2005-01-07 01:45:51 +00:00
|
|
|
/*-
|
1994-05-24 10:09:53 +00:00
|
|
|
* Copyright (c) 1982, 1986, 1991, 1993
|
|
|
|
* The Regents of the University of California. All rights reserved.
|
2004-11-13 17:05:40 +00:00
|
|
|
* Copyright (C) 2001 WIDE Project. All rights reserved.
|
1994-05-24 10:09:53 +00:00
|
|
|
*
|
|
|
|
* 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.
|
2017-02-28 23:42:47 +00:00
|
|
|
* 3. Neither the name of the University nor the names of its contributors
|
1994-05-24 10:09:53 +00:00
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
* without specific prior written permission.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
*
|
1995-09-21 17:50:45 +00:00
|
|
|
* @(#)in.c 8.4 (Berkeley) 1/9/95
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
|
2007-10-07 20:44:24 +00:00
|
|
|
#include <sys/cdefs.h>
|
|
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
|
2010-03-09 01:11:45 +00:00
|
|
|
#include "opt_mpath.h"
|
2005-03-20 10:27:17 +00:00
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/param.h>
|
2013-10-28 07:29:16 +00:00
|
|
|
#include <sys/eventhandler.h>
|
1994-05-25 09:21:21 +00:00
|
|
|
#include <sys/systm.h>
|
1997-03-24 11:33:46 +00:00
|
|
|
#include <sys/sockio.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/malloc.h>
|
2006-11-06 13:42:10 +00:00
|
|
|
#include <sys/priv.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/socket.h>
|
2009-01-09 13:06:56 +00:00
|
|
|
#include <sys/jail.h>
|
1995-12-09 20:43:53 +00:00
|
|
|
#include <sys/kernel.h>
|
2015-07-29 08:12:05 +00:00
|
|
|
#include <sys/lock.h>
|
2009-01-09 13:06:56 +00:00
|
|
|
#include <sys/proc.h>
|
2015-07-29 08:12:05 +00:00
|
|
|
#include <sys/rmlock.h>
|
1995-12-09 20:43:53 +00:00
|
|
|
#include <sys/sysctl.h>
|
2009-05-12 07:41:20 +00:00
|
|
|
#include <sys/syslog.h>
|
2013-11-05 07:44:15 +00:00
|
|
|
#include <sys/sx.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
#include <net/if.h>
|
2009-07-27 17:08:06 +00:00
|
|
|
#include <net/if_var.h>
|
2010-11-12 22:03:02 +00:00
|
|
|
#include <net/if_arp.h>
|
2009-05-12 07:41:20 +00:00
|
|
|
#include <net/if_dl.h>
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
#include <net/if_llatbl.h>
|
1999-12-22 19:13:38 +00:00
|
|
|
#include <net/if_types.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <net/route.h>
|
2009-05-12 07:41:20 +00:00
|
|
|
#include <net/vnet.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
#include <netinet/if_ether.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <netinet/in.h>
|
|
|
|
#include <netinet/in_var.h>
|
2001-08-04 17:10:14 +00:00
|
|
|
#include <netinet/in_pcb.h>
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
#include <netinet/ip_var.h>
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
#include <netinet/ip_carp.h>
|
2009-03-09 17:53:05 +00:00
|
|
|
#include <netinet/igmp_var.h>
|
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator
(DPCPU), as suggested by Peter Wemm, and implement a new per-virtual
network stack memory allocator. Modify vnet to use the allocator
instead of monolithic global container structures (vinet, ...). This
change solves many binary compatibility problems associated with
VIMAGE, and restores ELF symbols for virtualized global variables.
Each virtualized global variable exists as a "reference copy", and also
once per virtual network stack. Virtualized global variables are
tagged at compile-time, placing the in a special linker set, which is
loaded into a contiguous region of kernel memory. Virtualized global
variables in the base kernel are linked as normal, but those in modules
are copied and relocated to a reserved portion of the kernel's vnet
region with the help of a the kernel linker.
Virtualized global variables exist in per-vnet memory set up when the
network stack instance is created, and are initialized statically from
the reference copy. Run-time access occurs via an accessor macro, which
converts from the current vnet and requested symbol to a per-vnet
address. When "options VIMAGE" is not compiled into the kernel, normal
global ELF symbols will be used instead and indirection is avoided.
This change restores static initialization for network stack global
variables, restores support for non-global symbols and types, eliminates
the need for many subsystem constructors, eliminates large per-subsystem
structures that caused many binary compatibility issues both for
monitoring applications (netstat) and kernel modules, removes the
per-function INIT_VNET_*() macros throughout the stack, eliminates the
need for vnet_symmap ksym(2) munging, and eliminates duplicate
definitions of virtualized globals under VIMAGE_GLOBALS.
Bump __FreeBSD_version and update UPDATING.
Portions submitted by: bz
Reviewed by: bz, zec
Discussed with: gnn, jamie, jeff, jhb, julian, sam
Suggested by: peter
Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
|
|
|
#include <netinet/udp.h>
|
|
|
|
#include <netinet/udp_var.h>
|
1997-10-11 18:31:40 +00:00
|
|
|
|
2013-11-06 19:46:20 +00:00
|
|
|
static int in_aifaddr_ioctl(u_long, caddr_t, struct ifnet *, struct thread *);
|
2017-01-25 19:04:08 +00:00
|
|
|
static int in_difaddr_ioctl(u_long, caddr_t, struct ifnet *, struct thread *);
|
1999-12-22 19:13:38 +00:00
|
|
|
|
2002-03-19 21:25:46 +00:00
|
|
|
static void in_socktrim(struct sockaddr_in *);
|
2007-03-20 00:36:10 +00:00
|
|
|
static void in_purgemaddrs(struct ifnet *);
|
1994-05-24 10:09:53 +00:00
|
|
|
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
static VNET_DEFINE(int, nosameprefix);
|
|
|
|
#define V_nosameprefix VNET(nosameprefix)
|
2014-11-07 09:39:05 +00:00
|
|
|
SYSCTL_INT(_net_inet_ip, OID_AUTO, no_same_prefix, CTLFLAG_VNET | CTLFLAG_RW,
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
&VNET_NAME(nosameprefix), 0,
|
2005-08-18 10:34:30 +00:00
|
|
|
"Refuse to create same prefixes on different interfaces");
|
1997-01-13 21:26:53 +00:00
|
|
|
|
2010-04-29 11:52:42 +00:00
|
|
|
VNET_DECLARE(struct inpcbinfo, ripcbinfo);
|
|
|
|
#define V_ripcbinfo VNET(ripcbinfo)
|
|
|
|
|
2013-11-05 07:44:15 +00:00
|
|
|
static struct sx in_control_sx;
|
|
|
|
SX_SYSINIT(in_control_sx, &in_control_sx, "in_control");
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* Return 1 if an internet address is for a ``local'' host
|
2011-10-15 16:28:06 +00:00
|
|
|
* (one to which we have a connection).
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
2007-05-10 15:58:48 +00:00
|
|
|
in_localaddr(struct in_addr in)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2015-07-29 08:12:05 +00:00
|
|
|
struct rm_priotracker in_ifa_tracker;
|
2017-05-17 00:34:34 +00:00
|
|
|
u_long i = ntohl(in.s_addr);
|
|
|
|
struct in_ifaddr *ia;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RLOCK(&in_ifa_tracker);
|
2011-10-15 16:28:06 +00:00
|
|
|
TAILQ_FOREACH(ia, &V_in_ifaddrhead, ia_link) {
|
|
|
|
if ((i & ia->ia_subnetmask) == ia->ia_subnet) {
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2011-10-15 16:28:06 +00:00
|
|
|
return (1);
|
2009-06-25 11:52:33 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
1994-05-24 10:09:53 +00:00
|
|
|
return (0);
|
|
|
|
}
|
|
|
|
|
2004-08-11 11:49:48 +00:00
|
|
|
/*
|
|
|
|
* Return 1 if an internet address is for the local host and configured
|
|
|
|
* on one of its interfaces.
|
|
|
|
*/
|
|
|
|
int
|
2007-05-10 15:58:48 +00:00
|
|
|
in_localip(struct in_addr in)
|
2004-08-11 11:49:48 +00:00
|
|
|
{
|
2015-07-29 08:12:05 +00:00
|
|
|
struct rm_priotracker in_ifa_tracker;
|
2004-08-11 11:49:48 +00:00
|
|
|
struct in_ifaddr *ia;
|
|
|
|
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RLOCK(&in_ifa_tracker);
|
2004-08-11 11:49:48 +00:00
|
|
|
LIST_FOREACH(ia, INADDR_HASH(in.s_addr), ia_hash) {
|
2009-06-25 11:52:33 +00:00
|
|
|
if (IA_SIN(ia)->sin_addr.s_addr == in.s_addr) {
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2008-10-26 19:17:25 +00:00
|
|
|
return (1);
|
2009-06-25 11:52:33 +00:00
|
|
|
}
|
2004-08-11 11:49:48 +00:00
|
|
|
}
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2008-10-26 19:17:25 +00:00
|
|
|
return (0);
|
2004-08-11 11:49:48 +00:00
|
|
|
}
|
|
|
|
|
2015-04-17 11:57:06 +00:00
|
|
|
/*
|
|
|
|
* Return 1 if an internet address is configured on an interface.
|
|
|
|
*/
|
|
|
|
int
|
|
|
|
in_ifhasaddr(struct ifnet *ifp, struct in_addr in)
|
|
|
|
{
|
|
|
|
struct ifaddr *ifa;
|
|
|
|
struct in_ifaddr *ia;
|
|
|
|
|
|
|
|
IF_ADDR_RLOCK(ifp);
|
|
|
|
TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
|
|
|
|
if (ifa->ifa_addr->sa_family != AF_INET)
|
|
|
|
continue;
|
|
|
|
ia = (struct in_ifaddr *)ifa;
|
|
|
|
if (ia->ia_addr.sin_addr.s_addr == in.s_addr) {
|
|
|
|
IF_ADDR_RUNLOCK(ifp);
|
|
|
|
return (1);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
IF_ADDR_RUNLOCK(ifp);
|
|
|
|
|
|
|
|
return (0);
|
|
|
|
}
|
|
|
|
|
2013-11-05 07:44:15 +00:00
|
|
|
/*
|
|
|
|
* Return a reference to the interface address which is different to
|
|
|
|
* the supplied one but with same IP address value.
|
|
|
|
*/
|
|
|
|
static struct in_ifaddr *
|
|
|
|
in_localip_more(struct in_ifaddr *ia)
|
|
|
|
{
|
2015-07-29 08:12:05 +00:00
|
|
|
struct rm_priotracker in_ifa_tracker;
|
2013-11-05 07:44:15 +00:00
|
|
|
in_addr_t in = IA_SIN(ia)->sin_addr.s_addr;
|
|
|
|
struct in_ifaddr *it;
|
|
|
|
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RLOCK(&in_ifa_tracker);
|
2013-11-05 07:44:15 +00:00
|
|
|
LIST_FOREACH(it, INADDR_HASH(in), ia_hash) {
|
|
|
|
if (it != ia && IA_SIN(it)->sin_addr.s_addr == in) {
|
|
|
|
ifa_ref(&it->ia_ifa);
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2013-11-05 07:44:15 +00:00
|
|
|
return (it);
|
|
|
|
}
|
|
|
|
}
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2013-11-05 07:44:15 +00:00
|
|
|
|
|
|
|
return (NULL);
|
|
|
|
}
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* Determine whether an IP address is in a reserved set of addresses
|
|
|
|
* that may not be forwarded, or whether datagrams to that destination
|
|
|
|
* may be forwarded.
|
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
2007-05-10 15:58:48 +00:00
|
|
|
in_canforward(struct in_addr in)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2017-05-17 00:34:34 +00:00
|
|
|
u_long i = ntohl(in.s_addr);
|
|
|
|
u_long net;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2007-02-03 06:45:51 +00:00
|
|
|
if (IN_EXPERIMENTAL(i) || IN_MULTICAST(i) || IN_LINKLOCAL(i))
|
1994-05-24 10:09:53 +00:00
|
|
|
return (0);
|
|
|
|
if (IN_CLASSA(i)) {
|
|
|
|
net = i & IN_CLASSA_NET;
|
|
|
|
if (net == 0 || net == (IN_LOOPBACKNET << IN_CLASSA_NSHIFT))
|
|
|
|
return (0);
|
|
|
|
}
|
|
|
|
return (1);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Trim a mask in a sockaddr
|
|
|
|
*/
|
1995-11-14 20:34:56 +00:00
|
|
|
static void
|
2007-05-10 15:58:48 +00:00
|
|
|
in_socktrim(struct sockaddr_in *ap)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2017-05-17 00:34:34 +00:00
|
|
|
char *cplim = (char *) &ap->sin_addr;
|
|
|
|
char *cp = (char *) (&ap->sin_addr + 1);
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
ap->sin_len = 0;
|
1994-11-03 21:04:21 +00:00
|
|
|
while (--cp >= cplim)
|
2004-08-16 18:32:07 +00:00
|
|
|
if (*cp) {
|
1994-05-24 10:09:53 +00:00
|
|
|
(ap)->sin_len = cp - (char *) (ap) + 1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Generic internet control operations (ioctl's).
