2005-01-07 01:45:51 +00:00
|
|
|
/*-
|
1994-05-24 10:09:53 +00:00
|
|
|
* Copyright (c) 1982, 1986, 1988, 1993
|
2008-07-05 18:03:39 +00:00
|
|
|
* The Regents of the University of California.
|
|
|
|
* 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.
|
|
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
* without specific prior written permission.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
*
|
1995-09-21 19:59:43 +00:00
|
|
|
* @(#)raw_ip.c 8.7 (Berkeley) 5/15/95
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
|
2007-10-07 20:44:24 +00:00
|
|
|
#include <sys/cdefs.h>
|
|
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
|
1999-12-22 19:13:38 +00:00
|
|
|
#include "opt_inet6.h"
|
|
|
|
#include "opt_ipsec.h"
|
2002-07-31 18:30:34 +00:00
|
|
|
#include "opt_mac.h"
|
1999-12-22 19:13:38 +00:00
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/param.h>
|
2004-04-26 19:46:52 +00:00
|
|
|
#include <sys/jail.h>
|
1997-02-18 20:46:36 +00:00
|
|
|
#include <sys/kernel.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <sys/lock.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/malloc.h>
|
|
|
|
#include <sys/mbuf.h>
|
2006-11-06 13:42:10 +00:00
|
|
|
#include <sys/priv.h>
|
2001-10-05 07:06:32 +00:00
|
|
|
#include <sys/proc.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/protosw.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <sys/signalvar.h>
|
1997-02-18 20:46:36 +00:00
|
|
|
#include <sys/socket.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <sys/socketvar.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <sys/sx.h>
|
1997-02-18 20:46:36 +00:00
|
|
|
#include <sys/sysctl.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <sys/systm.h>
|
1998-03-28 10:18:26 +00:00
|
|
|
|
2002-03-20 05:48:55 +00:00
|
|
|
#include <vm/uma.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
#include <net/if.h>
|
|
|
|
#include <net/route.h>
|
|
|
|
|
|
|
|
#include <netinet/in.h>
|
|
|
|
#include <netinet/in_systm.h>
|
1995-10-21 02:12:20 +00:00
|
|
|
#include <netinet/in_pcb.h>
|
|
|
|
#include <netinet/in_var.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <netinet/ip.h>
|
1994-05-24 10:09:53 +00:00
|
|
|
#include <netinet/ip_var.h>
|
|
|
|
#include <netinet/ip_mroute.h>
|
|
|
|
|
1994-10-28 15:09:49 +00:00
|
|
|
#include <netinet/ip_fw.h>
|
2001-10-05 05:45:27 +00:00
|
|
|
#include <netinet/ip_dummynet.h>
|
1994-10-28 15:09:49 +00:00
|
|
|
|
2007-07-03 12:13:45 +00:00
|
|
|
#ifdef IPSEC
|
2002-10-16 02:25:05 +00:00
|
|
|
#include <netipsec/ipsec.h>
|
2007-07-03 12:13:45 +00:00
|
|
|
#endif /*IPSEC*/
|
2002-10-16 02:25:05 +00:00
|
|
|
|
2006-10-22 11:52:19 +00:00
|
|
|
#include <security/mac/mac_framework.h>
|
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
struct inpcbhead ripcb;
|
|
|
|
struct inpcbinfo ripcbinfo;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2001-10-05 05:45:27 +00:00
|
|
|
/* control hooks for ipfw and dummynet */
|
2004-08-17 22:05:54 +00:00
|
|
|
ip_fw_ctl_t *ip_fw_ctl_ptr = NULL;
|
|
|
|
ip_dn_ctl_t *ip_dn_ctl_ptr = NULL;
|
2001-10-05 05:45:27 +00:00
|
|
|
|
Massive cleanup of the ip_mroute code.
No functional changes, but:
+ the mrouting module now should behave the same as the compiled-in
version (it did not before, some of the rsvp code was not loaded
properly);
+ netinet/ip_mroute.c is now truly optional;
+ removed some redundant/unused code;
+ changed many instances of '0' to NULL and INADDR_ANY as appropriate;
+ removed several static variables to make the code more SMP-friendly;
+ fixed some minor bugs in the mrouting code (mostly, incorrect return
values from functions).
This commit is also a prerequisite to the addition of support for PIM,
which i would like to put in before DP2 (it does not change any of
the existing APIs, anyways).
Note, in the process we found out that some device drivers fail to
properly handle changes in IFF_ALLMULTI, leading to interesting
behaviour when a multicast router is started. This bug is not
corrected by this commit, and will be fixed with a separate commit.
Detailed changes:
--------------------
netinet/ip_mroute.c all the above.
conf/files make ip_mroute.c optional
net/route.c fix mrt_ioctl hook
netinet/ip_input.c fix ip_mforward hook, move rsvp_input() here
together with other rsvp code, and a couple
of indentation fixes.
netinet/ip_output.c fix ip_mforward and ip_mcast_src hooks
netinet/ip_var.h rsvp function hooks
netinet/raw_ip.c hooks for mrouting and rsvp functions, plus
interface cleanup.
netinet/ip_mroute.h remove an unused and optional field from a struct
Most of the code is from Pavlin Radoslavov and the XORP project
Reviewed by: sam
MFC after: 1 week
2002-11-15 22:53:53 +00:00
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* Hooks for multicast routing. They all default to NULL, so leave them not
|
|
|
|
* initialized and rely on BSS being set to 0.
|
Massive cleanup of the ip_mroute code.
No functional changes, but:
+ the mrouting module now should behave the same as the compiled-in
version (it did not before, some of the rsvp code was not loaded
properly);
+ netinet/ip_mroute.c is now truly optional;
+ removed some redundant/unused code;
+ changed many instances of '0' to NULL and INADDR_ANY as appropriate;
+ removed several static variables to make the code more SMP-friendly;
+ fixed some minor bugs in the mrouting code (mostly, incorrect return
values from functions).
This commit is also a prerequisite to the addition of support for PIM,
which i would like to put in before DP2 (it does not change any of
the existing APIs, anyways).
Note, in the process we found out that some device drivers fail to
properly handle changes in IFF_ALLMULTI, leading to interesting
behaviour when a multicast router is started. This bug is not
corrected by this commit, and will be fixed with a separate commit.
Detailed changes:
--------------------
netinet/ip_mroute.c all the above.
conf/files make ip_mroute.c optional
net/route.c fix mrt_ioctl hook
netinet/ip_input.c fix ip_mforward hook, move rsvp_input() here
together with other rsvp code, and a couple
of indentation fixes.
netinet/ip_output.c fix ip_mforward and ip_mcast_src hooks
netinet/ip_var.h rsvp function hooks
netinet/raw_ip.c hooks for mrouting and rsvp functions, plus
interface cleanup.
netinet/ip_mroute.h remove an unused and optional field from a struct
Most of the code is from Pavlin Radoslavov and the XORP project
Reviewed by: sam
MFC after: 1 week
2002-11-15 22:53:53 +00:00
|
|
|
*/
|
|
|
|
|
2008-07-05 18:03:39 +00:00
|
|
|
/*
|
|
|
|
* The socket used to communicate with the multicast routing daemon.
|
|
|
|
*/
|
Massive cleanup of the ip_mroute code.
No functional changes, but:
+ the mrouting module now should behave the same as the compiled-in
version (it did not before, some of the rsvp code was not loaded
properly);
+ netinet/ip_mroute.c is now truly optional;
+ removed some redundant/unused code;
+ changed many instances of '0' to NULL and INADDR_ANY as appropriate;
+ removed several static variables to make the code more SMP-friendly;
+ fixed some minor bugs in the mrouting code (mostly, incorrect return
values from functions).
This commit is also a prerequisite to the addition of support for PIM,
which i would like to put in before DP2 (it does not change any of
the existing APIs, anyways).
Note, in the process we found out that some device drivers fail to
properly handle changes in IFF_ALLMULTI, leading to interesting
behaviour when a multicast router is started. This bug is not
corrected by this commit, and will be fixed with a separate commit.
Detailed changes:
--------------------
netinet/ip_mroute.c all the above.
conf/files make ip_mroute.c optional
net/route.c fix mrt_ioctl hook
netinet/ip_input.c fix ip_mforward hook, move rsvp_input() here
together with other rsvp code, and a couple
of indentation fixes.
netinet/ip_output.c fix ip_mforward and ip_mcast_src hooks
netinet/ip_var.h rsvp function hooks
netinet/raw_ip.c hooks for mrouting and rsvp functions, plus
interface cleanup.
netinet/ip_mroute.h remove an unused and optional field from a struct
Most of the code is from Pavlin Radoslavov and the XORP project
Reviewed by: sam
MFC after: 1 week
2002-11-15 22:53:53 +00:00
|
|
|
struct socket *ip_mrouter;
|
|
|
|
|
2008-07-05 18:03:39 +00:00
|
|
|
/*
|
|
|
|
* The various mrouter and rsvp functions.
|
|
|
|
*/
|
Massive cleanup of the ip_mroute code.
No functional changes, but:
+ the mrouting module now should behave the same as the compiled-in
version (it did not before, some of the rsvp code was not loaded
properly);
+ netinet/ip_mroute.c is now truly optional;
+ removed some redundant/unused code;
+ changed many instances of '0' to NULL and INADDR_ANY as appropriate;
+ removed several static variables to make the code more SMP-friendly;
+ fixed some minor bugs in the mrouting code (mostly, incorrect return
values from functions).
This commit is also a prerequisite to the addition of support for PIM,
which i would like to put in before DP2 (it does not change any of
the existing APIs, anyways).
Note, in the process we found out that some device drivers fail to
properly handle changes in IFF_ALLMULTI, leading to interesting
behaviour when a multicast router is started. This bug is not
corrected by this commit, and will be fixed with a separate commit.
Detailed changes:
--------------------
netinet/ip_mroute.c all the above.
conf/files make ip_mroute.c optional
net/route.c fix mrt_ioctl hook
netinet/ip_input.c fix ip_mforward hook, move rsvp_input() here
together with other rsvp code, and a couple
of indentation fixes.
netinet/ip_output.c fix ip_mforward and ip_mcast_src hooks
netinet/ip_var.h rsvp function hooks
netinet/raw_ip.c hooks for mrouting and rsvp functions, plus
interface cleanup.
netinet/ip_mroute.h remove an unused and optional field from a struct
Most of the code is from Pavlin Radoslavov and the XORP project
Reviewed by: sam
MFC after: 1 week
2002-11-15 22:53:53 +00:00
|
|
|
int (*ip_mrouter_set)(struct socket *, struct sockopt *);
|
|
|
|
int (*ip_mrouter_get)(struct socket *, struct sockopt *);
|
|
|
|
int (*ip_mrouter_done)(void);
|
|
|
|
int (*ip_mforward)(struct ip *, struct ifnet *, struct mbuf *,
|
2004-08-16 18:32:07 +00:00
|
|
|
struct ip_moptions *);
|
Add code to allow the system to handle multiple routing tables.
This particular implementation is designed to be fully backwards compatible
and to be MFC-able to 7.x (and 6.x)
Currently the only protocol that can make use of the multiple tables is IPv4
Similar functionality exists in OpenBSD and Linux.
From my notes:
-----
One thing where FreeBSD has been falling behind, and which by chance I
have some time to work on is "policy based routing", which allows
different
packet streams to be routed by more than just the destination address.
Constraints:
------------
I want to make some form of this available in the 6.x tree
(and by extension 7.x) , but FreeBSD in general needs it so I might as
well do it in -current and back port the portions I need.
One of the ways that this can be done is to have the ability to
instantiate multiple kernel routing tables (which I will now
refer to as "Forwarding Information Bases" or "FIBs" for political
correctness reasons). Which FIB a particular packet uses to make
the next hop decision can be decided by a number of mechanisms.
