pv.pv_root_id;
Sync with the OpenBSD port of RSTP - use flags rather than sperate ioctls for edge, p2p - implement p2p and autop2p flags - define large pathcost constant as ULL - show bridgeid and rootid in ifconfig Obtained from: Reyk Floeter <reyk@openbsd.org>
2006-11-27 00:35:40 +00:00
req->ifbop_designated_bridge = bs->bs_root_pv.pv_dbridge_id;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
req->ifbop_last_tc_time.tv_sec = bs->bs_last_tc_time.tv_sec;
req->ifbop_last_tc_time.tv_usec = bs->bs_last_tc_time.tv_usec;
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
return (0);
}
static int
bridge_ioctl_grte(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
param->ifbrp_cexceeded = sc->sc_brtexceeded;
return (0);
}
static int
bridge_ioctl_gifsstp(struct bridge_softc *sc, void *arg)
{
struct ifbpstpconf *bifstp = arg;
struct bridge_iflist *bif;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
struct bstp_port *bp;
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
struct ifbpstpreq bpreq;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
char *buf, *outbuf;
int count, buflen, len, error = 0;
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
count = 0;
LIST_FOREACH(bif, &sc->sc_iflist, bif_next) {
if ((bif->bif_flags & IFBIF_STP) != 0)
count++;
}
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
buflen = sizeof(bpreq) * count;
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
if (bifstp->ifbpstp_len == 0) {
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
bifstp->ifbpstp_len = buflen;
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
return (0);
}
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
BRIDGE_UNLOCK(sc);
outbuf = malloc(buflen, M_TEMP, M_WAITOK | M_ZERO);
BRIDGE_LOCK(sc);
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
count = 0;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
buf = outbuf;
len = min(bifstp->ifbpstp_len, buflen);
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
bzero(&bpreq, sizeof(bpreq));
LIST_FOREACH(bif, &sc->sc_iflist, bif_next) {
if (len < sizeof(bpreq))
break;
if ((bif->bif_flags & IFBIF_STP) == 0)
continue;
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
bp = &bif->bif_stp;
bpreq.ifbp_portno = bif->bif_ifp->if_index & 0xfff;
bpreq.ifbp_fwd_trans = bp->bp_forward_transitions;
bpreq.ifbp_design_cost = bp->bp_desg_pv.pv_cost;
bpreq.ifbp_design_port = bp->bp_desg_pv.pv_port_id;
bpreq.ifbp_design_bridge = bp->bp_desg_pv.pv_dbridge_id;
bpreq.ifbp_design_root = bp->bp_desg_pv.pv_root_id;
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
memcpy(buf, &bpreq, sizeof(bpreq));
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
count++;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
buf += sizeof(bpreq);
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
len -= sizeof(bpreq);
}
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
BRIDGE_UNLOCK(sc);
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
bifstp->ifbpstp_len = sizeof(bpreq) * count;
Avoid holding the softc lock when using copyout(). Reported by: dfr Approved by: re (rwatson)
2007-07-26 20:30:18 +00:00
error = copyout(outbuf, bifstp->ifbpstp_req, bifstp->ifbpstp_len);
BRIDGE_LOCK(sc);
free(outbuf, M_TEMP);
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
return (error);
}
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
static int
bridge_ioctl_sproto(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
return (bstp_set_protocol(&sc->sc_stp, param->ifbrp_proto));
}
static int
bridge_ioctl_stxhc(struct bridge_softc *sc, void *arg)
{
struct ifbrparam *param = arg;
return (bstp_set_holdcount(&sc->sc_stp, param->ifbrp_txhc));
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* bridge_ifdetach:
*
* Detach an interface from a bridge. Called when a member
* interface is detaching.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
bridge_ifdetach(void *arg __unused, struct ifnet *ifp)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
struct bridge_softc *sc = ifp->if_bridge;
If we have been called from ether_ifdetach() then do not try and clear the promisc flag from the member interface, this is a no-op anyway since the interface is disappearing. The driver may have already released its resources such as miibus and this is likely to panic the kernel. Submitted and tested by: Wojciech A. Koszek MFC after: 2 weeks
2005-10-23 22:30:07 +00:00
struct bridge_iflist *bif;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
/* Check if the interface is a bridge member */
if (sc != NULL) {
BRIDGE_LOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
bif = bridge_lookup_member_if(sc, ifp);
if (bif != NULL)
bridge_delete_member(sc, bif, 1);
BRIDGE_UNLOCK(sc);
If we have been called from ether_ifdetach() then do not try and clear the promisc flag from the member interface, this is a no-op anyway since the interface is disappearing. The driver may have already released its resources such as miibus and this is likely to panic the kernel. Submitted and tested by: Wojciech A. Koszek MFC after: 2 weeks
2005-10-23 22:30:07 +00:00
return;
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
}
If we have been called from ether_ifdetach() then do not try and clear the promisc flag from the member interface, this is a no-op anyway since the interface is disappearing. The driver may have already released its resources such as miibus and this is likely to panic the kernel. Submitted and tested by: Wojciech A. Koszek MFC after: 2 weeks
2005-10-23 22:30:07 +00:00
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
/* Check if the interface is a span port */
mtx_lock(&bridge_list_mtx);
LIST_FOREACH(sc, &bridge_list, sc_list) {
BRIDGE_LOCK(sc);
LIST_FOREACH(bif, &sc->sc_spanlist, bif_next)
if (ifp == bif->bif_ifp) {
bridge_delete_span(sc, bif);
break;
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Change from a callback in if_ethersubr to using EVENTHANDLER in order to detach span ports when they disappear. The span port does not have a pointer to the softc so revert r1.31 and bring back the softc linked-list. MFC after: 2 weeks
2005-12-17 06:33:51 +00:00
BRIDGE_UNLOCK(sc);
}
mtx_unlock(&bridge_list_mtx);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
/*
* bridge_init:
*
* Initialize a bridge interface.
*/
static void
bridge_init(void *xsc)
{
struct bridge_softc *sc = (struct bridge_softc *)xsc;
Stop embedding struct ifnet at the top of driver softcs. Instead the struct ifnet or the layer 2 common structure it was embedded in have been replaced with a struct ifnet pointer to be filled by a call to the new function, if_alloc(). The layer 2 common structure is also allocated via if_alloc() based on the interface type. It is hung off the new struct ifnet member, if_l2com. This change removes the size of these structures from the kernel ABI and will allow us to better manage them as interfaces come and go. Other changes of note: - Struct arpcom is no longer referenced in normal interface code. Instead the Ethernet address is accessed via the IFP2ENADDR() macro. To enforce this ac_enaddr has been renamed to _ac_enaddr. - The second argument to ether_ifattach is now always the mac address from driver private storage rather than sometimes being ac_enaddr. Reviewed by: sobomax, sam
2005-06-10 16:49:24 +00:00
struct ifnet *ifp = sc->sc_ifp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
if (ifp->if_drv_flags & IFF_DRV_RUNNING)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return;
Mark the callouts as MPSAFE as if_bridge has been giant-free since day 1. Use the SMP friendly callout_init_mtx() while we are here. Approved by: mlaier (mentor) MFC after: 3 days
2005-08-18 20:17:00 +00:00
BRIDGE_LOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
callout_reset(&sc->sc_brcallout, bridge_rtable_prune_period * hz,
bridge_timer, sc);
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
ifp->if_drv_flags |= IFF_DRV_RUNNING;
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
bstp_init(&sc->sc_stp); /* Initialize Spanning Tree */
Ensure that we are holding the lock when initialising the bridge interface. We could initialise while unlocked if the bridge is not up when setting the inet address, ether_ioctl() would call bridge_init. Change it so bridge_init is always called unlocked and then locks before calling bstp_initialization(). Reported by: Michal Mertl Approved by: mlaier (mentor) MFC after: 3 days
2005-08-15 02:50:13 +00:00
BRIDGE_UNLOCK(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
/*
* bridge_stop:
*
* Stop the bridge interface.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_stop(struct ifnet *ifp, int disable)
{
struct bridge_softc *sc = ifp->if_softc;
Ensure that we are holding the lock when initialising the bridge interface. We could initialise while unlocked if the bridge is not up when setting the inet address, ether_ioctl() would call bridge_init. Change it so bridge_init is always called unlocked and then locks before calling bstp_initialization(). Reported by: Michal Mertl Approved by: mlaier (mentor) MFC after: 3 days
2005-08-15 02:50:13 +00:00
BRIDGE_LOCK_ASSERT(sc);
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
if ((ifp->if_drv_flags & IFF_DRV_RUNNING) == 0)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return;
callout_stop(&sc->sc_brcallout);
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
bstp_stop(&sc->sc_stp);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtflush(sc, IFBF_FLUSHDYN);
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
ifp->if_drv_flags &= ~IFF_DRV_RUNNING;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
/*
* bridge_enqueue:
*
* Enqueue a packet on a bridge member interface.
*
*/
Remove the dependency of bridgestp.h on if_bridgevar.h by moving a couple of private structures to if_bridge.c.
2006-07-27 21:01:48 +00:00
static void
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
bridge_enqueue(struct bridge_softc *sc, struct ifnet *dst_ifp, struct mbuf *m)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
int len, err = 0;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
short mflags;
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
struct mbuf *m0;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
len = m->m_pkthdr.len;
mflags = m->m_flags;
We check that all the member interfaces have the same MTU on attach to the bridge but the interface can still be changed afterwards. This falls under the 'dont do that' category but log an warning when INVARIANTS is defined. Approved by: mlaier (mentor) MFC after: 3 days
2005-07-25 02:22:37 +00:00
Fix spelling mistake in comment.
2006-06-19 02:25:11 +00:00
/* We may be sending a fragment so traverse the mbuf */
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
for (; m; m = m0) {
m0 = m->m_nextpkt;
m->m_nextpkt = NULL;
The bridging output function puts the mbuf directly on the interfaces send queue so the output network card must support the same tagging mechanism as how the frame was input (prepended Ethernet header tag or stripped HW mflag). Now the vlan Ethernet header is _always_ stripped in ether_input and the mbuf flagged, only only network cards with VLAN_HWTAGGING enabled would properly re-tag any outgoing vlan frames. If the outgoing interface does not support hardware tagging then readd the vlan header to the front of the frame. Move the common vlan encapsulation in to ether_vlanencap(). Reported by: Erik Osterholm, Jon Otterholm MFC after: 1 week
2007-10-18 21:22:15 +00:00
/*
* If underlying interface can not do VLAN tag insertion itself
* then attach a packet tag that holds it.
*/
if ((m->m_flags & M_VLANTAG) &&
(dst_ifp->if_capenable & IFCAP_VLAN_HWTAGGING) == 0) {
m = ether_vlanencap(m, m->m_pkthdr.ether_vtag);
if (m == NULL) {
if_printf(dst_ifp,
"unable to prepend VLAN header\n");
dst_ifp->if_oerrors++;
continue;
}
m->m_flags &= ~M_VLANTAG;
}
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
if (err == 0)
use if_transmit intead of direct frobbing of the if_snd q; this is no longer allowed Identified by: rwatson Reviewed by: kmacy
2009-04-27 22:06:49 +00:00
dst_ifp->if_transmit(dst_ifp, m);
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (err == 0) {
Stop embedding struct ifnet at the top of driver softcs. Instead the struct ifnet or the layer 2 common structure it was embedded in have been replaced with a struct ifnet pointer to be filled by a call to the new function, if_alloc(). The layer 2 common structure is also allocated via if_alloc() based on the interface type. It is hung off the new struct ifnet member, if_l2com. This change removes the size of these structures from the kernel ABI and will allow us to better manage them as interfaces come and go. Other changes of note: - Struct arpcom is no longer referenced in normal interface code. Instead the Ethernet address is accessed via the IFP2ENADDR() macro. To enforce this ac_enaddr has been renamed to _ac_enaddr. - The second argument to ether_ifattach is now always the mac address from driver private storage rather than sometimes being ac_enaddr. Reviewed by: sobomax, sam
2005-06-10 16:49:24 +00:00
sc->sc_ifp->if_opackets++;
sc->sc_ifp->if_obytes += len;
use if_transmit intead of direct frobbing of the if_snd q; this is no longer allowed Identified by: rwatson Reviewed by: kmacy
2009-04-27 22:06:49 +00:00
if (mflags & M_MCAST)
Stop embedding struct ifnet at the top of driver softcs. Instead the struct ifnet or the layer 2 common structure it was embedded in have been replaced with a struct ifnet pointer to be filled by a call to the new function, if_alloc(). The layer 2 common structure is also allocated via if_alloc() based on the interface type. It is hung off the new struct ifnet member, if_l2com. This change removes the size of these structures from the kernel ABI and will allow us to better manage them as interfaces come and go. Other changes of note: - Struct arpcom is no longer referenced in normal interface code. Instead the Ethernet address is accessed via the IFP2ENADDR() macro. To enforce this ac_enaddr has been renamed to _ac_enaddr. - The second argument to ether_ifattach is now always the mac address from driver private storage rather than sometimes being ac_enaddr. Reviewed by: sobomax, sam
2005-06-10 16:49:24 +00:00
sc->sc_ifp->if_omcasts++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
}
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
/*
* bridge_dummynet:
*
* Receive a queued packet from dummynet and pass it on to the output
* interface.
*
* The mbuf has the Ethernet header already attached.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
bridge_dummynet(struct mbuf *m, struct ifnet *ifp)
{
struct bridge_softc *sc;
sc = ifp->if_bridge;
/*
* The packet didnt originate from a member interface. This should only
* ever happen if a member interface is removed while packets are
* queued for it.
*/
Add missing {} in last commit.
2005-06-10 15:53:21 +00:00
if (sc == NULL) {
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
m_freem(m);
return;
Add missing {} in last commit.
2005-06-10 15:53:21 +00:00
}
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
if (PFIL_HOOKED(&V_inet_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#ifdef INET6
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
|| PFIL_HOOKED(&V_inet6_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#endif
) {
if (bridge_pfil(&m, sc->sc_ifp, ifp, PFIL_OUT) != 0)
return;
if (m == NULL)
return;
}
bridge_enqueue(sc, ifp, m);
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* bridge_output:
*
* Send output from a bridge member interface. This
* performs the bridging function for locally originated
* packets.
*
* The mbuf has the Ethernet header already attached. We must
* enqueue or free the mbuf before returning.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_output(struct ifnet *ifp, struct mbuf *m, struct sockaddr *sa,
struct rtentry *rt)
{
struct ether_header *eh;
struct ifnet *dst_if;
struct bridge_softc *sc;
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
uint16_t vlan;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (m->m_len < ETHER_HDR_LEN) {
m = m_pullup(m, ETHER_HDR_LEN);
if (m == NULL)
return (0);
}
eh = mtod(m, struct ether_header *);
sc = ifp->if_bridge;
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
vlan = VLANTAGOF(m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_LOCK(sc);
/*
* If bridge is down, but the original output interface is up,
* go ahead and send out that interface. Otherwise, the packet
* is dropped below.
*/
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
if ((sc->sc_ifp->if_drv_flags & IFF_DRV_RUNNING) == 0) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
dst_if = ifp;
goto sendunicast;
}
/*
* If the packet is a multicast, or we don't know a better way to
* get there, send to all interfaces.
