Commit Graph

1634 Commits

Author SHA1 Message Date
tuexen
6266aedc70 Whitespace changes.
The toolchain for processing the sources has been updated. No functional
change.

MFC after:	3 days
2016-12-26 11:06:41 +00:00
markj
dfc508adea Remove a bogus KASSERT from nd6_prefix_unlink().
The caller may unlink a prefix before purging referencing addresses. An
identical assertion in nd6_prefix_del() verifies that the addresses are
purged before the prefix is freed.

PR:		215372
X-MFC With:	r306829
2016-12-19 19:21:28 +00:00
ae
2c8179e485 ip[6]_tryforward does inbound and outbound packet firewall processing.
This can lead to change of mbuf pointer (packet filter could do m_pullup(),
NAT, etc). Also in case of change of destination address, tryforward can
decide that packet should be handled by local system. In this case modified
mbuf can be returned to the ip[6]_input(). To handle this correctly, check
M_FASTFWD_OURS flag after return from ip[6]_tryforward. And if it is present,
update variables that depend from mbuf pointer and skip another inbound
firewall processing.

No objection from:	#network
MFC after:	3 weeks
Sponsored by:	Yandex LLC
Differential Revision:	https://reviews.freebsd.org/D8764
2016-12-19 11:02:49 +00:00
ae
55258bdaf6 Modify IPv6 statistic accounting in ip6_input().
Add rcvif local variable to keep inbound interface pointer. Count
ifs6_in_discard errors in all "goto bad" cases. Now it will count
errors even if mbuf was freed. Modify all places where m->m_pkthdr.rcvif
is used to use local rcvif variable.

Obtained from:	Yandex LLC
MFC after:	1 month
2016-12-12 11:26:59 +00:00
ae
d9654156fd Add ip6_tryforward() - a run to completion forwarding implementation
for IPv6.

It gets performance benefits from reduced number of checks. It doesn't
copy mbuf to be able send ICMPv6 error message, because it keeps mbuf
unchanged until the moment, when the route decision has been made.
It doesn't do IPsec checks, and when some IPsec security policies present,
ip6_input() uses normal slow path.

Reviewed by:	bz, gnn
Obtained from:	Yandex LLC
MFC after:	1 month
Sponsored by:	Yandex LLC
Differential Revision:	https://reviews.freebsd.org/D8527
2016-12-12 10:57:32 +00:00
tuexen
ae1856036a Whitespace changes.
The tools using to generate the sources has been updated and produces
different whitespaces. Commit this seperately to avoid intermixing
these with real code changes.

MFC after:	3 days
2016-12-06 10:21:25 +00:00
tuexen
219874f57e Make ICMPv6 hard error handling for TCP consistent with the ICMPv4
handling. Ensure that:
* Protocol unreachable errors are handled by indicating ECONNREFUSED
  to the TCP user for both IPv4 and IPv6. These were ignored for IPv6.
* Communication prohibited errors are handled by indicating ECONNREFUSED
  to the TCP user for both IPv4 and IPv6. These were ignored for IPv6.
* Hop Limited exceeded errors are handled by indicating EHOSTUNREACH
  to the TCP user for both IPv4 and IPv6.
  For IPv6 the TCP connected was dropped but errno wasn't set.

Reviewed by: gallatin, rrs
MFC after: 1 month
Sponsored by: Netflix
Differential Revision: 7904
2016-10-21 10:32:57 +00:00
gnn
8c492572ae Limit the number of mbufs that can be allocated for IPV6_2292PKTOPTIONS
(and IPV6_PKTOPTIONS).

PR:		100219
Submitted by:	Joseph Kong
MFC after:	2 weeks
Differential Revision:	https://reviews.freebsd.org/D5157
2016-10-17 23:25:31 +00:00
glebius
444f6777c7 - Revert r300854, r303657 which tried to fix regression from r297225.
- Fix the regression proper way using RO_RTFREE().