|
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
2007-05-10 15:58:48 +00:00
|
|
|
in_control(struct socket *so, u_long cmd, caddr_t data, struct ifnet *ifp,
|
|
|
|
struct thread *td)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2013-11-05 07:44:15 +00:00
|
|
|
struct ifreq *ifr = (struct ifreq *)data;
|
|
|
|
struct sockaddr_in *addr = (struct sockaddr_in *)&ifr->ifr_addr;
|
2013-11-06 08:36:08 +00:00
|
|
|
struct ifaddr *ifa;
|
2013-11-05 07:44:15 +00:00
|
|
|
struct in_ifaddr *ia;
|
|
|
|
int error;
|
2001-11-30 14:00:55 +00:00
|
|
|
|
2013-11-05 07:44:15 +00:00
|
|
|
if (ifp == NULL)
|
|
|
|
return (EADDRNOTAVAIL);
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2009-04-23 21:41:37 +00:00
|
|
|
/*
|
2013-11-05 07:44:15 +00:00
|
|
|
* Filter out 4 ioctls we implement directly. Forward the rest
|
|
|
|
* to specific functions and ifp->if_ioctl().
|
2009-04-23 21:41:37 +00:00
|
|
|
*/
|
1999-12-22 19:13:38 +00:00
|
|
|
switch (cmd) {
|
2009-04-23 21:41:37 +00:00
|
|
|
case SIOCGIFADDR:
|
|
|
|
case SIOCGIFBRDADDR:
|
|
|
|
case SIOCGIFDSTADDR:
|
|
|
|
case SIOCGIFNETMASK:
|
2011-11-21 14:10:13 +00:00
|
|
|
break;
|
2013-11-05 07:44:15 +00:00
|
|
|
case SIOCDIFADDR:
|
|
|
|
sx_xlock(&in_control_sx);
|
2017-01-25 19:04:08 +00:00
|
|
|
error = in_difaddr_ioctl(cmd, data, ifp, td);
|
2013-11-05 07:44:15 +00:00
|
|
|
sx_xunlock(&in_control_sx);
|
|
|
|
return (error);
|
2013-11-06 19:46:20 +00:00
|
|
|
case OSIOCAIFADDR: /* 9.x compat */
|
2011-11-21 14:10:13 +00:00
|
|
|
case SIOCAIFADDR:
|
2013-11-05 07:44:15 +00:00
|
|
|
sx_xlock(&in_control_sx);
|
2013-11-06 19:46:20 +00:00
|
|
|
error = in_aifaddr_ioctl(cmd, data, ifp, td);
|
2013-11-05 07:44:15 +00:00
|
|
|
sx_xunlock(&in_control_sx);
|
|
|
|
return (error);
|
2009-04-23 21:41:37 +00:00
|
|
|
case SIOCSIFADDR:
|
|
|
|
case SIOCSIFBRDADDR:
|
|
|
|
case SIOCSIFDSTADDR:
|
|
|
|
case SIOCSIFNETMASK:
|
2012-01-16 09:53:24 +00:00
|
|
|
/* We no longer support that old commands. */
|
|
|
|
return (EINVAL);
|
2009-04-23 21:41:37 +00:00
|
|
|
default:
|
2013-11-05 07:44:15 +00:00
|
|
|
if (ifp->if_ioctl == NULL)
|
2009-04-23 21:41:37 +00:00
|
|
|
return (EOPNOTSUPP);
|
|
|
|
return ((*ifp->if_ioctl)(ifp, cmd, data));
|
1999-12-22 19:13:38 +00:00
|
|
|
}
|
|
|
|
|
2013-11-06 08:36:08 +00:00
|
|
|
if (addr->sin_addr.s_addr != INADDR_ANY &&
|
|
|
|
prison_check_ip4(td->td_ucred, &addr->sin_addr) != 0)
|
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
2014-08-22 19:08:12 +00:00
|
|
|
* Find address for this interface, if it exists. If an
|
|
|
|
* address was specified, find that one instead of the
|
|
|
|
* first one on the interface, if possible.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
2013-11-06 08:36:08 +00:00
|
|
|
IF_ADDR_RLOCK(ifp);
|
|
|
|
TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
|
2013-12-29 22:20:06 +00:00
|
|
|
if (ifa->ifa_addr->sa_family != AF_INET)
|
|
|
|
continue;
|
2013-11-06 08:36:08 +00:00
|
|
|
ia = (struct in_ifaddr *)ifa;
|
|
|
|
if (ia->ia_addr.sin_addr.s_addr == addr->sin_addr.s_addr)
|
2009-04-23 21:41:37 +00:00
|
|
|
break;
|
|
|
|
}
|
2014-08-22 19:08:12 +00:00
|
|
|
if (ifa == NULL)
|
|
|
|
TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link)
|
|
|
|
if (ifa->ifa_addr->sa_family == AF_INET) {
|
|
|
|
ia = (struct in_ifaddr *)ifa;
|
|
|
|
if (prison_check_ip4(td->td_ucred,
|
|
|
|
&ia->ia_addr.sin_addr) == 0)
|
|
|
|
break;
|
|
|
|
}
|
2013-11-05 07:44:15 +00:00
|
|
|
|
2013-11-06 08:36:08 +00:00
|
|
|
if (ifa == NULL) {
|
|
|
|
IF_ADDR_RUNLOCK(ifp);
|
2013-11-05 07:44:15 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
2001-09-29 04:34:11 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2009-04-25 23:02:57 +00:00
|
|
|
error = 0;
|
1994-05-24 10:09:53 +00:00
|
|
|
switch (cmd) {
|
|
|
|
case SIOCGIFADDR:
|
2013-11-05 07:44:15 +00:00
|
|
|
*addr = ia->ia_addr;
|
1994-05-24 10:09:53 +00:00
|
|
|
break;
|
|
|
|
|
|
|
|
case SIOCGIFBRDADDR:
|
2009-04-25 23:02:57 +00:00
|
|
|
if ((ifp->if_flags & IFF_BROADCAST) == 0) {
|
|
|
|
error = EINVAL;
|
2013-11-05 07:44:15 +00:00
|
|
|
break;
|
2009-04-25 23:02:57 +00:00
|
|
|
}
|
2013-11-05 07:44:15 +00:00
|
|
|
*addr = ia->ia_broadaddr;
|
|
|
|
break;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
case SIOCGIFDSTADDR:
|
2009-04-25 23:02:57 +00:00
|
|
|
if ((ifp->if_flags & IFF_POINTOPOINT) == 0) {
|
|
|
|
error = EINVAL;
|
2013-11-05 07:44:15 +00:00
|
|
|
break;
|
2009-04-25 23:02:57 +00:00
|
|
|
}
|
2013-11-05 07:44:15 +00:00
|
|
|
*addr = ia->ia_dstaddr;
|
|
|
|
break;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
case SIOCGIFNETMASK:
|
2013-11-05 07:44:15 +00:00
|
|
|
*addr = ia->ia_sockmask;
|
|
|
|
break;
|
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2013-11-06 08:36:08 +00:00
|
|
|
IF_ADDR_RUNLOCK(ifp);
|
2001-11-30 14:00:55 +00:00
|
|
|
|
2013-11-05 07:44:15 +00:00
|
|
|
return (error);
|
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2013-11-05 07:44:15 +00:00
|
|
|
static int
|
2013-11-06 19:46:20 +00:00
|
|
|
in_aifaddr_ioctl(u_long cmd, caddr_t data, struct ifnet *ifp, struct thread *td)
|
2013-11-05 07:44:15 +00:00
|
|
|
{
|
|
|
|
const struct in_aliasreq *ifra = (struct in_aliasreq *)data;
|
|
|
|
const struct sockaddr_in *addr = &ifra->ifra_addr;
|
|
|
|
const struct sockaddr_in *broadaddr = &ifra->ifra_broadaddr;
|
|
|
|
const struct sockaddr_in *mask = &ifra->ifra_mask;
|
|
|
|
const struct sockaddr_in *dstaddr = &ifra->ifra_dstaddr;
|
2013-11-06 19:46:20 +00:00
|
|
|
const int vhid = (cmd == SIOCAIFADDR) ? ifra->ifra_vhid : 0;
|
2013-11-05 07:44:15 +00:00
|
|
|
struct ifaddr *ifa;
|
|
|
|
struct in_ifaddr *ia;
|
|
|
|
bool iaIsFirst;
|
|
|
|
int error = 0;
|
|
|
|
|
|
|
|
error = priv_check(td, PRIV_NET_ADDIFADDR);
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* ifra_addr must be present and be of INET family.
|
|
|
|
* ifra_broadaddr/ifra_dstaddr and ifra_mask are optional.
|
|
|
|
*/
|
|
|
|
if (addr->sin_len != sizeof(struct sockaddr_in) ||
|
|
|
|
addr->sin_family != AF_INET)
|
|
|
|
return (EINVAL);
|
|
|
|
if (broadaddr->sin_len != 0 &&
|
|
|
|
(broadaddr->sin_len != sizeof(struct sockaddr_in) ||
|
|
|
|
broadaddr->sin_family != AF_INET))
|
|
|
|
return (EINVAL);
|
|
|
|
if (mask->sin_len != 0 &&
|
|
|
|
(mask->sin_len != sizeof(struct sockaddr_in) ||
|
|
|
|
mask->sin_family != AF_INET))
|
|
|
|
return (EINVAL);
|
|
|
|
if ((ifp->if_flags & IFF_POINTOPOINT) &&
|
|
|
|
(dstaddr->sin_len != sizeof(struct sockaddr_in) ||
|
|
|
|
dstaddr->sin_addr.s_addr == INADDR_ANY))
|
|
|
|
return (EDESTADDRREQ);
|
|
|
|
if (vhid > 0 && carp_attach_p == NULL)
|
|
|
|
return (EPROTONOSUPPORT);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* See whether address already exist.
|
|
|
|
*/
|
|
|
|
iaIsFirst = true;
|
|
|
|
ia = NULL;
|
|
|
|
IF_ADDR_RLOCK(ifp);
|
|
|
|
TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
|
2013-12-29 22:20:06 +00:00
|
|
|
struct in_ifaddr *it;
|
2013-11-05 07:44:15 +00:00
|
|
|
|
2013-12-29 22:20:06 +00:00
|
|
|
if (ifa->ifa_addr->sa_family != AF_INET)
|
2013-11-05 07:44:15 +00:00
|
|
|
continue;
|
|
|
|
|
2013-12-29 22:20:06 +00:00
|
|
|
it = (struct in_ifaddr *)ifa;
|
2013-11-05 07:44:15 +00:00
|
|
|
iaIsFirst = false;
|
|
|
|
if (it->ia_addr.sin_addr.s_addr == addr->sin_addr.s_addr &&
|
|
|
|
prison_check_ip4(td->td_ucred, &addr->sin_addr) == 0)
|
|
|
|
ia = it;
|
|
|
|
}
|
|
|
|
IF_ADDR_RUNLOCK(ifp);
|
|
|
|
|
|
|
|
if (ia != NULL)
|
2017-01-25 19:04:08 +00:00
|
|
|
(void )in_difaddr_ioctl(cmd, data, ifp, td);
|
2013-11-05 07:44:15 +00:00
|
|
|
|
|
|
|
ifa = ifa_alloc(sizeof(struct in_ifaddr), M_WAITOK);
|
|
|
|
ia = (struct in_ifaddr *)ifa;
|
|
|
|
ifa->ifa_addr = (struct sockaddr *)&ia->ia_addr;
|
|
|
|
ifa->ifa_dstaddr = (struct sockaddr *)&ia->ia_dstaddr;
|
|
|
|
ifa->ifa_netmask = (struct sockaddr *)&ia->ia_sockmask;
|
Add GARP retransmit capability
A single gratuitous ARP (GARP) is always transmitted when an IPv4
address is added to an interface, and that is usually sufficient.
However, in some circumstances, such as when a shared address is
passed between cluster nodes, this single GARP may occasionally be
dropped or lost. This can lead to neighbors on the network link
working with a stale ARP cache and sending packets destined for
that address to the node that previously owned the address, which
may not respond.
To avoid this situation, GARP retransmissions can be enabled by setting
the net.link.ether.inet.garp_rexmit_count sysctl to a value greater
than zero. The setting represents the maximum number of retransmissions.
The interval between retransmissions is calculated using an exponential
backoff algorithm, doubling each time, so the retransmission intervals
are: {1, 2, 4, 8, 16, ...} (seconds).
Due to the exponential backoff algorithm used for the interval
between GARP retransmissions, the maximum number of retransmissions
is limited to 16 for sanity. This limit corresponds to a maximum
interval between retransmissions of 2^16 seconds ~= 18 hours.
Increasing this limit is possible, but sending out GARPs spaced
days apart would be of little use.
Submitted by: David A. Bright <david.a.bright@dell.com>
MFC after: 1 month
Relnotes: yes
Sponsored by: Dell EMC
Differential Revision: https://reviews.freebsd.org/D7695
2016-10-02 01:42:45 +00:00
|
|
|
callout_init_rw(&ia->ia_garp_timer, &ifp->if_addr_lock,
|
|
|
|
CALLOUT_RETURNUNLOCKED);
|
2013-11-05 07:44:15 +00:00
|
|
|
|
|
|
|
ia->ia_ifp = ifp;
|
|
|
|
ia->ia_addr = *addr;
|
|
|
|
if (mask->sin_len != 0) {
|
|
|
|
ia->ia_sockmask = *mask;
|
|
|
|
ia->ia_subnetmask = ntohl(ia->ia_sockmask.sin_addr.s_addr);
|
|
|
|
} else {
|
|
|
|
in_addr_t i = ntohl(addr->sin_addr.s_addr);
|
2009-04-25 23:02:57 +00:00
|
|
|
|
net/route.c:
A route generated from an RTF_CLONING route had the RTF_WASCLONED flag
set but did not have a reference to the parent route, as documented in
the rtentry(9) manpage. This prevented such routes from being deleted
when their parent route is deleted.