*/
if (ETHER_IS_MULTICAST(eh->ether_dhost))
dst_if = NULL;
else
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
dst_if = bridge_rtlookup(sc, eh->ether_dhost, vlan);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (dst_if == NULL) {
struct bridge_iflist *bif;
struct mbuf *mc;
int error = 0, used = 0;
Call bridge_span before dropping the lock. MFC after: 5 days
2006-08-17 10:18:31 +00:00
bridge_span(sc, m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_LOCK2REF(sc, error);
if (error) {
m_freem(m);
return (0);
}
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
LIST_FOREACH(bif, &sc->sc_iflist, bif_next) {
dst_if = bif->bif_ifp;
Add RFC 3378 EtherIP support. This change makes it possible to add gif interfaces to bridges, which will then send and receive IP protocol 97 packets. Packets are Ethernet frames with an EtherIP header prepended. Obtained from: NetBSD MFC after: 2 weeks
2005-12-21 21:29:45 +00:00
if (dst_if->if_type == IFT_GIF)
continue;
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
continue;
/*
* If this is not the original output interface,
* and the interface is participating in spanning
* tree, make sure the port is in a state that
* allows forwarding.
*/
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
if (dst_if != ifp && (bif->bif_flags & IFBIF_STP) &&
bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING)
continue;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (LIST_NEXT(bif, bif_next) == NULL) {
used = 1;
mc = m;
} else {
Use m_copypacket() which is an optimization of the common case m_copym(m, 0, M_COPYALL, how). This is required for strict alignment architectures where we align the IP header in the input path but m_copym() will create an unaligned copy in bridge_broadcast(). m_copypacket() preserves alignment of the first mbuf. Noticed by: Petri Simolin Approved by: mlaier (mentor) MFC after: 3 days
2005-08-08 22:21:55 +00:00
mc = m_copypacket(m, M_DONTWAIT);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (mc == NULL) {
Stop embedding struct ifnet at the top of driver softcs. Instead the struct ifnet or the layer 2 common structure it was embedded in have been replaced with a struct ifnet pointer to be filled by a call to the new function, if_alloc(). The layer 2 common structure is also allocated via if_alloc() based on the interface type. It is hung off the new struct ifnet member, if_l2com. This change removes the size of these structures from the kernel ABI and will allow us to better manage them as interfaces come and go. Other changes of note: - Struct arpcom is no longer referenced in normal interface code. Instead the Ethernet address is accessed via the IFP2ENADDR() macro. To enforce this ac_enaddr has been renamed to _ac_enaddr. - The second argument to ether_ifattach is now always the mac address from driver private storage rather than sometimes being ac_enaddr. Reviewed by: sobomax, sam
2005-06-10 16:49:24 +00:00
sc->sc_ifp->if_oerrors++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
continue;
}
}
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
bridge_enqueue(sc, dst_if, mc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
if (used == 0)
m_freem(m);
BRIDGE_UNREF(sc);
return (0);
}
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
sendunicast:
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* XXX Spanning tree consideration here?
*/
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
bridge_span(sc, m);
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
m_freem(m);
BRIDGE_UNLOCK(sc);
return (0);
}
BRIDGE_UNLOCK(sc);
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
bridge_enqueue(sc, dst_if, m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (0);
}
/*
* bridge_start:
*
* Start output on a bridge.
*
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_start(struct ifnet *ifp)
{
struct bridge_softc *sc;
struct mbuf *m;
struct ether_header *eh;
struct ifnet *dst_if;
sc = ifp->if_softc;
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
ifp->if_drv_flags |= IFF_DRV_OACTIVE;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
for (;;) {
IFQ_DEQUEUE(&ifp->if_snd, m);
if (m == 0)
break;
Use ETHER_BPF_MTAP so that the vlan tags are visible to bpf(4) when bridging a vlan trunk. Discussed with: csjp MFC after: 3 days
2007-10-20 02:10:10 +00:00
ETHER_BPF_MTAP(ifp, m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
eh = mtod(m, struct ether_header *);
dst_if = NULL;
BRIDGE_LOCK(sc);
if ((m->m_flags & (M_BCAST|M_MCAST)) == 0) {
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
dst_if = bridge_rtlookup(sc, eh->ether_dhost, 1);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
if (dst_if == NULL)
Do not packet filter in the bridge_start() routine, locally generated packets are already filtered by the higher layers. Approved by: mlaier (mentor) MFC after: 3 days
2005-10-02 19:15:56 +00:00
bridge_broadcast(sc, ifp, m, 0);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
else {
BRIDGE_UNLOCK(sc);
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
bridge_enqueue(sc, dst_if, m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
}
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
ifp->if_drv_flags &= ~IFF_DRV_OACTIVE;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
/*
* bridge_forward:
*
* The forwarding function of the bridge.
*
* NOTE: Releases the lock on return.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
bridge_forward(struct bridge_softc *sc, struct bridge_iflist *sbif,
struct mbuf *m)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
struct bridge_iflist *dbif;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
struct ifnet *src_if, *dst_if, *ifp;
struct ether_header *eh;
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
uint16_t vlan;
IEEE 802.1D-2004 states, frames containing any of the group MAC Addresses specified in Table 7-10 in their destination address field shall not be relayed by the Bridge. Add a check in bridge_forward() to adhere to this. PR: kern/119744
2008-01-18 00:19:10 +00:00
uint8_t *dst;
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
int error;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
src_if = m->m_pkthdr.rcvif;
Stop embedding struct ifnet at the top of driver softcs. Instead the struct ifnet or the layer 2 common structure it was embedded in have been replaced with a struct ifnet pointer to be filled by a call to the new function, if_alloc(). The layer 2 common structure is also allocated via if_alloc() based on the interface type. It is hung off the new struct ifnet member, if_l2com. This change removes the size of these structures from the kernel ABI and will allow us to better manage them as interfaces come and go. Other changes of note: - Struct arpcom is no longer referenced in normal interface code. Instead the Ethernet address is accessed via the IFP2ENADDR() macro. To enforce this ac_enaddr has been renamed to _ac_enaddr. - The second argument to ether_ifattach is now always the mac address from driver private storage rather than sometimes being ac_enaddr. Reviewed by: sobomax, sam
2005-06-10 16:49:24 +00:00
ifp = sc->sc_ifp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
ifp->if_ipackets++;
ifp->if_ibytes += m->m_pkthdr.len;
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
vlan = VLANTAGOF(m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
if ((sbif->bif_flags & IFBIF_STP) &&
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
sbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING)
goto drop;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
eh = mtod(m, struct ether_header *);
IEEE 802.1D-2004 states, frames containing any of the group MAC Addresses specified in Table 7-10 in their destination address field shall not be relayed by the Bridge. Add a check in bridge_forward() to adhere to this. PR: kern/119744
2008-01-18 00:19:10 +00:00
dst = eh->ether_dhost;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
/* If the interface is learning, record the address. */
if (sbif->bif_flags & IFBIF_LEARNING) {
error = bridge_rtupdate(sc, eh->ether_shost, vlan,
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
sbif, 0, IFBAF_DYNAMIC);
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
/*
* If the interface has addresses limits then deny any source
* that is not in the cache.
*/
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
if (error && sbif->bif_addrmax)
goto drop;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
if ((sbif->bif_flags & IFBIF_STP) != 0 &&
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
sbif->bif_stp.bp_state == BSTP_IFSTATE_LEARNING)
goto drop;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* At this point, the port either doesn't participate
* in spanning tree or it is in the forwarding state.
*/
/*
* If the packet is unicast, destined for someone on
* "this" side of the bridge, drop it.
*/
if ((m->m_flags & (M_BCAST|M_MCAST)) == 0) {
IEEE 802.1D-2004 states, frames containing any of the group MAC Addresses specified in Table 7-10 in their destination address field shall not be relayed by the Bridge. Add a check in bridge_forward() to adhere to this. PR: kern/119744
2008-01-18 00:19:10 +00:00
dst_if = bridge_rtlookup(sc, dst, vlan);
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
if (src_if == dst_if)
goto drop;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
} else {
IEEE 802.1D-2004 states, frames containing any of the group MAC Addresses specified in Table 7-10 in their destination address field shall not be relayed by the Bridge. Add a check in bridge_forward() to adhere to this. PR: kern/119744
2008-01-18 00:19:10 +00:00
/*
* Check if its a reserved multicast address, any address
* listed in 802.1D section 7.12.6 may not be forwarded by the
* bridge.
* This is currently 01-80-C2-00-00-00 to 01-80-C2-00-00-0F
*/
if (dst[0] == 0x01 && dst[1] == 0x80 &&
dst[2] == 0xc2 && dst[3] == 0x00 &&
dst[4] == 0x00 && dst[5] <= 0x0f)
goto drop;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/* ...forward it to all interfaces. */
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
ifp->if_imcasts++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
dst_if = NULL;
}
Slightly re-worked bpf(4) code associated with bridging: if we have a destination interface as a member of our bridge or this is a unicast packet, push it through the bpf(4) machinery. For broadcast or multicast packets, don't bother with the bpf(4) because it will be re-injected into ether_input. We do this before we pass the packets through the pfil(9) framework, as it is possible that pfil(9) will drop the packet or possibly modify it, making it very difficult to debug firewall issues on the bridge. Further, implemented IFF_MONITOR for bridge interfaces. This does much the same thing that it does for regular network interfaces: it pushes the packet to any bpf(4) peers and then returns. This bypasses all of the bridge machinery, saving mutex acquisitions, list traversals, and other operations performed by the bridging code. This change to the bridging code is useful in situations where individuals use a bridge to multiplex RX/TX signals from two interfaces, as is required by some network taps for de-multiplexing links and transmitting the RX/TX signals out through two separate interfaces. This behaviour is quite common for network taps monitoring links, especially for certain manufacturers. Reviewed by: thompsa MFC after: 1 month Sponsored by: Seccuris Labs
2006-03-03 05:58:18 +00:00
/*
* If we have a destination interface which is a member of our bridge,
* OR this is a unicast packet, push it through the bpf(4) machinery.
* For broadcast or multicast packets, don't bother because it will
* be reinjected into ether_input. We do this before we pass the packets
* through the pfil(9) framework, as it is possible that pfil(9) will
* drop the packet, or possibly modify it, making it difficult to debug
* firewall issues on the bridge.
*/
if (dst_if != NULL || (m->m_flags & (M_BCAST | M_MCAST)) == 0)
Use ETHER_BPF_MTAP so that the vlan tags are visible to bpf(4) when bridging a vlan trunk. Discussed with: csjp MFC after: 3 days
2007-10-20 02:10:10 +00:00
ETHER_BPF_MTAP(ifp, m);
Slightly re-worked bpf(4) code associated with bridging: if we have a destination interface as a member of our bridge or this is a unicast packet, push it through the bpf(4) machinery. For broadcast or multicast packets, don't bother with the bpf(4) because it will be re-injected into ether_input. We do this before we pass the packets through the pfil(9) framework, as it is possible that pfil(9) will drop the packet or possibly modify it, making it very difficult to debug firewall issues on the bridge. Further, implemented IFF_MONITOR for bridge interfaces. This does much the same thing that it does for regular network interfaces: it pushes the packet to any bpf(4) peers and then returns. This bypasses all of the bridge machinery, saving mutex acquisitions, list traversals, and other operations performed by the bridging code. This change to the bridging code is useful in situations where individuals use a bridge to multiplex RX/TX signals from two interfaces, as is required by some network taps for de-multiplexing links and transmitting the RX/TX signals out through two separate interfaces. This behaviour is quite common for network taps monitoring links, especially for certain manufacturers. Reviewed by: thompsa MFC after: 1 month Sponsored by: Seccuris Labs
2006-03-03 05:58:18 +00:00
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/* run the packet filter */
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
if (PFIL_HOOKED(&V_inet_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#ifdef INET6
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
|| PFIL_HOOKED(&V_inet6_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#endif
) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_UNLOCK(sc);
if (bridge_pfil(&m, ifp, src_if, PFIL_IN) != 0)
return;
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
if (m == NULL)
return;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_LOCK(sc);
}
if (dst_if == NULL) {
Do not packet filter in the bridge_start() routine, locally generated packets are already filtered by the higher layers. Approved by: mlaier (mentor) MFC after: 3 days
2005-10-02 19:15:56 +00:00
bridge_broadcast(sc, src_if, m, 1);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return;
}
/*
* At this point, we're dealing with a unicast frame
* going to a different interface.
*/
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0)
goto drop;
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
dbif = bridge_lookup_member_if(sc, dst_if);
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
if (dbif == NULL)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/* Not a member of the bridge (anymore?) */
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
goto drop;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
/* Private segments can not talk to each other */
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
if (sbif->bif_flags & dbif->bif_flags & IFBIF_PRIVATE)
goto drop;
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
if ((dbif->bif_flags & IFBIF_STP) &&
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
dbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING)
goto drop;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_UNLOCK(sc);
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
if (PFIL_HOOKED(&V_inet_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#ifdef INET6
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
|| PFIL_HOOKED(&V_inet6_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#endif
) {
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
if (bridge_pfil(&m, ifp, dst_if, PFIL_OUT) != 0)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
return;
if (m == NULL)
return;
}
bridge_enqueue(sc, dst_if, m);
Simplify the error handling and use the dereferenced sc->sc_ifp pointer.
2007-12-18 09:13:04 +00:00
return;
drop:
BRIDGE_UNLOCK(sc);
m_freem(m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
/*
* bridge_input:
*
* Receive input from a member interface. Queue the packet for
* bridging if it is not for us.
*/
Make four more functions static that were missed in the last commit.
2005-10-14 20:57:02 +00:00
static struct mbuf *
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_input(struct ifnet *ifp, struct mbuf *m)
{
struct bridge_softc *sc = ifp->if_bridge;
Add a new address cache type called sticky. On an interface marked sticky any address learned by the bridge is made permanent, the address will not age out and most importantly will not migrate to another interface. This can be used to stop mac address poisoning or clients roaming in much the same way as static entries without the hassle of preloading the table.
2006-11-09 06:32:38 +00:00
struct bridge_iflist *bif, *bif2;
Add support for multicast to the bridge and allow inet6 addresses to be assigned to the interface. IPv6 auto-configuration is disabled. An IPv6 link-local address has a link-local scope within one link, the spec is unclear for the bridge case and it may cause scope violation. An address can be assigned in the usual way; ifconfig bridge0 inet6 xxxx:... Tested by: bmah Reviewed by: ume (netinet6) Approved by: mlaier (mentor) MFC after: 1 week
2005-09-06 21:11:59 +00:00
struct ifnet *bifp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
struct ether_header *eh;
Add support for multicast to the bridge and allow inet6 addresses to be assigned to the interface. IPv6 auto-configuration is disabled. An IPv6 link-local address has a link-local scope within one link, the spec is unclear for the bridge case and it may cause scope violation. An address can be assigned in the usual way; ifconfig bridge0 inet6 xxxx:... Tested by: bmah Reviewed by: ume (netinet6) Approved by: mlaier (mentor) MFC after: 1 week
2005-09-06 21:11:59 +00:00
struct mbuf *mc, *mc2;
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
uint16_t vlan;
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
int error;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
if ((sc->sc_ifp->if_drv_flags & IFF_DRV_RUNNING) == 0)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (m);
Add support for multicast to the bridge and allow inet6 addresses to be assigned to the interface. IPv6 auto-configuration is disabled. An IPv6 link-local address has a link-local scope within one link, the spec is unclear for the bridge case and it may cause scope violation. An address can be assigned in the usual way; ifconfig bridge0 inet6 xxxx:... Tested by: bmah Reviewed by: ume (netinet6) Approved by: mlaier (mentor) MFC after: 1 week
2005-09-06 21:11:59 +00:00
bifp = sc->sc_ifp;
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
vlan = VLANTAGOF(m);
Add support for multicast to the bridge and allow inet6 addresses to be assigned to the interface. IPv6 auto-configuration is disabled. An IPv6 link-local address has a link-local scope within one link, the spec is unclear for the bridge case and it may cause scope violation. An address can be assigned in the usual way; ifconfig bridge0 inet6 xxxx:... Tested by: bmah Reviewed by: ume (netinet6) Approved by: mlaier (mentor) MFC after: 1 week
2005-09-06 21:11:59 +00:00
Slightly re-worked bpf(4) code associated with bridging: if we have a destination interface as a member of our bridge or this is a unicast packet, push it through the bpf(4) machinery. For broadcast or multicast packets, don't bother with the bpf(4) because it will be re-injected into ether_input. We do this before we pass the packets through the pfil(9) framework, as it is possible that pfil(9) will drop the packet or possibly modify it, making it very difficult to debug firewall issues on the bridge. Further, implemented IFF_MONITOR for bridge interfaces. This does much the same thing that it does for regular network interfaces: it pushes the packet to any bpf(4) peers and then returns. This bypasses all of the bridge machinery, saving mutex acquisitions, list traversals, and other operations performed by the bridging code. This change to the bridging code is useful in situations where individuals use a bridge to multiplex RX/TX signals from two interfaces, as is required by some network taps for de-multiplexing links and transmitting the RX/TX signals out through two separate interfaces. This behaviour is quite common for network taps monitoring links, especially for certain manufacturers. Reviewed by: thompsa MFC after: 1 month Sponsored by: Seccuris Labs
2006-03-03 05:58:18 +00:00
/*
* Implement support for bridge monitoring. If this flag has been
* set on this interface, discard the packet once we push it through
* the bpf(4) machinery, but before we do, increment the byte and
* packet counters associated with this interface.