Submitted by:	ae
2016-10-13 20:15:47 +00:00
markj
0e9bbe2171 Lock the ND prefix list and add refcounting for prefixes.
This change extends the nd6 lock to protect the ND prefix list as well
as the list of advertising routers associated with each prefix. To handle
cases where the nd6 lock must be dropped while iterating over either the
prefix or default router lists, a generation counter is used to track
modifications to the lists. Additionally, a new mutex is used to serialize
prefix on-link/off-link transitions. This mutex must be acquired before
the nd6 lock and is held while updating the routing table in
nd6_prefix_onlink() and nd6_prefix_offlink().

Reviewed by:	ae, tuexen (SCTP bits)
Tested by:	Jason Wolfe <jason@llnw.com>,
		Larry Rosenman <ler@lerctr.org>
MFC after:	2 months
Differential Revision:	https://reviews.freebsd.org/D8125
2016-10-07 21:10:53 +00:00
markj
671e7fcca9 Reduce the number of conditional statements in nd6_prefix_onlink().
MFC after:	1 week
2016-10-07 21:03:18 +00:00
markj
d443c7f48a Combine several checks in nd6_prefix_offlink() into one.
MFC after:	1 week
2016-10-07 21:02:30 +00:00
markj
87c0df02e3 Fix whitespace around prototypes in nd6_rtr.c.
MFC after:	1 week
2016-10-07 00:36:18 +00:00
markj
9346a8b630 Fix a typo.
MFC after:	1 week
2016-10-07 00:35:28 +00:00
markj
71534a1a0c Shorten and simplify some of the loops in pfxlist_onlink_check().
No functional change intended.

MFC after:	1 week
2016-10-07 00:34:57 +00:00
markj
149b5ce5db Use a const reference to prefixes in nd6_is_new_addr_neighbor().
MFC after:	1 week
2016-10-07 00:26:36 +00:00
markj
b0fce0526c nd6_dad_timer(): don't assert that the address is tentative.
It appears that this assertion can be tripped in some cases when
multiple interfaces are on the same link. Until this is resolved, revert a
part of r306305 and simply log a message if the DAD timer fires on a
non-tentative address.

Reported by:	jhb
X-MFC With:	r306305
2016-10-01 01:30:34 +00:00
ae
f7637dcb59 Fix bug introduced in r274300.
In icmp6_reflect() use original source address of erroneous packet as
destination address for source selection algorithm when original
destination address is not one of our own.

Reported by:	Mark Kamichoff <prox at prolixium com>
Tested by:	Mark Kamichoff <prox at prolixium com>
MFC after:	1 week
2016-09-29 19:57:37 +00:00
markj
67747e91f9 Convert checks in nd6_dad_start() and nd6_dad_timer() to assertions.
In particular, these functions can assume they are operating on tentative
addresses.

MFC after:	2 weeks
2016-09-24 21:40:24 +00:00
markj
e89e3efaa6 Rename ndpr_refcnt to ndpr_addrcnt.
This field counts derived addresses and is not a true refcount for prefix
objects, so the previous name was misleading.

MFC after:	1 week
2016-09-24 01:14:25 +00:00
markj
bc48ae0071 Reduce code duplication around NDP message handlers in icmp6_input().
No functional change intended.

MFC after:	2 weeks
2016-09-20 18:08:17 +00:00
kevlo
518bc28463 Remove the 4.3BSD compatible macro m_copy(), use m_copym() instead.
Reviewed by:	gnn
Differential Revision:	https://reviews.freebsd.org/D7878
2016-09-15 07:41:48 +00:00
karels
821d0f98a3 Fix L2 caching for UDP over IPv6
ip6_output() was missing cache invalidation code analougous to
ip_output.c. r304545 disabled L2 caching for UDP/IPv6 as a workaround.
This change adds the missing cache invalidation code and reverts
r304545.