The policies these mechanisms implement are the "Policies" referred
to in "Policy based routing".
One of the constraints I have if I try to back port this work to
6.x is that it must be implemented as a EXTENSION to the existing
ABIs in 6.x so that third party applications do not need to be
recompiled in timespan of the branch.
This first version will not have some of the bells and whistles that
will come with later versions. It will, for example, be limited to 16
tables in the first commit.
Implementation method, Compatible version. (part 1)
-------------------------------
For this reason I have implemented a "sufficient subset" of a
multiple routing table solution in Perforce, and back-ported it
to 6.x. (also in Perforce though not always caught up with what I
have done in -current/P4). The subset allows a number of FIBs
to be defined at compile time (8 is sufficient for my purposes in 6.x)
and implements the changes needed to allow IPV4 to use them. I have not
done the changes for ipv6 simply because I do not need it, and I do not
have enough knowledge of ipv6 (e.g. neighbor discovery) needed to do it.
Other protocol families are left untouched and should there be
users with proprietary protocol families, they should continue to work
and be oblivious to the existence of the extra FIBs.
To understand how this is done, one must know that the current FIB
code starts everything off with a single dimensional array of
pointers to FIB head structures (One per protocol family), each of
which in turn points to the trie of routes available to that family.
The basic change in the ABI compatible version of the change is to
extent that array to be a 2 dimensional array, so that
instead of protocol family X looking at rt_tables[X] for the
table it needs, it looks at rt_tables[Y][X] when for all
protocol families except ipv4 Y is always 0.
Code that is unaware of the change always just sees the first row
of the table, which of course looks just like the one dimensional
array that existed before.
The entry points rtrequest(), rtalloc(), rtalloc1(), rtalloc_ign()
are all maintained, but refer only to the first row of the array,
so that existing callers in proprietary protocols can continue to
do the "right thing".
Some new entry points are added, for the exclusive use of ipv4 code
called in_rtrequest(), in_rtalloc(), in_rtalloc1() and in_rtalloc_ign(),
which have an extra argument which refers the code to the correct row.
In addition, there are some new entry points (currently called
rtalloc_fib() and friends) that check the Address family being
looked up and call either rtalloc() (and friends) if the protocol
is not IPv4 forcing the action to row 0 or to the appropriate row
if it IS IPv4 (and that info is available). These are for calling
from code that is not specific to any particular protocol. The way
these are implemented would change in the non ABI preserving code
to be added later.
One feature of the first version of the code is that for ipv4,
the interface routes show up automatically on all the FIBs, so
that no matter what FIB you select you always have the basic
direct attached hosts available to you. (rtinit() does this
automatically).
You CAN delete an interface route from one FIB should you want
to but by default it's there. ARP information is also available
in each FIB. It's assumed that the same machine would have the
same MAC address, regardless of which FIB you are using to get
to it.
This brings us as to how the correct FIB is selected for an outgoing
IPV4 packet.
Firstly, all packets have a FIB associated with them. if nothing
has been done to change it, it will be FIB 0. The FIB is changed
in the following ways.
Packets fall into one of a number of classes.
1/ locally generated packets, coming from a socket/PCB.
Such packets select a FIB from a number associated with the
socket/PCB. This in turn is inherited from the process,
but can be changed by a socket option. The process in turn
inherits it on fork. I have written a utility call setfib
that acts a bit like nice..
setfib -3 ping target.example.com # will use fib 3 for ping.
It is an obvious extension to make it a property of a jail
but I have not done so. It can be achieved by combining the setfib and
jail commands.
2/ packets received on an interface for forwarding.
By default these packets would use table 0,
(or possibly a number settable in a sysctl(not yet)).
but prior to routing the firewall can inspect them (see below).
(possibly in the future you may be able to associate a FIB
with packets received on an interface.. An ifconfig arg, but not yet.)
3/ packets inspected by a packet classifier, which can arbitrarily
associate a fib with it on a packet by packet basis.
A fib assigned to a packet by a packet classifier
(such as ipfw) would over-ride a fib associated by
a more default source. (such as cases 1 or 2).
4/ a tcp listen socket associated with a fib will generate
accept sockets that are associated with that same fib.
5/ Packets generated in response to some other packet (e.g. reset
or icmp packets). These should use the FIB associated with the
packet being reponded to.
6/ Packets generated during encapsulation.
gif, tun and other tunnel interfaces will encapsulate using the FIB
that was in effect withthe proces that set up the tunnel.
thus setfib 1 ifconfig gif0 [tunnel instructions]
will set the fib for the tunnel to use to be fib 1.
Routing messages would be associated with their
process, and thus select one FIB or another.
messages from the kernel would be associated with the fib they
refer to and would only be received by a routing socket associated
with that fib. (not yet implemented)
In addition Netstat has been edited to be able to cope with the
fact that the array is now 2 dimensional. (It looks in system
memory using libkvm (!)). Old versions of netstat see only the first FIB.
In addition two sysctls are added to give:
a) the number of FIBs compiled in (active)
b) the default FIB of the calling process.
Early testing experience:
-------------------------
Basically our (IronPort's) appliance does this functionality already
using ipfw fwd but that method has some drawbacks.
For example,
It can't fully simulate a routing table because it can't influence the
socket's choice of local address when a connect() is done.
Testing during the generating of these changes has been
remarkably smooth so far. Multiple tables have co-existed
with no notable side effects, and packets have been routes
accordingly.
ipfw has grown 2 new keywords:
setfib N ip from anay to any
count ip from any to any fib N
In pf there seems to be a requirement to be able to give symbolic names to the
fibs but I do not have that capacity. I am not sure if it is required.
SCTP has interestingly enough built in support for this, called VRFs
in Cisco parlance. it will be interesting to see how that handles it
when it suddenly actually does something.
Where to next:
--------------------
After committing the ABI compatible version and MFCing it, I'd
like to proceed in a forward direction in -current. this will
result in some roto-tilling in the routing code.
Firstly: the current code's idea of having a separate tree per
protocol family, all of the same format, and pointed to by the
1 dimensional array is a bit silly. Especially when one considers that
there is code that makes assumptions about every protocol having the
same internal structures there. Some protocols don't WANT that
sort of structure. (for example the whole idea of a netmask is foreign
to appletalk). This needs to be made opaque to the external code.
My suggested first change is to add routing method pointers to the
'domain' structure, along with information pointing the data.
instead of having an array of pointers to uniform structures,
there would be an array pointing to the 'domain' structures
for each protocol address domain (protocol family),
and the methods this reached would be called. The methods would have
an argument that gives FIB number, but the protocol would be free
to ignore it.
When the ABI can be changed it raises the possibilty of the
addition of a fib entry into the "struct route". Currently,
the structure contains the sockaddr of the desination, and the resulting
fib entry. To make this work fully, one could add a fib number
so that given an address and a fib, one can find the third element, the
fib entry.
Interaction with the ARP layer/ LL layer would need to be
revisited as well. Qing Li has been working on this already.
This work was sponsored by Ironport Systems/Cisco
Reviewed by: several including rwatson, bz and mlair (parts each)
Obtained from: Ironport systems/Cisco
2008-05-09 23:03:00 +00:00
|
|
|
int (*mrt_ioctl)(int, caddr_t, int);
|
Massive cleanup of the ip_mroute code.
No functional changes, but:
+ the mrouting module now should behave the same as the compiled-in
version (it did not before, some of the rsvp code was not loaded
properly);
+ netinet/ip_mroute.c is now truly optional;
+ removed some redundant/unused code;
+ changed many instances of '0' to NULL and INADDR_ANY as appropriate;
+ removed several static variables to make the code more SMP-friendly;
+ fixed some minor bugs in the mrouting code (mostly, incorrect return
values from functions).
This commit is also a prerequisite to the addition of support for PIM,
which i would like to put in before DP2 (it does not change any of
the existing APIs, anyways).
Note, in the process we found out that some device drivers fail to
properly handle changes in IFF_ALLMULTI, leading to interesting
behaviour when a multicast router is started. This bug is not
corrected by this commit, and will be fixed with a separate commit.
Detailed changes:
--------------------
netinet/ip_mroute.c all the above.
conf/files make ip_mroute.c optional
net/route.c fix mrt_ioctl hook
netinet/ip_input.c fix ip_mforward hook, move rsvp_input() here
together with other rsvp code, and a couple
of indentation fixes.
netinet/ip_output.c fix ip_mforward and ip_mcast_src hooks
netinet/ip_var.h rsvp function hooks
netinet/raw_ip.c hooks for mrouting and rsvp functions, plus
interface cleanup.