*/
if ((bifp->if_flags & IFF_MONITOR) != 0) {
m->m_pkthdr.rcvif = bifp;
Use ETHER_BPF_MTAP so that the vlan tags are visible to bpf(4) when bridging a vlan trunk. Discussed with: csjp MFC after: 3 days
2007-10-20 02:10:10 +00:00
ETHER_BPF_MTAP(bifp, m);
Slightly re-worked bpf(4) code associated with bridging: if we have a destination interface as a member of our bridge or this is a unicast packet, push it through the bpf(4) machinery. For broadcast or multicast packets, don't bother with the bpf(4) because it will be re-injected into ether_input. We do this before we pass the packets through the pfil(9) framework, as it is possible that pfil(9) will drop the packet or possibly modify it, making it very difficult to debug firewall issues on the bridge. Further, implemented IFF_MONITOR for bridge interfaces. This does much the same thing that it does for regular network interfaces: it pushes the packet to any bpf(4) peers and then returns. This bypasses all of the bridge machinery, saving mutex acquisitions, list traversals, and other operations performed by the bridging code. This change to the bridging code is useful in situations where individuals use a bridge to multiplex RX/TX signals from two interfaces, as is required by some network taps for de-multiplexing links and transmitting the RX/TX signals out through two separate interfaces. This behaviour is quite common for network taps monitoring links, especially for certain manufacturers. Reviewed by: thompsa MFC after: 1 month Sponsored by: Seccuris Labs
2006-03-03 05:58:18 +00:00
bifp->if_ipackets++;
bifp->if_ibytes += m->m_pkthdr.len;
Fix possible leak when bridge is in monitor mode. Use m_freem() which will free the entire chain, instead of using m_free() which will free just the mbuf that was passed. Discussed with: thompsa MFC after: 3 days
2006-11-05 17:56:25 +00:00
m_freem(m);
Slightly re-worked bpf(4) code associated with bridging: if we have a destination interface as a member of our bridge or this is a unicast packet, push it through the bpf(4) machinery. For broadcast or multicast packets, don't bother with the bpf(4) because it will be re-injected into ether_input. We do this before we pass the packets through the pfil(9) framework, as it is possible that pfil(9) will drop the packet or possibly modify it, making it very difficult to debug firewall issues on the bridge. Further, implemented IFF_MONITOR for bridge interfaces. This does much the same thing that it does for regular network interfaces: it pushes the packet to any bpf(4) peers and then returns. This bypasses all of the bridge machinery, saving mutex acquisitions, list traversals, and other operations performed by the bridging code. This change to the bridging code is useful in situations where individuals use a bridge to multiplex RX/TX signals from two interfaces, as is required by some network taps for de-multiplexing links and transmitting the RX/TX signals out through two separate interfaces. This behaviour is quite common for network taps monitoring links, especially for certain manufacturers. Reviewed by: thompsa MFC after: 1 month Sponsored by: Seccuris Labs
2006-03-03 05:58:18 +00:00
return (NULL);
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_LOCK(sc);
bif = bridge_lookup_member_if(sc, ifp);
if (bif == NULL) {
BRIDGE_UNLOCK(sc);
return (m);
}
eh = mtod(m, struct ether_header *);
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
bridge_span(sc, m);
M_BCAST & M_MCAST are now set by ether_input before passing to the bridge.
2007-03-19 19:13:02 +00:00
if (m->m_flags & (M_BCAST|M_MCAST)) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/* Tap off 802.1D packets; they do not get forwarded. */
if (memcmp(eh->ether_dhost, bstp_etheraddr,
ETHER_ADDR_LEN) == 0) {
/tmp/cvsuusTrc
2006-07-26 10:43:02 +00:00
m = bstp_input(&bif->bif_stp, ifp, m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (m == NULL) {
BRIDGE_UNLOCK(sc);
return (NULL);
}
}
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
if ((bif->bif_flags & IFBIF_STP) &&
bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) {
BRIDGE_UNLOCK(sc);
return (m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
/*
* Make a deep copy of the packet and enqueue the copy
* for bridge processing; return the original packet for
* local processing.
*/
mc = m_dup(m, M_DONTWAIT);
if (mc == NULL) {
BRIDGE_UNLOCK(sc);
return (m);
}
/* Perform the bridge forwarding function with the copy. */
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
bridge_forward(sc, bif, mc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add support for multicast to the bridge and allow inet6 addresses to be assigned to the interface. IPv6 auto-configuration is disabled. An IPv6 link-local address has a link-local scope within one link, the spec is unclear for the bridge case and it may cause scope violation. An address can be assigned in the usual way; ifconfig bridge0 inet6 xxxx:... Tested by: bmah Reviewed by: ume (netinet6) Approved by: mlaier (mentor) MFC after: 1 week
2005-09-06 21:11:59 +00:00
/*
* Reinject the mbuf as arriving on the bridge so we have a
* chance at claiming multicast packets. We can not loop back
* here from ether_input as a bridge is never a member of a
* bridge.
*/
KASSERT(bifp->if_bridge == NULL,
("loop created in bridge_input"));
It is not safe to use m_copypacket() here as the returned mbuf is readonly, change to m_dup and keep the alignment on the layer3 header. MFC after: 1 week
2005-12-15 19:34:39 +00:00
mc2 = m_dup(m, M_DONTWAIT);
if (mc2 != NULL) {
/* Keep the layer3 header aligned */
int i = min(mc2->m_pkthdr.len, max_protohdr);
mc2 = m_copyup(mc2, i, ETHER_ALIGN);
}
Add support for multicast to the bridge and allow inet6 addresses to be assigned to the interface. IPv6 auto-configuration is disabled. An IPv6 link-local address has a link-local scope within one link, the spec is unclear for the bridge case and it may cause scope violation. An address can be assigned in the usual way; ifconfig bridge0 inet6 xxxx:... Tested by: bmah Reviewed by: ume (netinet6) Approved by: mlaier (mentor) MFC after: 1 week
2005-09-06 21:11:59 +00:00
if (mc2 != NULL) {
mc2->m_pkthdr.rcvif = bifp;
(*bifp->if_input)(bifp, mc2);
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/* Return the original packet for local processing. */
return (m);
}
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
if ((bif->bif_flags & IFBIF_STP) &&
bif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING) {
BRIDGE_UNLOCK(sc);
return (m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Allow carp(4) to be loaded as a kernel module. Follow precedent set by bridge(4), lagg(4) etc. and make use of function pointers and pf_proto_register() to hook carp into the network stack. Currently, because of the uncertainty about whether the unload path is free of race condition panics, unloads are disallowed by default. Compiling with CARPMOD_CAN_UNLOAD in CFLAGS removes this anti foot shooting measure. This commit requires IP6PROTOSPACER, introduced in r211115. Reviewed by: bz, simon Approved by: ken (mentor) MFC after: 2 weeks
2010-08-11 00:51:50 +00:00
#if (defined(INET) || defined(INET6))
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
# define OR_CARP_CHECK_WE_ARE_DST(iface) \
|| ((iface)->if_carp \
Unbreak LINT by moving all carp hooks to net/if.c / netinet/ip_carp.h, with the appropriate ifdefs. Reviewed by: bz Approved by: ken (mentor)
2010-08-11 20:18:19 +00:00
&& (*carp_forus_p)((iface), eh->ether_dhost))
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
# define OR_CARP_CHECK_WE_ARE_SRC(iface) \
|| ((iface)->if_carp \
Unbreak LINT by moving all carp hooks to net/if.c / netinet/ip_carp.h, with the appropriate ifdefs. Reviewed by: bz Approved by: ken (mentor)
2010-08-11 20:18:19 +00:00
&& (*carp_forus_p)((iface), eh->ether_shost))
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
#else
# define OR_CARP_CHECK_WE_ARE_DST(iface)
# define OR_CARP_CHECK_WE_ARE_SRC(iface)
#endif
Remove a chunk of duplicated code, test the destination address against the bridge the same way we check member interfaces.
2008-01-18 09:34:09 +00:00
#ifdef INET6
# define OR_PFIL_HOOKED_INET6 \
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
|| PFIL_HOOKED(&V_inet6_pfil_hook)
Remove a chunk of duplicated code, test the destination address against the bridge the same way we check member interfaces.
2008-01-18 09:34:09 +00:00
#else
# define OR_PFIL_HOOKED_INET6
#endif
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
#define GRAB_OUR_PACKETS(iface) \
if ((iface)->if_type == IFT_GIF) \
continue; \
/* It is destined for us. */ \
if (memcmp(IF_LLADDR((iface)), eh->ether_dhost, ETHER_ADDR_LEN) == 0 \
OR_CARP_CHECK_WE_ARE_DST((iface)) \
) { \
Remove a chunk of duplicated code, test the destination address against the bridge the same way we check member interfaces.
2008-01-18 09:34:09 +00:00
if ((iface)->if_type == IFT_BRIDGE) { \
ETHER_BPF_MTAP(iface, m); \
iface->if_ipackets++; \
/* Filter on the physical interface. */ \
if (pfil_local_phys && \
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
(PFIL_HOOKED(&V_inet_pfil_hook) \
Remove a chunk of duplicated code, test the destination address against the bridge the same way we check member interfaces.
2008-01-18 09:34:09 +00:00
OR_PFIL_HOOKED_INET6)) { \
if (bridge_pfil(&m, NULL, ifp, \
PFIL_IN) != 0 || m == NULL) { \
BRIDGE_UNLOCK(sc); \
return (NULL); \
} \
} \
} \
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
if (bif->bif_flags & IFBIF_LEARNING) { \
error = bridge_rtupdate(sc, eh->ether_shost, \
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
vlan, bif, 0, IFBAF_DYNAMIC); \
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
if (error && bif->bif_addrmax) { \
BRIDGE_UNLOCK(sc); \
m_freem(m); \
return (NULL); \
} \
} \
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
m->m_pkthdr.rcvif = iface; \
BRIDGE_UNLOCK(sc); \
return (m); \
} \
\
/* We just received a packet that we sent out. */ \
if (memcmp(IF_LLADDR((iface)), eh->ether_shost, ETHER_ADDR_LEN) == 0 \
OR_CARP_CHECK_WE_ARE_SRC((iface)) \
) { \
BRIDGE_UNLOCK(sc); \
m_freem(m); \
return (NULL); \
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
Remove a chunk of duplicated code, test the destination address against the bridge the same way we check member interfaces.
2008-01-18 09:34:09 +00:00
* Unicast. Make sure it's not for the bridge.
*/
do { GRAB_OUR_PACKETS(bifp) } while (0);
/*
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
* Give a chance for ifp at first priority. This will help when the
* packet comes through the interface like VLAN's with the same MACs
* on several interfaces from the same bridge. This also will save
* some CPU cycles in case the destination interface and the input
* interface (eq ifp) are the same.
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
*/
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
do { GRAB_OUR_PACKETS(ifp) } while (0);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
/* Now check the all bridge members. */
LIST_FOREACH(bif2, &sc->sc_iflist, bif_next) {
GRAB_OUR_PACKETS(bif2->bif_ifp)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
#undef OR_CARP_CHECK_WE_ARE_DST
#undef OR_CARP_CHECK_WE_ARE_SRC
Remove a chunk of duplicated code, test the destination address against the bridge the same way we check member interfaces.
2008-01-18 09:34:09 +00:00
#undef OR_PFIL_HOOKED_INET6
Give a chance for packet to appear with a correct input interfaces in case of multiple interfaces with the same MAC in the same bridge. This commit do not solve the entire problem. Only case where packet arrived from such interface. PR: kern/109815 MFC after: 7 days Submitted by: Eygene Ryabinkin and rik@ Discussed with: bms@, thompsa@, yar@
2007-03-18 23:28:53 +00:00
#undef GRAB_OUR_PACKETS
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/* Perform the bridge forwarding function. */
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
bridge_forward(sc, bif, m);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (NULL);
}
/*
* bridge_broadcast:
*
* Send a frame to all interfaces that are members of
* the bridge, except for the one on which the packet
* arrived.
*
* NOTE: Releases the lock on return.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_broadcast(struct bridge_softc *sc, struct ifnet *src_if,
Do not packet filter in the bridge_start() routine, locally generated packets are already filtered by the higher layers. Approved by: mlaier (mentor) MFC after: 3 days
2005-10-02 19:15:56 +00:00
struct mbuf *m, int runfilt)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
struct bridge_iflist *dbif, *sbif;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
struct mbuf *mc;
struct ifnet *dst_if;
m_dup () packet not m_copypacket () since we will modify it. For more details see PR kern/94448. PR: kern/94448 Original patch: Eygene A. Ryabinkin <rea-fbsd at rea dot mbslab dot kiae dot ru>Final patch: thompsa@ Tested by: thompsa@, Eygene A. Ryabinkin MFC after: 7 days
2006-03-23 22:57:10 +00:00
int error = 0, used = 0, i;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
sbif = bridge_lookup_member_if(sc, src_if);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_LOCK2REF(sc, error);
if (error) {
m_freem(m);
return;
}
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
/* Filter on the bridge interface before broadcasting */
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
if (runfilt && (PFIL_HOOKED(&V_inet_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#ifdef INET6
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
|| PFIL_HOOKED(&V_inet6_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#endif
Do not packet filter in the bridge_start() routine, locally generated packets are already filtered by the higher layers. Approved by: mlaier (mentor) MFC after: 3 days
2005-10-02 19:15:56 +00:00
)) {
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
if (bridge_pfil(&m, sc->sc_ifp, NULL, PFIL_OUT) != 0)
Fix a second missed case where the refcount is not decremented. MFC after: 3 days
2005-11-13 20:26:19 +00:00
goto out;
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
if (m == NULL)
Fix a second missed case where the refcount is not decremented. MFC after: 3 days
2005-11-13 20:26:19 +00:00
goto out;
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
}
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
LIST_FOREACH(dbif, &sc->sc_iflist, bif_next) {
dst_if = dbif->bif_ifp;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (dst_if == src_if)
continue;
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
/* Private segments can not talk to each other */
if (sbif && (sbif->bif_flags & dbif->bif_flags & IFBIF_PRIVATE))
continue;
if ((dbif->bif_flags & IFBIF_STP) &&
dbif->bif_stp.bp_state == BSTP_IFSTATE_DISCARDING)
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
continue;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
if ((dbif->bif_flags & IFBIF_DISCOVER) == 0 &&
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
(m->m_flags & (M_BCAST|M_MCAST)) == 0)
continue;
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
continue;
Add a bridge interface flag called PRIVATE where any private port can not communicate with another private port. All unicast/broadcast/multicast layer2 traffic is blocked so it works much the same way as using firewall rules but scales better and is generally easier as firewall packages usually do not allow ARP blocking. An example usage would be having a number of customers on separate vlans bridged with a server network. All the vlans are marked private, they can all communicate with the server network unhindered, but can not exchange any traffic whatsoever with each other. Approved by: re (rwatson)
2007-08-01 00:33:52 +00:00
if (LIST_NEXT(dbif, bif_next) == NULL) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
mc = m;
used = 1;
} else {
m_dup () packet not m_copypacket () since we will modify it. For more details see PR kern/94448. PR: kern/94448 Original patch: Eygene A. Ryabinkin <rea-fbsd at rea dot mbslab dot kiae dot ru>Final patch: thompsa@ Tested by: thompsa@, Eygene A. Ryabinkin MFC after: 7 days
2006-03-23 22:57:10 +00:00
mc = m_dup(m, M_DONTWAIT);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (mc == NULL) {
Stop embedding struct ifnet at the top of driver softcs. Instead the struct ifnet or the layer 2 common structure it was embedded in have been replaced with a struct ifnet pointer to be filled by a call to the new function, if_alloc(). The layer 2 common structure is also allocated via if_alloc() based on the interface type. It is hung off the new struct ifnet member, if_l2com. This change removes the size of these structures from the kernel ABI and will allow us to better manage them as interfaces come and go. Other changes of note: - Struct arpcom is no longer referenced in normal interface code. Instead the Ethernet address is accessed via the IFP2ENADDR() macro. To enforce this ac_enaddr has been renamed to _ac_enaddr. - The second argument to ether_ifattach is now always the mac address from driver private storage rather than sometimes being ac_enaddr. Reviewed by: sobomax, sam
2005-06-10 16:49:24 +00:00
sc->sc_ifp->if_oerrors++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
continue;
}
}
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
/*
* Filter on the output interface. Pass a NULL bridge interface
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
* pointer so we do not redundantly filter on the bridge for
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
* each interface we broadcast on.