Reviewed by:	gnn
Approved by:	gnn (mentor)
Tested by:	peter@, Mike Andrews
MFC after:	3 weeks
Differential Revision:	https://reviews.freebsd.org/D7591
2016-08-24 00:52:30 +00:00
bz
55cbdc7ad3 Remove the kernel optoion for IPSEC_FILTERTUNNEL, which was deprecated
more than 7 years ago in favour of a sysctl in r192648.
2016-08-21 18:55:30 +00:00
karels
2fb476f05a Disable L2 caching for UDP over IPv6
The ip6_output routine is missing L2 cache invalication as done
in ip_output.  Even with that code, some problems with UDP over
IPv6 have been reported.  Diabling L2 cache for that problem works
around the problem for now.

PR:		211872 211926
Reviewed by:	gnn
Approved by:	gnn (mentor)
MFC after:	immediate
2016-08-20 20:46:53 +00:00
ae
8c03d2551f Add ipfw_nat64 module that implements stateless and stateful NAT64.
The module works together with ipfw(4) and implemented as its external
action module.

Stateless NAT64 registers external action with name nat64stl. This
keyword should be used to create NAT64 instance and to address this
instance in rules. Stateless NAT64 uses two lookup tables with mapped
IPv4->IPv6 and IPv6->IPv4 addresses to perform translation.

A configuration of instance should looks like this:
 1. Create lookup tables:
 # ipfw table T46 create type addr valtype ipv6
 # ipfw table T64 create type addr valtype ipv4
 2. Fill T46 and T64 tables.
 3. Add rule to allow neighbor solicitation and advertisement:
 # ipfw add allow icmp6 from any to any icmp6types 135,136
 4. Create NAT64 instance:
 # ipfw nat64stl NAT create table4 T46 table6 T64
 5. Add rules that matches the traffic:
 # ipfw add nat64stl NAT ip from any to table(T46)
 # ipfw add nat64stl NAT ip from table(T64) to 64:ff9b::/96
 6. Configure DNS64 for IPv6 clients and add route to 64:ff9b::/96
    via NAT64 host.

Stateful NAT64 registers external action with name nat64lsn. The only
one option required to create nat64lsn instance - prefix4. It defines
the pool of IPv4 addresses used for translation.

A configuration of instance should looks like this:
 1. Add rule to allow neighbor solicitation and advertisement:
 # ipfw add allow icmp6 from any to any icmp6types 135,136
 2. Create NAT64 instance:
 # ipfw nat64lsn NAT create prefix4 A.B.C.D/28
 3. Add rules that matches the traffic:
 # ipfw add nat64lsn NAT ip from any to A.B.C.D/28
 # ipfw add nat64lsn NAT ip6 from any to 64:ff9b::/96
 4. Configure DNS64 for IPv6 clients and add route to 64:ff9b::/96
    via NAT64 host.

Obtained from:	Yandex LLC
Relnotes:	yes
Sponsored by:	Yandex LLC
Differential Revision:	https://reviews.freebsd.org/D6434
2016-08-13 16:09:49 +00:00
stevek
12aa5d6087 Move IPv4-specific jail functions to new file netinet/in_jail.c
_prison_check_ip4 renamed to prison_check_ip4_locked

Move IPv6-specific jail functions to new file netinet6/in6_jail.c
_prison_check_ip6 renamed to prison_check_ip6_locked

Add appropriate prototypes to sys/sys/jail.h

Adjust kern_jail.c to call prison_check_ip4_locked and
prison_check_ip6_locked accordingly.

Add netinet/in_jail.c and netinet6/in6_jail.c to the list of files that
need to be built when INET and INET6, respectively, are configured in the
kernel configuration file.

Reviewed by:	jtl
Approved by:	sjg (mentor)
Sponsored by:	Juniper Networks, Inc.
Differential Revision:	https://reviews.freebsd.org/D6799
2016-08-09 02:16:21 +00:00
ae
a937127683 Fix NULL pointer dereference.
ro pointer can be NULL when IPSec consumes mbuf.