netinet/ip_mroute.h remove an unused and optional field from a struct
Most of the code is from Pavlin Radoslavov and the XORP project
Reviewed by: sam
MFC after: 1 week
2002-11-15 22:53:53 +00:00
|
|
|
int (*legal_vif_num)(int);
|
|
|
|
u_long (*ip_mcast_src)(int);
|
|
|
|
|
|
|
|
void (*rsvp_input_p)(struct mbuf *m, int off);
|
|
|
|
int (*ip_rsvp_vif)(struct socket *, struct sockopt *);
|
|
|
|
void (*ip_rsvp_force_done)(struct socket *);
|
|
|
|
|
2008-07-26 17:32:15 +00:00
|
|
|
/*
|
|
|
|
* Hash functions
|
|
|
|
*/
|
|
|
|
|
|
|
|
#define INP_PCBHASH_RAW_SIZE 256
|
|
|
|
#define INP_PCBHASH_RAW(proto, laddr, faddr, mask) \
|
|
|
|
(((proto) + (laddr) + (faddr)) % (mask) + 1)
|
|
|
|
|
|
|
|
static void
|
|
|
|
rip_inshash(struct inpcb *inp)
|
|
|
|
{
|
|
|
|
struct inpcbinfo *pcbinfo = inp->inp_pcbinfo;
|
|
|
|
struct inpcbhead *pcbhash;
|
|
|
|
int hash;
|
|
|
|
|
|
|
|
INP_INFO_WLOCK_ASSERT(pcbinfo);
|
|
|
|
INP_WLOCK_ASSERT(inp);
|
|
|
|
|
|
|
|
if (inp->inp_ip_p && inp->inp_laddr.s_addr && inp->inp_faddr.s_addr) {
|
|
|
|
hash = INP_PCBHASH_RAW(inp->inp_ip_p, inp->inp_laddr.s_addr,
|
|
|
|
inp->inp_faddr.s_addr, pcbinfo->ipi_hashmask);
|
|
|
|
} else {
|
|
|
|
hash = 0;
|
|
|
|
}
|
|
|
|
pcbhash = &pcbinfo->ipi_hashbase[hash];
|
|
|
|
LIST_INSERT_HEAD(pcbhash, inp, inp_hash);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
rip_delhash(struct inpcb *inp)
|
|
|
|
{
|
|
|
|
INP_WLOCK_ASSERT(inp);
|
|
|
|
LIST_REMOVE(inp, inp_hash);
|
|
|
|
}
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
|
|
|
* Raw interface to IP protocol.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
2002-11-20 19:00:54 +00:00
|
|
|
* Initialize raw connection block q.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
2006-04-21 09:25:40 +00:00
|
|
|
static void
|
|
|
|
rip_zone_change(void *tag)
|
|
|
|
{
|
|
|
|
|
|
|
|
uma_zone_set_max(ripcbinfo.ipi_zone, maxsockets);
|
|
|
|
}
|
|
|
|
|
2006-07-18 22:34:27 +00:00
|
|
|
static int
|
|
|
|
rip_inpcb_init(void *mem, int size, int flags)
|
|
|
|
{
|
2006-12-29 14:58:18 +00:00
|
|
|
struct inpcb *inp = mem;
|
|
|
|
|
2006-07-18 22:34:27 +00:00
|
|
|
INP_LOCK_INIT(inp, "inp", "rawinp");
|
|
|
|
return (0);
|
|
|
|
}
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
void
|
2007-05-10 15:58:48 +00:00
|
|
|
rip_init(void)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2007-05-10 15:58:48 +00:00
|
|
|
|
2002-06-11 10:58:57 +00:00
|
|
|
INP_INFO_LOCK_INIT(&ripcbinfo, "rip");
|
1995-04-09 01:29:31 +00:00
|
|
|
LIST_INIT(&ripcb);
|
2007-04-30 23:12:05 +00:00
|
|
|
ripcbinfo.ipi_listhead = &ripcb;
|
2008-07-26 17:32:15 +00:00
|
|
|
ripcbinfo.ipi_hashbase = hashinit(INP_PCBHASH_RAW_SIZE, M_PCB,
|
|
|
|
&ripcbinfo.ipi_hashmask);
|
2007-04-30 23:12:05 +00:00
|
|
|
ripcbinfo.ipi_porthashbase = hashinit(1, M_PCB,
|
|
|
|
&ripcbinfo.ipi_porthashmask);
|
2002-03-20 05:48:55 +00:00
|
|
|
ripcbinfo.ipi_zone = uma_zcreate("ripcb", sizeof(struct inpcb),
|
2006-07-18 22:34:27 +00:00
|
|
|
NULL, NULL, rip_inpcb_init, NULL, UMA_ALIGN_PTR, UMA_ZONE_NOFREE);
|
2002-03-20 05:48:55 +00:00
|
|
|
uma_zone_set_max(ripcbinfo.ipi_zone, maxsockets);
|
2008-07-05 18:03:39 +00:00
|
|
|
EVENTHANDLER_REGISTER(maxsockets_change, rip_zone_change, NULL,
|
|
|
|
EVENTHANDLER_PRI_ANY);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
2003-09-01 04:23:48 +00:00
|
|
|
static int
|
2008-07-18 10:47:07 +00:00
|
|
|
rip_append(struct inpcb *last, struct ip *ip, struct mbuf *n,
|
|
|
|
struct sockaddr_in *ripsrc)
|
2003-09-01 04:23:48 +00:00
|
|
|
{
|
|
|
|
int policyfail = 0;
|
|
|
|
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RLOCK_ASSERT(last);
|
2004-05-04 00:10:16 +00:00
|
|
|
|
2007-07-03 12:13:45 +00:00
|
|
|
#ifdef IPSEC
|
2003-09-01 04:23:48 +00:00
|
|
|
/* check AH/ESP integrity. */
|
2004-02-03 18:20:55 +00:00
|
|
|
if (ipsec4_in_reject(n, last)) {
|
2003-09-01 04:23:48 +00:00
|
|
|
policyfail = 1;
|
|
|
|
}
|
2007-07-03 12:13:45 +00:00
|
|
|
#endif /* IPSEC */
|
2003-09-01 04:23:48 +00:00
|
|
|
#ifdef MAC
|
2007-10-24 19:04:04 +00:00
|
|
|
if (!policyfail && mac_inpcb_check_deliver(last, n) != 0)
|
2003-09-01 04:23:48 +00:00
|
|
|
policyfail = 1;
|
|
|
|
#endif
|
2005-08-22 16:13:08 +00:00
|
|
|
/* Check the minimum TTL for socket. */
|
|
|
|
if (last->inp_ip_minttl && last->inp_ip_minttl > ip->ip_ttl)
|
|
|
|
policyfail = 1;
|
2003-09-01 04:23:48 +00:00
|
|
|
if (!policyfail) {
|
|
|
|
struct mbuf *opts = NULL;
|
Reduce the number of unnecessary unlock-relocks on socket buffer mutexes
associated with performing a wakeup on the socket buffer:
- When performing an sbappend*() followed by a so[rw]wakeup(), explicitly
acquire the socket buffer lock and use the _locked() variants of both
calls. Note that the _locked() sowakeup() versions unlock the mutex on
return. This is done in uipc_send(), divert_packet(), mroute
socket_send(), raw_append(), tcp_reass(), tcp_input(), and udp_append().
- When the socket buffer lock is dropped before a sowakeup(), remove the
explicit unlock and use the _locked() sowakeup() variant. This is done
in soisdisconnecting(), soisdisconnected() when setting the can't send/
receive flags and dropping data, and in uipc_rcvd() which adjusting
back-pressure on the sockets.
For UNIX domain sockets running mpsafe with a contention-intensive SMP
mysql benchmark, this results in a 1.6% query rate improvement due to
reduce mutex costs.
2004-06-26 19:10:39 +00:00
|
|
|
struct socket *so;
|
2003-09-01 04:23:48 +00:00
|
|
|
|
Reduce the number of unnecessary unlock-relocks on socket buffer mutexes
associated with performing a wakeup on the socket buffer:
- When performing an sbappend*() followed by a so[rw]wakeup(), explicitly
acquire the socket buffer lock and use the _locked() variants of both
calls. Note that the _locked() sowakeup() versions unlock the mutex on
return. This is done in uipc_send(), divert_packet(), mroute
socket_send(), raw_append(), tcp_reass(), tcp_input(), and udp_append().
- When the socket buffer lock is dropped before a sowakeup(), remove the
explicit unlock and use the _locked() sowakeup() variant. This is done
in soisdisconnecting(), soisdisconnected() when setting the can't send/
receive flags and dropping data, and in uipc_rcvd() which adjusting
back-pressure on the sockets.
For UNIX domain sockets running mpsafe with a contention-intensive SMP
mysql benchmark, this results in a 1.6% query rate improvement due to
reduce mutex costs.
2004-06-26 19:10:39 +00:00
|
|
|
so = last->inp_socket;
|
2003-09-01 04:23:48 +00:00
|
|
|
if ((last->inp_flags & INP_CONTROLOPTS) ||
|
2005-10-04 18:19:21 +00:00
|
|
|
(so->so_options & (SO_TIMESTAMP | SO_BINTIME)))
|
2003-09-01 04:23:48 +00:00
|
|
|
ip_savecontrol(last, &opts, ip, n);
|
Reduce the number of unnecessary unlock-relocks on socket buffer mutexes
associated with performing a wakeup on the socket buffer:
- When performing an sbappend*() followed by a so[rw]wakeup(), explicitly
acquire the socket buffer lock and use the _locked() variants of both
calls. Note that the _locked() sowakeup() versions unlock the mutex on
return. This is done in uipc_send(), divert_packet(), mroute
socket_send(), raw_append(), tcp_reass(), tcp_input(), and udp_append().
- When the socket buffer lock is dropped before a sowakeup(), remove the
explicit unlock and use the _locked() sowakeup() variant. This is done
in soisdisconnecting(), soisdisconnected() when setting the can't send/
receive flags and dropping data, and in uipc_rcvd() which adjusting
back-pressure on the sockets.
For UNIX domain sockets running mpsafe with a contention-intensive SMP
mysql benchmark, this results in a 1.6% query rate improvement due to
reduce mutex costs.
2004-06-26 19:10:39 +00:00
|
|
|
SOCKBUF_LOCK(&so->so_rcv);
|
|
|
|
if (sbappendaddr_locked(&so->so_rcv,
|
2008-07-18 10:47:07 +00:00
|
|
|
(struct sockaddr *)ripsrc, n, opts) == 0) {
|
2003-09-01 04:23:48 +00:00
|
|
|
/* should notify about lost packet */
|
|
|
|
m_freem(n);
|
|
|
|
if (opts)
|
|
|
|
m_freem(opts);
|
Reduce the number of unnecessary unlock-relocks on socket buffer mutexes
associated with performing a wakeup on the socket buffer:
- When performing an sbappend*() followed by a so[rw]wakeup(), explicitly
acquire the socket buffer lock and use the _locked() variants of both
calls. Note that the _locked() sowakeup() versions unlock the mutex on
return. This is done in uipc_send(), divert_packet(), mroute
socket_send(), raw_append(), tcp_reass(), tcp_input(), and udp_append().
- When the socket buffer lock is dropped before a sowakeup(), remove the
explicit unlock and use the _locked() sowakeup() variant. This is done
in soisdisconnecting(), soisdisconnected() when setting the can't send/
receive flags and dropping data, and in uipc_rcvd() which adjusting
back-pressure on the sockets.
For UNIX domain sockets running mpsafe with a contention-intensive SMP
mysql benchmark, this results in a 1.6% query rate improvement due to
reduce mutex costs.
2004-06-26 19:10:39 +00:00
|
|
|
SOCKBUF_UNLOCK(&so->so_rcv);
|
2003-09-01 04:23:48 +00:00
|
|
|
} else
|
Reduce the number of unnecessary unlock-relocks on socket buffer mutexes
associated with performing a wakeup on the socket buffer:
- When performing an sbappend*() followed by a so[rw]wakeup(), explicitly
acquire the socket buffer lock and use the _locked() variants of both
calls. Note that the _locked() sowakeup() versions unlock the mutex on
return. This is done in uipc_send(), divert_packet(), mroute
socket_send(), raw_append(), tcp_reass(), tcp_input(), and udp_append().
- When the socket buffer lock is dropped before a sowakeup(), remove the
explicit unlock and use the _locked() sowakeup() variant. This is done
in soisdisconnecting(), soisdisconnected() when setting the can't send/
receive flags and dropping data, and in uipc_rcvd() which adjusting
back-pressure on the sockets.
For UNIX domain sockets running mpsafe with a contention-intensive SMP
mysql benchmark, this results in a 1.6% query rate improvement due to
reduce mutex costs.
