CVS log for kame/kame/sys/netinet6/ip6_input.c

Up to [KAME snap] / kame / kame / sys / netinet6

Request diff between arbitrary revisions - Display revisions graphically

Keyword substitution: kv
Default branch: MAIN

ANSIfy, to reduce diffs with *BSDs.
based on diff from dunceor at gmail.com.
XXX not tested yet, sorry, we need to ressurect buildlab configs.

random thoughts:
- IGMPv3 support needs a serious cleanup.  argument passing is horrible, style
  does not meet style(9) at all.  i'm afraid there could be a lot of bugs.
- DCCP/SCTP/NAT-PT need someone to maintain them, otherwise we should drop
  them at once.
- we should pick a platform (one of *BSD) and try to get KAME tree in sync
  with more recent version of *BSD.  otherwise, KAME tree gets more and more
  out-of-date.

clarifying some points in the previous code comment as well as some minor
style changes

comment on "drop packets with more than 1 routing header" code.
suggested by jinmei.

drop packets with more than 1 routing headers.
from claudio@openbsd

ignore "unrecognized" router alert options so that the router can
forward packets containing such a router alert option.  Actually, this code
only treats IP6OPT_RTALERT_MLD as "recognized".

pointed out by Andrew McDonald

cleanup SCOPEDROUTING (unifdef -USCOPEDROUTING)

removed an unexpected white character in an indentation.

It is preferable to use NIST > 0 rather than IFT_IST to know isatap is used.

Bug fix for ISATAP

added temporary debug printf for the new scope code

style

in ip6_input() make sure that the original source or destination address
is not in our internal form for scoped addresses.  Such addresses
are not necessarily invalid spec-wise, but we cannot accept them due
to the usage conflict.

revised scope-related interfaces, clarifying the relationship between
the kernel-internal form and the public sockaddr_in6 form.

changed the word '2292bis' to 'RFC3542' when appropriate.

removed a freebsd4-specific include file

remove freebsd[234]-specific code, since they are no longer supported in KAME-SNAP

make it compilable on netbsd/openbsd

added a knob to enable path MTU discovery for multicast packets.
(by default, it is disabled)

fixed PF malfunction for inbound IPv6 traffic on freebsd5

remove a duplicated block that was introduced in v1.349 when merging
freebsd-5.3.
reported by Joerg Sonnenberger <joerg@britannica.bec.de>

- sys/netipsec (FAST_IPSEC) has #define for time_second
- use pfil_hook for PF

sync w/ 2.0.
needs checking/fixing: igmp3, pf, altq, dccp, sctp
FAST_IPSEC path is not considered, intentionally.

NIST > 0 would be better to know ISATAP is configured.

New Mobile IPv6 code which also supports NEMO.

makes ISATAP compilable on OpenBSD

sync with freebsd53 and makes it compilable
(their behavior is not confirmed yet...)

another clarification of rfc2462bis:
disable IPv6 operation if DAD fails for some EUI-64 link-local addresses.
export get_hw_ifid() (and rename it) as a subroutine for this process.

remove support for freebsd[23], simplify #ifdef

comment wording

do not check super user privilege in ip6_savecontrol.  It is meaningless and
can even be harmful.

nuke some bsdi #ifdef.  make pf.c compatible w/ nd6_output() arg.

de-support freebsd2

simplify #ifdef by de-support bsdi[34] and freebsd[23].

make it compilable on freebsd521+kame

ToDo: reflect changes in freebsd521, test, ...

- Removed an unnecessary if-condition.

- Applies per-interface MSF at ICMPv6 layer (instead of IP layer).
  This is to prevent MLDv2 packets from being filtered by per-interface MSF.
- Applies per-socket MSF for raw-socket/ICMPv6 socket

restructure of header files based on draft-ietf-mip6-mipext-advapi-00.

reorder to meet w/ old code

knf

makes MLDv2 compilable
(behavior is not confirmed yet...)

move away from in6_{get,set}pktaddrs framework, go back to embedded
scopeid in s6_addr16[1].  this is to make *BSD merge easier.

TODO: check missing/removed scope border code
TODO: variable name re-sync w/ *BSD platforms
TODO: fix non-netbsd platforms
TODO: fix mip6, sctp, dccp, mldv2

increase nxt stat after packet filtering ops

changed mobility header protocol no. and MIP6 related ICMPv6 type no.
based on IANA assignment.

fixed an endian bug on fragment header scanning

Reported by  Masahito Endo <masaxmasa@tahi.org>

a 2292bis cleanup:
  now ip6_savecontrol() simply takes mbuf ** just like ip_savecontrol(), since
  RFC3542 doesn't require to keep history of control information.
  removed ip6_recvpktopts{} accordingly.

freebsd[345] calls ip6_savecontrol with IPv4 packet, weird!