PR:		211486
MFC after:	3 days
2016-08-02 12:18:06 +00:00
gallatin
11f6fcfd28 Rework IPV6 TCP path MTU discovery to match IPv4
- Re-write tcp_ctlinput6() to closely mimic the IPv4 tcp_ctlinput()

- Now that tcp_ctlinput6() updates t_maxseg, we can allow ip6_output()
  to send TCP packets without looking at the tcp host cache for every
  single transmit.

- Make the icmp6 code mimic the IPv4 code & avoid returning
  PRC_HOSTDEAD because it is so expensive.

Without these changes in place, every TCP6 pmtu discovery or host
unreachable ICMP resulted in a call to in6_pcbnotify() which walks the
tcbinfo table with the write lock held.  Because the tcbinfo table is
shared between IPv4 and IPv6, this causes huge scalabilty issues on
servers with lots of (~100K) TCP connections, to the point where even
a small percent of IPv6 traffic had a disproportionate impact on
overall throughput.

Reviewed by:	bz, rrs, ae (all earlier versions), lstewart (in Netflix's tree)
Sponsored by:		Netflix
Differential Revision:	https://reviews.freebsd.org/D7272
2016-08-01 17:02:21 +00:00
stevek
3acd4a25e6 Prepare for network stack as a module
- Move cr_canseeinpcb to sys/netinet/in_prot.c in order to separate the
   INET and INET6-specific code from the rest of the prot code (It is only
   used by the network stack, so it makes sense for it to live with the
   other network stack code.)
 - Move cr_canseeinpcb prototype from sys/systm.h to netinet/in_systm.h
 - Rename cr_seeotheruids to cr_canseeotheruids and cr_seeothergids to
   cr_canseeothergids, make them non-static, and add prototypes (so they
   can be seen/called by in_prot.c functions.)
 - Remove sw_csum variable from ip6_forward in ip6_forward.c, as it is an
   unused variable.

Reviewed by:	gnn, jtl
Approved by:	sjg (mentor)
Sponsored by:	Juniper Networks, Inc.
Differential Revision:	https://reviews.freebsd.org/D2901
2016-07-27 20:34:09 +00:00
karels
e31cf93024 Fix per-connection L2 caching in fast path
r301217 re-added per-connection L2 caching from a previous change,
but it omitted caching in the fast path.  Add it.

Reviewed By: gallatin
Approved by: gnn (mentor)
Differential Revision: https://reviews.freebsd.org/D7239
2016-07-22 02:11:49 +00:00
ae
2c47439b3f Add ipfw_nptv6 module that implements Network Prefix Translation for IPv6
as defined in RFC 6296. The module works together with ipfw(4) and
implemented as its external action module. When it is loaded, it registers
as eaction and can be used in rules. The usage pattern is similar to
ipfw_nat(4). All matched by rule traffic goes to the NPT module.

Reviewed by:	hrs
Obtained from:	Yandex LLC
MFC after:	1 month
Relnotes:	yes
Sponsored by:	Yandex LLC
Differential Revision:	https://reviews.freebsd.org/D6420
2016-07-18 19:46:31 +00:00
ae
7a18a4b316 Add net.inet6.ip6.intr_queue_maxlen sysctl. It can be used to
change netisr queue limit for IPv6 at runtime.

Obtained from:	Yandex LLC
MFC after:	2 weeks
Sponsored by:	Yandex LLC
2016-07-15 17:09:30 +00:00
dim
8a8ea0466a Fix a page fault in ip6_setpktopt(), occurring when the pflog module is
loaded, and syncthing is started, which uses setsockopt(IPV6_PKGINFO).

This is because pflog interfaces do not normally have an IPv6 address,
causing the ND_IFINFO() macro to dereference a NULL pointer.

Reviewed by:	ae
PR:		210943
MFC after:	3 days
2016-07-13 19:41:19 +00:00
tuexen
c2c8b26056 Don't consider the socket when processing an incoming ICMP/ICMP6 packet,
which was triggered by an SCTP packet. Whether a socket exists, is just
not relevant.