2004-06-26 19:10:39 +00:00
|
|
|
sorwakeup_locked(so);
|
2003-09-01 04:23:48 +00:00
|
|
|
} else
|
|
|
|
m_freem(n);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (policyfail);
|
2003-09-01 04:23:48 +00:00
|
|
|
}
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* Setup generic address and protocol structures for raw_input routine, then
|
|
|
|
* pass them along with mbuf chain.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
void
|
2003-09-01 04:23:48 +00:00
|
|
|
rip_input(struct mbuf *m, int off)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2003-09-01 04:23:48 +00:00
|
|
|
struct ip *ip = mtod(m, struct ip *);
|
2001-09-03 20:03:55 +00:00
|
|
|
int proto = ip->ip_p;
|
2003-09-01 04:23:48 +00:00
|
|
|
struct inpcb *inp, *last;
|
2008-07-18 10:47:07 +00:00
|
|
|
struct sockaddr_in ripsrc;
|
2008-07-26 17:32:15 +00:00
|
|
|
int hash;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2008-07-18 10:47:07 +00:00
|
|
|
bzero(&ripsrc, sizeof(ripsrc));
|
|
|
|
ripsrc.sin_len = sizeof(ripsrc);
|
|
|
|
ripsrc.sin_family = AF_INET;
|
1994-05-24 10:09:53 +00:00
|
|
|
ripsrc.sin_addr = ip->ip_src;
|
2003-09-01 04:23:48 +00:00
|
|
|
last = NULL;
|
2008-07-18 10:47:07 +00:00
|
|
|
INP_INFO_RLOCK(&ripcbinfo);
|
2008-07-26 17:32:15 +00:00
|
|
|
hash = INP_PCBHASH_RAW(proto, ip->ip_src.s_addr,
|
|
|
|
ip->ip_dst.s_addr, ripcbinfo.ipi_hashmask);
|
|
|
|
LIST_FOREACH(inp, &ripcbinfo.ipi_hashbase[hash], inp_hash) {
|
|
|
|
INP_RLOCK(inp);
|
|
|
|
if (inp->inp_ip_p != proto) {
|
|
|
|
docontinue1:
|
|
|
|
INP_RUNLOCK(inp);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
#ifdef INET6
|
|
|
|
if ((inp->inp_vflag & INP_IPV4) == 0)
|
|
|
|
goto docontinue1;
|
|
|
|
#endif
|
|
|
|
if (inp->inp_laddr.s_addr != ip->ip_dst.s_addr)
|
|
|
|
goto docontinue1;
|
|
|
|
if (inp->inp_faddr.s_addr != ip->ip_src.s_addr)
|
|
|
|
goto docontinue1;
|
|
|
|
if (jailed(inp->inp_socket->so_cred) &&
|
|
|
|
(htonl(prison_getip(inp->inp_socket->so_cred)) !=
|
|
|
|
ip->ip_dst.s_addr))
|
|
|
|
goto docontinue1;
|
|
|
|
if (last) {
|
|
|
|
struct mbuf *n;
|
|
|
|
|
|
|
|
n = m_copy(m, 0, (int)M_COPYALL);
|
|
|
|
if (n != NULL)
|
|
|
|
(void) rip_append(last, ip, n, &ripsrc);
|
|
|
|
/* XXX count dropped packet */
|
|
|
|
INP_RUNLOCK(last);
|
|
|
|
}
|
|
|
|
last = inp;
|
|
|
|
}
|
|
|
|
LIST_FOREACH(inp, &ripcbinfo.ipi_hashbase[0], inp_hash) {
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RLOCK(inp);
|
2003-09-01 04:23:48 +00:00
|
|
|
if (inp->inp_ip_p && inp->inp_ip_p != proto) {
|
|
|
|
docontinue:
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(inp);
|
2003-09-01 04:23:48 +00:00
|
|
|
continue;
|
|
|
|
}
|
1999-12-22 19:13:38 +00:00
|
|
|
#ifdef INET6
|
|
|
|
if ((inp->inp_vflag & INP_IPV4) == 0)
|
2003-09-01 04:23:48 +00:00
|
|
|
goto docontinue;
|
1999-12-22 19:13:38 +00:00
|
|
|
#endif
|
2002-11-20 19:00:54 +00:00
|
|
|
if (inp->inp_laddr.s_addr &&
|
2004-08-16 18:32:07 +00:00
|
|
|
inp->inp_laddr.s_addr != ip->ip_dst.s_addr)
|
2003-09-01 04:23:48 +00:00
|
|
|
goto docontinue;
|
2002-11-20 19:00:54 +00:00
|
|
|
if (inp->inp_faddr.s_addr &&
|
2004-08-16 18:32:07 +00:00
|
|
|
inp->inp_faddr.s_addr != ip->ip_src.s_addr)
|
2003-09-01 04:23:48 +00:00
|
|
|
goto docontinue;
|
2004-04-26 19:46:52 +00:00
|
|
|
if (jailed(inp->inp_socket->so_cred))
|
2004-05-03 17:15:34 +00:00
|
|
|
if (htonl(prison_getip(inp->inp_socket->so_cred)) !=
|
|
|
|
ip->ip_dst.s_addr)
|
2004-04-26 19:46:52 +00:00
|
|
|
goto docontinue;
|
1994-05-24 10:09:53 +00:00
|
|
|
if (last) {
|
2003-09-01 04:23:48 +00:00
|
|
|
struct mbuf *n;
|
2001-06-11 12:39:29 +00:00
|
|
|
|
2003-09-01 04:23:48 +00:00
|
|
|
n = m_copy(m, 0, (int)M_COPYALL);
|
|
|
|
if (n != NULL)
|
2008-07-18 10:47:07 +00:00
|
|
|
(void) rip_append(last, ip, n, &ripsrc);
|
2003-09-01 04:23:48 +00:00
|
|
|
/* XXX count dropped packet */
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(last);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
1996-11-11 04:56:32 +00:00
|
|
|
last = inp;
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
2003-09-01 04:23:48 +00:00
|
|
|
if (last != NULL) {
|
2008-07-18 10:47:07 +00:00
|
|
|
if (rip_append(last, ip, m, &ripsrc) != 0)
|
2002-07-31 18:30:34 +00:00
|
|
|
ipstat.ips_delivered--;
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(last);
|
1994-05-24 10:09:53 +00:00
|
|
|
} else {
|
|
|
|
m_freem(m);
|
2001-06-11 12:39:29 +00:00
|
|
|
ipstat.ips_noproto++;
|
|
|
|
ipstat.ips_delivered--;
|
|
|
|
}
|
2003-09-01 04:23:48 +00:00
|
|
|
INP_INFO_RUNLOCK(&ripcbinfo);
|
1995-03-16 16:25:55 +00:00
|
|
|
}
|
|
|
|
|
1994-05-24 10:09:53 +00:00
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* Generate IP header and pass packet to ip_output. Tack on options user may
|
|
|
|
* have setup with control call.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
int
|
2003-09-01 04:23:48 +00:00
|
|
|
rip_output(struct mbuf *m, struct socket *so, u_long dst)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
2003-09-01 04:23:48 +00:00
|
|
|
struct ip *ip;
|
2004-06-03 03:15:29 +00:00
|
|
|
int error;
|
2003-09-01 04:23:48 +00:00
|
|
|
struct inpcb *inp = sotoinpcb(so);
|
2004-09-05 02:34:12 +00:00
|
|
|
int flags = ((so->so_options & SO_DONTROUTE) ? IP_ROUTETOIF : 0) |
|
|
|
|
IP_ALLOWBROADCAST;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* If the user handed us a complete IP packet, use it. Otherwise,
|
|
|
|
* allocate an mbuf for a header and fill it in.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
if ((inp->inp_flags & INP_HDRINCL) == 0) {
|
1996-10-25 17:57:53 +00:00
|
|
|
if (m->m_pkthdr.len + sizeof(struct ip) > IP_MAXPACKET) {
|
|
|
|
m_freem(m);
|
|
|
|
return(EMSGSIZE);
|
|
|
|
}
|
2004-07-20 20:52:30 +00:00
|
|
|
M_PREPEND(m, sizeof(struct ip), M_DONTWAIT);
|
2003-08-26 14:11:48 +00:00
|
|
|
if (m == NULL)
|
|
|
|
return(ENOBUFS);
|
2004-06-03 03:15:29 +00:00
|
|
|
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RLOCK(inp);
|
1994-05-24 10:09:53 +00:00
|
|
|
ip = mtod(m, struct ip *);
|
2001-03-09 12:22:51 +00:00
|
|
|
ip->ip_tos = inp->inp_ip_tos;
|
2005-09-26 20:25:16 +00:00
|
|
|
if (inp->inp_flags & INP_DONTFRAG)
|
|
|
|
ip->ip_off = IP_DF;
|
|
|
|
else
|
|
|
|
ip->ip_off = 0;
|
1997-04-03 05:14:45 +00:00
|
|
|
ip->ip_p = inp->inp_ip_p;
|
1994-05-24 10:09:53 +00:00
|
|
|
ip->ip_len = m->m_pkthdr.len;
|
2004-04-26 19:46:52 +00:00
|
|
|
if (jailed(inp->inp_socket->so_cred))
|
|
|
|
ip->ip_src.s_addr =
|
2004-05-03 17:15:34 +00:00
|
|
|
htonl(prison_getip(inp->inp_socket->so_cred));
|
2004-04-26 19:46:52 +00:00
|
|
|
else
|
|
|
|
ip->ip_src = inp->inp_laddr;
|
1994-05-24 10:09:53 +00:00
|
|
|
ip->ip_dst.s_addr = dst;
|
2001-03-09 12:22:51 +00:00
|
|
|
ip->ip_ttl = inp->inp_ip_ttl;
|
1994-05-24 10:09:53 +00:00
|
|
|
} else {
|
1996-10-25 17:57:53 +00:00
|
|
|
if (m->m_pkthdr.len > IP_MAXPACKET) {
|
|
|
|
m_freem(m);
|
|
|
|
return(EMSGSIZE);
|
|
|
|
}
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RLOCK(inp);
|
1994-05-24 10:09:53 +00:00
|
|
|
ip = mtod(m, struct ip *);
|
2004-04-26 19:46:52 +00:00
|
|
|
if (jailed(inp->inp_socket->so_cred)) {
|
|
|
|
if (ip->ip_src.s_addr !=
|
2004-05-03 17:15:34 +00:00
|
|
|
htonl(prison_getip(inp->inp_socket->so_cred))) {
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(inp);
|
2004-04-26 19:46:52 +00:00
|
|
|
m_freem(m);
|
|
|
|
return (EPERM);
|
|
|
|
}
|
|
|
|
}
|
2008-07-05 18:03:39 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Don't allow both user specified and setsockopt options,
|
|
|
|
* and don't allow packet length sizes that will crash.
|
|
|
|
*/
|
|
|
|
if (((ip->ip_hl != (sizeof (*ip) >> 2)) && inp->inp_options)
|
1997-05-22 20:52:56 +00:00
|
|
|
|| (ip->ip_len > m->m_pkthdr.len)
|
2002-10-20 22:52:07 +00:00
|
|
|
|| (ip->ip_len < (ip->ip_hl << 2))) {
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(inp);
|
1996-03-13 08:02:45 +00:00
|
|
|
m_freem(m);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EINVAL);
|
1996-03-13 08:02:45 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
if (ip->ip_id == 0)
|
2004-08-14 15:32:40 +00:00
|
|
|
ip->ip_id = ip_newid();
|
2008-07-05 18:03:39 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* XXX prevent ip_output from overwriting header fields.
|
|
|
|
*/
|
1994-05-24 10:09:53 +00:00
|
|
|
flags |= IP_RAWOUTPUT;
|
|
|
|
ipstat.ips_rawout++;
|
|
|
|
}
|
1999-12-22 19:13:38 +00:00
|
|
|
|
2006-09-06 19:04:36 +00:00
|
|
|
if (inp->inp_flags & INP_ONESBCAST)
|
2003-08-20 14:46:40 +00:00
|
|
|
flags |= IP_SENDONES;
|
|
|
|
|
2004-06-03 03:15:29 +00:00
|
|
|
#ifdef MAC
|
2007-10-24 19:04:04 +00:00
|
|
|
mac_inpcb_create_mbuf(inp, m);
|
2004-06-03 03:15:29 +00:00
|
|
|
#endif
|
|
|
|
|
|
|
|
error = ip_output(m, inp->inp_options, NULL, flags,
|
|
|
|
inp->inp_moptions, inp);
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(inp);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Raw IP socket option processing.
|
2003-07-18 16:10:36 +00:00
|
|
|
*
|
When the access control on creating raw sockets was modified so that
processes in jail could create raw sockets, additional access control
checks were added to raw IP sockets to limit the ways in which those
sockets could be used. Specifically, only the socket option IP_HDRINCL
was permitted in rip_ctloutput(). Other socket options were protected
by a call to suser(). This change was required to prevent processes
in a Jail from modifying system properties such as multicast routing
and firewall rule sets.
However, it also introduced a regression: processes that create a raw
socket with root privilege, but then downgraded credential (i.e., a
daemon giving up root, or a setuid process switching back to the real
uid) could no longer issue other unprivileged generic IP socket option
operations, such as IP_TOS, IP_TTL, and the multicast group membership
options, which prevented multicast routing daemons (and some other
tools) from operating correctly.
This change pushes the access control decision down to the granularity
of individual socket options, rather than all socket options, on raw
IP sockets. When rip_ctloutput() doesn't implement an option, it will
now pass the request directly to in_control() without an access
control check. This should restore the functionality of the generic
IP socket options for raw sockets in the above-described scenarios,
which may be confirmed with the ipsockopt regression test.
RELENG_5 candidate.