*/
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
if (runfilt && (PFIL_HOOKED(&V_inet_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#ifdef INET6
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
|| PFIL_HOOKED(&V_inet6_pfil_hook)
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
#endif
Do not packet filter in the bridge_start() routine, locally generated packets are already filtered by the higher layers. Approved by: mlaier (mentor) MFC after: 3 days
2005-10-02 19:15:56 +00:00
)) {
m_dup () packet not m_copypacket () since we will modify it. For more details see PR kern/94448. PR: kern/94448 Original patch: Eygene A. Ryabinkin <rea-fbsd at rea dot mbslab dot kiae dot ru>Final patch: thompsa@ Tested by: thompsa@, Eygene A. Ryabinkin MFC after: 7 days
2006-03-23 22:57:10 +00:00
if (used == 0) {
/* Keep the layer3 header aligned */
i = min(mc->m_pkthdr.len, max_protohdr);
mc = m_copyup(mc, i, ETHER_ALIGN);
if (mc == NULL) {
sc->sc_ifp->if_oerrors++;
continue;
}
}
Fix a mbuf and refcnt leak in the broadcast code. If the packet is rejected from pfil(9) then continue the loop rather than returning, this means that we can still try to send it out the remaining interfaces but more importantly the mbuf is freed and refcount decremented on exit.
2005-11-13 19:36:59 +00:00
if (bridge_pfil(&mc, NULL, dst_if, PFIL_OUT) != 0)
continue;
if (mc == NULL)
continue;
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
}
Further clean up the bridge hooks in if_ethersubr.c and ng_ether.c - move the function pointer definitions to if_bridgevar.h - move most of the logic to the new BRIDGE_INPUT and BRIDGE_OUTPUT macros - remove unneeded functions from if_bridgevar.h and sort a little.
2005-10-14 02:38:47 +00:00
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
bridge_enqueue(sc, dst_if, mc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
if (used == 0)
m_freem(m);
Fix a second missed case where the refcount is not decremented. MFC after: 3 days
2005-11-13 20:26:19 +00:00
out:
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_UNREF(sc);
}
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
/*
* bridge_span:
*
* Duplicate a packet out one or more interfaces that are in span mode,
* the original mbuf is unmodified.
*/
static void
bridge_span(struct bridge_softc *sc, struct mbuf *m)
{
struct bridge_iflist *bif;
struct ifnet *dst_if;
struct mbuf *mc;
if (LIST_EMPTY(&sc->sc_spanlist))
return;
LIST_FOREACH(bif, &sc->sc_spanlist, bif_next) {
dst_if = bif->bif_ifp;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
Add support for creating span ports so that one can snoop bridged traffic from another interface/machine/network. Obtained from: OpenBSD MFC after: 2 weeks
2005-12-14 02:52:13 +00:00
if ((dst_if->if_drv_flags & IFF_DRV_RUNNING) == 0)
continue;
mc = m_copypacket(m, M_DONTWAIT);
if (mc == NULL) {
sc->sc_ifp->if_oerrors++;
continue;
}
bridge_enqueue(sc, dst_if, mc);
}
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* bridge_rtupdate:
*
* Add a bridge routing entry.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
bridge_rtupdate(struct bridge_softc *sc, const uint8_t *dst, uint16_t vlan,
Add a new address cache type called sticky. On an interface marked sticky any address learned by the bridge is made permanent, the address will not age out and most importantly will not migrate to another interface. This can be used to stop mac address poisoning or clients roaming in much the same way as static entries without the hassle of preloading the table.
2006-11-09 06:32:38 +00:00
struct bridge_iflist *bif, int setflags, uint8_t flags)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
struct bridge_rtnode *brt;
int error;
BRIDGE_LOCK_ASSERT(sc);
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
/* Check the source address is valid and not multicast. */
if (ETHER_IS_MULTICAST(dst) ||
(dst[0] == 0 && dst[1] == 0 && dst[2] == 0 &&
dst[3] == 0 && dst[4] == 0 && dst[5] == 0) != 0)
return (EINVAL);
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
/* 802.1p frames map to vlan 1 */
if (vlan == 0)
vlan = 1;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* A route for this destination might already exist. If so,
* update it, otherwise create a new one.
*/
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
if ((brt = bridge_rtnode_lookup(sc, dst, vlan)) == NULL) {
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
if (sc->sc_brtcnt >= sc->sc_brtmax) {
sc->sc_brtexceeded++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (ENOSPC);
Add some statistics that are needed to support RFC4188 as part of the SoC2006 work on a bridge monitoring module for BSNMP. Submitted by: shteryana (SoC 2006)
2006-07-31 20:24:46 +00:00
}
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
/* Check per interface address limits (if enabled) */
if (bif->bif_addrmax && bif->bif_addrcnt >= bif->bif_addrmax) {
bif->bif_addrexceeded++;
return (ENOSPC);
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* Allocate a new bridge forwarding node, and
* initialize the expiration time and Ethernet
* address.
*/
brt = uma_zalloc(bridge_rtnode_zone, M_NOWAIT | M_ZERO);
if (brt == NULL)
return (ENOMEM);
Add a new address cache type called sticky. On an interface marked sticky any address learned by the bridge is made permanent, the address will not age out and most importantly will not migrate to another interface. This can be used to stop mac address poisoning or clients roaming in much the same way as static entries without the hassle of preloading the table.
2006-11-09 06:32:38 +00:00
if (bif->bif_flags & IFBIF_STICKY)
brt->brt_flags = IFBAF_STICKY;
else
brt->brt_flags = IFBAF_DYNAMIC;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
memcpy(brt->brt_addr, dst, ETHER_ADDR_LEN);
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
brt->brt_vlan = vlan;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if ((error = bridge_rtnode_insert(sc, brt)) != 0) {
uma_zfree(bridge_rtnode_zone, brt);
return (error);
}
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
brt->brt_dst = bif;
bif->bif_addrcnt++;
}
if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC &&
brt->brt_dst != bif) {
brt->brt_dst->bif_addrcnt--;
brt->brt_dst = bif;
brt->brt_dst->bif_addrcnt++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Fix a brain-o in the last commit, the conditional was always false.
2006-01-02 23:02:43 +00:00
if ((flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC)
Reorganise bridge_rtupdate slightly to reduce duplication.
2006-01-02 22:44:54 +00:00
brt->brt_expire = time_uptime + sc->sc_brttimeout;
Reset the route expiry time on each update rather than always letting them get GC'd and recreated.
2006-01-02 22:29:41 +00:00
if (setflags)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
brt->brt_flags = flags;
return (0);
}
/*
* bridge_rtlookup:
*
* Lookup the destination interface for an address.
*/
Make four more functions static that were missed in the last commit.
2005-10-14 20:57:02 +00:00
static struct ifnet *
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
bridge_rtlookup(struct bridge_softc *sc, const uint8_t *addr, uint16_t vlan)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
struct bridge_rtnode *brt;
BRIDGE_LOCK_ASSERT(sc);
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
if ((brt = bridge_rtnode_lookup(sc, addr, vlan)) == NULL)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (NULL);
return (brt->brt_ifp);
}
/*
* bridge_rttrim:
*
* Trim the routine table so that we have a number
* of routing entries less than or equal to the
* maximum number.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rttrim(struct bridge_softc *sc)
{
struct bridge_rtnode *brt, *nbrt;
BRIDGE_LOCK_ASSERT(sc);
/* Make sure we actually need to do this. */
if (sc->sc_brtcnt <= sc->sc_brtmax)
return;
/* Force an aging cycle; this might trim enough addresses. */
bridge_rtage(sc);
if (sc->sc_brtcnt <= sc->sc_brtmax)
return;
Use LIST_FOREACH_SAFE instead of a hand rolled version.
2006-10-09 00:49:57 +00:00
LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) {
bridge_rtnode_destroy(sc, brt);
if (sc->sc_brtcnt <= sc->sc_brtmax)
return;
}
}
}
/*
* bridge_timer:
*
* Aging timer for the bridge.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_timer(void *arg)
{
struct bridge_softc *sc = arg;
Mark the callouts as MPSAFE as if_bridge has been giant-free since day 1. Use the SMP friendly callout_init_mtx() while we are here. Approved by: mlaier (mentor) MFC after: 3 days
2005-08-18 20:17:00 +00:00
BRIDGE_LOCK_ASSERT(sc);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtage(sc);
Propagate rename of IFF_OACTIVE and IFF_RUNNING to IFF_DRV_OACTIVE and IFF_DRV_RUNNING, as well as the move from ifnet.if_flags to ifnet.if_drv_flags. Device drivers are now responsible for synchronizing access to these flags, as they are in if_drv_flags. This helps prevent races between the network stack and device driver in maintaining the interface flags field. Many __FreeBSD__ and __FreeBSD_version checks maintained and continued; some less so. Reviewed by: pjd, bz MFC after: 7 days
2005-08-09 10:20:02 +00:00
if (sc->sc_ifp->if_drv_flags & IFF_DRV_RUNNING)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
callout_reset(&sc->sc_brcallout,
bridge_rtable_prune_period * hz, bridge_timer, sc);
}
/*
* bridge_rtage:
*
* Perform an aging cycle.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtage(struct bridge_softc *sc)
{
struct bridge_rtnode *brt, *nbrt;
BRIDGE_LOCK_ASSERT(sc);
Use LIST_FOREACH_SAFE instead of a hand rolled version.
2006-10-09 00:49:57 +00:00
LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if ((brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC) {
It is better to use time_uptime here since it is monotonic. Pointed out by: glebius
2006-01-02 22:23:03 +00:00
if (time_uptime >= brt->brt_expire)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtnode_destroy(sc, brt);
}
}
}
/*
* bridge_rtflush:
*
* Remove all dynamic addresses from the bridge.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtflush(struct bridge_softc *sc, int full)
{
struct bridge_rtnode *brt, *nbrt;
BRIDGE_LOCK_ASSERT(sc);
Use LIST_FOREACH_SAFE instead of a hand rolled version.
2006-10-09 00:49:57 +00:00
LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (full || (brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC)
bridge_rtnode_destroy(sc, brt);
}
}
/*
* bridge_rtdaddr:
*
* Remove an address from the table.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
bridge_rtdaddr(struct bridge_softc *sc, const uint8_t *addr, uint16_t vlan)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
struct bridge_rtnode *brt;
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
int found = 0;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
BRIDGE_LOCK_ASSERT(sc);
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
/*
* If vlan is zero then we want to delete for all vlans so the lookup
* may return more than one.
*/
while ((brt = bridge_rtnode_lookup(sc, addr, vlan)) != NULL) {
bridge_rtnode_destroy(sc, brt);
found = 1;
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
return (found ? 0 : ENOENT);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
/*
* bridge_rtdelete:
*
* Delete routes to a speicifc member interface.
*/
Remove the dependency of bridgestp.h on if_bridgevar.h by moving a couple of private structures to if_bridge.c.
2006-07-27 21:01:48 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtdelete(struct bridge_softc *sc, struct ifnet *ifp, int full)
{
struct bridge_rtnode *brt, *nbrt;
BRIDGE_LOCK_ASSERT(sc);
Use LIST_FOREACH_SAFE instead of a hand rolled version.
2006-10-09 00:49:57 +00:00
LIST_FOREACH_SAFE(brt, &sc->sc_rtlist, brt_list, nbrt) {
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
if (brt->brt_ifp == ifp && (full ||
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
(brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC))
bridge_rtnode_destroy(sc, brt);
}
}
/*
* bridge_rtable_init:
*
* Initialize the route table for this bridge.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtable_init(struct bridge_softc *sc)
{
int i;
sc->sc_rthash = malloc(sizeof(*sc->sc_rthash) * BRIDGE_RTHASH_SIZE,
M_DEVBUF, M_NOWAIT);
if (sc->sc_rthash == NULL)
return (ENOMEM);
for (i = 0; i < BRIDGE_RTHASH_SIZE; i++)
LIST_INIT(&sc->sc_rthash[i]);
sc->sc_rthash_key = arc4random();
LIST_INIT(&sc->sc_rtlist);
return (0);
}
/*
* bridge_rtable_fini:
*
* Deconstruct the route table for this bridge.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtable_fini(struct bridge_softc *sc)
{
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
KASSERT(sc->sc_brtcnt == 0,
("%s: %d bridge routes referenced", __func__, sc->sc_brtcnt));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
free(sc->sc_rthash, M_DEVBUF);
}
/*
* The following hash function is adapted from "Hash Functions" by Bob Jenkins
* ("Algorithm Alley", Dr. Dobbs Journal, September 1997).
*/
#define mix(a, b, c) \
do { \
a -= b; a -= c; a ^= (c >> 13); \
b -= c; b -= a; b ^= (a << 8); \
c -= a; c -= b; c ^= (b >> 13); \
a -= b; a -= c; a ^= (c >> 12); \
b -= c; b -= a; b ^= (a << 16); \
c -= a; c -= b; c ^= (b >> 5); \
a -= b; a -= c; a ^= (c >> 3); \
b -= c; b -= a; b ^= (a << 10); \
c -= a; c -= b; c ^= (b >> 15); \
} while (/*CONSTCOND*/0)
static __inline uint32_t
bridge_rthash(struct bridge_softc *sc, const uint8_t *addr)
{
uint32_t a = 0x9e3779b9, b = 0x9e3779b9, c = sc->sc_rthash_key;
b += addr[5] << 8;
b += addr[4];
a += addr[3] << 24;
a += addr[2] << 16;
a += addr[1] << 8;
a += addr[0];
mix(a, b, c);
return (c & BRIDGE_RTHASH_MASK);
}
#undef mix
Fix two bugs with the bridge - code expects memcmp() to return a signed value, our memcmp() returns 0 if args are equal and > 0 if not. - It's possible to hijack interface for static entry. If bridge recieves packet from interface marked as learning it will replace the bridge_rtnode entry for the source address even if such entry marked as static. Submitted by: Gleb Kurtsov <k-gleb yandex.ru> MFC after: 3 days
2006-01-31 21:21:28 +00:00
static int
bridge_rtnode_addr_cmp(const uint8_t *a, const uint8_t *b)
{
int i, d;
for (i = 0, d = 0; i < ETHER_ADDR_LEN && d == 0; i++) {
d = ((int)a[i]) - ((int)b[i]);
}
return (d);
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* bridge_rtnode_lookup:
*
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
* Look up a bridge route node for the specified destination. Compare the
* vlan id or if zero then just return the first match.