Approved by: re (kib)
MFC after: 1 week
2016-06-23 09:13:15 +00:00
ae
7cdbaef028 Fix the NULL pointer dereference for unresolved link layer entries in
the netinet6 code. Copy link layer address only when corresponding entry
has LLE_VALID flag.

PR:		210379
Approved by:	re (kib)
2016-06-22 11:29:21 +00:00
bz
7a1c0b1ad1 Get closer to a VIMAGE network stack teardown from top to bottom rather
than removing the network interfaces first. This change is rather larger
and convoluted as the ordering requirements cannot be separated.

Move the pfil(9) framework to SI_SUB_PROTO_PFIL, move Firewalls and
related modules to their own SI_SUB_PROTO_FIREWALL.
Move initialization of "physical" interfaces to SI_SUB_DRIVERS,
move virtual (cloned) interfaces to SI_SUB_PSEUDO.
Move Multicast to SI_SUB_PROTO_MC.

Re-work parts of multicast initialisation and teardown, not taking the
huge amount of memory into account if used as a module yet.

For interface teardown we try to do as many of them as we can on
SI_SUB_INIT_IF, but for some this makes no sense, e.g., when tunnelling
over a higher layer protocol such as IP. In that case the interface
has to go along (or before) the higher layer protocol is shutdown.

Kernel hhooks need to go last on teardown as they may be used at various
higher layers and we cannot remove them before we cleaned up the higher
layers.

For interface teardown there are multiple paths:
(a) a cloned interface is destroyed (inside a VIMAGE or in the base system),
(b) any interface is moved from a virtual network stack to a different
network stack ("vmove"), or (c) a virtual network stack is being shut down.
All code paths go through if_detach_internal() where we, depending on the
vmove flag or the vnet state, make a decision on how much to shut down;
in case we are destroying a VNET the individual protocol layers will
cleanup their own parts thus we cannot do so again for each interface as
we end up with, e.g., double-frees, destroying locks twice or acquiring
already destroyed locks.
When calling into protocol cleanups we equally have to tell them
whether they need to detach upper layer protocols ("ulp") or not
(e.g., in6_ifdetach()).

Provide or enahnce helper functions to do proper cleanup at a protocol
rather than at an interface level.

Approved by:		re (hrs)
Obtained from:		projects/vnet
Reviewed by:		gnn, jhb
Sponsored by:		The FreeBSD Foundation
MFC after:		2 weeks
Differential Revision:	https://reviews.freebsd.org/D6747
2016-06-21 13:48:49 +00:00
pfg
0f12e1993e Remove the SIOCSIFALIFETIME_IN6 ioctl.
The SIOCSIFALIFETIME_IN6 provided by the kame project is unused,
it can't really be used safely and has been completely removed from
NetBSD and OpenBSD.

Obtained from:	NetBSD (kern/35897)
PR:		210148 (exp-run)
Reviewed by:	ae, hrs
Relnotes:	yes
Approved by:	re (glebius)
Differential Revision:	https://reviews.freebsd.org/D5491
2016-06-13 22:31:16 +00:00
ae
48b268cd67 Cleanup unneded include "opt_ipfw.h".
It was used for conditional build IPFIREWALL_FORWARD support.
But IPFIREWALL_FORWARD option was removed a long time ago.
2016-06-09 05:48:34 +00:00
bz
5baf25edd0 Make KASSERT message more useful by printing the variables on which
we assert.

Obtained from:	projects/vnet
MFC after:	2 weeks
Sponsored by:	The FreeBSD Foundation
2016-06-06 22:34:12 +00:00
bz
aaac6c5a13 Move the callout_reset() to the end of the work not having it stick
before we do anything.