Reviewed by: csjp
2004-10-12 16:47:25 +00:00
|
|
|
* IMPORTANT NOTE regarding access control: Traditionally, raw sockets could
|
|
|
|
* only be created by a privileged process, and as such, socket option
|
|
|
|
* operations to manage system properties on any raw socket were allowed to
|
|
|
|
* take place without explicit additional access control checks. However,
|
|
|
|
* raw sockets can now also be created in jail(), and therefore explicit
|
|
|
|
* checks are now required. Likewise, raw sockets can be used by a process
|
|
|
|
* after it gives up privilege, so some caution is required. For options
|
|
|
|
* passed down to the IP layer via ip_ctloutput(), checks are assumed to be
|
|
|
|
* performed in ip_ctloutput() and therefore no check occurs here.
|
2007-07-02 15:44:30 +00:00
|
|
|
* Unilaterally checking priv_check() here breaks normal IP socket option
|
When the access control on creating raw sockets was modified so that
processes in jail could create raw sockets, additional access control
checks were added to raw IP sockets to limit the ways in which those
sockets could be used. Specifically, only the socket option IP_HDRINCL
was permitted in rip_ctloutput(). Other socket options were protected
by a call to suser(). This change was required to prevent processes
in a Jail from modifying system properties such as multicast routing
and firewall rule sets.
However, it also introduced a regression: processes that create a raw
socket with root privilege, but then downgraded credential (i.e., a
daemon giving up root, or a setuid process switching back to the real
uid) could no longer issue other unprivileged generic IP socket option
operations, such as IP_TOS, IP_TTL, and the multicast group membership
options, which prevented multicast routing daemons (and some other
tools) from operating correctly.
This change pushes the access control decision down to the granularity
of individual socket options, rather than all socket options, on raw
IP sockets. When rip_ctloutput() doesn't implement an option, it will
now pass the request directly to in_control() without an access
control check. This should restore the functionality of the generic
IP socket options for raw sockets in the above-described scenarios,
which may be confirmed with the ipsockopt regression test.
RELENG_5 candidate.
Reviewed by: csjp
2004-10-12 16:47:25 +00:00
|
|
|
* operations on raw sockets.
|
|
|
|
*
|
|
|
|
* When adding new socket options here, make sure to add access control
|
|
|
|
* checks here as necessary.
|
1994-05-24 10:09:53 +00:00
|
|
|
*/
|
|
|
|
int
|
2003-09-01 04:23:48 +00:00
|
|
|
rip_ctloutput(struct socket *so, struct sockopt *sopt)
|
1994-05-24 10:09:53 +00:00
|
|
|
{
|
1998-08-23 03:07:17 +00:00
|
|
|
struct inpcb *inp = sotoinpcb(so);
|
|
|
|
int error, optval;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1998-08-23 03:07:17 +00:00
|
|
|
if (sopt->sopt_level != IPPROTO_IP)
|
1994-05-24 10:09:53 +00:00
|
|
|
return (EINVAL);
|
|
|
|
|
1998-08-23 03:07:17 +00:00
|
|
|
error = 0;
|
|
|
|
switch (sopt->sopt_dir) {
|
|
|
|
case SOPT_GET:
|
|
|
|
switch (sopt->sopt_name) {
|
|
|
|
case IP_HDRINCL:
|
|
|
|
optval = inp->inp_flags & INP_HDRINCL;
|
|
|
|
error = sooptcopyout(sopt, &optval, sizeof optval);
|
|
|
|
break;
|
|
|
|
|
2001-11-04 22:56:25 +00:00
|
|
|
case IP_FW_ADD: /* ADD actually returns the body... */
|
1998-08-23 03:07:17 +00:00
|
|
|
case IP_FW_GET:
|
2004-06-09 20:10:38 +00:00
|
|
|
case IP_FW_TABLE_GETSIZE:
|
|
|
|
case IP_FW_TABLE_LIST:
|
2006-12-29 21:59:17 +00:00
|
|
|
case IP_FW_NAT_GET_CONFIG:
|
|
|
|
case IP_FW_NAT_GET_LOG:
|
2004-08-17 22:05:54 +00:00
|
|
|
if (ip_fw_ctl_ptr != NULL)
|
1998-08-23 03:07:17 +00:00
|
|
|
error = ip_fw_ctl_ptr(sopt);
|
2001-11-04 22:56:25 +00:00
|
|
|
else
|
|
|
|
error = ENOPROTOOPT;
|
1998-08-23 03:07:17 +00:00
|
|
|
break;
|
|
|
|
|
1998-12-14 18:09:13 +00:00
|
|
|
case IP_DUMMYNET_GET:
|
2004-08-17 22:05:54 +00:00
|
|
|
if (ip_dn_ctl_ptr != NULL)
|
1998-12-14 18:09:13 +00:00
|
|
|
error = ip_dn_ctl_ptr(sopt);
|
2001-11-04 22:56:25 +00:00
|
|
|
else
|
|
|
|
error = ENOPROTOOPT;
|
1998-12-14 18:09:13 +00:00
|
|
|
break ;
|
1998-08-23 03:07:17 +00:00
|
|
|
|
|
|
|
case MRT_INIT:
|
|
|
|
case MRT_DONE:
|
|
|
|
case MRT_ADD_VIF:
|
|
|
|
case MRT_DEL_VIF:
|
|
|
|
case MRT_ADD_MFC:
|
|
|
|
case MRT_DEL_MFC:
|
|
|
|
case MRT_VERSION:
|
|
|
|
case MRT_ASSERT:
|
2003-08-07 18:16:59 +00:00
|
|
|
case MRT_API_SUPPORT:
|
|
|
|
case MRT_API_CONFIG:
|
|
|
|
case MRT_ADD_BW_UPCALL:
|
|
|
|
case MRT_DEL_BW_UPCALL:
|
2006-11-06 13:42:10 +00:00
|
|
|
error = priv_check(curthread, PRIV_NETINET_MROUTE);
|
When the access control on creating raw sockets was modified so that
processes in jail could create raw sockets, additional access control
checks were added to raw IP sockets to limit the ways in which those
sockets could be used. Specifically, only the socket option IP_HDRINCL
was permitted in rip_ctloutput(). Other socket options were protected
by a call to suser(). This change was required to prevent processes
in a Jail from modifying system properties such as multicast routing
and firewall rule sets.
However, it also introduced a regression: processes that create a raw
socket with root privilege, but then downgraded credential (i.e., a
daemon giving up root, or a setuid process switching back to the real
uid) could no longer issue other unprivileged generic IP socket option
operations, such as IP_TOS, IP_TTL, and the multicast group membership
options, which prevented multicast routing daemons (and some other
tools) from operating correctly.
This change pushes the access control decision down to the granularity
of individual socket options, rather than all socket options, on raw
IP sockets. When rip_ctloutput() doesn't implement an option, it will
now pass the request directly to in_control() without an access
control check. This should restore the functionality of the generic
IP socket options for raw sockets in the above-described scenarios,
which may be confirmed with the ipsockopt regression test.
RELENG_5 candidate.
Reviewed by: csjp
2004-10-12 16:47:25 +00:00
|
|
|
if (error != 0)
|
|
|
|
return (error);
|
Massive cleanup of the ip_mroute code.
No functional changes, but:
+ the mrouting module now should behave the same as the compiled-in
version (it did not before, some of the rsvp code was not loaded
properly);
+ netinet/ip_mroute.c is now truly optional;
+ removed some redundant/unused code;
+ changed many instances of '0' to NULL and INADDR_ANY as appropriate;
+ removed several static variables to make the code more SMP-friendly;
+ fixed some minor bugs in the mrouting code (mostly, incorrect return
values from functions).
This commit is also a prerequisite to the addition of support for PIM,
which i would like to put in before DP2 (it does not change any of
the existing APIs, anyways).
Note, in the process we found out that some device drivers fail to
properly handle changes in IFF_ALLMULTI, leading to interesting
behaviour when a multicast router is started. This bug is not
corrected by this commit, and will be fixed with a separate commit.
Detailed changes:
--------------------
netinet/ip_mroute.c all the above.
conf/files make ip_mroute.c optional
net/route.c fix mrt_ioctl hook
netinet/ip_input.c fix ip_mforward hook, move rsvp_input() here
together with other rsvp code, and a couple
of indentation fixes.
netinet/ip_output.c fix ip_mforward and ip_mcast_src hooks
netinet/ip_var.h rsvp function hooks
netinet/raw_ip.c hooks for mrouting and rsvp functions, plus
interface cleanup.
netinet/ip_mroute.h remove an unused and optional field from a struct
Most of the code is from Pavlin Radoslavov and the XORP project
Reviewed by: sam
MFC after: 1 week
2002-11-15 22:53:53 +00:00
|
|
|
error = ip_mrouter_get ? ip_mrouter_get(so, sopt) :
|
|
|
|
EOPNOTSUPP;
|
1998-08-23 03:07:17 +00:00
|
|
|
break;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1998-08-23 03:07:17 +00:00
|
|
|
default:
|
|
|
|
error = ip_ctloutput(so, sopt);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case SOPT_SET:
|
|
|
|
switch (sopt->sopt_name) {
|
|
|
|
case IP_HDRINCL:
|
|
|
|
error = sooptcopyin(sopt, &optval, sizeof optval,
|
|
|
|
sizeof optval);
|
|
|
|
if (error)
|
|
|
|
break;
|
|
|
|
if (optval)
|
1995-09-21 19:59:43 +00:00
|
|
|
inp->inp_flags |= INP_HDRINCL;
|
|
|
|
else
|
|
|
|
inp->inp_flags &= ~INP_HDRINCL;
|
1998-08-23 03:07:17 +00:00
|
|
|
break;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2001-11-26 10:05:58 +00:00
|
|
|
case IP_FW_ADD:
|
1998-08-23 03:07:17 +00:00
|
|
|
case IP_FW_DEL:
|
|
|
|
case IP_FW_FLUSH:
|
|
|
|
case IP_FW_ZERO:
|
1999-08-01 16:57:24 +00:00
|
|
|
case IP_FW_RESETLOG:
|
2004-06-09 20:10:38 +00:00
|
|
|
case IP_FW_TABLE_ADD:
|
|
|
|
case IP_FW_TABLE_DEL:
|
|
|
|
case IP_FW_TABLE_FLUSH:
|
2006-12-29 21:59:17 +00:00
|
|
|
case IP_FW_NAT_CFG:
|
|
|
|
case IP_FW_NAT_DEL:
|
2004-08-17 22:05:54 +00:00
|
|
|
if (ip_fw_ctl_ptr != NULL)
|
1998-08-23 03:07:17 +00:00
|
|
|
error = ip_fw_ctl_ptr(sopt);
|
2001-11-04 22:56:25 +00:00
|
|
|
else
|
|
|
|
error = ENOPROTOOPT;
|
1998-08-23 03:07:17 +00:00
|
|
|
break;
|
1995-01-12 13:06:32 +00:00
|
|
|
|
1998-12-14 18:09:13 +00:00
|
|
|
case IP_DUMMYNET_CONFIGURE:
|
|
|
|
case IP_DUMMYNET_DEL:
|
|
|
|
case IP_DUMMYNET_FLUSH:
|
2004-08-17 22:05:54 +00:00
|
|
|
if (ip_dn_ctl_ptr != NULL)
|
1998-12-14 18:09:13 +00:00
|
|
|
error = ip_dn_ctl_ptr(sopt);
|
2001-11-04 22:56:25 +00:00
|
|
|
else
|
|
|
|
error = ENOPROTOOPT ;
|
1998-12-14 18:09:13 +00:00
|
|
|
break ;
|
1996-08-21 21:37:07 +00:00
|
|
|
|
1998-08-23 03:07:17 +00:00
|
|
|
case IP_RSVP_ON:
|
2006-11-06 13:42:10 +00:00
|
|
|
error = priv_check(curthread, PRIV_NETINET_MROUTE);
|
When the access control on creating raw sockets was modified so that
processes in jail could create raw sockets, additional access control
checks were added to raw IP sockets to limit the ways in which those
sockets could be used. Specifically, only the socket option IP_HDRINCL
was permitted in rip_ctloutput(). Other socket options were protected
by a call to suser(). This change was required to prevent processes
in a Jail from modifying system properties such as multicast routing
and firewall rule sets.