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
*/
Make four more functions static that were missed in the last commit.
2005-10-14 20:57:02 +00:00
static struct bridge_rtnode *
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
bridge_rtnode_lookup(struct bridge_softc *sc, const uint8_t *addr, uint16_t vlan)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
struct bridge_rtnode *brt;
uint32_t hash;
int dir;
BRIDGE_LOCK_ASSERT(sc);
hash = bridge_rthash(sc, addr);
LIST_FOREACH(brt, &sc->sc_rthash[hash], brt_hash) {
Fix two bugs with the bridge - code expects memcmp() to return a signed value, our memcmp() returns 0 if args are equal and > 0 if not. - It's possible to hijack interface for static entry. If bridge recieves packet from interface marked as learning it will replace the bridge_rtnode entry for the source address even if such entry marked as static. Submitted by: Gleb Kurtsov <k-gleb yandex.ru> MFC after: 3 days
2006-01-31 21:21:28 +00:00
dir = bridge_rtnode_addr_cmp(addr, brt->brt_addr);
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
if (dir == 0 && (brt->brt_vlan == vlan || vlan == 0))
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (brt);
if (dir > 0)
return (NULL);
}
return (NULL);
}
/*
* bridge_rtnode_insert:
*
* Insert the specified bridge node into the route table. We
* assume the entry is not already in the table.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static int
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtnode_insert(struct bridge_softc *sc, struct bridge_rtnode *brt)
{
struct bridge_rtnode *lbrt;
uint32_t hash;
int dir;
BRIDGE_LOCK_ASSERT(sc);
hash = bridge_rthash(sc, brt->brt_addr);
lbrt = LIST_FIRST(&sc->sc_rthash[hash]);
if (lbrt == NULL) {
LIST_INSERT_HEAD(&sc->sc_rthash[hash], brt, brt_hash);
goto out;
}
do {
Fix two bugs with the bridge - code expects memcmp() to return a signed value, our memcmp() returns 0 if args are equal and > 0 if not. - It's possible to hijack interface for static entry. If bridge recieves packet from interface marked as learning it will replace the bridge_rtnode entry for the source address even if such entry marked as static. Submitted by: Gleb Kurtsov <k-gleb yandex.ru> MFC after: 3 days
2006-01-31 21:21:28 +00:00
dir = bridge_rtnode_addr_cmp(brt->brt_addr, lbrt->brt_addr);
Add the vlan tag to the bridge route table. This allows a vlan trunk to be bridged, previously legitimate traffic was not passed as the bridge could not tell that it was on a different Ethernet segment. All non-tagged traffic is treated as vlan1 as per IEEE 802.1Q-2003
2007-06-13 18:58:04 +00:00
if (dir == 0 && brt->brt_vlan == lbrt->brt_vlan)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
return (EEXIST);
if (dir > 0) {
LIST_INSERT_BEFORE(lbrt, brt, brt_hash);
goto out;
}
if (LIST_NEXT(lbrt, brt_hash) == NULL) {
LIST_INSERT_AFTER(lbrt, brt, brt_hash);
goto out;
}
lbrt = LIST_NEXT(lbrt, brt_hash);
} while (lbrt != NULL);
#ifdef DIAGNOSTIC
panic("bridge_rtnode_insert: impossible");
#endif
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
out:
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
LIST_INSERT_HEAD(&sc->sc_rtlist, brt, brt_list);
sc->sc_brtcnt++;
return (0);
}
/*
* bridge_rtnode_destroy:
*
* Destroy a bridge rtnode.
*/
Change most of the bridge and stp funtions to static. This has highlighted that the following funtions are not used, wrap in '#ifdef noused' for the moment. bstp_enable_change_detection bstp_disable_change_detection bstp_set_bridge_priority bstp_set_port_priority bstp_set_path_cost
2005-10-14 10:38:12 +00:00
static void
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bridge_rtnode_destroy(struct bridge_softc *sc, struct bridge_rtnode *brt)
{
BRIDGE_LOCK_ASSERT(sc);
LIST_REMOVE(brt, brt_hash);
LIST_REMOVE(brt, brt_list);
sc->sc_brtcnt--;
Add an option to limit the number of source MACs that can be behind a bridge interface. Once the limit is reached packets with unknown source addresses are dropped until an existing host cache entry expires or is removed. Useful to use with the STICKY cache option. Sponsored by: miniSuperHappyDevHouse NZ
2007-11-04 08:32:27 +00:00
brt->brt_dst->bif_addrcnt--;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
uma_zfree(bridge_rtnode_zone, brt);
}
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
/*
* bridge_rtable_expire:
*
* Set the expiry time for all routes on an interface.
*/
static void
bridge_rtable_expire(struct ifnet *ifp, int age)
{
struct bridge_softc *sc = ifp->if_bridge;
struct bridge_rtnode *brt;
BRIDGE_LOCK(sc);
/*
* If the age is zero then flush, otherwise set all the expiry times to
* age for the interface
*/
if (age == 0)
bridge_rtdelete(sc, ifp, IFBF_FLUSHDYN);
else {
LIST_FOREACH(brt, &sc->sc_rtlist, brt_list) {
/* Cap the expiry time to 'age' */
if (brt->brt_ifp == ifp &&
brt->brt_expire > time_uptime + age &&
(brt->brt_flags & IFBAF_TYPEMASK) == IFBAF_DYNAMIC)
brt->brt_expire = time_uptime + age;
}
}
BRIDGE_UNLOCK(sc);
}
- Use the new bridgestp callback to once again flush our bridge routes when an interface is disabled. - Log port changes to syslog, defaulting to off
2006-08-02 03:54:28 +00:00
/*
* bridge_state_change:
*
* Callback from the bridgestp code when a port changes states.
*/
static void
bridge_state_change(struct ifnet *ifp, int state)
{
struct bridge_softc *sc = ifp->if_bridge;
static const char *stpstates[] = {
"disabled",
"listening",
"learning",
"forwarding",
"blocking",
Bring in support for the Rapid Spanning Tree Protocol (802.1w). RSTP provides faster spanning tree convergence, the protocol will exchange information with neighboring switches to quickly transition to forwarding without creating loops. The code will default to RSTP mode but will downgrade any port connected to a legacy STP network so is fully backward compatible. Reviewed by: syrinx Tested by: syrinx
2006-11-01 09:07:47 +00:00
"discarding"
- Use the new bridgestp callback to once again flush our bridge routes when an interface is disabled. - Log port changes to syslog, defaulting to off
2006-08-02 03:54:28 +00:00
};
if (log_stp)
log(LOG_NOTICE, "%s: state changed to %s on %s\n",
sc->sc_ifp->if_xname, stpstates[state], ifp->if_xname);
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* Send bridge packets through pfil if they are one of the types pfil can deal
* with, or if they are ARP or REVARP. (pfil will pass ARP and REVARP without
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
* question.) If *bifp or *ifp are NULL then packet filtering is skipped for
* that interface.
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
*/
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
static int
bridge_pfil(struct mbuf **mp, struct ifnet *bifp, struct ifnet *ifp, int dir)
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
{
Recalculate IP checksum after running pfil hooks. Reviewed by: thompsa Tested by: Adam McDougall <mcdouga9@egr.msu.edu>
2006-05-15 11:49:01 +00:00
int snap, error, i, hlen;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
struct ether_header *eh1, eh2;
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
struct ip_fw_args args;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
struct ip *ip;
Sync if_bridge to NetBSD r1.31 Rename conflicting variables when handling SNAP Ethernet frames. Obtained from: NetBSD Approved by: mlaier (mentor) Approved by: re (blanket)
2005-06-29 19:23:32 +00:00
struct llc llc1;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
u_int16_t ether_type;
snap = 0;
error = -1; /* Default error if not error == 0 */
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
Remove a KASSERT intended to help the developer, the condition is no longer valid since the span code was added. PR: kern/113170 MFC after: 1 week
2007-05-30 19:39:55 +00:00
#if 0
Assert that the mbuf is not shared to ensure problems like the last commit are not reintroduced.
2006-03-26 20:52:47 +00:00
/* we may return with the IP fields swapped, ensure its not shared */
KASSERT(M_WRITABLE(*mp), ("%s: modifying a shared mbuf", __func__));
Remove a KASSERT intended to help the developer, the condition is no longer valid since the span code was added. PR: kern/113170 MFC after: 1 week
2007-05-30 19:39:55 +00:00
#endif
Assert that the mbuf is not shared to ensure problems like the last commit are not reintroduced.
2006-03-26 20:52:47 +00:00
When pfil(9) is enabled the bridge only considers ETHERTYPE_ARP, ETHERTYPE_IP and ETHERTYPE_IPV6 frames. Change this to be a sysctl knob so that is able to still bridge non-IP packets if desired. Also return early if all pfil_* sysctls are turned off, the user obviously does not want to filter on the bridge.
2005-12-29 09:39:15 +00:00
if (pfil_bridge == 0 && pfil_member == 0 && pfil_ipfw == 0)
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (0); /* filtering is disabled */
When pfil(9) is enabled the bridge only considers ETHERTYPE_ARP, ETHERTYPE_IP and ETHERTYPE_IPV6 frames. Change this to be a sysctl knob so that is able to still bridge non-IP packets if desired. Also return early if all pfil_* sysctls are turned off, the user obviously does not want to filter on the bridge.
2005-12-29 09:39:15 +00:00
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
i = min((*mp)->m_pkthdr.len, max_protohdr);
if ((*mp)->m_len < i) {
*mp = m_pullup(*mp, i);
if (*mp == NULL) {
printf("%s: m_pullup failed\n", __func__);
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (-1);
Bring in IPFW layer2 filtering from bridge.c, this allows Ethernet filtering using the layer2, mac and mac-type keywords. This is one of the last features that bridge.c has over if_bridge and gets us very close to a full functional replacement. Approved by: mlaier (mentor)
2005-06-07 21:20:18 +00:00
}
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
eh1 = mtod(*mp, struct ether_header *);
ether_type = ntohs(eh1->ether_type);
/*
* Check for SNAP/LLC.
*/
if (ether_type < ETHERMTU) {
Sync if_bridge to NetBSD r1.31 Rename conflicting variables when handling SNAP Ethernet frames. Obtained from: NetBSD Approved by: mlaier (mentor) Approved by: re (blanket)
2005-06-29 19:23:32 +00:00
struct llc *llc2 = (struct llc *)(eh1 + 1);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if ((*mp)->m_len >= ETHER_HDR_LEN + 8 &&
Sync if_bridge to NetBSD r1.31 Rename conflicting variables when handling SNAP Ethernet frames. Obtained from: NetBSD Approved by: mlaier (mentor) Approved by: re (blanket)
2005-06-29 19:23:32 +00:00
llc2->llc_dsap == LLC_SNAP_LSAP &&
llc2->llc_ssap == LLC_SNAP_LSAP &&
llc2->llc_control == LLC_UI) {
ether_type = htons(llc2->llc_un.type_snap.ether_type);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
snap = 1;
}
}
Revert r1.80 as the ethernet header was inadvertently stripped from ARP packets. Reimplement this correctly and use a sysctl that defaults to off so the user doesnt get any suprises if ipfw blocks the ARP packet. MFC after: 3 days
2006-09-22 21:57:52 +00:00
/*
* If we're trying to filter bridge traffic, don't look at anything
* other than IP and ARP traffic. If the filter doesn't understand
* IPv6, don't allow IPv6 through the bridge either. This is lame
* since if we really wanted, say, an AppleTalk filter, we are hosed,
* but of course we don't have an AppleTalk filter to begin with.
* (Note that since pfil doesn't understand ARP it will pass *ALL*
* ARP traffic.)
*/
switch (ether_type) {
case ETHERTYPE_ARP:
case ETHERTYPE_REVARP:
if (pfil_ipfw_arp == 0)
return (0); /* Automatically pass */
break;
case ETHERTYPE_IP:
#ifdef INET6
case ETHERTYPE_IPV6:
#endif /* INET6 */
break;
default:
/*
* Check to see if the user wants to pass non-ip
* packets, these will not be checked by pfil(9) and
* passed unconditionally so the default is to drop.
*/
if (pfil_onlyip)
goto bad;
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/* Strip off the Ethernet header and keep a copy. */
m_copydata(*mp, 0, ETHER_HDR_LEN, (caddr_t) &eh2);
m_adj(*mp, ETHER_HDR_LEN);
/* Strip off snap header, if present */
if (snap) {
Sync if_bridge to NetBSD r1.31 Rename conflicting variables when handling SNAP Ethernet frames. Obtained from: NetBSD Approved by: mlaier (mentor) Approved by: re (blanket)
2005-06-29 19:23:32 +00:00
m_copydata(*mp, 0, sizeof(struct llc), (caddr_t) &llc1);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
m_adj(*mp, sizeof(struct llc));
}
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
/*
* Check the IP header for alignment and errors
*/
if (dir == PFIL_IN) {
switch (ether_type) {
case ETHERTYPE_IP:
error = bridge_ip_checkbasic(mp);
break;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
#ifdef INET6
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
case ETHERTYPE_IPV6:
error = bridge_ip6_checkbasic(mp);
break;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
#endif /* INET6 */
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
default:
error = 0;
}
if (error)
goto bad;
}
merge code from ipfw3-head to reduce contention on the ipfw lock and remove all O(N) sequences from kernel critical sections in ipfw. In detail: 1. introduce a IPFW_UH_LOCK to arbitrate requests from the upper half of the kernel. Some things, such as 'ipfw show', can be done holding this lock in read mode, whereas insert and delete require IPFW_UH_WLOCK. 2. introduce a mapping structure to keep rules together. This replaces the 'next' chain currently used in ipfw rules. At the moment the map is a simple array (sorted by rule number and then rule_id), so we can find a rule quickly instead of having to scan the list. This reduces many expensive lookups from O(N) to O(log N). 3. when an expensive operation (such as insert or delete) is done by userland, we grab IPFW_UH_WLOCK, create a new copy of the map without blocking the bottom half of the kernel, then acquire IPFW_WLOCK and quickly update pointers to the map and related info. After dropping IPFW_LOCK we can then continue the cleanup protected by IPFW_UH_LOCK. So userland still costs O(N) but the kernel side is only blocked for O(1). 4. do not pass pointers to rules through dummynet, netgraph, divert etc, but rather pass a <slot, chain_id, rulenum, rule_id> tuple. We validate the slot index (in the array of #2) with chain_id, and if successful do a O(1) dereference; otherwise, we can find the rule in O(log N) through <rulenum, rule_id> All the above does not change the userland/kernel ABI, though there are some disgusting casts between pointers and uint32_t Operation costs now are as follows: Function Old Now Planned ------------------------------------------------------------------- + skipto X, non cached O(N) O(log N) + skipto X, cached O(1) O(1) XXX dynamic rule lookup O(1) O(log N) O(1) + skipto tablearg O(N) O(1) + reinject, non cached O(N) O(log N) + reinject, cached O(1) O(1) + kernel blocked during setsockopt() O(N) O(1) ------------------------------------------------------------------- The only (very small) regression is on dynamic rule lookup and this will be fixed in a day or two, without changing the userland/kernel ABI Supported by: Valeria Paoli MFC after: 1 month
2009-12-22 19:01:47 +00:00
/* XXX this section is also in if_ethersubr.c */
Various cleanup done in ipfw3-head branch including: - use a uniform mtag format for all packets that exit and re-enter the firewall in the middle of a rulechain. On reentry, all tags containing reinject info are renamed to MTAG_IPFW_RULE so the processing is simpler. - make ipfw and dummynet use ip_len and ip_off in network format everywhere. Conversion is done only once instead of tracking the format in every place. - use a macro FREE_PKT to dispose of mbufs. This eases portability. On passing i also removed a few typos, staticise or localise variables, remove useless declarations and other minor things. Overall the code shrinks a bit and is hopefully more readable. I have tested functionality for all but ng_ipfw and if_bridge/if_ethersubr. For ng_ipfw i am actually waiting for feedback from glebius@ because we might have some small changes to make. For if_bridge and if_ethersubr feedback would be welcome (there are still some redundant parts in these two modules that I would like to remove, but first i need to check functionality).