Now, for example, if you delete an IP address from a network interface,
all ARP entries that were cloned from this interface route are flushed.
This also has an impact on netstat(1) output. Previously, dynamically
created ARP cache entries (RTF_STATIC flag is unset) were displayed as
part of the routing table display (-r). Now, they are only printed if
the -a option is given.
netinet/in.c, netinet/in_rmx.c:
When address is removed from an interface, also delete all routes that
point to this interface and address. Previously, for example, if you
changed the address on an interface, outgoing IP datagrams might still
use the old address. The only solution was to delete and re-add some
routes. (The problem is easily observed with the route(8) command.)
Note, that if the socket was already bound to the local address before
this address is removed, new datagrams generated from this socket will
still be sent from the old address.
PR: kern/20785, kern/21914
Reviewed by: wollman (the idea)
2001-03-15 14:52:12 +00:00
|
|
|
/*
|
2013-11-05 07:44:15 +00:00
|
|
|
* Be compatible with network classes, if netmask isn't
|
|
|
|
* supplied, guess it based on classes.
|
|
|
|
*/
|
|
|
|
if (IN_CLASSA(i))
|
|
|
|
ia->ia_subnetmask = IN_CLASSA_NET;
|
|
|
|
else if (IN_CLASSB(i))
|
|
|
|
ia->ia_subnetmask = IN_CLASSB_NET;
|
|
|
|
else
|
|
|
|
ia->ia_subnetmask = IN_CLASSC_NET;
|
|
|
|
ia->ia_sockmask.sin_addr.s_addr = htonl(ia->ia_subnetmask);
|
|
|
|
}
|
|
|
|
ia->ia_subnet = ntohl(addr->sin_addr.s_addr) & ia->ia_subnetmask;
|
|
|
|
in_socktrim(&ia->ia_sockmask);
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2013-11-05 07:44:15 +00:00
|
|
|
if (ifp->if_flags & IFF_BROADCAST) {
|
|
|
|
if (broadaddr->sin_len != 0) {
|
|
|
|
ia->ia_broadaddr = *broadaddr;
|
|
|
|
} else if (ia->ia_subnetmask == IN_RFC3021_MASK) {
|
|
|
|
ia->ia_broadaddr.sin_addr.s_addr = INADDR_BROADCAST;
|
|
|
|
ia->ia_broadaddr.sin_len = sizeof(struct sockaddr_in);
|
|
|
|
ia->ia_broadaddr.sin_family = AF_INET;
|
|
|
|
} else {
|
|
|
|
ia->ia_broadaddr.sin_addr.s_addr =
|
|
|
|
htonl(ia->ia_subnet | ~ia->ia_subnetmask);
|
|
|
|
ia->ia_broadaddr.sin_len = sizeof(struct sockaddr_in);
|
|
|
|
ia->ia_broadaddr.sin_family = AF_INET;
|
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
2001-11-30 14:00:55 +00:00
|
|
|
|
2013-11-05 07:44:15 +00:00
|
|
|
if (ifp->if_flags & IFF_POINTOPOINT)
|
|
|
|
ia->ia_dstaddr = *dstaddr;
|
|
|
|
|
|
|
|
/* XXXGL: rtinit() needs this strange assignment. */
|
|
|
|
if (ifp->if_flags & IFF_LOOPBACK)
|
|
|
|
ia->ia_dstaddr = ia->ia_addr;
|
|
|
|
|
2014-07-31 16:43:56 +00:00
|
|
|
if (vhid != 0) {
|
|
|
|
error = (*carp_attach_p)(&ia->ia_ifa, vhid);
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
2014-01-16 12:35:18 +00:00
|
|
|
/* if_addrhead is already referenced by ifa_alloc() */
|
2013-11-05 07:44:15 +00:00
|
|
|
IF_ADDR_WLOCK(ifp);
|
|
|
|
TAILQ_INSERT_TAIL(&ifp->if_addrhead, ifa, ifa_link);
|
|
|
|
IF_ADDR_WUNLOCK(ifp);
|
|
|
|
|
|
|
|
ifa_ref(ifa); /* in_ifaddrhead */
|
|
|
|
IN_IFADDR_WLOCK();
|
|
|
|
TAILQ_INSERT_TAIL(&V_in_ifaddrhead, ia, ia_link);
|
|
|
|
LIST_INSERT_HEAD(INADDR_HASH(ia->ia_addr.sin_addr.s_addr), ia, ia_hash);
|
|
|
|
IN_IFADDR_WUNLOCK();
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Give the interface a chance to initialize
|
|
|
|
* if this is its first address,
|
|
|
|
* and to validate the address if necessary.
|
|
|
|
*/
|
2014-07-31 09:18:29 +00:00
|
|
|
if (ifp->if_ioctl != NULL) {
|
2013-11-05 07:44:15 +00:00
|
|
|
error = (*ifp->if_ioctl)(ifp, SIOCSIFADDR, (caddr_t)ia);
|
2014-07-31 09:18:29 +00:00
|
|
|
if (error)
|
2014-07-31 16:43:56 +00:00
|
|
|
goto fail1;
|
2014-07-31 09:18:29 +00:00
|
|
|
}
|
2013-11-05 07:44:15 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Add route for the network.
|
|
|
|
*/
|
|
|
|
if (vhid == 0) {
|
|
|
|
int flags = RTF_UP;
|
|
|
|
|
|
|
|
if (ifp->if_flags & (IFF_LOOPBACK|IFF_POINTOPOINT))
|
|
|
|
flags |= RTF_HOST;
|
|
|
|
|
|
|
|
error = in_addprefix(ia, flags);
|
|
|
|
if (error)
|
2014-07-31 16:43:56 +00:00
|
|
|
goto fail1;
|
2013-11-05 07:44:15 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Add a loopback route to self.
|
|
|
|
*/
|
|
|
|
if (vhid == 0 && (ifp->if_flags & IFF_LOOPBACK) == 0 &&
|
2013-11-17 15:14:07 +00:00
|
|
|
ia->ia_addr.sin_addr.s_addr != INADDR_ANY &&
|
|
|
|
!((ifp->if_flags & IFF_POINTOPOINT) &&
|
|
|
|
ia->ia_dstaddr.sin_addr.s_addr == ia->ia_addr.sin_addr.s_addr)) {
|
2013-11-05 07:44:15 +00:00
|
|
|
struct in_ifaddr *eia;
|
|
|
|
|
|
|
|
eia = in_localip_more(ia);
|
|
|
|
|
|
|
|
if (eia == NULL) {
|
|
|
|
error = ifa_add_loopback_route((struct ifaddr *)ia,
|
|
|
|
(struct sockaddr *)&ia->ia_addr);
|
|
|
|
if (error)
|
2014-07-31 16:43:56 +00:00
|
|
|
goto fail2;
|
2013-11-05 07:44:15 +00:00
|
|
|
} else
|
|
|
|
ifa_free(&eia->ia_ifa);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (iaIsFirst && (ifp->if_flags & IFF_MULTICAST)) {
|
|
|
|
struct in_addr allhosts_addr;
|
|
|
|
struct in_ifinfo *ii;
|
|
|
|
|
|
|
|
ii = ((struct in_ifinfo *)ifp->if_afdata[AF_INET]);
|
|
|
|
allhosts_addr.s_addr = htonl(INADDR_ALLHOSTS_GROUP);
|
|
|
|
|
|
|
|
error = in_joingroup(ifp, &allhosts_addr, NULL,
|
|
|
|
&ii->ii_allhosts);
|
|
|
|
}
|
|
|
|
|
|
|
|
EVENTHANDLER_INVOKE(ifaddr_event, ifp);
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
|
2014-07-31 16:43:56 +00:00
|
|
|
fail2:
|
2013-11-05 07:44:15 +00:00
|
|
|
if (vhid == 0)
|
|
|
|
(void )in_scrubprefix(ia, LLE_STATIC);
|
|
|
|
|
2014-07-31 16:43:56 +00:00
|
|
|
fail1:
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
if (ia->ia_ifa.ifa_carp)
|
2017-01-25 19:04:08 +00:00
|
|
|
(*carp_detach_p)(&ia->ia_ifa, false);
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
|
2013-11-05 07:44:15 +00:00
|
|
|
IF_ADDR_WLOCK(ifp);
|
|
|
|
TAILQ_REMOVE(&ifp->if_addrhead, &ia->ia_ifa, ifa_link);
|
|
|
|
IF_ADDR_WUNLOCK(ifp);
|
2014-01-16 12:35:18 +00:00
|
|
|
ifa_free(&ia->ia_ifa); /* if_addrhead */
|
2013-11-05 07:44:15 +00:00
|
|
|
|
|
|
|
IN_IFADDR_WLOCK();
|
|
|
|
TAILQ_REMOVE(&V_in_ifaddrhead, ia, ia_link);
|
|
|
|
LIST_REMOVE(ia, ia_hash);
|
|
|
|
IN_IFADDR_WUNLOCK();
|
2014-01-16 12:35:18 +00:00
|
|
|
ifa_free(&ia->ia_ifa); /* in_ifaddrhead */
|
2013-11-05 07:44:15 +00:00
|
|
|
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
2017-01-25 19:04:08 +00:00
|
|
|
in_difaddr_ioctl(u_long cmd, caddr_t data, struct ifnet *ifp, struct thread *td)
|
2013-11-05 07:44:15 +00:00
|
|
|
{
|
|
|
|
const struct ifreq *ifr = (struct ifreq *)data;
|
2013-11-06 01:14:00 +00:00
|
|
|
const struct sockaddr_in *addr = (const struct sockaddr_in *)
|
|
|
|
&ifr->ifr_addr;
|
2013-11-05 07:44:15 +00:00
|
|
|
struct ifaddr *ifa;
|
|
|
|
struct in_ifaddr *ia;
|
|
|
|
bool deleteAny, iaIsLast;
|
|
|
|
int error;
|
|
|
|
|
|
|
|
if (td != NULL) {
|
|
|
|
error = priv_check(td, PRIV_NET_DELIFADDR);
|
|
|
|
if (error)
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (addr->sin_len != sizeof(struct sockaddr_in) ||
|
|
|
|
addr->sin_family != AF_INET)
|
|
|
|
deleteAny = true;
|
|
|
|
else
|
|
|
|
deleteAny = false;
|
|
|
|
|
|
|
|
iaIsLast = true;
|
|
|
|
ia = NULL;
|
2012-01-05 19:00:36 +00:00
|
|
|
IF_ADDR_WLOCK(ifp);
|
2010-10-16 19:53:22 +00:00
|
|
|
TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
|
2013-12-29 22:20:06 +00:00
|
|
|
struct in_ifaddr *it;
|
2013-11-05 07:44:15 +00:00
|
|
|
|
2013-12-29 22:20:06 +00:00
|
|
|
if (ifa->ifa_addr->sa_family != AF_INET)
|
2013-11-05 07:44:15 +00:00
|
|
|
continue;
|
|
|
|
|
2013-12-29 22:20:06 +00:00
|
|
|
it = (struct in_ifaddr *)ifa;
|
2013-11-05 07:44:15 +00:00
|
|
|
if (deleteAny && ia == NULL && (td == NULL ||
|
|
|
|
prison_check_ip4(td->td_ucred, &it->ia_addr.sin_addr) == 0))
|
|
|
|
ia = it;
|
|
|
|
|
|
|
|
if (it->ia_addr.sin_addr.s_addr == addr->sin_addr.s_addr &&
|
|
|
|
(td == NULL || prison_check_ip4(td->td_ucred,
|
|
|
|
&addr->sin_addr) == 0))
|
|
|
|
ia = it;
|
|
|
|
|
|
|
|
if (it != ia)
|
|
|
|
iaIsLast = false;
|
2010-10-16 19:53:22 +00:00
|
|
|
}
|
2013-11-05 07:44:15 +00:00
|
|
|
|
|
|
|
if (ia == NULL) {
|
2012-01-05 19:00:36 +00:00
|
|
|
IF_ADDR_WUNLOCK(ifp);
|
2013-11-05 07:44:15 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
2010-10-16 19:53:22 +00:00
|
|
|
}
|
2013-11-05 07:44:15 +00:00
|
|
|
|
2001-11-30 14:00:55 +00:00
|
|
|
TAILQ_REMOVE(&ifp->if_addrhead, &ia->ia_ifa, ifa_link);
|
2012-01-05 19:00:36 +00:00
|
|
|
IF_ADDR_WUNLOCK(ifp);
|
2013-11-05 07:44:15 +00:00
|
|
|
ifa_free(&ia->ia_ifa); /* if_addrhead */
|
2009-06-25 11:52:33 +00:00
|
|
|
|
|
|
|
IN_IFADDR_WLOCK();
|
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
|
|
|
TAILQ_REMOVE(&V_in_ifaddrhead, ia, ia_link);
|
2011-11-21 14:10:13 +00:00
|
|
|
LIST_REMOVE(ia, ia_hash);
|
|
|
|
IN_IFADDR_WUNLOCK();
|
2013-11-05 07:44:15 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* in_scrubprefix() kills the interface route.