- remove a couple of NULL checks (should panic)
- DIAGNOSTIC is not for printf()
- IN6P_HOPOPTS should only be set on privileged in6p (ip6_output.c)
  so panic if mismatch happens
- ip6_savecontrol() argument is redundant

use 2292bis def for router alert opt

randomize flowlabel/fragment ID.

more comment wording

comment wording

- sync with Freebsd 5.1-RELEASE
- removed FreeBSD-specific ifdef for M_MOVE_PKTHDR(), since all *BSDs have M_MOVE_PKTHDR() now

suppress ICMPv6 redirect if packet filter rewrote ip6_dst to some address
on the incoming interface.  cedric@openbsd

enable PF for freebsd4

deal w/ openbsd ipsec

byebye new_struct_route

don't inject ipsec-tunnelled packet to PF again

cleanup for netbsd PF

drop NEW_STRUCT_ROUTE for netbsd

KNF

answers to MLD query for link-local multicast group address.
(stopped embedding the ifindex of group address in struct in6_multi)

Reported by: Kentaro Ohara <Kentarou.Oohara@jp.yokogawa.com>
(users@jp.ipv6.org #3055)

- swap HAO and ip6src before processing AH/ESP/MH.
  this is just a careless mistake when restructuring mip6 code.

- restructured mip6 code.
  divided cn/ha/mn part to separate files.  this makes it easier to
  merge a paticular part (ex. CN) to original BSD or to integrate
  a paticular part (ex. HA) to routers.

makes it compilable on freebsd-4.8

typo in a comment line.

ip6_getpktaddrs() should return -1 if the stored addresses are invalid.

remove unneeded mtod() for ip6

- introduced M_MOVE_PKTHDR to netbsd.
  TODO: freebsd5

Make it compilable on NetBSD (and maybe other BSDs) along the changes
regarding freebsd5 and ip6_getpktaddrs().

makes it compilable on freebsd5, too

changed 2nd and 3rd argument of ip6_getpktaddrs() from sockaddr_in6 **
to sockaddr_in6 * in order to avoid sharing a pointer in an mbuf.

changes for *bsd-specific files will follow.

looks okay for freebsd4.  will compilation check for others.

link m_tag to mbuf

transition from m_aux framework to m_tag framework.
XXX freebsd4/bsdi3 won't compile for a while

When listening to a source-specific-multicast group, receives IPv6 packets
(e.g. ICMPv6 ECHO) as well as IPv6 UDP packets if source and group are matched

fixed a bug that isatap router list cannot be obtained under NetBSD/OpenBSD

make it compilable under other platforms than FreeBSD4 by
	- avoiding a use of "ifnet.if_physical"
	- enabling ISATAP features appropriately

ISATAP implementation using stf device (see "man stf" for further detail)

ToDo: test test test ....

- removed an unused variable.

- for NATPT support.  ip_forward() needs 3 argument.

sync w/ 1.6.  GENERIC.KAME/i386 compiles, not tested.
- NATPT and IP6FW support were dropped (you can use ipfilter).
- altq, ecn, RTF_CACHE, SCOPEDROUTING need checking

correct signedness mixup in pointer passing.

KNF - return is not a function.

MLDv2 host-side implementation for NetBSD and FreeBSD-4,
based on draft-vida-mld-v2-03.txt.

ToDo: porting to OpenBSD & BSDI4, much more test

typo

MIP6: fix bug a HAO and not work frag. packet. discard mh->pproto != NONE

MIP6: mip6 id-18 supported (todo: L/S bit, refresh opt, BA authdata and etc...)

remove redundant extern decl

MIPv6: MN-MN work w/o HA-MN ESP tunnel

MIP6:
-mip6 tunnel and framgent problem
-bug fix proxy ND(not work unicast NS)

check if n is null or not.

whitespace cleanup

whitespace

KNF ip6_process_hopopts

minor cleanups

0 is not NULL

minor KNF

use arc4random except bsdi

limit number of fragments (not fragment queues) in kernel.

nd6_ifattach(loif) shouldn't be necessary here

added a new function "scope6_init" to inialize scope zone IDs
for as many interfaces as possible.
this function is called from ip6_init() during the boot procedure.

sync with 3.1.  not tested yet
- some sysctls were renamed/renumbered
- sctp syscall have renumbered
- some of IFT_xxx are renumbered
- sctp does not work (some of sys/kern changes had conflict and backed out)