However, it also introduced a regression: processes that create a raw
socket with root privilege, but then downgraded credential (i.e., a
daemon giving up root, or a setuid process switching back to the real
uid) could no longer issue other unprivileged generic IP socket option
operations, such as IP_TOS, IP_TTL, and the multicast group membership
options, which prevented multicast routing daemons (and some other
tools) from operating correctly.
This change pushes the access control decision down to the granularity
of individual socket options, rather than all socket options, on raw
IP sockets. When rip_ctloutput() doesn't implement an option, it will
now pass the request directly to in_control() without an access
control check. This should restore the functionality of the generic
IP socket options for raw sockets in the above-described scenarios,
which may be confirmed with the ipsockopt regression test.
RELENG_5 candidate.
Reviewed by: csjp
2004-10-12 16:47:25 +00:00
|
|
|
if (error != 0)
|
|
|
|
return (error);
|
1998-08-23 03:07:17 +00:00
|
|
|
error = ip_rsvp_init(so);
|
|
|
|
break;
|
1994-09-06 22:42:31 +00:00
|
|
|
|
1998-08-23 03:07:17 +00:00
|
|
|
case IP_RSVP_OFF:
|
2006-11-06 13:42:10 +00:00
|
|
|
error = priv_check(curthread, PRIV_NETINET_MROUTE);
|
When the access control on creating raw sockets was modified so that
processes in jail could create raw sockets, additional access control
checks were added to raw IP sockets to limit the ways in which those
sockets could be used. Specifically, only the socket option IP_HDRINCL
was permitted in rip_ctloutput(). Other socket options were protected
by a call to suser(). This change was required to prevent processes
in a Jail from modifying system properties such as multicast routing
and firewall rule sets.
However, it also introduced a regression: processes that create a raw
socket with root privilege, but then downgraded credential (i.e., a
daemon giving up root, or a setuid process switching back to the real
uid) could no longer issue other unprivileged generic IP socket option
operations, such as IP_TOS, IP_TTL, and the multicast group membership
options, which prevented multicast routing daemons (and some other
tools) from operating correctly.
This change pushes the access control decision down to the granularity
of individual socket options, rather than all socket options, on raw
IP sockets. When rip_ctloutput() doesn't implement an option, it will
now pass the request directly to in_control() without an access
control check. This should restore the functionality of the generic
IP socket options for raw sockets in the above-described scenarios,
which may be confirmed with the ipsockopt regression test.
RELENG_5 candidate.
Reviewed by: csjp
2004-10-12 16:47:25 +00:00
|
|
|
if (error != 0)
|
|
|
|
return (error);
|
1998-08-23 03:07:17 +00:00
|
|
|
error = ip_rsvp_done();
|
|
|
|
break;
|
1994-09-06 22:42:31 +00:00
|
|
|
|
1998-08-23 03:07:17 +00:00
|
|
|
case IP_RSVP_VIF_ON:
|
|
|
|
case IP_RSVP_VIF_OFF:
|
2006-11-06 13:42:10 +00:00
|
|
|
error = priv_check(curthread, PRIV_NETINET_MROUTE);
|
When the access control on creating raw sockets was modified so that
processes in jail could create raw sockets, additional access control
checks were added to raw IP sockets to limit the ways in which those
sockets could be used. Specifically, only the socket option IP_HDRINCL
was permitted in rip_ctloutput(). Other socket options were protected
by a call to suser(). This change was required to prevent processes
in a Jail from modifying system properties such as multicast routing
and firewall rule sets.
However, it also introduced a regression: processes that create a raw
socket with root privilege, but then downgraded credential (i.e., a
daemon giving up root, or a setuid process switching back to the real
uid) could no longer issue other unprivileged generic IP socket option
operations, such as IP_TOS, IP_TTL, and the multicast group membership
options, which prevented multicast routing daemons (and some other
tools) from operating correctly.
This change pushes the access control decision down to the granularity
of individual socket options, rather than all socket options, on raw
IP sockets. When rip_ctloutput() doesn't implement an option, it will
now pass the request directly to in_control() without an access
control check. This should restore the functionality of the generic
IP socket options for raw sockets in the above-described scenarios,
which may be confirmed with the ipsockopt regression test.
RELENG_5 candidate.
Reviewed by: csjp
2004-10-12 16:47:25 +00:00
|
|
|
if (error != 0)
|
|
|
|
return (error);
|
Massive cleanup of the ip_mroute code.
No functional changes, but:
+ the mrouting module now should behave the same as the compiled-in
version (it did not before, some of the rsvp code was not loaded
properly);
+ netinet/ip_mroute.c is now truly optional;
+ removed some redundant/unused code;
+ changed many instances of '0' to NULL and INADDR_ANY as appropriate;
+ removed several static variables to make the code more SMP-friendly;
+ fixed some minor bugs in the mrouting code (mostly, incorrect return
values from functions).
This commit is also a prerequisite to the addition of support for PIM,
which i would like to put in before DP2 (it does not change any of
the existing APIs, anyways).
Note, in the process we found out that some device drivers fail to
properly handle changes in IFF_ALLMULTI, leading to interesting
behaviour when a multicast router is started. This bug is not
corrected by this commit, and will be fixed with a separate commit.
Detailed changes:
--------------------
netinet/ip_mroute.c all the above.
conf/files make ip_mroute.c optional
net/route.c fix mrt_ioctl hook
netinet/ip_input.c fix ip_mforward hook, move rsvp_input() here
together with other rsvp code, and a couple
of indentation fixes.
netinet/ip_output.c fix ip_mforward and ip_mcast_src hooks
netinet/ip_var.h rsvp function hooks
netinet/raw_ip.c hooks for mrouting and rsvp functions, plus
interface cleanup.
netinet/ip_mroute.h remove an unused and optional field from a struct
Most of the code is from Pavlin Radoslavov and the XORP project
Reviewed by: sam
MFC after: 1 week
2002-11-15 22:53:53 +00:00
|
|
|
error = ip_rsvp_vif ?
|
|
|
|
ip_rsvp_vif(so, sopt) : EINVAL;
|
1998-08-23 03:07:17 +00:00
|
|
|
break;
|
|
|
|
|
|
|
|
case MRT_INIT:
|
|
|
|
case MRT_DONE:
|
|
|
|
case MRT_ADD_VIF:
|
|
|
|
case MRT_DEL_VIF:
|
|
|
|
case MRT_ADD_MFC:
|
|
|
|
case MRT_DEL_MFC:
|
|
|
|
case MRT_VERSION:
|
|
|
|
case MRT_ASSERT:
|
2003-08-07 18:16:59 +00:00
|
|
|
case MRT_API_SUPPORT:
|
|
|
|
case MRT_API_CONFIG:
|
|
|
|
case MRT_ADD_BW_UPCALL:
|
|
|
|
case MRT_DEL_BW_UPCALL:
|
2006-11-06 13:42:10 +00:00
|
|
|
error = priv_check(curthread, PRIV_NETINET_MROUTE);
|
When the access control on creating raw sockets was modified so that
processes in jail could create raw sockets, additional access control
checks were added to raw IP sockets to limit the ways in which those
sockets could be used. Specifically, only the socket option IP_HDRINCL
was permitted in rip_ctloutput(). Other socket options were protected
by a call to suser(). This change was required to prevent processes
in a Jail from modifying system properties such as multicast routing
and firewall rule sets.
However, it also introduced a regression: processes that create a raw
socket with root privilege, but then downgraded credential (i.e., a
daemon giving up root, or a setuid process switching back to the real
uid) could no longer issue other unprivileged generic IP socket option
operations, such as IP_TOS, IP_TTL, and the multicast group membership
options, which prevented multicast routing daemons (and some other
tools) from operating correctly.
This change pushes the access control decision down to the granularity
of individual socket options, rather than all socket options, on raw
IP sockets. When rip_ctloutput() doesn't implement an option, it will
now pass the request directly to in_control() without an access
control check. This should restore the functionality of the generic
IP socket options for raw sockets in the above-described scenarios,
which may be confirmed with the ipsockopt regression test.
RELENG_5 candidate.
Reviewed by: csjp
2004-10-12 16:47:25 +00:00
|
|
|
if (error != 0)
|
|
|
|
return (error);
|
Massive cleanup of the ip_mroute code.
No functional changes, but:
+ the mrouting module now should behave the same as the compiled-in
version (it did not before, some of the rsvp code was not loaded
properly);
+ netinet/ip_mroute.c is now truly optional;
+ removed some redundant/unused code;
+ changed many instances of '0' to NULL and INADDR_ANY as appropriate;
+ removed several static variables to make the code more SMP-friendly;
+ fixed some minor bugs in the mrouting code (mostly, incorrect return
values from functions).
This commit is also a prerequisite to the addition of support for PIM,
which i would like to put in before DP2 (it does not change any of
the existing APIs, anyways).
Note, in the process we found out that some device drivers fail to
properly handle changes in IFF_ALLMULTI, leading to interesting
behaviour when a multicast router is started. This bug is not
corrected by this commit, and will be fixed with a separate commit.
Detailed changes:
--------------------
netinet/ip_mroute.c all the above.
conf/files make ip_mroute.c optional
net/route.c fix mrt_ioctl hook
netinet/ip_input.c fix ip_mforward hook, move rsvp_input() here
together with other rsvp code, and a couple
of indentation fixes.
netinet/ip_output.c fix ip_mforward and ip_mcast_src hooks
netinet/ip_var.h rsvp function hooks
netinet/raw_ip.c hooks for mrouting and rsvp functions, plus
interface cleanup.
netinet/ip_mroute.h remove an unused and optional field from a struct
Most of the code is from Pavlin Radoslavov and the XORP project
Reviewed by: sam
MFC after: 1 week
2002-11-15 22:53:53 +00:00
|
|
|
error = ip_mrouter_set ? ip_mrouter_set(so, sopt) :
|
|
|
|
EOPNOTSUPP;
|
1998-08-23 03:07:17 +00:00
|
|
|
break;
|
|
|
|
|
|
|
|
default:
|
|
|
|
error = ip_ctloutput(so, sopt);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
break;
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
1998-08-23 03:07:17 +00:00
|
|
|
|
|
|
|
return (error);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
|
|
|
|
1997-02-13 19:46:45 +00:00
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* This function exists solely to receive the PRC_IFDOWN messages which are
|
|
|
|
* sent by if_down(). It looks for an ifaddr whose ifa_addr is sa, and calls
|
|
|
|
* in_ifadown() to remove all routes corresponding to that address. It also
|
|
|
|
* receives the PRC_IFUP messages from if_up() and reinstalls the interface
|
|
|
|
* routes.
|
1997-02-13 19:46:45 +00:00
|
|
|
*/
|
|
|
|
void
|
2003-09-01 04:23:48 +00:00
|
|
|
rip_ctlinput(int cmd, struct sockaddr *sa, void *vip)
|
1997-02-13 19:46:45 +00:00
|
|
|
{
|
|
|
|
struct in_ifaddr *ia;
|
|
|
|
struct ifnet *ifp;
|
|
|
|
int err;
|
|
|
|
int flags;
|
|
|
|
|
1998-08-23 03:07:17 +00:00
|
|
|
switch (cmd) {
|
1997-02-13 19:46:45 +00:00
|
|
|
case PRC_IFDOWN:
|
2001-03-16 20:00:53 +00:00
|
|
|
TAILQ_FOREACH(ia, &in_ifaddrhead, ia_link) {
|
1997-02-13 19:46:45 +00:00
|
|
|
if (ia->ia_ifa.ifa_addr == sa
|
|
|
|
&& (ia->ia_flags & IFA_ROUTE)) {
|
|
|
|
/*
|
|
|
|
* in_ifscrub kills the interface route.
|
|
|
|
*/
|
|
|
|
in_ifscrub(ia->ia_ifp, ia);
|
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* 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.