|
|
|
|
*/
|
|
|
|
in_scrubprefix(ia, LLE_STATIC);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* in_ifadown gets rid of all the rest of
|
|
|
|
* the routes. This is not quite the right
|
|
|
|
* thing to do, but at least if we are running
|
|
|
|
* a routing process they will come back.
|
|
|
|
*/
|
|
|
|
in_ifadown(&ia->ia_ifa, 1);
|
|
|
|
|
|
|
|
if (ia->ia_ifa.ifa_carp)
|
2017-01-25 19:04:08 +00:00
|
|
|
(*carp_detach_p)(&ia->ia_ifa,
|
|
|
|
(cmd == SIOCDIFADDR) ? false : true);
|
2013-11-05 07:44:15 +00:00
|
|
|
|
2011-11-21 14:10:13 +00:00
|
|
|
/*
|
|
|
|
* If this is the last IPv4 address configured on this
|
|
|
|
* interface, leave the all-hosts group.
|
|
|
|
* No state-change report need be transmitted.
|
|
|
|
*/
|
2013-11-05 07:44:15 +00:00
|
|
|
if (iaIsLast && (ifp->if_flags & IFF_MULTICAST)) {
|
|
|
|
struct in_ifinfo *ii;
|
|
|
|
|
2011-11-21 14:10:13 +00:00
|
|
|
ii = ((struct in_ifinfo *)ifp->if_afdata[AF_INET]);
|
|
|
|
IN_MULTI_LOCK();
|
|
|
|
if (ii->ii_allhosts) {
|
|
|
|
(void)in_leavegroup_locked(ii->ii_allhosts, NULL);
|
|
|
|
ii->ii_allhosts = NULL;
|
|
|
|
}
|
|
|
|
IN_MULTI_UNLOCK();
|
2013-11-05 07:44:15 +00:00
|
|
|
}
|
2011-11-21 14:10:13 +00:00
|
|
|
|
Add GARP retransmit capability
A single gratuitous ARP (GARP) is always transmitted when an IPv4
address is added to an interface, and that is usually sufficient.
However, in some circumstances, such as when a shared address is
passed between cluster nodes, this single GARP may occasionally be
dropped or lost. This can lead to neighbors on the network link
working with a stale ARP cache and sending packets destined for
that address to the node that previously owned the address, which
may not respond.
To avoid this situation, GARP retransmissions can be enabled by setting
the net.link.ether.inet.garp_rexmit_count sysctl to a value greater
than zero. The setting represents the maximum number of retransmissions.
The interval between retransmissions is calculated using an exponential
backoff algorithm, doubling each time, so the retransmission intervals
are: {1, 2, 4, 8, 16, ...} (seconds).
Due to the exponential backoff algorithm used for the interval
between GARP retransmissions, the maximum number of retransmissions
is limited to 16 for sanity. This limit corresponds to a maximum
interval between retransmissions of 2^16 seconds ~= 18 hours.
Increasing this limit is possible, but sending out GARPs spaced
days apart would be of little use.
Submitted by: David A. Bright <david.a.bright@dell.com>
MFC after: 1 month
Relnotes: yes
Sponsored by: Dell EMC
Differential Revision: https://reviews.freebsd.org/D7695
2016-10-02 01:42:45 +00:00
|
|
|
IF_ADDR_WLOCK(ifp);
|
|
|
|
if (callout_stop(&ia->ia_garp_timer) == 1) {
|
|
|
|
ifa_free(&ia->ia_ifa);
|
|
|
|
}
|
|
|
|
IF_ADDR_WUNLOCK(ifp);
|
|
|
|
|
2013-11-05 07:44:15 +00:00
|
|
|
EVENTHANDLER_INVOKE(ifaddr_event, ifp);
|
2014-01-16 12:35:18 +00:00
|
|
|
ifa_free(&ia->ia_ifa); /* in_ifaddrhead */
|
2013-11-05 07:44:15 +00:00
|
|
|
|
|
|
|
return (0);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
2004-11-12 20:53:51 +00:00
|
|
|
#define rtinitflags(x) \
|
|
|
|
((((x)->ia_ifp->if_flags & (IFF_LOOPBACK | IFF_POINTOPOINT)) != 0) \
|
|
|
|
? RTF_HOST : 0)
|
2009-12-30 22:13:01 +00:00
|
|
|
|
2004-11-12 20:53:51 +00:00
|
|
|
/*
|
2007-06-17 00:31:24 +00:00
|
|
|
* Check if we have a route for the given prefix already or add one accordingly.
|
2004-11-12 20:53:51 +00:00
|
|
|
*/
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
int
|
2007-05-10 15:58:48 +00:00
|
|
|
in_addprefix(struct in_ifaddr *target, int flags)
|
2004-11-12 20:53:51 +00:00
|
|
|
{
|
2015-07-29 08:12:05 +00:00
|
|
|
struct rm_priotracker in_ifa_tracker;
|
2004-11-12 20:53:51 +00:00
|
|
|
struct in_ifaddr *ia;
|
2005-10-22 14:50:27 +00:00
|
|
|
struct in_addr prefix, mask, p, m;
|
2014-04-24 23:56:56 +00:00
|
|
|
int error;
|
2004-11-12 20:53:51 +00:00
|
|
|
|
2007-06-17 00:31:24 +00:00
|
|
|
if ((flags & RTF_HOST) != 0) {
|
2004-11-12 20:53:51 +00:00
|
|
|
prefix = target->ia_dstaddr.sin_addr;
|
2007-06-17 00:31:24 +00:00
|
|
|
mask.s_addr = 0;
|
|
|
|
} else {
|
2004-11-12 20:53:51 +00:00
|
|
|
prefix = target->ia_addr.sin_addr;
|
|
|
|
mask = target->ia_sockmask.sin_addr;
|
|
|
|
prefix.s_addr &= mask.s_addr;
|
|
|
|
}
|
|
|
|
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RLOCK(&in_ifa_tracker);
|
2014-04-24 23:56:56 +00:00
|
|
|
/* Look for an existing address with the same prefix, mask, and fib */
|
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
|
|
|
TAILQ_FOREACH(ia, &V_in_ifaddrhead, ia_link) {
|
2005-10-22 14:50:27 +00:00
|
|
|
if (rtinitflags(ia)) {
|
2011-08-10 06:17:06 +00:00
|
|
|
p = ia->ia_dstaddr.sin_addr;
|
2004-11-12 20:53:51 +00:00
|
|
|
|
2005-10-22 14:50:27 +00:00
|
|
|
if (prefix.s_addr != p.s_addr)
|
|
|
|
continue;
|
|
|
|
} else {
|
|
|
|
p = ia->ia_addr.sin_addr;
|
|
|
|
m = ia->ia_sockmask.sin_addr;
|
|
|
|
p.s_addr &= m.s_addr;
|
|
|
|
|
|
|
|
if (prefix.s_addr != p.s_addr ||
|
|
|
|
mask.s_addr != m.s_addr)
|
|
|
|
continue;
|
|
|
|
}
|
2014-04-24 23:56:56 +00:00
|
|
|
if (target->ia_ifp->if_fib != ia->ia_ifp->if_fib)
|
|
|
|
continue;
|
2004-11-12 20:53:51 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If we got a matching prefix route inserted by other
|
|
|
|
* interface address, we are done here.
|
|
|
|
*/
|
2005-08-18 10:34:30 +00:00
|
|
|
if (ia->ia_flags & IFA_ROUTE) {
|
2010-03-09 01:11:45 +00:00
|
|
|
#ifdef RADIX_MPATH
|
2012-07-31 11:31:12 +00:00
|
|
|
if (ia->ia_addr.sin_addr.s_addr ==
|
2010-09-04 16:06:01 +00:00
|
|
|
target->ia_addr.sin_addr.s_addr) {
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2010-03-09 01:11:45 +00:00
|
|
|
return (EEXIST);
|
2010-09-04 16:06:01 +00:00
|
|
|
} else
|
2010-03-09 01:11:45 +00:00
|
|
|
break;
|
|
|
|
#endif
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
if (V_nosameprefix) {
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2005-08-18 10:34:30 +00:00
|
|
|
return (EEXIST);
|
2009-06-25 11:52:33 +00:00
|
|
|
} else {
|
2014-04-24 23:56:56 +00:00
|
|
|
int fibnum;
|
|
|
|
|
2014-09-21 04:38:50 +00:00
|
|
|
fibnum = V_rt_add_addr_allfibs ? RT_ALL_FIBS :
|
2014-04-24 23:56:56 +00:00
|
|
|
target->ia_ifp->if_fib;
|
Simplify inet alias handling code: if we're adding/removing alias which
has the same prefix as some other alias on the same interface, use
newly-added rt_addrmsg() instead of hand-rolled in_addralias_rtmsg().
This eliminates the following rtsock messages:
Pinned RTM_ADD for prefix (for alias addition).
Pinned RTM_DELETE for prefix (for alias withdrawal).
Example (got 10.0.0.1/24 on vlan4, playing with 10.0.0.2/24):
before commit, addition:
got message of size 116 on Fri Jan 10 14:13:15 2014
RTM_NEWADDR: address being added to iface: len 116, metric 0, flags:
sockaddrs: <NETMASK,IFP,IFA,BRD>
255.255.255.0 vlan4:8.0.27.c5.29.d4 10.0.0.2 10.0.0.255
got message of size 192 on Fri Jan 10 14:13:15 2014
RTM_ADD: Add Route: len 192, pid: 0, seq 0, errno 0, flags:<UP,PINNED>
locks: inits:
sockaddrs: <DST,GATEWAY,NETMASK>
10.0.0.0 10.0.0.2 (255) ffff ffff ff
after commit, addition:
got message of size 116 on Fri Jan 10 13:56:26 2014
RTM_NEWADDR: address being added to iface: len 116, metric 0, flags:
sockaddrs: <NETMASK,IFP,IFA,BRD>
255.255.255.0 vlan4:8.0.27.c5.29.d4 14.0.0.2 14.0.0.255
before commit, wihdrawal:
got message of size 192 on Fri Jan 10 13:58:59 2014
RTM_DELETE: Delete Route: len 192, pid: 0, seq 0, errno 0, flags:<UP,PINNED>
locks: inits:
sockaddrs: <DST,GATEWAY,NETMASK>
10.0.0.0 10.0.0.2 (255) ffff ffff ff
got message of size 116 on Fri Jan 10 13:58:59 2014
RTM_DELADDR: address being removed from iface: len 116, metric 0, flags:
sockaddrs: <NETMASK,IFP,IFA,BRD>
255.255.255.0 vlan4:8.0.27.c5.29.d4 10.0.0.2 10.0.0.255
adter commit, withdrawal:
got message of size 116 on Fri Jan 10 14:14:11 2014
RTM_DELADDR: address being removed from iface: len 116, metric 0, flags:
sockaddrs: <NETMASK,IFP,IFA,BRD>
255.255.255.0 vlan4:8.0.27.c5.29.d4 10.0.0.2 10.0.0.255
Sending both RTM_ADD/RTM_DELETE messages to rtsock is completely wrong
(and requires some hacks to keep prefix in route table on RTM_DELETE).
I've tested this change with quagga (no change) and bird (*).
bird alias handling is already broken in *BSD sysdep code, so nothing
changes here, too.
I'm going to MFC this change if there will be no complains about behavior
change.
While here, fix some style(9) bugs introduced by r260488
(pointed by glebius and bde).