2010-01-04 19:01:22 +00:00
// XXX PFIL_OUT or DIR_OUT ?
if (V_ip_fw_chk_ptr && pfil_ipfw != 0 &&
dir == PFIL_OUT && ifp != NULL) {
bring in several cleanups tested in ipfw3-head branch, namely: r201011 - move most of ng_ipfw.h into ip_fw_private.h, as this code is ipfw-specific. This removes a dependency on ng_ipfw.h from some files. - move many equivalent definitions of direction (IN, OUT) for reinjected packets into ip_fw_private.h - document the structure of the packet tags used for dummynet and netgraph; r201049 - merge some common code to attach/detach hooks into a single function. r201055 - remove some duplicated code in ip_fw_pfil. The input and output processing uses almost exactly the same code so there is no need to use two separate hooks. ip_fw_pfil.o goes from 2096 to 1382 bytes of .text r201057 (see the svn log for full details) - macros to make the conversion of ip_len and ip_off between host and network format more explicit r201113 (the remaining parts) - readability fixes -- put braces around some large for() blocks, localize variables so the compiler does not think they are uninitialized, do not insist on precise allocation size if we have more than we need. r201119 - when doing a lookup, keys must be in big endian format because this is what the radix code expects (this fixes a bug in the recently-introduced 'lookup' option) No ABI changes in this commit. MFC after: 1 week
2009-12-28 10:47:04 +00:00
struct m_tag *mtag;
Step 1.5 of importing the network stack virtualization infrastructure from the vimage project, as per plan established at devsummit 08/08: http://wiki.freebsd.org/Image/Notes200808DevSummit Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator macros, and CURVNET_SET() context setting macros, all currently resolving to NOPs. Prepare for virtualization of selected SYSCTL objects by introducing a family of SYSCTL_V_*() macros, currently resolving to their global counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT(). Move selected #defines from sys/sys/vimage.h to newly introduced header files specific to virtualized subsystems (sys/net/vnet.h, sys/netinet/vinet.h etc.). All the changes are verified to have zero functional impact at this point in time by doing MD5 comparision between pre- and post-change object files(*). (*) netipsec/keysock.c did not validate depending on compile time options. Implemented by: julian, bz, brooks, zec Reviewed by: julian, bz, brooks, kris, rwatson, ... Approved by: julian (mentor) Obtained from: //depot/projects/vimage-commit2/... X-MFC after: never Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
error = -1;
Various cleanup done in ipfw3-head branch including: - use a uniform mtag format for all packets that exit and re-enter the firewall in the middle of a rulechain. On reentry, all tags containing reinject info are renamed to MTAG_IPFW_RULE so the processing is simpler. - make ipfw and dummynet use ip_len and ip_off in network format everywhere. Conversion is done only once instead of tracking the format in every place. - use a macro FREE_PKT to dispose of mbufs. This eases portability. On passing i also removed a few typos, staticise or localise variables, remove useless declarations and other minor things. Overall the code shrinks a bit and is hopefully more readable. I have tested functionality for all but ng_ipfw and if_bridge/if_ethersubr. For ng_ipfw i am actually waiting for feedback from glebius@ because we might have some small changes to make. For if_bridge and if_ethersubr feedback would be welcome (there are still some redundant parts in these two modules that I would like to remove, but first i need to check functionality).
2010-01-04 19:01:22 +00:00
/* fetch the start point from existing tags, if any */
mtag = m_tag_locate(*mp, MTAG_IPFW_RULE, 0, NULL);
bring in several cleanups tested in ipfw3-head branch, namely: r201011 - move most of ng_ipfw.h into ip_fw_private.h, as this code is ipfw-specific. This removes a dependency on ng_ipfw.h from some files. - move many equivalent definitions of direction (IN, OUT) for reinjected packets into ip_fw_private.h - document the structure of the packet tags used for dummynet and netgraph; r201049 - merge some common code to attach/detach hooks into a single function. r201055 - remove some duplicated code in ip_fw_pfil. The input and output processing uses almost exactly the same code so there is no need to use two separate hooks. ip_fw_pfil.o goes from 2096 to 1382 bytes of .text r201057 (see the svn log for full details) - macros to make the conversion of ip_len and ip_off between host and network format more explicit r201113 (the remaining parts) - readability fixes -- put braces around some large for() blocks, localize variables so the compiler does not think they are uninitialized, do not insist on precise allocation size if we have more than we need. r201119 - when doing a lookup, keys must be in big endian format because this is what the radix code expects (this fixes a bug in the recently-introduced 'lookup' option) No ABI changes in this commit. MFC after: 1 week
2009-12-28 10:47:04 +00:00
if (mtag == NULL) {
Various cleanup done in ipfw3-head branch including: - use a uniform mtag format for all packets that exit and re-enter the firewall in the middle of a rulechain. On reentry, all tags containing reinject info are renamed to MTAG_IPFW_RULE so the processing is simpler. - make ipfw and dummynet use ip_len and ip_off in network format everywhere. Conversion is done only once instead of tracking the format in every place. - use a macro FREE_PKT to dispose of mbufs. This eases portability. On passing i also removed a few typos, staticise or localise variables, remove useless declarations and other minor things. Overall the code shrinks a bit and is hopefully more readable. I have tested functionality for all but ng_ipfw and if_bridge/if_ethersubr. For ng_ipfw i am actually waiting for feedback from glebius@ because we might have some small changes to make. For if_bridge and if_ethersubr feedback would be welcome (there are still some redundant parts in these two modules that I would like to remove, but first i need to check functionality).
2010-01-04 19:01:22 +00:00
args.rule.slot = 0;
merge code from ipfw3-head to reduce contention on the ipfw lock and remove all O(N) sequences from kernel critical sections in ipfw. In detail: 1. introduce a IPFW_UH_LOCK to arbitrate requests from the upper half of the kernel. Some things, such as 'ipfw show', can be done holding this lock in read mode, whereas insert and delete require IPFW_UH_WLOCK. 2. introduce a mapping structure to keep rules together. This replaces the 'next' chain currently used in ipfw rules. At the moment the map is a simple array (sorted by rule number and then rule_id), so we can find a rule quickly instead of having to scan the list. This reduces many expensive lookups from O(N) to O(log N). 3. when an expensive operation (such as insert or delete) is done by userland, we grab IPFW_UH_WLOCK, create a new copy of the map without blocking the bottom half of the kernel, then acquire IPFW_WLOCK and quickly update pointers to the map and related info. After dropping IPFW_LOCK we can then continue the cleanup protected by IPFW_UH_LOCK. So userland still costs O(N) but the kernel side is only blocked for O(1). 4. do not pass pointers to rules through dummynet, netgraph, divert etc, but rather pass a <slot, chain_id, rulenum, rule_id> tuple. We validate the slot index (in the array of #2) with chain_id, and if successful do a O(1) dereference; otherwise, we can find the rule in O(log N) through <rulenum, rule_id> All the above does not change the userland/kernel ABI, though there are some disgusting casts between pointers and uint32_t Operation costs now are as follows: Function Old Now Planned ------------------------------------------------------------------- + skipto X, non cached O(N) O(log N) + skipto X, cached O(1) O(1) XXX dynamic rule lookup O(1) O(log N) O(1) + skipto tablearg O(N) O(1) + reinject, non cached O(N) O(log N) + reinject, cached O(1) O(1) + kernel blocked during setsockopt() O(N) O(1) ------------------------------------------------------------------- The only (very small) regression is on dynamic rule lookup and this will be fixed in a day or two, without changing the userland/kernel ABI Supported by: Valeria Paoli MFC after: 1 month
2009-12-22 19:01:47 +00:00
} else {
Bring in the most recent version of ipfw and dummynet, developed and tested over the past two months in the ipfw3-head branch. This also happens to be the same code available in the Linux and Windows ports of ipfw and dummynet. The major enhancement is a completely restructured version of dummynet, with support for different packet scheduling algorithms (loadable at runtime), faster queue/pipe lookup, and a much cleaner internal architecture and kernel/userland ABI which simplifies future extensions. In addition to the existing schedulers (FIFO and WF2Q+), we include a Deficit Round Robin (DRR or RR for brevity) scheduler, and a new, very fast version of WF2Q+ called QFQ. Some test code is also present (in sys/netinet/ipfw/test) that lets you build and test schedulers in userland. Also, we have added a compatibility layer that understands requests from the RELENG_7 and RELENG_8 versions of the /sbin/ipfw binaries, and replies correctly (at least, it does its best; sometimes you just cannot tell who sent the request and how to answer). The compatibility layer should make it possible to MFC this code in a relatively short time. Some minor glitches (e.g. handling of ipfw set enable/disable, and a workaround for a bug in RELENG_7's /sbin/ipfw) will be fixed with separate commits. CREDITS: This work has been partly supported by the ONELAB2 project, and mostly developed by Riccardo Panicucci and myself. The code for the qfq scheduler is mostly from Fabio Checconi, and Marta Carbone and Francesco Magno have helped with testing, debugging and some bug fixes.
2010-03-02 17:40:48 +00:00
struct ipfw_rule_ref *r;
bring in several cleanups tested in ipfw3-head branch, namely: r201011 - move most of ng_ipfw.h into ip_fw_private.h, as this code is ipfw-specific. This removes a dependency on ng_ipfw.h from some files. - move many equivalent definitions of direction (IN, OUT) for reinjected packets into ip_fw_private.h - document the structure of the packet tags used for dummynet and netgraph; r201049 - merge some common code to attach/detach hooks into a single function. r201055 - remove some duplicated code in ip_fw_pfil. The input and output processing uses almost exactly the same code so there is no need to use two separate hooks. ip_fw_pfil.o goes from 2096 to 1382 bytes of .text r201057 (see the svn log for full details) - macros to make the conversion of ip_len and ip_off between host and network format more explicit r201113 (the remaining parts) - readability fixes -- put braces around some large for() blocks, localize variables so the compiler does not think they are uninitialized, do not insist on precise allocation size if we have more than we need. r201119 - when doing a lookup, keys must be in big endian format because this is what the radix code expects (this fixes a bug in the recently-introduced 'lookup' option) No ABI changes in this commit. MFC after: 1 week
2009-12-28 10:47:04 +00:00
Various cleanup done in ipfw3-head branch including: - use a uniform mtag format for all packets that exit and re-enter the firewall in the middle of a rulechain. On reentry, all tags containing reinject info are renamed to MTAG_IPFW_RULE so the processing is simpler. - make ipfw and dummynet use ip_len and ip_off in network format everywhere. Conversion is done only once instead of tracking the format in every place. - use a macro FREE_PKT to dispose of mbufs. This eases portability. On passing i also removed a few typos, staticise or localise variables, remove useless declarations and other minor things. Overall the code shrinks a bit and is hopefully more readable. I have tested functionality for all but ng_ipfw and if_bridge/if_ethersubr. For ng_ipfw i am actually waiting for feedback from glebius@ because we might have some small changes to make. For if_bridge and if_ethersubr feedback would be welcome (there are still some redundant parts in these two modules that I would like to remove, but first i need to check functionality).
2010-01-04 19:01:22 +00:00
/* XXX can we free the tag after use ? */
bring in several cleanups tested in ipfw3-head branch, namely: r201011 - move most of ng_ipfw.h into ip_fw_private.h, as this code is ipfw-specific. This removes a dependency on ng_ipfw.h from some files. - move many equivalent definitions of direction (IN, OUT) for reinjected packets into ip_fw_private.h - document the structure of the packet tags used for dummynet and netgraph; r201049 - merge some common code to attach/detach hooks into a single function. r201055 - remove some duplicated code in ip_fw_pfil. The input and output processing uses almost exactly the same code so there is no need to use two separate hooks. ip_fw_pfil.o goes from 2096 to 1382 bytes of .text r201057 (see the svn log for full details) - macros to make the conversion of ip_len and ip_off between host and network format more explicit r201113 (the remaining parts) - readability fixes -- put braces around some large for() blocks, localize variables so the compiler does not think they are uninitialized, do not insist on precise allocation size if we have more than we need. r201119 - when doing a lookup, keys must be in big endian format because this is what the radix code expects (this fixes a bug in the recently-introduced 'lookup' option) No ABI changes in this commit. MFC after: 1 week
2009-12-28 10:47:04 +00:00
mtag->m_tag_id = PACKET_TAG_NONE;
Bring in the most recent version of ipfw and dummynet, developed and tested over the past two months in the ipfw3-head branch. This also happens to be the same code available in the Linux and Windows ports of ipfw and dummynet. The major enhancement is a completely restructured version of dummynet, with support for different packet scheduling algorithms (loadable at runtime), faster queue/pipe lookup, and a much cleaner internal architecture and kernel/userland ABI which simplifies future extensions. In addition to the existing schedulers (FIFO and WF2Q+), we include a Deficit Round Robin (DRR or RR for brevity) scheduler, and a new, very fast version of WF2Q+ called QFQ. Some test code is also present (in sys/netinet/ipfw/test) that lets you build and test schedulers in userland. Also, we have added a compatibility layer that understands requests from the RELENG_7 and RELENG_8 versions of the /sbin/ipfw binaries, and replies correctly (at least, it does its best; sometimes you just cannot tell who sent the request and how to answer). The compatibility layer should make it possible to MFC this code in a relatively short time. Some minor glitches (e.g. handling of ipfw set enable/disable, and a workaround for a bug in RELENG_7's /sbin/ipfw) will be fixed with separate commits. CREDITS: This work has been partly supported by the ONELAB2 project, and mostly developed by Riccardo Panicucci and myself. The code for the qfq scheduler is mostly from Fabio Checconi, and Marta Carbone and Francesco Magno have helped with testing, debugging and some bug fixes.
2010-03-02 17:40:48 +00:00
r = (struct ipfw_rule_ref *)(mtag + 1);
Various cleanup done in ipfw3-head branch including: - use a uniform mtag format for all packets that exit and re-enter the firewall in the middle of a rulechain. On reentry, all tags containing reinject info are renamed to MTAG_IPFW_RULE so the processing is simpler. - make ipfw and dummynet use ip_len and ip_off in network format everywhere. Conversion is done only once instead of tracking the format in every place. - use a macro FREE_PKT to dispose of mbufs. This eases portability. On passing i also removed a few typos, staticise or localise variables, remove useless declarations and other minor things. Overall the code shrinks a bit and is hopefully more readable. I have tested functionality for all but ng_ipfw and if_bridge/if_ethersubr. For ng_ipfw i am actually waiting for feedback from glebius@ because we might have some small changes to make. For if_bridge and if_ethersubr feedback would be welcome (there are still some redundant parts in these two modules that I would like to remove, but first i need to check functionality).