|
1997-02-13 19:46:45 +00:00
|
|
|
*/
|
2001-05-11 14:37:34 +00:00
|
|
|
in_ifadown(&ia->ia_ifa, 0);
|
1997-02-13 19:46:45 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case PRC_IFUP:
|
2001-03-16 20:00:53 +00:00
|
|
|
TAILQ_FOREACH(ia, &in_ifaddrhead, ia_link) {
|
1997-02-13 19:46:45 +00:00
|
|
|
if (ia->ia_ifa.ifa_addr == sa)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (ia == 0 || (ia->ia_flags & IFA_ROUTE))
|
|
|
|
return;
|
|
|
|
flags = RTF_UP;
|
|
|
|
ifp = ia->ia_ifa.ifa_ifp;
|
|
|
|
|
|
|
|
if ((ifp->if_flags & IFF_LOOPBACK)
|
|
|
|
|| (ifp->if_flags & IFF_POINTOPOINT))
|
|
|
|
flags |= RTF_HOST;
|
|
|
|
|
|
|
|
err = rtinit(&ia->ia_ifa, RTM_ADD, flags);
|
|
|
|
if (err == 0)
|
|
|
|
ia->ia_flags |= IFA_ROUTE;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2007-03-20 13:15:20 +00:00
|
|
|
u_long rip_sendspace = 9216;
|
|
|
|
u_long rip_recvspace = 9216;
|
1995-10-21 02:12:20 +00:00
|
|
|
|
2005-12-14 22:27:48 +00:00
|
|
|
SYSCTL_ULONG(_net_inet_raw, OID_AUTO, maxdgram, CTLFLAG_RW,
|
1999-05-03 23:57:32 +00:00
|
|
|
&rip_sendspace, 0, "Maximum outgoing raw IP datagram size");
|
2005-12-14 22:27:48 +00:00
|
|
|
SYSCTL_ULONG(_net_inet_raw, OID_AUTO, recvspace, CTLFLAG_RW,
|
2004-01-27 22:17:39 +00:00
|
|
|
&rip_recvspace, 0, "Maximum space for incoming raw IP datagrams");
|
1995-10-21 02:12:20 +00:00
|
|
|
|
1997-02-18 20:46:36 +00:00
|
|
|
static int
|
2001-09-12 08:38:13 +00:00
|
|
|
rip_attach(struct socket *so, int proto, struct thread *td)
|
1997-02-18 20:46:36 +00:00
|
|
|
{
|
|
|
|
struct inpcb *inp;
|
2003-09-01 04:23:48 +00:00
|
|
|
int error;
|
1997-02-18 20:46:36 +00:00
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp == NULL, ("rip_attach: inp != NULL"));
|
2007-06-12 00:12:01 +00:00
|
|
|
|
|
|
|
error = priv_check(td, PRIV_NETINET_RAW);
|
2006-11-06 13:42:10 +00:00
|
|
|
if (error)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
if (proto >= IPPROTO_MAX || proto < 0)
|
2002-10-29 16:46:13 +00:00
|
|
|
return EPROTONOSUPPORT;
|
1999-12-22 19:13:38 +00:00
|
|
|
error = soreserve(so, rip_sendspace, rip_recvspace);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
if (error)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
INP_INFO_WLOCK(&ripcbinfo);
|
2006-07-18 22:34:27 +00:00
|
|
|
error = in_pcballoc(so, &ripcbinfo);
|
2003-09-01 04:23:48 +00:00
|
|
|
if (error) {
|
|
|
|
INP_INFO_WUNLOCK(&ripcbinfo);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
2003-09-01 04:23:48 +00:00
|
|
|
}
|
1997-02-18 20:46:36 +00:00
|
|
|
inp = (struct inpcb *)so->so_pcb;
|
1999-12-22 19:13:38 +00:00
|
|
|
inp->inp_vflag |= INP_IPV4;
|
1997-04-03 05:14:45 +00:00
|
|
|
inp->inp_ip_p = proto;
|
2001-03-09 12:22:51 +00:00
|
|
|
inp->inp_ip_ttl = ip_defttl;
|
2008-07-26 17:32:15 +00:00
|
|
|
rip_inshash(inp);
|
|
|
|
INP_INFO_WUNLOCK(&ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (0);
|
1997-02-18 20:46:36 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2003-11-08 22:53:41 +00:00
|
|
|
static void
|
2006-07-21 17:11:15 +00:00
|
|
|
rip_detach(struct socket *so)
|
2003-11-08 22:53:41 +00:00
|
|
|
{
|
2006-07-21 17:11:15 +00:00
|
|
|
struct inpcb *inp;
|
2005-06-01 11:38:35 +00:00
|
|
|
|
2006-07-21 17:11:15 +00:00
|
|
|
inp = sotoinpcb(so);
|
|
|
|
KASSERT(inp != NULL, ("rip_detach: inp == NULL"));
|
|
|
|
KASSERT(inp->inp_faddr.s_addr == INADDR_ANY,
|
|
|
|
("rip_detach: not closed"));
|
2003-11-08 22:53:41 +00:00
|
|
|
|
2006-07-21 17:11:15 +00:00
|
|
|
INP_INFO_WLOCK(&ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
2008-07-26 17:32:15 +00:00
|
|
|
rip_delhash(inp);
|
2003-11-08 22:53:41 +00:00
|
|
|
if (so == ip_mrouter && ip_mrouter_done)
|
|
|
|
ip_mrouter_done();
|
|
|
|
if (ip_rsvp_force_done)
|
|
|
|
ip_rsvp_force_done(so);
|
|
|
|
if (so == ip_rsvpd)
|
|
|
|
ip_rsvp_done();
|
|
|
|
in_pcbdetach(inp);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
in_pcbfree(inp);
|
2006-07-21 17:11:15 +00:00
|
|
|
INP_INFO_WUNLOCK(&ripcbinfo);
|
2003-11-08 22:53:41 +00:00
|
|
|
}
|
|
|
|
|
Chance protocol switch method pru_detach() so that it returns void
rather than an error. Detaches do not "fail", they other occur or
the protocol flags SS_PROTOREF to take ownership of the socket.
soclose() no longer looks at so_pcb to see if it's NULL, relying
entirely on the protocol to decide whether it's time to free the
socket or not using SS_PROTOREF. so_pcb is now entirely owned and
managed by the protocol code. Likewise, no longer test so_pcb in
other socket functions, such as soreceive(), which have no business
digging into protocol internals.
Protocol detach routines no longer try to free the socket on detach,
this is performed in the socket code if the protocol permits it.
In rts_detach(), no longer test for rp != NULL in detach, and
likewise in other protocols that don't permit a NULL so_pcb, reduce
the incidence of testing for it during detach.
netinet and netinet6 are not fully updated to this change, which
will be in an upcoming commit. In their current state they may leak
memory or panic.
MFC after: 3 months
2006-04-01 15:42:02 +00:00
|
|
|
static void
|
2006-07-21 17:11:15 +00:00
|
|
|
rip_dodisconnect(struct socket *so, struct inpcb *inp)
|
|
|
|
{
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK_ASSERT(inp);
|
2006-07-21 17:11:15 +00:00
|
|
|
|
2008-07-26 17:32:15 +00:00
|
|
|
rip_delhash(inp);
|
2006-07-21 17:11:15 +00:00
|
|
|
inp->inp_faddr.s_addr = INADDR_ANY;
|
2008-07-26 17:32:15 +00:00
|
|
|
rip_inshash(inp);
|
2006-07-21 17:11:15 +00:00
|
|
|
SOCK_LOCK(so);
|
|
|
|
so->so_state &= ~SS_ISCONNECTED;
|
|
|
|
SOCK_UNLOCK(so);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
rip_abort(struct socket *so)
|
1997-02-18 20:46:36 +00:00
|
|
|
{
|
|
|
|
struct inpcb *inp;
|
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
2006-07-21 17:11:15 +00:00
|
|
|
KASSERT(inp != NULL, ("rip_abort: inp == NULL"));
|
|
|
|
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
INP_INFO_WLOCK(&ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
2006-07-21 17:11:15 +00:00
|
|
|
rip_dodisconnect(so, inp);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
2003-09-01 04:23:48 +00:00
|
|
|
INP_INFO_WUNLOCK(&ripcbinfo);
|
1997-02-18 20:46:36 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2006-04-01 15:15:05 +00:00
|
|
|
static void
|
2006-07-21 17:11:15 +00:00
|
|
|
rip_close(struct socket *so)
|
1997-02-18 20:46:36 +00:00
|
|
|
{
|
2003-11-08 22:53:41 +00:00
|
|
|
struct inpcb *inp;
|
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
2006-07-21 17:11:15 +00:00
|
|
|
KASSERT(inp != NULL, ("rip_close: inp == NULL"));
|
|
|
|
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
INP_INFO_WLOCK(&ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
2006-07-21 17:11:15 +00:00
|
|
|
rip_dodisconnect(so, inp);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
2003-11-08 22:53:41 +00:00
|
|
|
INP_INFO_WUNLOCK(&ripcbinfo);
|
1997-02-18 20:46:36 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1997-02-18 20:46:36 +00:00
|
|
|
static int
|
|
|
|
rip_disconnect(struct socket *so)
|
|
|
|
{
|
2006-05-15 09:28:57 +00:00
|
|
|
struct inpcb *inp;
|
|
|
|
|
2002-05-31 11:52:35 +00:00
|
|
|
if ((so->so_state & SS_ISCONNECTED) == 0)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (ENOTCONN);
|
2006-05-15 09:28:57 +00:00
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
|
|
|
KASSERT(inp != NULL, ("rip_disconnect: inp == NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
2006-05-15 09:28:57 +00:00
|
|
|
INP_INFO_WLOCK(&ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
2006-07-21 17:11:15 +00:00
|
|
|
rip_dodisconnect(so, inp);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
2006-05-15 09:28:57 +00:00
|
|
|
INP_INFO_WUNLOCK(&ripcbinfo);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
return (0);
|
1997-02-18 20:46:36 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1997-02-18 20:46:36 +00:00
|
|
|
static int
|
2001-09-12 08:38:13 +00:00
|
|
|
rip_bind(struct socket *so, struct sockaddr *nam, struct thread *td)
|
1997-02-18 20:46:36 +00:00
|
|
|
{
|
1997-08-16 19:16:27 +00:00
|
|
|
struct sockaddr_in *addr = (struct sockaddr_in *)nam;
|
2003-11-08 22:53:41 +00:00
|
|
|
struct inpcb *inp;
|
1997-02-18 20:46:36 +00:00
|
|
|
|
1997-08-16 19:16:27 +00:00
|
|
|
if (nam->sa_len != sizeof(*addr))
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EINVAL);
|
1997-02-18 20:46:36 +00:00
|
|
|
|
2004-04-26 19:46:52 +00:00
|
|
|
if (jailed(td->td_ucred)) {
|
2004-05-03 17:15:34 +00:00
|
|
|
if (addr->sin_addr.s_addr == INADDR_ANY)
|
2004-04-26 19:46:52 +00:00
|
|
|
addr->sin_addr.s_addr =
|
2004-05-03 17:15:34 +00:00
|
|
|
htonl(prison_getip(td->td_ucred));
|
|
|
|
if (htonl(prison_getip(td->td_ucred)) != addr->sin_addr.s_addr)
|
2004-04-26 19:46:52 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
}
|
|
|
|
|
2003-11-08 22:53:41 +00:00
|
|
|
if (TAILQ_EMPTY(&ifnet) ||
|
|
|
|
(addr->sin_family != AF_INET && addr->sin_family != AF_IMPLINK) ||
|
2002-11-20 19:00:54 +00:00
|
|
|
(addr->sin_addr.s_addr &&
|
1997-02-18 20:46:36 +00:00
|
|
|
ifa_ifwithaddr((struct sockaddr *)addr) == 0))
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
2003-11-08 22:53:41 +00:00
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp != NULL, ("rip_bind: inp == NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
INP_INFO_WLOCK(&ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
2008-07-26 17:32:15 +00:00
|
|
|
rip_delhash(inp);
|
1997-02-18 20:46:36 +00:00
|
|
|
inp->inp_laddr = addr->sin_addr;
|
2008-07-26 17:32:15 +00:00
|
|
|
rip_inshash(inp);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
2003-11-08 22:53:41 +00:00
|
|
|
INP_INFO_WUNLOCK(&ripcbinfo);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (0);
|
1997-02-18 20:46:36 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1997-02-18 20:46:36 +00:00
|
|
|
static int
|
2001-09-12 08:38:13 +00:00
|
|
|
rip_connect(struct socket *so, struct sockaddr *nam, struct thread *td)
|
1997-02-18 20:46:36 +00:00
|
|
|
{
|
1997-08-16 19:16:27 +00:00
|
|
|
struct sockaddr_in *addr = (struct sockaddr_in *)nam;
|
2003-11-08 22:53:41 +00:00
|
|
|
struct inpcb *inp;
|
1997-02-18 20:46:36 +00:00
|
|
|
|
1997-08-16 19:16:27 +00:00
|
|
|
if (nam->sa_len != sizeof(*addr))
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EINVAL);
|
1997-02-18 20:46:36 +00:00
|
|
|
if (TAILQ_EMPTY(&ifnet))
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
2003-11-08 22:53:41 +00:00
|
|
|
if (addr->sin_family != AF_INET && addr->sin_family != AF_IMPLINK)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EAFNOSUPPORT);
|
2003-11-08 22:53:41 +00:00
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp != NULL, ("rip_connect: inp == NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
INP_INFO_WLOCK(&ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
2008-07-26 17:32:15 +00:00
|
|
|
rip_delhash(inp);
|
1997-02-18 20:46:36 +00:00
|
|
|
inp->inp_faddr = addr->sin_addr;
|
2008-07-26 17:32:15 +00:00
|
|
|
rip_inshash(inp);
|
1997-02-18 20:46:36 +00:00
|
|
|
soisconnected(so);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
2003-11-08 22:53:41 +00:00
|
|
|
INP_INFO_WUNLOCK(&ripcbinfo);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (0);
|
1997-02-18 20:46:36 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1997-02-18 20:46:36 +00:00
|
|
|
static int
|
|
|
|
rip_shutdown(struct socket *so)
|
|
|
|
{
|
2003-11-08 22:53:41 +00:00
|
|
|
struct inpcb *inp;
|
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp != NULL, ("rip_shutdown: inp == NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
1997-02-18 20:46:36 +00:00
|
|
|
socantsendmore(so);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (0);
|
1997-02-18 20:46:36 +00:00
|
|
|
}
|
1994-05-24 10:09:53 +00:00
|
|
|
|
1997-02-18 20:46:36 +00:00
|
|
|
static int
|
1997-08-16 19:16:27 +00:00
|
|
|
rip_send(struct socket *so, int flags, struct mbuf *m, struct sockaddr *nam,
|
2007-05-10 15:58:48 +00:00
|
|
|
struct mbuf *control, struct thread *td)
|
1997-02-18 20:46:36 +00:00
|
|
|
{
|
2003-11-08 22:53:41 +00:00
|
|
|
struct inpcb *inp;
|
|
|
|
u_long dst;
|
1994-05-24 10:09:53 +00:00
|
|
|
|
2003-11-08 22:53:41 +00:00
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp != NULL, ("rip_send: inp == NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
/*
|
|
|
|
* Note: 'dst' reads below are unlocked.