Sponsored by: Yandex LLC
MFC after: 4 weeks
2014-01-10 12:13:55 +00:00
|
|
|
rt_addrmsg(RTM_ADD, &target->ia_ifa, fibnum);
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2005-08-18 10:34:30 +00:00
|
|
|
return (0);
|
2009-06-25 11:52:33 +00:00
|
|
|
}
|
2005-08-18 10:34:30 +00:00
|
|
|
}
|
2004-11-12 20:53:51 +00:00
|
|
|
}
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2004-11-12 20:53:51 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* No-one seem to have this prefix route, so we try to insert it.
|
|
|
|
*/
|
|
|
|
error = rtinit(&target->ia_ifa, (int)RTM_ADD, flags);
|
|
|
|
if (!error)
|
|
|
|
target->ia_flags |= IFA_ROUTE;
|
2008-10-26 19:17:25 +00:00
|
|
|
return (error);
|
2004-11-12 20:53:51 +00:00
|
|
|
}
|
|
|
|
|
2015-09-14 16:48:19 +00:00
|
|
|
/*
|
|
|
|
* Removes either all lle entries for given @ia, or lle
|
|
|
|
* corresponding to @ia address.
|
|
|
|
*/
|
|
|
|
static void
|
|
|
|
in_scrubprefixlle(struct in_ifaddr *ia, int all, u_int flags)
|
|
|
|
{
|
|
|
|
struct sockaddr_in addr, mask;
|
|
|
|
struct sockaddr *saddr, *smask;
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
|
|
|
saddr = (struct sockaddr *)&addr;
|
|
|
|
bzero(&addr, sizeof(addr));
|
|
|
|
addr.sin_len = sizeof(addr);
|
|
|
|
addr.sin_family = AF_INET;
|
|
|
|
smask = (struct sockaddr *)&mask;
|
|
|
|
bzero(&mask, sizeof(mask));
|
|
|
|
mask.sin_len = sizeof(mask);
|
|
|
|
mask.sin_family = AF_INET;
|
|
|
|
mask.sin_addr.s_addr = ia->ia_subnetmask;
|
|
|
|
ifp = ia->ia_ifp;
|
|
|
|
|
2015-10-18 12:26:25 +00:00
|
|
|
if (all) {
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Remove all L2 entries matching given prefix.
|
|
|
|
* Convert address to host representation to avoid
|
|
|
|
* doing this on every callback. ia_subnetmask is already
|
|
|
|
* stored in host representation.
|
|
|
|
*/
|
|
|
|
addr.sin_addr.s_addr = ntohl(ia->ia_addr.sin_addr.s_addr);
|
2015-09-14 16:48:19 +00:00
|
|
|
lltable_prefix_free(AF_INET, saddr, smask, flags);
|
2015-10-18 12:26:25 +00:00
|
|
|
} else {
|
|
|
|
/* Remove interface address only */
|
|
|
|
addr.sin_addr.s_addr = ia->ia_addr.sin_addr.s_addr;
|
2015-09-14 16:48:19 +00:00
|
|
|
lltable_delete_addr(LLTABLE(ifp), LLE_IFADDR, saddr);
|
2015-10-18 12:26:25 +00:00
|
|
|
}
|
2015-09-14 16:48:19 +00:00
|
|
|
}
|
|
|
|
|
2004-11-12 20:53:51 +00:00
|
|
|
/*
|
|
|
|
* If there is no other address in the system that can serve a route to the
|
|
|
|
* same prefix, remove the route. Hand over the route to the new address
|
|
|
|
* otherwise.
|
|
|
|
*/
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
int
|
2011-05-20 19:12:20 +00:00
|
|
|
in_scrubprefix(struct in_ifaddr *target, u_int flags)
|
2004-11-12 20:53:51 +00:00
|
|
|
{
|
2015-07-29 08:12:05 +00:00
|
|
|
struct rm_priotracker in_ifa_tracker;
|
2004-11-12 20:53:51 +00:00
|
|
|
struct in_ifaddr *ia;
|
2011-12-13 06:56:43 +00:00
|
|
|
struct in_addr prefix, mask, p, m;
|
2014-04-29 14:46:45 +00:00
|
|
|
int error = 0;
|
2004-11-12 20:53:51 +00:00
|
|
|
|
2009-07-27 17:08:06 +00:00
|
|
|
/*
|
|
|
|
* Remove the loopback route to the interface address.
|
|
|
|
*/
|
2009-05-14 05:27:09 +00:00
|
|
|
if ((target->ia_addr.sin_addr.s_addr != INADDR_ANY) &&
|
2009-12-30 21:35:34 +00:00
|
|
|
!(target->ia_ifp->if_flags & IFF_LOOPBACK) &&
|
2013-11-05 07:44:15 +00:00
|
|
|
(flags & LLE_STATIC)) {
|
|
|
|
struct in_ifaddr *eia;
|
|
|
|
|
2015-09-16 06:23:15 +00:00
|
|
|
/*
|
|
|
|
* XXXME: add fib-aware in_localip.
|
|
|
|
* We definitely don't want to switch between
|
|
|
|
* prefixes in different fibs.
|
|
|
|
*/
|
2013-11-05 07:44:15 +00:00
|
|
|
eia = in_localip_more(target);
|
|
|
|
|
|
|
|
if (eia != NULL) {
|
|
|
|
error = ifa_switch_loopback_route((struct ifaddr *)eia,
|
2015-09-16 06:23:15 +00:00
|
|
|
(struct sockaddr *)&target->ia_addr);
|
2013-11-05 07:44:15 +00:00
|
|
|
ifa_free(&eia->ia_ifa);
|
|
|
|
} else {
|
2009-12-30 21:35:34 +00:00
|
|
|
error = ifa_del_loopback_route((struct ifaddr *)target,
|
2012-07-31 11:31:12 +00:00
|
|
|
(struct sockaddr *)&target->ia_addr);
|
2011-05-20 19:12:20 +00:00
|
|
|
}
|
2009-05-12 07:41:20 +00:00
|
|
|
}
|
|
|
|
|
2011-12-13 06:56:43 +00:00
|
|
|
if (rtinitflags(target)) {
|
2004-11-12 20:53:51 +00:00
|
|
|
prefix = target->ia_dstaddr.sin_addr;
|
2011-12-13 06:56:43 +00:00
|
|
|
mask.s_addr = 0;
|
|
|
|
} else {
|
2004-11-12 20:53:51 +00:00
|
|
|
prefix = target->ia_addr.sin_addr;
|
|
|
|
mask = target->ia_sockmask.sin_addr;
|
|
|
|
prefix.s_addr &= mask.s_addr;
|
|
|
|
}
|
|
|
|
|
2009-12-30 22:13:01 +00:00
|
|
|
if ((target->ia_flags & IFA_ROUTE) == 0) {
|
2014-04-29 14:46:45 +00:00
|
|
|
int fibnum;
|
|
|
|
|
2014-09-21 04:38:50 +00:00
|
|
|
fibnum = V_rt_add_addr_allfibs ? RT_ALL_FIBS :
|
2014-04-29 14:46:45 +00:00
|
|
|
target->ia_ifp->if_fib;
|
Simplify inet alias handling code: if we're adding/removing alias which
has the same prefix as some other alias on the same interface, use
newly-added rt_addrmsg() instead of hand-rolled in_addralias_rtmsg().
This eliminates the following rtsock messages:
Pinned RTM_ADD for prefix (for alias addition).
Pinned RTM_DELETE for prefix (for alias withdrawal).
Example (got 10.0.0.1/24 on vlan4, playing with 10.0.0.2/24):
before commit, addition:
got message of size 116 on Fri Jan 10 14:13:15 2014
RTM_NEWADDR: address being added to iface: len 116, metric 0, flags:
sockaddrs: <NETMASK,IFP,IFA,BRD>
255.255.255.0 vlan4:8.0.27.c5.29.d4 10.0.0.2 10.0.0.255
got message of size 192 on Fri Jan 10 14:13:15 2014
RTM_ADD: Add Route: len 192, pid: 0, seq 0, errno 0, flags:<UP,PINNED>
locks: inits:
sockaddrs: <DST,GATEWAY,NETMASK>
10.0.0.0 10.0.0.2 (255) ffff ffff ff
after commit, addition:
got message of size 116 on Fri Jan 10 13:56:26 2014
RTM_NEWADDR: address being added to iface: len 116, metric 0, flags:
sockaddrs: <NETMASK,IFP,IFA,BRD>
255.255.255.0 vlan4:8.0.27.c5.29.d4 14.0.0.2 14.0.0.255
before commit, wihdrawal:
got message of size 192 on Fri Jan 10 13:58:59 2014
RTM_DELETE: Delete Route: len 192, pid: 0, seq 0, errno 0, flags:<UP,PINNED>
locks: inits:
sockaddrs: <DST,GATEWAY,NETMASK>
10.0.0.0 10.0.0.2 (255) ffff ffff ff
got message of size 116 on Fri Jan 10 13:58:59 2014
RTM_DELADDR: address being removed from iface: len 116, metric 0, flags:
sockaddrs: <NETMASK,IFP,IFA,BRD>
255.255.255.0 vlan4:8.0.27.c5.29.d4 10.0.0.2 10.0.0.255
adter commit, withdrawal:
got message of size 116 on Fri Jan 10 14:14:11 2014
RTM_DELADDR: address being removed from iface: len 116, metric 0, flags:
sockaddrs: <NETMASK,IFP,IFA,BRD>
255.255.255.0 vlan4:8.0.27.c5.29.d4 10.0.0.2 10.0.0.255
Sending both RTM_ADD/RTM_DELETE messages to rtsock is completely wrong
(and requires some hacks to keep prefix in route table on RTM_DELETE).
I've tested this change with quagga (no change) and bird (*).
bird alias handling is already broken in *BSD sysdep code, so nothing
changes here, too.
I'm going to MFC this change if there will be no complains about behavior
change.
While here, fix some style(9) bugs introduced by r260488
(pointed by glebius and bde).
Sponsored by: Yandex LLC
MFC after: 4 weeks
2014-01-10 12:13:55 +00:00
|
|
|
rt_addrmsg(RTM_DELETE, &target->ia_ifa, fibnum);
|
2015-10-18 12:26:25 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Removing address from !IFF_UP interface or
|
|
|
|
* prefix which exists on other interface (along with route).
|
|
|
|
* No entries should exist here except target addr.
|
|
|
|
* Given that, delete this entry only.
|
|
|
|
*/
|
|
|
|
in_scrubprefixlle(target, 0, flags);
|
2009-12-30 22:13:01 +00:00
|
|
|
return (0);
|
|
|
|
}
|
|
|
|
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RLOCK(&in_ifa_tracker);
|
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
|
|
|
TAILQ_FOREACH(ia, &V_in_ifaddrhead, ia_link) {
|
2011-12-13 06:56:43 +00:00
|
|
|
if (rtinitflags(ia)) {
|
2004-11-12 20:53:51 +00:00
|
|
|
p = ia->ia_dstaddr.sin_addr;
|
2011-12-13 06:56:43 +00:00
|
|
|
|
|
|
|
if (prefix.s_addr != p.s_addr)
|
|
|
|
continue;
|
|
|
|
} else {
|
2004-11-12 20:53:51 +00:00
|
|
|
p = ia->ia_addr.sin_addr;
|
2011-12-13 06:56:43 +00:00
|
|
|
m = ia->ia_sockmask.sin_addr;
|
|
|
|
p.s_addr &= m.s_addr;
|
|
|
|
|
|
|
|
if (prefix.s_addr != p.s_addr ||
|
|
|
|
mask.s_addr != m.s_addr)
|
|
|
|
continue;
|
2004-11-12 20:53:51 +00:00
|
|
|
}
|
|
|
|
|
2011-12-13 06:56:43 +00:00
|
|
|
if ((ia->ia_ifp->if_flags & IFF_UP) == 0)
|
2004-11-12 20:53:51 +00:00
|
|
|
continue;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If we got a matching prefix address, move IFA_ROUTE and
|
|
|
|
* the route itself to it. Make sure that routing daemons
|
|
|
|
* get a heads-up.
|
|
|
|
*/
|
A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.
The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.
ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.
To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]
The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.
Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!
PR: kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by: bz
Submitted by: bz [1]
2011-12-16 12:16:56 +00:00
|
|
|
if ((ia->ia_flags & IFA_ROUTE) == 0) {
|
2011-03-21 14:19:40 +00:00
|
|
|
ifa_ref(&ia->ia_ifa);
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2011-05-29 02:21:35 +00:00
|
|
|
error = rtinit(&(target->ia_ifa), (int)RTM_DELETE,
|
2004-11-12 20:53:51 +00:00
|
|
|
rtinitflags(target));
|
2011-05-29 02:21:35 +00:00
|
|
|
if (error == 0)
|
|
|
|
target->ia_flags &= ~IFA_ROUTE;
|
|
|
|
else
|
|
|
|
log(LOG_INFO, "in_scrubprefix: err=%d, old prefix delete failed\n",
|
|
|
|
error);
|
2015-09-14 16:48:19 +00:00
|
|
|
/* Scrub all entries IFF interface is different */
|
|
|
|
in_scrubprefixlle(target, target->ia_ifp != ia->ia_ifp,
|
|
|
|
flags);
|
2004-11-12 20:53:51 +00:00
|
|
|
error = rtinit(&ia->ia_ifa, (int)RTM_ADD,
|
|
|
|
rtinitflags(ia) | RTF_UP);
|
|
|
|
if (error == 0)
|
|
|
|
ia->ia_flags |= IFA_ROUTE;
|
2011-05-29 02:21:35 +00:00
|
|
|
else
|
|
|
|
log(LOG_INFO, "in_scrubprefix: err=%d, new prefix add failed\n",
|
|
|
|
error);
|
2011-03-21 14:19:40 +00:00
|
|
|
ifa_free(&ia->ia_ifa);
|
2008-10-26 19:17:25 +00:00
|
|
|
return (error);
|
2004-11-12 20:53:51 +00:00
|
|
|
}
|
|
|
|
}
|
2015-07-29 08:12:05 +00:00
|
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
2004-11-12 20:53:51 +00:00
|
|
|
|
2009-05-20 21:07:15 +00:00
|
|
|
/*
|
|
|
|
* remove all L2 entries on the given prefix
|
|
|
|
*/
|
2015-09-14 16:48:19 +00:00
|
|
|
in_scrubprefixlle(target, 1, flags);
|
2009-05-20 21:07:15 +00:00
|
|
|
|
2004-11-12 20:53:51 +00:00
|
|
|
/*
|
|
|
|
* As no-one seem to have this prefix, we can remove the route.
|
|
|
|
*/
|
2011-05-29 02:21:35 +00:00
|
|
|
error = rtinit(&(target->ia_ifa), (int)RTM_DELETE, rtinitflags(target));
|
|
|
|
if (error == 0)
|
|
|
|
target->ia_flags &= ~IFA_ROUTE;
|
|
|
|
else
|
|
|
|
log(LOG_INFO, "in_scrubprefix: err=%d, prefix delete failed\n", error);
|
|
|
|
return (error);
|
2004-11-12 20:53:51 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
#undef rtinitflags
|
1994-05-24 10:09:53 +00:00
|
|
|
|
Get closer to a VIMAGE network stack teardown from top to bottom rather
than removing the network interfaces first. This change is rather larger
and convoluted as the ordering requirements cannot be separated.