2010-01-04 19:01:22 +00:00
/* packet already partially processed ? */
Bring in the most recent version of ipfw and dummynet, developed and tested over the past two months in the ipfw3-head branch. This also happens to be the same code available in the Linux and Windows ports of ipfw and dummynet. The major enhancement is a completely restructured version of dummynet, with support for different packet scheduling algorithms (loadable at runtime), faster queue/pipe lookup, and a much cleaner internal architecture and kernel/userland ABI which simplifies future extensions. In addition to the existing schedulers (FIFO and WF2Q+), we include a Deficit Round Robin (DRR or RR for brevity) scheduler, and a new, very fast version of WF2Q+ called QFQ. Some test code is also present (in sys/netinet/ipfw/test) that lets you build and test schedulers in userland. Also, we have added a compatibility layer that understands requests from the RELENG_7 and RELENG_8 versions of the /sbin/ipfw binaries, and replies correctly (at least, it does its best; sometimes you just cannot tell who sent the request and how to answer). The compatibility layer should make it possible to MFC this code in a relatively short time. Some minor glitches (e.g. handling of ipfw set enable/disable, and a workaround for a bug in RELENG_7's /sbin/ipfw) will be fixed with separate commits. CREDITS: This work has been partly supported by the ONELAB2 project, and mostly developed by Riccardo Panicucci and myself. The code for the qfq scheduler is mostly from Fabio Checconi, and Marta Carbone and Francesco Magno have helped with testing, debugging and some bug fixes.
2010-03-02 17:40:48 +00:00
if (r->info & IPFW_ONEPASS)
Close long existed race with net.inet.ip.fw.one_pass = 0: If packet leaves ipfw to other kernel subsystem (dummynet, netgraph, etc) it carries pointer to matching ipfw rule. If this packet then reinjected back to ipfw, ruleset processing starts from that rule. If rule was deleted meanwhile, due to existed race condition panic was possible (as well as other odd effects like parsing rules in 'reap list'). P.S. this commit changes ABI so userland ipfw related binaries should be recompiled. MFC after: 1 month Tested by: Mikolaj Golub
2009-06-09 21:27:11 +00:00
goto ipfwpass;
Bring in the most recent version of ipfw and dummynet, developed and tested over the past two months in the ipfw3-head branch. This also happens to be the same code available in the Linux and Windows ports of ipfw and dummynet. The major enhancement is a completely restructured version of dummynet, with support for different packet scheduling algorithms (loadable at runtime), faster queue/pipe lookup, and a much cleaner internal architecture and kernel/userland ABI which simplifies future extensions. In addition to the existing schedulers (FIFO and WF2Q+), we include a Deficit Round Robin (DRR or RR for brevity) scheduler, and a new, very fast version of WF2Q+ called QFQ. Some test code is also present (in sys/netinet/ipfw/test) that lets you build and test schedulers in userland. Also, we have added a compatibility layer that understands requests from the RELENG_7 and RELENG_8 versions of the /sbin/ipfw binaries, and replies correctly (at least, it does its best; sometimes you just cannot tell who sent the request and how to answer). The compatibility layer should make it possible to MFC this code in a relatively short time. Some minor glitches (e.g. handling of ipfw set enable/disable, and a workaround for a bug in RELENG_7's /sbin/ipfw) will be fixed with separate commits. CREDITS: This work has been partly supported by the ONELAB2 project, and mostly developed by Riccardo Panicucci and myself. The code for the qfq scheduler is mostly from Fabio Checconi, and Marta Carbone and Francesco Magno have helped with testing, debugging and some bug fixes.
2010-03-02 17:40:48 +00:00
args.rule = *r;
merge code from ipfw3-head to reduce contention on the ipfw lock and remove all O(N) sequences from kernel critical sections in ipfw. In detail: 1. introduce a IPFW_UH_LOCK to arbitrate requests from the upper half of the kernel. Some things, such as 'ipfw show', can be done holding this lock in read mode, whereas insert and delete require IPFW_UH_WLOCK. 2. introduce a mapping structure to keep rules together. This replaces the 'next' chain currently used in ipfw rules. At the moment the map is a simple array (sorted by rule number and then rule_id), so we can find a rule quickly instead of having to scan the list. This reduces many expensive lookups from O(N) to O(log N). 3. when an expensive operation (such as insert or delete) is done by userland, we grab IPFW_UH_WLOCK, create a new copy of the map without blocking the bottom half of the kernel, then acquire IPFW_WLOCK and quickly update pointers to the map and related info. After dropping IPFW_LOCK we can then continue the cleanup protected by IPFW_UH_LOCK. So userland still costs O(N) but the kernel side is only blocked for O(1). 4. do not pass pointers to rules through dummynet, netgraph, divert etc, but rather pass a <slot, chain_id, rulenum, rule_id> tuple. We validate the slot index (in the array of #2) with chain_id, and if successful do a O(1) dereference; otherwise, we can find the rule in O(log N) through <rulenum, rule_id> All the above does not change the userland/kernel ABI, though there are some disgusting casts between pointers and uint32_t Operation costs now are as follows: Function Old Now Planned ------------------------------------------------------------------- + skipto X, non cached O(N) O(log N) + skipto X, cached O(1) O(1) XXX dynamic rule lookup O(1) O(log N) O(1) + skipto tablearg O(N) O(1) + reinject, non cached O(N) O(log N) + reinject, cached O(1) O(1) + kernel blocked during setsockopt() O(N) O(1) ------------------------------------------------------------------- The only (very small) regression is on dynamic rule lookup and this will be fixed in a day or two, without changing the userland/kernel ABI Supported by: Valeria Paoli MFC after: 1 month
2009-12-22 19:01:47 +00:00
}
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
args.m = *mp;
args.oif = ifp;
args.next_hop = NULL;
args.eh = &eh2;
Properly initialize args structure before passing it to ipfw_chk(): having uninitialized args.inp is unhealthy for uid/gid/jail ipfw rules. PR: kern/92589 Approved by: glebius (mentor) MFC after: 1 week
2006-02-03 23:03:07 +00:00
args.inp = NULL; /* used by ipfw uid/gid/jail rules */
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
i = V_ip_fw_chk_ptr(&args);
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
*mp = args.m;
if (*mp == NULL)
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (error);
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
More cleanup in preparation of ipfw relocation (no actual code change): + move ipfw and dummynet hooks declarations to raw_ip.c (definitions in ip_var.h) same as for most other global variables. This removes some dependencies from ip_input.c; + remove the IPFW_LOADED macro, just test ip_fw_chk_ptr directly; + remove the DUMMYNET_LOADED macro, just test ip_dn_io_ptr directly; + move ip_dn_ruledel_ptr to ip_fw2.c which is the only file using it; To be merged together with rev 193497 MFC after: 5 days
2009-06-05 13:44:30 +00:00
if (ip_dn_io_ptr && (i == IP_FW_DUMMYNET)) {
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
/* put the Ethernet header back on */
M_PREPEND(*mp, ETHER_HDR_LEN, M_DONTWAIT);
if (*mp == NULL)
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (error);
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
bcopy(&eh2, mtod(*mp, caddr_t), ETHER_HDR_LEN);
/*
* Pass the pkt to dummynet, which consumes it. The
* packet will return to us via bridge_dummynet().
*/
args.oif = ifp;
bring in several cleanups tested in ipfw3-head branch, namely: r201011 - move most of ng_ipfw.h into ip_fw_private.h, as this code is ipfw-specific. This removes a dependency on ng_ipfw.h from some files. - move many equivalent definitions of direction (IN, OUT) for reinjected packets into ip_fw_private.h - document the structure of the packet tags used for dummynet and netgraph; r201049 - merge some common code to attach/detach hooks into a single function. r201055 - remove some duplicated code in ip_fw_pfil. The input and output processing uses almost exactly the same code so there is no need to use two separate hooks. ip_fw_pfil.o goes from 2096 to 1382 bytes of .text r201057 (see the svn log for full details) - macros to make the conversion of ip_len and ip_off between host and network format more explicit r201113 (the remaining parts) - readability fixes -- put braces around some large for() blocks, localize variables so the compiler does not think they are uninitialized, do not insist on precise allocation size if we have more than we need. r201119 - when doing a lookup, keys must be in big endian format because this is what the radix code expects (this fixes a bug in the recently-introduced 'lookup' option) No ABI changes in this commit. MFC after: 1 week
2009-12-28 10:47:04 +00:00
ip_dn_io_ptr(mp, DIR_FWD | PROTO_IFB, &args);
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (error);
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
}
if (i != IP_FW_PASS) /* drop */
goto bad;
}
ipfwpass:
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
error = 0;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
Revert r1.80 as the ethernet header was inadvertently stripped from ARP packets. Reimplement this correctly and use a sysctl that defaults to off so the user doesnt get any suprises if ipfw blocks the ARP packet. MFC after: 3 days
2006-09-22 21:57:52 +00:00
* Run the packet through pfil
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
*/
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
switch (ether_type) {
case ETHERTYPE_IP:
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* before calling the firewall, swap fields the same as
* IP does. here we assume the header is contiguous
*/
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
ip = mtod(*mp, struct ip *);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
ip->ip_len = ntohs(ip->ip_len);
ip->ip_off = ntohs(ip->ip_off);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* Run pfil on the member interface and the bridge, both can
* be skipped by clearing pfil_member or pfil_bridge.
*
* Keep the order:
* in_if -> bridge_if -> out_if
*/
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
if (pfil_bridge && dir == PFIL_OUT && bifp != NULL)
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
error = pfil_run_hooks(&V_inet_pfil_hook, mp, bifp,
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
dir, NULL);
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
if (*mp == NULL || error != 0) /* filter may consume */
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
break;
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
if (pfil_member && ifp != NULL)
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
error = pfil_run_hooks(&V_inet_pfil_hook, mp, ifp,
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
dir, NULL);
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
if (*mp == NULL || error != 0) /* filter may consume */
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
break;
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
if (pfil_bridge && dir == PFIL_IN && bifp != NULL)
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
error = pfil_run_hooks(&V_inet_pfil_hook, mp, bifp,
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
dir, NULL);
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
if (*mp == NULL || error != 0) /* filter may consume */
break;
/* check if we need to fragment the packet */
if (pfil_member && ifp != NULL && dir == PFIL_OUT) {
i = (*mp)->m_pkthdr.len;
if (i > ifp->if_mtu) {
error = bridge_fragment(ifp, *mp, &eh2, snap,
&llc1);
return (error);
}
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Recalculate IP checksum after running pfil hooks. Reviewed by: thompsa Tested by: Adam McDougall <mcdouga9@egr.msu.edu>
2006-05-15 11:49:01 +00:00
/* Recalculate the ip checksum and restore byte ordering */
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
ip = mtod(*mp, struct ip *);
Recalculate IP checksum after running pfil hooks. Reviewed by: thompsa Tested by: Adam McDougall <mcdouga9@egr.msu.edu>
2006-05-15 11:49:01 +00:00
hlen = ip->ip_hl << 2;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
if (hlen < sizeof(struct ip))
goto bad;
if (hlen > (*mp)->m_len) {
if ((*mp = m_pullup(*mp, hlen)) == 0)
goto bad;
ip = mtod(*mp, struct ip *);
if (ip == NULL)
goto bad;
}
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
ip->ip_len = htons(ip->ip_len);
ip->ip_off = htons(ip->ip_off);
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
ip->ip_sum = 0;
if (hlen == sizeof(struct ip))
ip->ip_sum = in_cksum_hdr(ip);
else
ip->ip_sum = in_cksum(*mp, hlen);
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
break;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
#ifdef INET6
case ETHERTYPE_IPV6:
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
if (pfil_bridge && dir == PFIL_OUT && bifp != NULL)
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
error = pfil_run_hooks(&V_inet6_pfil_hook, mp, bifp,
Change ipv6 packet filtering to match ipv4. It now checks pfil_member and pfil_bridge to determine which interfaces to filter on. Approved by: mlaier (mentor)
2005-06-06 02:41:29 +00:00
dir, NULL);
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
if (*mp == NULL || error != 0) /* filter may consume */
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
break;
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
if (pfil_member && ifp != NULL)
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
error = pfil_run_hooks(&V_inet6_pfil_hook, mp, ifp,
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
dir, NULL);
Change ipv6 packet filtering to match ipv4. It now checks pfil_member and pfil_bridge to determine which interfaces to filter on. Approved by: mlaier (mentor)
2005-06-06 02:41:29 +00:00
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
if (*mp == NULL || error != 0) /* filter may consume */
Add dummynet(4) support to if_bridge, this code is largely based on bridge.c. This is the final piece to match bridge.c in functionality, we can now be a drop-in replacement. Approved by: mlaier (mentor)
2005-06-10 01:25:22 +00:00
break;
- Previously when broadcasting to N number of interfaces we would run pfil hooks for each outgoing interface but also run pfil hooks _N times_ on the bridge interface. This is changed so pfil hooks are run once for the bridge interface (bridge0) and then only on the outgoing interfaces in the broadcast loop. - Simplify bridge_enqueue() by moving bridge_pfil() to the callers. - Check (inet6_pfil_hook.ph_busy_count >= 0), it may be possible to have a packet filter hooked for only ipv6 but we were only checking if ipv4 hooks were busy. - Minor optimisation for null mbuf check after bridge_pfil(), move it into the if-block as it couldnt possibly be null outside. Prodded by: mlaier Approved by: re (scottl), mlaier (mentor)
2005-07-06 01:24:45 +00:00
if (pfil_bridge && dir == PFIL_IN && bifp != NULL)
Virtualize the pfil hooks so that different jails may chose different packet filters. ALso allows ipfw to be enabled on on ejail and disabled on another. In 8.0 it's a global setting. Sitting aroung in tree waiting to commit for: 2 months MFC after: 2 months
2009-10-11 05:59:43 +00:00
error = pfil_run_hooks(&V_inet6_pfil_hook, mp, bifp,
Change ipv6 packet filtering to match ipv4. It now checks pfil_member and pfil_bridge to determine which interfaces to filter on. Approved by: mlaier (mentor)
2005-06-06 02:41:29 +00:00
dir, NULL);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
break;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
#endif
default:
Revert r1.80 as the ethernet header was inadvertently stripped from ARP packets. Reimplement this correctly and use a sysctl that defaults to off so the user doesnt get any suprises if ipfw blocks the ARP packet. MFC after: 3 days
2006-09-22 21:57:52 +00:00
error = 0;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
break;
}
if (*mp == NULL)
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (error);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (error != 0)
goto bad;
error = -1;
/*
* Finally, put everything back the way it was and return
*/
if (snap) {
M_PREPEND(*mp, sizeof(struct llc), M_DONTWAIT);
if (*mp == NULL)
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (error);
Sync if_bridge to NetBSD r1.31 Rename conflicting variables when handling SNAP Ethernet frames. Obtained from: NetBSD Approved by: mlaier (mentor) Approved by: re (blanket)
2005-06-29 19:23:32 +00:00
bcopy(&llc1, mtod(*mp, caddr_t), sizeof(struct llc));
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
M_PREPEND(*mp, ETHER_HDR_LEN, M_DONTWAIT);
if (*mp == NULL)
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (error);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
bcopy(&eh2, mtod(*mp, caddr_t), ETHER_HDR_LEN);
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (0);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
bad:
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
m_freem(*mp);
*mp = NULL;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (error);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
/*
* Perform basic checks on header size since
* pfil assumes ip_input has already processed
* it for it. Cut-and-pasted from ip_input.c.
* Given how simple the IPv6 version is,
* does the IPv4 version really need to be
* this complicated?
*
* XXX Should we update ipstat here, or not?
* XXX Right now we update ipstat but not
* XXX csum_counter.