|
|
|
|
*/
|
1997-02-18 20:46:36 +00:00
|
|
|
if (so->so_state & SS_ISCONNECTED) {
|
|
|
|
if (nam) {
|
|
|
|
m_freem(m);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EISCONN);
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
dst = inp->inp_faddr.s_addr; /* Unlocked read. */
|
1997-02-18 20:46:36 +00:00
|
|
|
} else {
|
|
|
|
if (nam == NULL) {
|
|
|
|
m_freem(m);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (ENOTCONN);
|
1997-02-18 20:46:36 +00:00
|
|
|
}
|
1997-08-16 19:16:27 +00:00
|
|
|
dst = ((struct sockaddr_in *)nam)->sin_addr.s_addr;
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
2008-07-05 18:03:39 +00:00
|
|
|
return (rip_output(m, so, dst));
|
1994-05-24 10:09:53 +00:00
|
|
|
}
|
1997-02-18 20:46:36 +00:00
|
|
|
|
1998-05-15 20:11:40 +00:00
|
|
|
static int
|
2000-07-04 11:25:35 +00:00
|
|
|
rip_pcblist(SYSCTL_HANDLER_ARGS)
|
1998-05-15 20:11:40 +00:00
|
|
|
{
|
2003-09-01 04:23:48 +00:00
|
|
|
int error, i, n;
|
1998-05-15 20:11:40 +00:00
|
|
|
struct inpcb *inp, **inp_list;
|
|
|
|
inp_gen_t gencnt;
|
|
|
|
struct xinpgen xig;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The process of preparing the TCB list is too time-consuming and
|
|
|
|
* resource-intensive to repeat twice on every request.
|
|
|
|
*/
|
|
|
|
if (req->oldptr == 0) {
|
|
|
|
n = ripcbinfo.ipi_count;
|
|
|
|
req->oldidx = 2 * (sizeof xig)
|
2008-07-05 18:03:39 +00:00
|
|
|
+ (n + n/8) * sizeof(struct xinpcb);
|
|
|
|
return (0);
|
1998-05-15 20:11:40 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
if (req->newptr != 0)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EPERM);
|
1998-05-15 20:11:40 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* OK, now we're committed to doing something.
|
|
|
|
*/
|
2003-09-01 04:23:48 +00:00
|
|
|
INP_INFO_RLOCK(&ripcbinfo);
|
1998-05-15 20:11:40 +00:00
|
|
|
gencnt = ripcbinfo.ipi_gencnt;
|
|
|
|
n = ripcbinfo.ipi_count;
|
2003-09-01 04:23:48 +00:00
|
|
|
INP_INFO_RUNLOCK(&ripcbinfo);
|
1998-05-15 20:11:40 +00:00
|
|
|
|
|
|
|
xig.xig_len = sizeof xig;
|
|
|
|
xig.xig_count = n;
|
|
|
|
xig.xig_gen = gencnt;
|
|
|
|
xig.xig_sogen = so_gencnt;
|
|
|
|
error = SYSCTL_OUT(req, &xig, sizeof xig);
|
|
|
|
if (error)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
1998-05-15 20:11:40 +00:00
|
|
|
|
2003-02-19 05:47:46 +00:00
|
|
|
inp_list = malloc(n * sizeof *inp_list, M_TEMP, M_WAITOK);
|
1998-05-15 20:11:40 +00:00
|
|
|
if (inp_list == 0)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (ENOMEM);
|
2008-04-17 21:38:18 +00:00
|
|
|
|
2003-09-01 04:23:48 +00:00
|
|
|
INP_INFO_RLOCK(&ripcbinfo);
|
2007-04-30 23:12:05 +00:00
|
|
|
for (inp = LIST_FIRST(ripcbinfo.ipi_listhead), i = 0; inp && i < n;
|
2001-02-04 13:13:25 +00:00
|
|
|
inp = LIST_NEXT(inp, inp_list)) {
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RLOCK(inp);
|
2003-09-17 21:13:16 +00:00
|
|
|
if (inp->inp_gencnt <= gencnt &&
|
|
|
|
cr_canseesocket(req->td->td_ucred, inp->inp_socket) == 0) {
|
2003-09-01 04:23:48 +00:00
|
|
|
/* XXX held references? */
|
1998-05-15 20:11:40 +00:00
|
|
|
inp_list[i++] = inp;
|
2001-10-05 07:06:32 +00:00
|
|
|
}
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(inp);
|
1998-05-15 20:11:40 +00:00
|
|
|
}
|
2003-09-01 04:23:48 +00:00
|
|
|
INP_INFO_RUNLOCK(&ripcbinfo);
|
1998-05-15 20:11:40 +00:00
|
|
|
n = i;
|
|
|
|
|
|
|
|
error = 0;
|
|
|
|
for (i = 0; i < n; i++) {
|
|
|
|
inp = inp_list[i];
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RLOCK(inp);
|
1998-05-15 20:11:40 +00:00
|
|
|
if (inp->inp_gencnt <= gencnt) {
|
|
|
|
struct xinpcb xi;
|
2005-05-06 02:50:00 +00:00
|
|
|
bzero(&xi, sizeof(xi));
|
1998-05-15 20:11:40 +00:00
|
|
|
xi.xi_len = sizeof xi;
|
|
|
|
/* XXX should avoid extra copy */
|
|
|
|
bcopy(inp, &xi.xi_inp, sizeof *inp);
|
|
|
|
if (inp->inp_socket)
|
|
|
|
sotoxsocket(inp->inp_socket, &xi.xi_socket);
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(inp);
|
1998-05-15 20:11:40 +00:00
|
|
|
error = SYSCTL_OUT(req, &xi, sizeof xi);
|
2006-07-18 22:34:27 +00:00
|
|
|
} else
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(inp);
|
1998-05-15 20:11:40 +00:00
|
|
|
}
|
|
|
|
if (!error) {
|
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* Give the user an updated idea of our state. If the
|
|
|
|
* generation differs from what we told her before, she knows
|
|
|
|
* that something happened while we were processing this
|
|
|
|
* request, and it might be necessary to retry.
|
1998-05-15 20:11:40 +00:00
|
|
|
*/
|
2003-09-01 04:23:48 +00:00
|
|
|
INP_INFO_RLOCK(&ripcbinfo);
|
1998-05-15 20:11:40 +00:00
|
|
|
xig.xig_gen = ripcbinfo.ipi_gencnt;
|
|
|
|
xig.xig_sogen = so_gencnt;
|
|
|
|
xig.xig_count = ripcbinfo.ipi_count;
|
2003-09-01 04:23:48 +00:00
|
|
|
INP_INFO_RUNLOCK(&ripcbinfo);
|
1998-05-15 20:11:40 +00:00
|
|
|
error = SYSCTL_OUT(req, &xig, sizeof xig);
|
|
|
|
}
|
|
|
|
free(inp_list, M_TEMP);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
1998-05-15 20:11:40 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
SYSCTL_PROC(_net_inet_raw, OID_AUTO/*XXX*/, pcblist, CTLFLAG_RD, 0, 0,
|
2008-07-05 18:03:39 +00:00
|
|
|
rip_pcblist, "S,xinpcb", "List of active raw IP sockets");
|
1998-05-15 20:11:40 +00:00
|
|
|
|
1997-02-18 20:46:36 +00:00
|
|
|
struct pr_usrreqs rip_usrreqs = {
|
2004-11-08 14:44:54 +00:00
|
|
|
.pru_abort = rip_abort,
|
|
|
|
.pru_attach = rip_attach,
|
|
|
|
.pru_bind = rip_bind,
|
|
|
|
.pru_connect = rip_connect,
|
|
|
|
.pru_control = in_control,
|
|
|
|
.pru_detach = rip_detach,
|
|
|
|
.pru_disconnect = rip_disconnect,
|
2007-05-11 10:20:51 +00:00
|
|
|
.pru_peeraddr = in_getpeeraddr,
|
2004-11-08 14:44:54 +00:00
|
|
|
.pru_send = rip_send,
|
|
|
|
.pru_shutdown = rip_shutdown,
|
2007-05-11 10:20:51 +00:00
|
|
|
.pru_sockaddr = in_getsockaddr,
|
2006-07-21 17:11:15 +00:00
|
|
|
.pru_sosetlabel = in_pcbsosetlabel,
|
|
|
|
.pru_close = rip_close,
|
1997-02-18 20:46:36 +00:00
|
|
|
};
|