Move the pfil(9) framework to SI_SUB_PROTO_PFIL, move Firewalls and
related modules to their own SI_SUB_PROTO_FIREWALL.
Move initialization of "physical" interfaces to SI_SUB_DRIVERS,
move virtual (cloned) interfaces to SI_SUB_PSEUDO.
Move Multicast to SI_SUB_PROTO_MC.
Re-work parts of multicast initialisation and teardown, not taking the
huge amount of memory into account if used as a module yet.
For interface teardown we try to do as many of them as we can on
SI_SUB_INIT_IF, but for some this makes no sense, e.g., when tunnelling
over a higher layer protocol such as IP. In that case the interface
has to go along (or before) the higher layer protocol is shutdown.
Kernel hhooks need to go last on teardown as they may be used at various
higher layers and we cannot remove them before we cleaned up the higher
layers.
For interface teardown there are multiple paths:
(a) a cloned interface is destroyed (inside a VIMAGE or in the base system),
(b) any interface is moved from a virtual network stack to a different
network stack ("vmove"), or (c) a virtual network stack is being shut down.
All code paths go through if_detach_internal() where we, depending on the
vmove flag or the vnet state, make a decision on how much to shut down;
in case we are destroying a VNET the individual protocol layers will
cleanup their own parts thus we cannot do so again for each interface as
we end up with, e.g., double-frees, destroying locks twice or acquiring
already destroyed locks.
When calling into protocol cleanups we equally have to tell them
whether they need to detach upper layer protocols ("ulp") or not
(e.g., in6_ifdetach()).
Provide or enahnce helper functions to do proper cleanup at a protocol
rather than at an interface level.
Approved by: re (hrs)
Obtained from: projects/vnet
Reviewed by: gnn, jhb
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D6747
2016-06-21 13:48:49 +00:00
|
|
|
void
|
|
|
|
in_ifscrub_all(void)
|
|
|
|
{
|
|
|
|
struct ifnet *ifp;
|
|
|
|
struct ifaddr *ifa, *nifa;
|
|
|
|
struct ifaliasreq ifr;
|
|
|
|
|
|
|
|
IFNET_RLOCK();
|
|
|
|
TAILQ_FOREACH(ifp, &V_ifnet, if_link) {
|
|
|
|
/* Cannot lock here - lock recursion. */
|
|
|
|
/* IF_ADDR_RLOCK(ifp); */
|
|
|
|
TAILQ_FOREACH_SAFE(ifa, &ifp->if_addrhead, ifa_link, nifa) {
|
|
|
|
if (ifa->ifa_addr->sa_family != AF_INET)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This is ugly but the only way for legacy IP to
|
|
|
|
* cleanly remove addresses and everything attached.
|
|
|
|
*/
|
|
|
|
bzero(&ifr, sizeof(ifr));
|
|
|
|
ifr.ifra_addr = *ifa->ifa_addr;
|
|
|
|
if (ifa->ifa_dstaddr)
|
|
|
|
ifr.ifra_broadaddr = *ifa->ifa_dstaddr;
|
|
|
|
(void)in_control(NULL, SIOCDIFADDR, (caddr_t)&ifr,
|
|
|
|
ifp, NULL);
|
|
|
|
}
|
|
|
|
/* IF_ADDR_RUNLOCK(ifp); */
|
|
|
|
in_purgemaddrs(ifp);
|
|
|
|
igmp_domifdetach(ifp);
|
|
|
|
}
|
|
|
|
IFNET_RUNLOCK();
|
|
|
|
}
|
|
|
|
|
2016-08-18 22:59:00 +00:00
|
|
|
int
|
|
|
|
in_ifaddr_broadcast(struct in_addr in, struct in_ifaddr *ia)
|
|
|
|
{
|
|
|
|
|
|
|
|
return ((in.s_addr == ia->ia_broadaddr.sin_addr.s_addr ||
|
|
|
|
/*
|
|
|
|
* Check for old-style (host 0) broadcast, but
|
|
|
|
* taking into account that RFC 3021 obsoletes it.
|
|
|
|
*/
|
|
|
|
(ia->ia_subnetmask != IN_RFC3021_MASK &&
|
|
|
|
ntohl(in.s_addr) == ia->ia_subnet)) &&
|
|
|
|
/*
|
|
|
|
* Check for an all one subnetmask. These
|
|
|
|
* only exist when an interface gets a secondary
|
|
|
|
* address.
|
|
|
|
*/
|
|
|
|
ia->ia_subnetmask != (u_long)0xffffffff);
|
|
|
|
}
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* Return 1 if the address might be a local broadcast address.
|
|
|
|
*/
|
1994-05-25 09:21:21 +00:00
|
|
|
int
|
2007-05-10 15:58:48 +00:00
|
|
|
in_broadcast(struct in_addr in, struct ifnet *ifp)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2017-05-17 00:34:34 +00:00
|
|
|
struct ifaddr *ifa;
|
2016-08-18 22:59:10 +00:00
|
|
|
int found;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
if (in.s_addr == INADDR_BROADCAST ||
|
|
|
|
in.s_addr == INADDR_ANY)
|
2008-10-26 19:17:25 +00:00
|
|
|
return (1);
|
1994-05-24 10:09:53 +00:00
|
|
|
if ((ifp->if_flags & IFF_BROADCAST) == 0)
|
2008-10-26 19:17:25 +00:00
|
|
|
return (0);
|
2016-08-18 22:59:10 +00:00
|
|
|
found = 0;
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* Look through the list of addresses for a match
|
|
|
|
* with a broadcast address.
|
|
|
|
*/
|
2016-08-18 22:59:10 +00:00
|
|
|
IF_ADDR_RLOCK(ifp);
|
2001-03-16 20:00:53 +00:00
|
|
|
TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link)
|
1994-05-24 10:09:53 +00:00
|
|
|
if (ifa->ifa_addr->sa_family == AF_INET &&
|
2016-08-18 22:59:10 +00:00
|
|
|
in_ifaddr_broadcast(in, (struct in_ifaddr *)ifa)) {
|
|
|
|
found = 1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
IF_ADDR_RUNLOCK(ifp);
|
|
|
|
return (found);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
2007-03-20 00:36:10 +00:00
|
|
|
|
2009-03-09 17:53:05 +00:00
|
|
|
/*
|
|
|
|
* On interface removal, clean up IPv4 data structures hung off of the ifnet.
|
|
|
|
*/
|
|
|
|
void
|
|
|
|
in_ifdetach(struct ifnet *ifp)
|
|
|
|
{
|
|
|
|
|
|
|
|
in_pcbpurgeif0(&V_ripcbinfo, ifp);
|
|
|
|
in_pcbpurgeif0(&V_udbinfo, ifp);
|
2014-04-07 01:53:03 +00:00
|
|
|
in_pcbpurgeif0(&V_ulitecbinfo, ifp);
|
2009-03-09 17:53:05 +00:00
|
|
|
in_purgemaddrs(ifp);
|
|
|
|
}
|
|
|
|
|
2006-09-28 10:04:07 +00:00
|
|
|
/*
|
2007-03-20 00:36:10 +00:00
|
|
|
* Delete all IPv4 multicast address records, and associated link-layer
|
|
|
|
* multicast address records, associated with ifp.
|
2009-03-09 17:53:05 +00:00
|
|
|
* XXX It looks like domifdetach runs AFTER the link layer cleanup.
|
2009-03-17 14:41:54 +00:00
|
|
|
* XXX This should not race with ifma_protospec being set during
|
|
|
|
* a new allocation, if it does, we have bigger problems.
|
2006-09-28 10:04:07 +00:00
|
|
|
*/
|
2007-03-20 00:36:10 +00:00
|
|
|
static void
|
|
|
|
in_purgemaddrs(struct ifnet *ifp)
|
2006-09-28 10:04:07 +00:00
|
|
|
{
|
2009-03-09 17:53:05 +00:00
|
|
|
LIST_HEAD(,in_multi) purgeinms;
|
|
|
|
struct in_multi *inm, *tinm;
|
|
|
|
struct ifmultiaddr *ifma;
|
2006-09-28 10:04:07 +00:00
|
|
|
|
2009-03-09 17:53:05 +00:00
|
|
|
LIST_INIT(&purgeinms);
|
2006-09-28 10:04:07 +00:00
|
|
|
IN_MULTI_LOCK();
|
2009-03-09 17:53:05 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Extract list of in_multi associated with the detaching ifp
|
|
|
|
* which the PF_INET layer is about to release.
|
|
|
|
* We need to do this as IF_ADDR_LOCK() may be re-acquired
|
|
|
|
* by code further down.
|
|
|
|
*/
|
2012-01-05 19:00:36 +00:00
|
|
|
IF_ADDR_RLOCK(ifp);
|
2009-03-09 17:53:05 +00:00
|
|
|
TAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link) {
|
2009-03-17 14:41:54 +00:00
|
|
|
if (ifma->ifma_addr->sa_family != AF_INET ||
|
|
|
|
ifma->ifma_protospec == NULL)
|
2009-03-09 17:53:05 +00:00
|
|
|
continue;
|
2009-03-17 14:41:54 +00:00
|
|
|
#if 0
|
|
|
|
KASSERT(ifma->ifma_protospec != NULL,
|
|
|
|
("%s: ifma_protospec is NULL", __func__));
|
|
|
|
#endif
|
2009-03-09 17:53:05 +00:00
|
|
|
inm = (struct in_multi *)ifma->ifma_protospec;
|
|
|
|
LIST_INSERT_HEAD(&purgeinms, inm, inm_link);
|
2006-09-28 10:04:07 +00:00
|
|
|
}
|
2012-01-05 19:00:36 +00:00
|
|
|
IF_ADDR_RUNLOCK(ifp);
|
2005-09-18 17:36:28 +00:00
|
|
|
|
2009-03-09 17:53:05 +00:00
|
|
|
LIST_FOREACH_SAFE(inm, &purgeinms, inm_link, tinm) {
|
|
|
|
LIST_REMOVE(inm, inm_link);
|
2009-04-24 22:11:53 +00:00
|
|
|
inm_release_locked(inm);
|
2009-03-09 17:53:05 +00:00
|
|
|
}
|
|
|
|
igmp_ifdetach(ifp);
|
2005-09-18 17:36:28 +00:00
|
|
|
|
2009-03-09 17:53:05 +00:00
|
|
|
IN_MULTI_UNLOCK();
|
2005-09-18 17:36:28 +00:00
|
|
|
}
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
|
|
|
|
struct in_llentry {
|
|
|
|
struct llentry base;
|
|
|
|
};
|
|
|
|
|
2015-08-10 12:03:59 +00:00
|
|
|
#define IN_LLTBL_DEFAULT_HSIZE 32
|
|
|
|
#define IN_LLTBL_HASH(k, h) \
|
|
|
|
(((((((k >> 8) ^ k) >> 8) ^ k) >> 8) ^ k) & ((h) - 1))
|
|
|
|
|
2012-02-23 18:21:37 +00:00
|
|
|
/*
|
2015-08-10 12:03:59 +00:00
|
|
|
* Do actual deallocation of @lle.
|
2016-04-26 23:13:48 +00:00
|
|
|
*/
|
|
|
|
static void
|
|
|
|
in_lltable_destroy_lle_unlocked(struct llentry *lle)
|
|
|
|
{
|
|
|
|
|
|
|
|
LLE_LOCK_DESTROY(lle);
|
|
|
|
LLE_REQ_DESTROY(lle);
|
|
|
|
free(lle, M_LLTABLE);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2015-08-10 12:03:59 +00:00
|
|
|
* Called by LLE_FREE_LOCKED when number of references
|
|
|
|
* drops to zero.
|
2012-02-23 18:21:37 +00:00
|
|
|
*/
|
|
|
|
static void
|
2015-08-10 12:03:59 +00:00
|
|
|
in_lltable_destroy_lle(struct llentry *lle)
|
2012-02-23 18:21:37 +00:00
|
|
|
{
|
2015-08-10 12:03:59 +00:00
|
|
|
|
2012-02-23 18:21:37 +00:00
|
|
|
LLE_WUNLOCK(lle);
|
2016-04-26 23:13:48 +00:00
|
|
|
in_lltable_destroy_lle_unlocked(lle);
|
2012-02-23 18:21:37 +00:00
|
|
|
}
|
|
|
|
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
static struct llentry *
|
2015-08-11 09:26:11 +00:00
|
|
|
in_lltable_new(struct in_addr addr4, u_int flags)
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
{
|
|
|
|
struct in_llentry *lle;
|
|
|
|
|
2012-04-10 06:52:39 +00:00
|
|
|
lle = malloc(sizeof(struct in_llentry), M_LLTABLE, M_NOWAIT | M_ZERO);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
if (lle == NULL) /* NB: caller generates msg */
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* For IPv4 this will trigger "arpresolve" to generate
|
|
|
|
* an ARP request.