*/
static int
bridge_ip_checkbasic(struct mbuf **mp)
{
struct mbuf *m = *mp;
struct ip *ip;
int len, hlen;
u_short sum;
if (*mp == NULL)
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (-1);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
if (IP_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) {
if ((m = m_copyup(m, sizeof(struct ip),
(max_linkhdr + 3) & ~3)) == NULL) {
/* XXXJRT new stat, please */
Many network stack subsystems use a single global data structure to hold all pertinent statatistics for the subsystem. These structures are sometimes "borrowed" by kernel modules that require a place to store statistics for similar events. Add KPI accessor functions for statistics structures referenced by kernel modules so that they no longer encode certain specifics of how the data structures are named and stored. This change is intended to make it easier to move to per-CPU network stats following 8.0-RELEASE. The following modules are affected by this change: if_bridge if_cxgb if_gif ip_mroute ipdivert pf In practice, most of these statistics consumers should, in fact, maintain their own statistics data structures rather than borrowing structures from the base network stack. However, that change is too agressive for this point in the release cycle. Reviewed by: bz Approved by: re (kib)
2009-08-02 19:43:32 +00:00
KMOD_IPSTAT_INC(ips_toosmall);
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
goto bad;
}
} else if (__predict_false(m->m_len < sizeof (struct ip))) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if ((m = m_pullup(m, sizeof (struct ip))) == NULL) {
Many network stack subsystems use a single global data structure to hold all pertinent statatistics for the subsystem. These structures are sometimes "borrowed" by kernel modules that require a place to store statistics for similar events. Add KPI accessor functions for statistics structures referenced by kernel modules so that they no longer encode certain specifics of how the data structures are named and stored. This change is intended to make it easier to move to per-CPU network stats following 8.0-RELEASE. The following modules are affected by this change: if_bridge if_cxgb if_gif ip_mroute ipdivert pf In practice, most of these statistics consumers should, in fact, maintain their own statistics data structures rather than borrowing structures from the base network stack. However, that change is too agressive for this point in the release cycle. Reviewed by: bz Approved by: re (kib)
2009-08-02 19:43:32 +00:00
KMOD_IPSTAT_INC(ips_toosmall);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
goto bad;
}
}
ip = mtod(m, struct ip *);
if (ip == NULL) goto bad;
if (ip->ip_v != IPVERSION) {
Many network stack subsystems use a single global data structure to hold all pertinent statatistics for the subsystem. These structures are sometimes "borrowed" by kernel modules that require a place to store statistics for similar events. Add KPI accessor functions for statistics structures referenced by kernel modules so that they no longer encode certain specifics of how the data structures are named and stored. This change is intended to make it easier to move to per-CPU network stats following 8.0-RELEASE. The following modules are affected by this change: if_bridge if_cxgb if_gif ip_mroute ipdivert pf In practice, most of these statistics consumers should, in fact, maintain their own statistics data structures rather than borrowing structures from the base network stack. However, that change is too agressive for this point in the release cycle. Reviewed by: bz Approved by: re (kib)
2009-08-02 19:43:32 +00:00
KMOD_IPSTAT_INC(ips_badvers);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
goto bad;
}
hlen = ip->ip_hl << 2;
if (hlen < sizeof(struct ip)) { /* minimum header length */
Many network stack subsystems use a single global data structure to hold all pertinent statatistics for the subsystem. These structures are sometimes "borrowed" by kernel modules that require a place to store statistics for similar events. Add KPI accessor functions for statistics structures referenced by kernel modules so that they no longer encode certain specifics of how the data structures are named and stored. This change is intended to make it easier to move to per-CPU network stats following 8.0-RELEASE. The following modules are affected by this change: if_bridge if_cxgb if_gif ip_mroute ipdivert pf In practice, most of these statistics consumers should, in fact, maintain their own statistics data structures rather than borrowing structures from the base network stack. However, that change is too agressive for this point in the release cycle. Reviewed by: bz Approved by: re (kib)
2009-08-02 19:43:32 +00:00
KMOD_IPSTAT_INC(ips_badhlen);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
goto bad;
}
if (hlen > m->m_len) {
if ((m = m_pullup(m, hlen)) == 0) {
Many network stack subsystems use a single global data structure to hold all pertinent statatistics for the subsystem. These structures are sometimes "borrowed" by kernel modules that require a place to store statistics for similar events. Add KPI accessor functions for statistics structures referenced by kernel modules so that they no longer encode certain specifics of how the data structures are named and stored. This change is intended to make it easier to move to per-CPU network stats following 8.0-RELEASE. The following modules are affected by this change: if_bridge if_cxgb if_gif ip_mroute ipdivert pf In practice, most of these statistics consumers should, in fact, maintain their own statistics data structures rather than borrowing structures from the base network stack. However, that change is too agressive for this point in the release cycle. Reviewed by: bz Approved by: re (kib)
2009-08-02 19:43:32 +00:00
KMOD_IPSTAT_INC(ips_badhlen);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
goto bad;
}
ip = mtod(m, struct ip *);
if (ip == NULL) goto bad;
}
if (m->m_pkthdr.csum_flags & CSUM_IP_CHECKED) {
sum = !(m->m_pkthdr.csum_flags & CSUM_IP_VALID);
} else {
if (hlen == sizeof(struct ip)) {
sum = in_cksum_hdr(ip);
} else {
sum = in_cksum(m, hlen);
}
}
if (sum) {
Many network stack subsystems use a single global data structure to hold all pertinent statatistics for the subsystem. These structures are sometimes "borrowed" by kernel modules that require a place to store statistics for similar events. Add KPI accessor functions for statistics structures referenced by kernel modules so that they no longer encode certain specifics of how the data structures are named and stored. This change is intended to make it easier to move to per-CPU network stats following 8.0-RELEASE. The following modules are affected by this change: if_bridge if_cxgb if_gif ip_mroute ipdivert pf In practice, most of these statistics consumers should, in fact, maintain their own statistics data structures rather than borrowing structures from the base network stack. However, that change is too agressive for this point in the release cycle. Reviewed by: bz Approved by: re (kib)
2009-08-02 19:43:32 +00:00
KMOD_IPSTAT_INC(ips_badsum);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
goto bad;
}
/* Retrieve the packet length. */
len = ntohs(ip->ip_len);
/*
* Check for additional length bogosity
*/
if (len < hlen) {
Many network stack subsystems use a single global data structure to hold all pertinent statatistics for the subsystem. These structures are sometimes "borrowed" by kernel modules that require a place to store statistics for similar events. Add KPI accessor functions for statistics structures referenced by kernel modules so that they no longer encode certain specifics of how the data structures are named and stored. This change is intended to make it easier to move to per-CPU network stats following 8.0-RELEASE. The following modules are affected by this change: if_bridge if_cxgb if_gif ip_mroute ipdivert pf In practice, most of these statistics consumers should, in fact, maintain their own statistics data structures rather than borrowing structures from the base network stack. However, that change is too agressive for this point in the release cycle. Reviewed by: bz Approved by: re (kib)
2009-08-02 19:43:32 +00:00
KMOD_IPSTAT_INC(ips_badlen);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
goto bad;
}
/*
* Check that the amount of data in the buffers
* is as at least much as the IP header would have us expect.
* Drop packet if shorter than we expect.
*/
if (m->m_pkthdr.len < len) {
Many network stack subsystems use a single global data structure to hold all pertinent statatistics for the subsystem. These structures are sometimes "borrowed" by kernel modules that require a place to store statistics for similar events. Add KPI accessor functions for statistics structures referenced by kernel modules so that they no longer encode certain specifics of how the data structures are named and stored. This change is intended to make it easier to move to per-CPU network stats following 8.0-RELEASE. The following modules are affected by this change: if_bridge if_cxgb if_gif ip_mroute ipdivert pf In practice, most of these statistics consumers should, in fact, maintain their own statistics data structures rather than borrowing structures from the base network stack. However, that change is too agressive for this point in the release cycle. Reviewed by: bz Approved by: re (kib)
2009-08-02 19:43:32 +00:00
KMOD_IPSTAT_INC(ips_tooshort);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
goto bad;
}
/* Checks out, proceed */
*mp = m;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (0);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
bad:
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
*mp = m;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (-1);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
#ifdef INET6
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
/*
* Same as above, but for IPv6.
* Cut-and-pasted from ip6_input.c.
* XXX Should we update ip6stat, or not?
*/
static int
bridge_ip6_checkbasic(struct mbuf **mp)
{
struct mbuf *m = *mp;
struct ip6_hdr *ip6;
/*
* If the IPv6 header is not aligned, slurp it up into a new
* mbuf with space for link headers, in the event we forward
* it. Otherwise, if it is aligned, make sure the entire base
* IPv6 header is in the first mbuf of the chain.
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
*/
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
if (IP6_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) {
struct ifnet *inifp = m->m_pkthdr.rcvif;
if ((m = m_copyup(m, sizeof(struct ip6_hdr),
(max_linkhdr + 3) & ~3)) == NULL) {
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
/* XXXJRT new stat, please */
Commit step 1 of the vimage project, (network stack) virtualization work done by Marko Zec (zec@). This is the first in a series of commits over the course of the next few weeks. Mark all uses of global variables to be virtualized with a V_ prefix. Use macros to map them back to their global names for now, so this is a NOP change only. We hope to have caught at least 85-90% of what is needed so we do not invalidate a lot of outstanding patches again. Obtained from: //depot/projects/vimage-commit2/... Reviewed by: brooks, des, ed, mav, julian, jamie, kris, rwatson, zec, ... (various people I forgot, different versions) md5 (with a bit of help) Sponsored by: NLnet Foundation, The FreeBSD Foundation X-MFC after: never V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
V_ip6stat.ip6s_toosmall++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
in6_ifstat_inc(inifp, ifs6_in_hdrerr);
goto bad;
}
Check the alignment of the IP header before passing the packet up to the packet filter. This would cause a panic on architectures that require strict alignment such as sparc64 (tier1) and ia64/ppc (tier2). This adds two new macros that check the alignment, these are compile time dependent on __NO_STRICT_ALIGNMENT which is set for i386 and amd64 where alignment isn't need so the cost is avoided. IP_HDR_ALIGNED_P() IP6_HDR_ALIGNED_P() Move bridge_ip_checkbasic()/bridge_ip6_checkbasic() up so that the alignment is checked for ipfw and dummynet too. PR: ia64/81284 Obtained from: NetBSD Approved by: re (dwhite), mlaier (mentor)
2005-07-02 23:13:31 +00:00
} else if (__predict_false(m->m_len < sizeof(struct ip6_hdr))) {
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
struct ifnet *inifp = m->m_pkthdr.rcvif;
if ((m = m_pullup(m, sizeof(struct ip6_hdr))) == NULL) {
Commit step 1 of the vimage project, (network stack) virtualization work done by Marko Zec (zec@). This is the first in a series of commits over the course of the next few weeks. Mark all uses of global variables to be virtualized with a V_ prefix. Use macros to map them back to their global names for now, so this is a NOP change only. We hope to have caught at least 85-90% of what is needed so we do not invalidate a lot of outstanding patches again. Obtained from: //depot/projects/vimage-commit2/... Reviewed by: brooks, des, ed, mav, julian, jamie, kris, rwatson, zec, ... (various people I forgot, different versions) md5 (with a bit of help) Sponsored by: NLnet Foundation, The FreeBSD Foundation X-MFC after: never V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
V_ip6stat.ip6s_toosmall++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
in6_ifstat_inc(inifp, ifs6_in_hdrerr);
goto bad;
}
}
ip6 = mtod(m, struct ip6_hdr *);
if ((ip6->ip6_vfc & IPV6_VERSION_MASK) != IPV6_VERSION) {
Commit step 1 of the vimage project, (network stack) virtualization work done by Marko Zec (zec@). This is the first in a series of commits over the course of the next few weeks. Mark all uses of global variables to be virtualized with a V_ prefix. Use macros to map them back to their global names for now, so this is a NOP change only. We hope to have caught at least 85-90% of what is needed so we do not invalidate a lot of outstanding patches again. Obtained from: //depot/projects/vimage-commit2/... Reviewed by: brooks, des, ed, mav, julian, jamie, kris, rwatson, zec, ... (various people I forgot, different versions) md5 (with a bit of help) Sponsored by: NLnet Foundation, The FreeBSD Foundation X-MFC after: never V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
V_ip6stat.ip6s_badvers++;
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
in6_ifstat_inc(m->m_pkthdr.rcvif, ifs6_in_hdrerr);
goto bad;
}
/* Checks out, proceed */
*mp = m;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (0);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
Minor whitespace cleanup.
2005-12-17 10:03:48 +00:00
bad:
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
*mp = m;
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
return (-1);
Add if_bridge, which provides more advanced Ethernet bridging and 802.1d spanning tree support. Based on Jason Wright's bridge driver from OpenBSD, and modified by Jason R. Thorpe in NetBSD. Reviewed by: mlaier, bms, green Silence from: -net Approved by: mlaier (mentor) Obtained from: NetBSD
2005-06-05 02:59:26 +00:00
}
Fix style(9) nits, whitespace and parentheses.
2006-05-16 22:50:41 +00:00
#endif /* INET6 */
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
/*
* bridge_fragment:
*
* Return a fragmented mbuf chain.
*/
static int
bridge_fragment(struct ifnet *ifp, struct mbuf *m, struct ether_header *eh,
int snap, struct llc *llc)
{
struct mbuf *m0;
struct ip *ip;
int error = -1;
if (m->m_len < sizeof(struct ip) &&
(m = m_pullup(m, sizeof(struct ip))) == NULL)
goto out;
ip = mtod(m, struct ip *);
error = ip_fragment(ip, &m, ifp->if_mtu, ifp->if_hwassist,
CSUM_DELAY_IP);
if (error)
goto out;
/* walk the chain and re-add the Ethernet header */
for (m0 = m; m0; m0 = m0->m_nextpkt) {
if (error == 0) {
if (snap) {
M_PREPEND(m0, sizeof(struct llc), M_DONTWAIT);
if (m0 == NULL) {
error = ENOBUFS;
continue;
}
bcopy(llc, mtod(m0, caddr_t),
sizeof(struct llc));
}
M_PREPEND(m0, ETHER_HDR_LEN, M_DONTWAIT);
if (m0 == NULL) {
error = ENOBUFS;
continue;
}
bcopy(eh, mtod(m0, caddr_t), ETHER_HDR_LEN);
} else
m_freem(m);
}
if (error == 0)
Many network stack subsystems use a single global data structure to hold all pertinent statatistics for the subsystem. These structures are sometimes "borrowed" by kernel modules that require a place to store statistics for similar events. Add KPI accessor functions for statistics structures referenced by kernel modules so that they no longer encode certain specifics of how the data structures are named and stored. This change is intended to make it easier to move to per-CPU network stats following 8.0-RELEASE. The following modules are affected by this change: if_bridge if_cxgb if_gif ip_mroute ipdivert pf In practice, most of these statistics consumers should, in fact, maintain their own statistics data structures rather than borrowing structures from the base network stack. However, that change is too agressive for this point in the release cycle. Reviewed by: bz Approved by: re (kib)
2009-08-02 19:43:32 +00:00
KMOD_IPSTAT_INC(ips_fragmented);
Add support for fragmenting ipv4 packets. The packet filter may reassemble the ip fragments and return a packet that is larger than the MTU of the sending interface. There is no check for DF or icmp replies as we can only get a large packet to fragment by reassembling a previous fragment, and this only happens after a call to pfil(9). Obtained from: OpenBSD (mostly) Glanced at by: mlaier MFC after: 1 month
2006-04-29 05:37:25 +00:00
return (error);
out:
if (m != NULL)
m_freem(m);
return (error);
}
Reference in New Issue Copy Permalink