Obtained from:	projects/vnet
MFC after:	2 week
Sponsored by:	The FreeBSD Foundation
2016-06-06 14:01:09 +00:00
bz
69cdb2137c Introduce a per-VNET flag to enable/disable netisr prcessing on that VNET.
Add accessor functions to toggle the state per VNET.
The base system (vnet0) will always enable itself with the normal
registration. We will share the registered protocol handlers in all
VNETs minimising duplication and management.
Upon disabling netisr processing for a VNET drain the netisr queue from
packets for that VNET.

Update netisr consumers to (de)register on a per-VNET start/teardown using
VNET_SYS(UN)INIT functionality.

The change should be transparent for non-VIMAGE kernels.

Reviewed by:	gnn (, hiren)
Obtained from:	projects/vnet
MFC after:	2 weeks
Sponsored by:	The FreeBSD Foundation
Differential Revision:	https://reviews.freebsd.org/D6691
2016-06-03 13:57:10 +00:00
gnn
d75e0c471e This change re-adds L2 caching for TCP and UDP, as originally added in D4306
but removed due to other changes in the system. Restore the llentry pointer
to the "struct route", and use it to cache the L2 lookup (ARP or ND6) as
appropriate.

Submitted by:	Mike Karels
Differential Revision:	https://reviews.freebsd.org/D6262
2016-06-02 17:51:29 +00:00
markj
8b17712ca6 Exploit r301213 to fix in6 ifaddr locking in pfxlist_onlink_check().
Reviewed by:	ae, hrs
MFC after:	2 weeks
Differential Revision:	https://reviews.freebsd.org/D6639
2016-06-02 17:21:57 +00:00
markj
71ff51c027 Always start IPv6 DAD asynchronously.
Otherwise we transmit the first neighbour solicitation in the context of the
caller of nd6_dad_start(), which can easily result in lock recursion. When
DAD is to be started after some delay, we send the first NS from the DAD
callout handler, so just change the implementation to do this in the
non-delayed case as well.

Reviewed by:	ae, hrs
MFC after:	2 weeks
Differential Revision:	https://reviews.freebsd.org/D6639
2016-06-02 17:17:15 +00:00
bz
fac944a70a The pr_destroy field does not allow us to run the teardown code in a
specific order.  VNET_SYSUNINITs however are doing exactly that.
Thus remove the VIMAGE conditional field from the domain(9) protosw
structure and replace it with VNET_SYSUNINITs.
This also allows us to change some order and to make the teardown functions
file local static.
Also convert divert(4) as it uses the same mechanism ip(4) and ip6(4) use
internally.

Slightly reshuffle the SI_SUB_* fields in kernel.h and add a new ones, e.g.,
for pfil consumers (firewalls), partially for this commit and for others
to come.

Reviewed by:		gnn, tuexen (sctp), jhb (kernel.h)
Obtained from:		projects/vnet
MFC after:		2 weeks
X-MFC:			do not remove pr_destroy
Sponsored by:		The FreeBSD Foundation
Differential Revision:	https://reviews.freebsd.org/D6652
2016-06-01 10:14:04 +00:00
tuexen
525f16040f Add PR_CONNREQUIRED for SOCK_STREAM sockets using SCTP.
This is required to signal connetion setup on non-blocking sockets
via becoming writable. This still allows for implicit connection
setup.

MFC after:	1 week
2016-05-30 18:24:23 +00:00
glebius
bbfa6d2853 Plug route reference underleak that happens with FLOWTABLE after r297225.
Submitted by:	Mike Karels <mike karels.net>
2016-05-27 17:31:02 +00:00
markj
9cb221bdb0 Mark the prefix and default router list sysctl handlers MPSAFE.
MFC after:	2 weeks
2016-05-23 20:18:11 +00:00
markj
86c15ee95b Acquire the nd6 lock in the prefix list sysctl handler.
The nd6 lock will be used to synchronize access to the NDP prefix list.

MFC after:	2 weeks
Tested by:	Jason Wolfe (as part of a larger change)
2016-05-23 20:15:08 +00:00