|
|
|
|
*/
|
2010-11-30 15:57:00 +00:00
|
|
|
lle->base.la_expire = time_uptime; /* mark expired */
|
2015-08-11 09:26:11 +00:00
|
|
|
lle->base.r_l3addr.addr4 = addr4;
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
lle->base.lle_refcnt = 1;
|
2015-08-10 12:03:59 +00:00
|
|
|
lle->base.lle_free = in_lltable_destroy_lle;
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
LLE_LOCK_INIT(&lle->base);
|
2015-12-05 09:50:37 +00:00
|
|
|
LLE_REQ_INIT(&lle->base);
|
2015-08-11 12:38:54 +00:00
|
|
|
callout_init(&lle->base.lle_timer, 1);
|
2012-08-02 13:57:49 +00:00
|
|
|
|
|
|
|
return (&lle->base);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
}
|
|
|
|
|
2015-09-14 16:48:19 +00:00
|
|
|
#define IN_ARE_MASKED_ADDR_EQUAL(d, a, m) ( \
|
|
|
|
((((d).s_addr ^ (a).s_addr) & (m).s_addr)) == 0 )
|
2009-05-20 21:07:15 +00:00
|
|
|
|
2015-08-10 12:03:59 +00:00
|
|
|
static int
|
2015-09-14 16:48:19 +00:00
|
|
|
in_lltable_match_prefix(const struct sockaddr *saddr,
|
|
|
|
const struct sockaddr *smask, u_int flags, struct llentry *lle)
|
2009-05-20 21:07:15 +00:00
|
|
|
{
|
2015-09-14 16:48:19 +00:00
|
|
|
struct in_addr addr, mask, lle_addr;
|
2015-08-10 12:03:59 +00:00
|
|
|
|
2015-09-14 16:48:19 +00:00
|
|
|
addr = ((const struct sockaddr_in *)saddr)->sin_addr;
|
|
|
|
mask = ((const struct sockaddr_in *)smask)->sin_addr;
|
|
|
|
lle_addr.s_addr = ntohl(lle->r_l3addr.addr4.s_addr);
|
|
|
|
|
|
|
|
if (IN_ARE_MASKED_ADDR_EQUAL(lle_addr, addr, mask) == 0)
|
|
|
|
return (0);
|
|
|
|
|
|
|
|
if (lle->la_flags & LLE_IFADDR) {
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Delete LLE_IFADDR records IFF address & flag matches.
|
|
|
|
* Note that addr is the interface address within prefix
|
|
|
|
* being matched.
|
|
|
|
* Note also we should handle 'ifdown' cases without removing
|
|
|
|
* ifaddr macs.
|
|
|
|
*/
|
|
|
|
if (addr.s_addr == lle_addr.s_addr && (flags & LLE_STATIC) != 0)
|
|
|
|
return (1);
|
|
|
|
return (0);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* flags & LLE_STATIC means deleting both dynamic and static entries */
|
|
|
|
if ((flags & LLE_STATIC) || !(lle->la_flags & LLE_STATIC))
|
2015-08-10 12:03:59 +00:00
|
|
|
return (1);
|
|
|
|
|
|
|
|
return (0);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
in_lltable_free_entry(struct lltable *llt, struct llentry *lle)
|
|
|
|
{
|
|
|
|
struct ifnet *ifp;
|
2010-11-12 22:03:02 +00:00
|
|
|
size_t pkts_dropped;
|
2009-05-20 21:07:15 +00:00
|
|
|
|
2015-08-10 12:03:59 +00:00
|
|
|
LLE_WLOCK_ASSERT(lle);
|
|
|
|
KASSERT(llt != NULL, ("lltable is NULL"));
|
|
|
|
|
|
|
|
/* Unlink entry from table if not already */
|
|
|
|
if ((lle->la_flags & LLE_LINKED) != 0) {
|
|
|
|
ifp = llt->llt_ifp;
|
|
|
|
IF_AFDATA_WLOCK_ASSERT(ifp);
|
|
|
|
lltable_unlink_entry(llt, lle);
|
2009-05-20 21:07:15 +00:00
|
|
|
}
|
|
|
|
|
2015-08-10 12:03:59 +00:00
|
|
|
/* cancel timer */
|
2015-11-13 22:51:35 +00:00
|
|
|
if (callout_stop(&lle->lle_timer) > 0)
|
2015-08-10 12:03:59 +00:00
|
|
|
LLE_REMREF(lle);
|
|
|
|
|
|
|
|
/* Drop hold queue */
|
|
|
|
pkts_dropped = llentry_free(lle);
|
|
|
|
ARPSTAT_ADD(dropped, pkts_dropped);
|
|
|
|
}
|
2009-05-20 21:07:15 +00:00
|
|
|
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
static int
|
2009-12-30 21:35:34 +00:00
|
|
|
in_lltable_rtcheck(struct ifnet *ifp, u_int flags, const struct sockaddr *l3addr)
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
{
|
2016-01-04 15:03:20 +00:00
|
|
|
struct rt_addrinfo info;
|
|
|
|
struct sockaddr_in rt_key, rt_mask;
|
|
|
|
struct sockaddr rt_gateway;
|
|
|
|
int rt_flags;
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
|
|
|
|
KASSERT(l3addr->sa_family == AF_INET,
|
|
|
|
("sin_family %d", l3addr->sa_family));
|
|
|
|
|
2016-01-04 15:03:20 +00:00
|
|
|
bzero(&rt_key, sizeof(rt_key));
|
|
|
|
rt_key.sin_len = sizeof(rt_key);
|
|
|
|
bzero(&rt_mask, sizeof(rt_mask));
|
|
|
|
rt_mask.sin_len = sizeof(rt_mask);
|
|
|
|
bzero(&rt_gateway, sizeof(rt_gateway));
|
|
|
|
rt_gateway.sa_len = sizeof(rt_gateway);
|
2011-07-08 09:38:33 +00:00
|
|
|
|
2016-01-04 15:03:20 +00:00
|
|
|
bzero(&info, sizeof(info));
|
|
|
|
info.rti_info[RTAX_DST] = (struct sockaddr *)&rt_key;
|
|
|
|
info.rti_info[RTAX_NETMASK] = (struct sockaddr *)&rt_mask;
|
|
|
|
info.rti_info[RTAX_GATEWAY] = (struct sockaddr *)&rt_gateway;
|
|
|
|
|
|
|
|
if (rib_lookup_info(ifp->if_fib, l3addr, NHR_REF, 0, &info) != 0)
|
2011-10-03 19:06:55 +00:00
|
|
|
return (EINVAL);
|
|
|
|
|
2016-01-04 15:03:20 +00:00
|
|
|
rt_flags = info.rti_flags;
|
|
|
|
|
2011-07-08 09:38:33 +00:00
|
|
|
/*
|
|
|
|
* If the gateway for an existing host route matches the target L3
|
2011-10-03 19:06:55 +00:00
|
|
|
* address, which is a special route inserted by some implementation
|
|
|
|
* such as MANET, and the interface is of the correct type, then
|
|
|
|
* allow for ARP to proceed.
|
2011-07-08 09:38:33 +00:00
|
|
|
*/
|
2016-01-04 15:03:20 +00:00
|
|
|
if (rt_flags & RTF_GATEWAY) {
|
|
|
|
if (!(rt_flags & RTF_HOST) || !info.rti_ifp ||
|
|
|
|
info.rti_ifp->if_type != IFT_ETHER ||
|
|
|
|
(info.rti_ifp->if_flags & (IFF_NOARP | IFF_STATICARP)) != 0 ||
|
|
|
|
memcmp(rt_gateway.sa_data, l3addr->sa_data,
|
2012-07-31 11:31:12 +00:00
|
|
|
sizeof(in_addr_t)) != 0) {
|
2016-01-04 15:03:20 +00:00
|
|
|
rib_free_info(&info);
|
2011-10-10 17:41:11 +00:00
|
|
|
return (EINVAL);
|
|
|
|
}
|
2011-10-03 19:51:18 +00:00
|
|
|
}
|
2016-01-04 15:03:20 +00:00
|
|
|
rib_free_info(&info);
|
2011-10-03 19:51:18 +00:00
|
|
|
|
|
|
|
/*
|
2012-07-31 11:31:12 +00:00
|
|
|
* Make sure that at least the destination address is covered
|
|
|
|
* by the route. This is for handling the case where 2 or more
|
2011-10-03 19:51:18 +00:00
|
|
|
* interfaces have the same prefix. An incoming packet arrives
|
|
|
|
* on one interface and the corresponding outgoing packet leaves
|
|
|
|
* another interface.
|
|
|
|
*/
|
2016-01-04 15:03:20 +00:00
|
|
|
if (!(rt_flags & RTF_HOST) && info.rti_ifp != ifp) {
|
2011-10-10 17:41:11 +00:00
|
|
|
const char *sa, *mask, *addr, *lim;
|
2017-02-16 20:47:41 +00:00
|
|
|
const struct sockaddr_in *l3sin;
|
2011-10-03 19:51:18 +00:00
|
|
|
|
2016-01-04 15:03:20 +00:00
|
|
|
mask = (const char *)&rt_mask;
|
2011-10-25 04:06:29 +00:00
|
|
|
/*
|
|
|
|
* Just being extra cautious to avoid some custom
|
|
|
|
* code getting into trouble.
|
|
|
|
*/
|
2016-01-04 15:03:20 +00:00
|
|
|
if ((info.rti_addrs & RTA_NETMASK) == 0)
|
2011-10-25 04:06:29 +00:00
|
|
|
return (EINVAL);
|
|
|
|
|
2016-01-04 15:03:20 +00:00
|
|
|
sa = (const char *)&rt_key;
|
2011-10-10 17:41:11 +00:00
|
|
|
addr = (const char *)l3addr;
|
2017-02-16 20:47:41 +00:00
|
|
|
l3sin = (const struct sockaddr_in *)l3addr;
|
|
|
|
lim = addr + l3sin->sin_len;
|
2011-10-03 19:51:18 +00:00
|
|
|
|
|
|
|
for ( ; addr < lim; sa++, mask++, addr++) {
|
|
|
|
if ((*sa ^ *addr) & *mask) {
|
2009-10-23 18:27:34 +00:00
|
|
|
#ifdef DIAGNOSTIC
|
2017-02-16 20:47:41 +00:00
|
|
|
char addrbuf[INET_ADDRSTRLEN];
|
|
|
|
|
|
|
|
log(LOG_INFO, "IPv4 address: \"%s\" "
|
|
|
|
"is not on the network\n",
|
|
|
|
inet_ntoa_r(l3sin->sin_addr, addrbuf));
|
2009-10-02 01:45:11 +00:00
|
|
|
#endif
|
2011-10-10 17:41:11 +00:00
|
|
|
return (EINVAL);
|
2011-10-03 19:51:18 +00:00
|
|
|
}
|
|
|
|
}
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
}
|
2011-10-03 19:51:18 +00:00
|
|
|
|
2011-10-10 17:41:11 +00:00
|
|
|
return (0);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
}
|
|
|
|
|
2015-08-10 12:03:59 +00:00
|
|
|
static inline uint32_t
|
|
|
|
in_lltable_hash_dst(const struct in_addr dst, uint32_t hsize)
|
|
|
|
{
|
|
|
|
|
|
|
|
return (IN_LLTBL_HASH(dst.s_addr, hsize));
|
|
|
|
}
|
|
|
|
|
|
|
|
static uint32_t
|
|
|
|
in_lltable_hash(const struct llentry *lle, uint32_t hsize)
|
|
|
|
{
|
|
|
|
|
2015-08-11 09:26:11 +00:00
|
|
|
return (in_lltable_hash_dst(lle->r_l3addr.addr4, hsize));
|
2015-08-10 12:03:59 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
in_lltable_fill_sa_entry(const struct llentry *lle, struct sockaddr *sa)
|
|
|
|
{
|
|
|
|
struct sockaddr_in *sin;
|
|
|
|
|
|
|
|
sin = (struct sockaddr_in *)sa;
|
|
|
|
bzero(sin, sizeof(*sin));
|
|
|
|
sin->sin_family = AF_INET;
|
|
|
|
sin->sin_len = sizeof(*sin);
|
2015-08-11 09:26:11 +00:00
|
|
|
sin->sin_addr = lle->r_l3addr.addr4;
|
2015-08-10 12:03:59 +00:00
|
|
|
}
|
|
|
|
|
2014-11-15 18:54:07 +00:00
|
|
|
static inline struct llentry *
|
|
|
|
in_lltable_find_dst(struct lltable *llt, struct in_addr dst)
|
|
|
|
{
|
|
|
|
struct llentry *lle;
|
|
|
|
struct llentries *lleh;
|
2015-08-10 12:03:59 +00:00
|
|
|
u_int hashidx;
|
2014-11-15 18:54:07 +00:00
|
|
|
|
2015-01-05 17:23:02 +00:00
|
|
|
hashidx = in_lltable_hash_dst(dst, llt->llt_hsize);
|
2015-08-10 12:03:59 +00:00
|
|
|
lleh = &llt->lle_head[hashidx];
|
2014-11-15 18:54:07 +00:00
|
|
|
LIST_FOREACH(lle, lleh, lle_next) {
|
|
|
|
if (lle->la_flags & LLE_DELETED)
|
|
|
|
continue;
|
2015-08-11 09:26:11 +00:00
|
|
|
if (lle->r_l3addr.addr4.s_addr == dst.s_addr)
|
2014-11-15 18:54:07 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
return (lle);
|
|
|
|
}
|
|
|
|
|
2015-09-14 16:48:19 +00:00
|
|
|
static void
|
|
|
|
in_lltable_delete_entry(struct lltable *llt, struct llentry *lle)
|
2014-11-15 18:54:07 +00:00
|
|
|
{
|
|
|
|
|
2015-09-14 16:48:19 +00:00
|
|
|
lle->la_flags |= LLE_DELETED;
|
|
|
|
EVENTHANDLER_INVOKE(lle_event, lle, LLENTRY_DELETED);
|
2014-11-15 18:54:07 +00:00
|
|
|
#ifdef DIAGNOSTIC
|
2015-09-14 16:48:19 +00:00
|
|
|
log(LOG_INFO, "ifaddr cache = %p is deleted\n", lle);
|
2014-11-15 18:54:07 +00:00
|
|
|
#endif
|
2015-09-14 16:48:19 +00:00
|
|
|
llentry_free(lle);
|
2014-11-15 18:54:07 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct llentry *
|
2015-08-20 12:05:17 +00:00
|
|
|
in_lltable_alloc(struct lltable *llt, u_int flags, const struct sockaddr *l3addr)
|
2014-11-15 18:54:07 +00:00
|
|
|
{
|
|
|
|
const struct sockaddr_in *sin = (const struct sockaddr_in *)l3addr;
|
|
|
|
struct ifnet *ifp = llt->llt_ifp;
|
|
|
|
struct llentry *lle;
|
2015-12-31 05:03:27 +00:00
|
|
|
char linkhdr[LLE_MAX_LINKHDR];
|
|
|
|
size_t linkhdrsize;
|
|
|
|
int lladdr_off;
|
2014-11-15 18:54:07 +00:00
|
|
|
|
|
|
|
KASSERT(l3addr->sa_family == AF_INET,
|
|
|
|
("sin_family %d", l3addr->sa_family));
|
|
|
|
|
|
|
|
/*
|
|
|
|
* A route that covers the given address must have
|
|
|
|
* been installed 1st because we are doing a resolution,
|
|
|
|
* verify this.
|
|
|
|
*/
|
|
|
|
if (!(flags & LLE_IFADDR) &&
|
|
|
|
in_lltable_rtcheck(ifp, flags, l3addr) != 0)
|
|
|
|
return (NULL);
|
|
|
|
|
2015-08-11 09:26:11 +00:00
|
|
|
lle = in_lltable_new(sin->sin_addr, flags);
|
2014-11-15 18:54:07 +00:00
|
|
|
if (lle == NULL) {
|
|
|
|
log(LOG_INFO, "lla_lookup: new lle malloc failed\n");
|
|
|
|
return (NULL);
|
|
|
|
}
|
|
|
|
lle->la_flags = flags;
|
2015-12-05 09:50:37 +00:00
|
|
|
if (flags & LLE_STATIC)
|
|
|
|
lle->r_flags |= RLLE_VALID;
|
2014-11-15 18:54:07 +00:00
|
|
|
if ((flags & LLE_IFADDR) == LLE_IFADDR) {
|
2015-12-31 05:03:27 +00:00
|
|
|
linkhdrsize = LLE_MAX_LINKHDR;
|
|
|
|
if (lltable_calc_llheader(ifp, AF_INET, IF_LLADDR(ifp),
|
2016-04-26 23:13:48 +00:00
|
|
|
linkhdr, &linkhdrsize, &lladdr_off) != 0) {
|
|
|
|
in_lltable_destroy_lle_unlocked(lle);
|
2015-12-31 05:03:27 +00:00
|
|
|
return (NULL);
|
2016-04-26 23:13:48 +00:00
|
|
|
}
|
2015-12-31 05:03:27 +00:00
|
|
|
lltable_set_entry_addr(ifp, lle, linkhdr, linkhdrsize,
|
|
|
|
lladdr_off);
|
2015-11-07 11:12:00 +00:00
|
|
|
lle->la_flags |= LLE_STATIC;
|
2015-12-05 09:50:37 +00:00
|
|
|
lle->r_flags |= (RLLE_VALID | RLLE_IFADDR);
|
2014-11-15 18:54:07 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return (lle);
|
|
|
|
}
|
|
|
|
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
/*
|
|
|
|
* Return NULL if not found or marked for deletion.
|
|
|
|
* If found return lle read locked.
|
|
|
|
*/
|
|
|
|
static struct llentry *
|
|
|
|
in_lltable_lookup(struct lltable *llt, u_int flags, const struct sockaddr *l3addr)
|
|
|
|
{
|
|
|
|
const struct sockaddr_in *sin = (const struct sockaddr_in *)l3addr;
|
|
|
|
struct llentry *lle;
|
|
|
|
|
2015-08-08 21:41:59 +00:00
|
|
|
IF_AFDATA_LOCK_ASSERT(llt->llt_ifp);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
KASSERT(l3addr->sa_family == AF_INET,
|
|
|
|
("sin_family %d", l3addr->sa_family));
|
2015-08-10 12:03:59 +00:00
|
|
|
lle = in_lltable_find_dst(llt, sin->sin_addr);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
|
2014-11-15 18:54:07 +00:00
|
|
|
if (lle == NULL)
|
|
|
|
return (NULL);
|
|
|
|
|
2015-12-05 09:50:37 +00:00
|
|
|
KASSERT((flags & (LLE_UNLOCKED|LLE_EXCLUSIVE)) !=
|
|
|
|
(LLE_UNLOCKED|LLE_EXCLUSIVE),("wrong lle request flags: 0x%X",
|
|
|
|
flags));
|
|
|
|
|
|
|
|
if (flags & LLE_UNLOCKED)
|
|
|
|
return (lle);
|
|
|
|
|
2014-11-15 18:54:07 +00:00
|
|
|
if (flags & LLE_EXCLUSIVE)
|
|
|
|
LLE_WLOCK(lle);
|
|
|
|
else
|
|
|
|
LLE_RLOCK(lle);
|
2012-08-01 09:00:26 +00:00
|
|
|
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
return (lle);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
2015-08-10 12:03:59 +00:00
|
|
|
in_lltable_dump_entry(struct lltable *llt, struct llentry *lle,
|
|
|
|
struct sysctl_req *wr)
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
{
|
|
|
|
struct ifnet *ifp = llt->llt_ifp;
|
|
|
|
/* XXX stack use */
|
|
|
|
struct {
|
|
|
|
struct rt_msghdr rtm;
|
2013-01-31 08:55:21 +00:00
|
|
|
struct sockaddr_in sin;
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
struct sockaddr_dl sdl;
|
|
|
|
} arpc;
|
2015-08-10 12:03:59 +00:00
|
|
|
struct sockaddr_dl *sdl;
|
|
|
|
int error;
|
2012-08-01 09:00:26 +00:00
|
|
|
|
2015-08-10 12:03:59 +00:00
|
|
|
bzero(&arpc, sizeof(arpc));
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
/* skip deleted entries */
|
2009-10-15 06:12:04 +00:00
|
|
|
if ((lle->la_flags & LLE_DELETED) == LLE_DELETED)
|
2015-08-10 12:03:59 +00:00
|
|
|
return (0);
|
2009-01-09 21:57:49 +00:00
|
|
|
/* Skip if jailed and not a valid IP of the prison. */
|
2015-08-10 12:03:59 +00:00
|
|
|
lltable_fill_sa_entry(lle,(struct sockaddr *)&arpc.sin);
|
|
|
|
if (prison_if(wr->td->td_ucred,
|
|
|
|
(struct sockaddr *)&arpc.sin) != 0)
|
|
|
|
return (0);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
/*
|
|
|
|
* produce a msg made of:
|
|
|
|
* struct rt_msghdr;
|
2013-01-31 08:55:21 +00:00
|
|
|
* struct sockaddr_in; (IPv4)
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
* struct sockaddr_dl;
|
|
|
|
*/
|
|
|
|
arpc.rtm.rtm_msglen = sizeof(arpc);
|
2009-01-09 10:58:59 +00:00
|
|
|
arpc.rtm.rtm_version = RTM_VERSION;
|
|
|
|
arpc.rtm.rtm_type = RTM_GET;
|
|
|
|
arpc.rtm.rtm_flags = RTF_UP;
|
|
|
|
arpc.rtm.rtm_addrs = RTA_DST | RTA_GATEWAY;
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
|
|
|
|
/* publish */
|
2013-01-31 08:55:21 +00:00
|
|
|
if (lle->la_flags & LLE_PUB)
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
arpc.rtm.rtm_flags |= RTF_ANNOUNCE;
|
|
|
|
|
|
|
|
sdl = &arpc.sdl;
|
|
|
|
sdl->sdl_family = AF_LINK;
|
|
|
|
sdl->sdl_len = sizeof(*sdl);
|
|
|
|
sdl->sdl_index = ifp->if_index;
|
|
|
|
sdl->sdl_type = ifp->if_type;
|
2009-10-15 06:12:04 +00:00
|
|
|
if ((lle->la_flags & LLE_VALID) == LLE_VALID) {
|
|
|
|
sdl->sdl_alen = ifp->if_addrlen;
|
2015-12-31 05:03:27 +00:00
|
|
|
bcopy(lle->ll_addr, LLADDR(sdl), ifp->if_addrlen);
|
2009-10-15 06:12:04 +00:00
|
|
|
} else {
|
|
|
|
sdl->sdl_alen = 0;
|
|
|
|
bzero(LLADDR(sdl), ifp->if_addrlen);
|
|
|
|
}
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
|
|
|
|
arpc.rtm.rtm_rmx.rmx_expire =
|
|
|
|
lle->la_flags & LLE_STATIC ? 0 : lle->la_expire;
|
2008-12-26 19:45:24 +00:00
|
|
|
arpc.rtm.rtm_flags |= (RTF_HOST | RTF_LLDATA);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
if (lle->la_flags & LLE_STATIC)
|
|
|
|
arpc.rtm.rtm_flags |= RTF_STATIC;
|
2015-09-27 04:54:29 +00:00
|
|
|
if (lle->la_flags & LLE_IFADDR)
|
|
|
|
arpc.rtm.rtm_flags |= RTF_PINNED;
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
arpc.rtm.rtm_index = ifp->if_index;
|
|
|
|
error = SYSCTL_OUT(wr, &arpc, sizeof(arpc));
|
2015-08-10 12:03:59 +00:00
|
|
|
|
|
|
|
return (error);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
}
|
|
|
|
|
2015-01-05 17:23:02 +00:00
|
|
|
static struct lltable *
|
|
|
|
in_lltattach(struct ifnet *ifp)
|
2009-03-09 17:53:05 +00:00
|
|
|
{
|
|
|
|
struct lltable *llt;
|
2014-12-07 17:32:06 +00:00
|
|
|
|
2015-01-05 17:23:02 +00:00
|
|
|
llt = lltable_allocate_htbl(IN_LLTBL_DEFAULT_HSIZE);
|
2015-08-11 05:51:00 +00:00
|
|
|
llt->llt_af = AF_INET;
|
|
|
|
llt->llt_ifp = ifp;
|
2014-12-07 17:32:06 +00:00
|
|
|
|
|
|
|
llt->llt_lookup = in_lltable_lookup;
|
2015-08-20 12:05:17 +00:00
|
|
|
llt->llt_alloc_entry = in_lltable_alloc;
|
2015-09-14 16:48:19 +00:00
|
|
|
llt->llt_delete_entry = in_lltable_delete_entry;
|
2014-12-07 17:32:06 +00:00
|
|
|
llt->llt_dump_entry = in_lltable_dump_entry;
|
|
|
|
llt->llt_hash = in_lltable_hash;
|
2014-12-09 00:48:08 +00:00
|
|
|
llt->llt_fill_sa_entry = in_lltable_fill_sa_entry;
|
2015-08-11 05:51:00 +00:00
|
|
|
llt->llt_free_entry = in_lltable_free_entry;
|
2014-12-07 17:32:06 +00:00
|
|
|
llt->llt_match_prefix = in_lltable_match_prefix;
|
2015-08-11 05:51:00 +00:00
|
|
|
lltable_link(llt);
|
2009-03-09 17:53:05 +00:00
|
|
|
|
2015-01-05 17:23:02 +00:00
|
|
|
return (llt);
|
|
|
|
}
|
|
|
|
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
void *
|
|
|
|
in_domifattach(struct ifnet *ifp)
|
2009-03-09 17:53:05 +00:00
|
|
|
{
|
|
|
|
struct in_ifinfo *ii;
|
|
|
|
|
|
|
|
ii = malloc(sizeof(struct in_ifinfo), M_IFADDR, M_WAITOK|M_ZERO);
|
|
|
|
|
2015-01-05 17:23:02 +00:00
|
|
|
ii->ii_llt = in_lltattach(ifp);
|
2009-03-09 17:53:05 +00:00
|
|
|
ii->ii_igmp = igmp_domifattach(ifp);
|
|
|
|
|
2015-08-11 05:51:00 +00:00
|
|
|
return (ii);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void
|
2009-03-09 17:53:05 +00:00
|
|
|
in_domifdetach(struct ifnet *ifp, void *aux)
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
{
|
2009-03-09 17:53:05 +00:00
|
|
|
struct in_ifinfo *ii = (struct in_ifinfo *)aux;
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
|
2009-03-09 17:53:05 +00:00
|
|
|
igmp_domifdetach(ifp);
|
|
|
|
lltable_free(ii->ii_llt);
|
|
|
|
free(ii, M_IFADDR);
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
}
|