- the first step for mip6 id17 support.
  this commit is for the KAME mip6 developpers.  do not try to use this.

removed redundant space at the end of line

typo

use lo0ifp for openbsd (XXX)

folded a long line.

call nd6_ifattach(lo0) from ip6_init2() in order to initialize nd_ifinfo{}
for lo0.

cleanup; removed a duplicated check for mapped src/dst addr from ip6_input().
there seemed to be a confusion at the revision 1.99.

do not call in6_ifattach(lo0) from ip6_init2().
we in fact do not need to call in6_ifattach() at this stage.
additionally, in6_ifattach() has a bad effect of sending packets even though
some parts of the kernel is not ready for sending.

typo

changed the type of the address part of in6_multi{} to sockaddr_in6
so that it would contain scope zone ID field.
related functions are modified accordingly.

clean-up; removed code to measure kernel performance, which is not
necessary any more.

clean-up.

do not embed link IDs to the incoming packet, because we now
maintain full sockaddr_in6 structures and always refer to them to qualify
addresses.

totally revised IPv6 scoped address architecture in the kernel.
- sticked to use sockaddr_in6, not in6_addr, as much as possible.
- did not refer to ip6_src/ip6_dst (which may have ambiguity on scope zones),
  but used full sockaddr_in6 structures attached to the packet.
- replaced special cases for link-local addresses with code that used
  generic functions such as in6_addr2zoneid().
- additional cleanups mainly for scoped address handling.

allow ip6_setpktaddrs() to take a NULL address so that the caller can
update a (one) address already set before.

added a new function ip6_getpktaddrs() to get addresses set by ip6_setpktaddrs.
used the new function in ip6_forward().

disabled calling ip6_forward() for NATPT.  this should currently not happen.

- made ip6aux{} scope-aware; added sockaddr_in6 structures to record
  source and destinaion addresses with scope information.
- recorded the addresses in ip6_input().
- used the recorded addresses to check scope breakage and to get a route
  in ip6_forward().

added a generic scope check, and removed checks for loopback src/dst addresses.

style: added a white space before an open bracket.

comment wording

restricted the behavior that rejected link-local src/dst addresses with
the "link ID" field portion being filled in to !SCOPEDROUTING cases.

- Adjust variable name to a NATPT related name.
  s/ip6_protocol_tr/natpt_enable/

removed if-0'ed code that does linear search for goto-ours.
the code has been there (with being disabled) since July '99, but our
rtalloc approach is now mature enough.

do not accept packets to a destination that has a rejected or blackhole
host route.

pf_test() was called too early.  sync w/openbsd-current

[cleanup] removed meaningless functions "ip6_update_recvpcbopt" and
"ip6_reset_rcvopt".

made it compile on netbsd

removed the "prevctl" argument of ip6_savecontrol().

in ip6_savecontrol(), do not compare the previous values.
this part has been removed from the adv API spec.

remove opt_inet6.h,mip6_var.h&etc. includes from if_hif.h&mip6.h

variable initialization

in ip6_savecontrol(), do not separate the case for IPV6_RTHDRDSTOPTS and
the case for IPV6_DSTOPTS, according to rfc2292bis-03.

removed an obsoleted comment.

renamed sysctl name for ip6addrctl

make it compile even if NGIF=0

whitespace

include netinet6/in6_gif.h

missing callout_init() for in6_tmpaddrtimer_ch

recorded counter for permance evaluation *before* calling ip6_setdstifaddr().
otherwise, the evaluation result would unintentionally worse.

sync with openbsd 3.0.  compiles but not tested.  ALTQ does not work for now.

duplicate comment

count per-ifa stats presented on netstat -is (bsdi4)

made the MEASURE_PERFORMANCE case compilable on freebsd4

supported use_defaultzone sysctl

KNF for NAT-PT

check new-offset overrun in ip6_nexthdr()

whitespace fix

whitespace

sync with freebsd44.

make IPV6CTL_ADDRSELPOLICY available on net/openbsd.

implemented the policy table for address selection according to
draft-ietf-ipngwg-default-addr-select.
"To be considered" items:
  - using sysctl to set/get the policy is really a good way?
  - port to FreeBSD

add missing include mip6.h

missing prototype on -DMIP6

make IPV6_TCLASS to take int, not u_int8_t.  -1 means the default value.
follows 2292bis.  fix a couple of #ifdef bugs while i'm here.
XXX we need to keep backward binary compat for IPV6_OTCLASS as bsdi ships
with IPV6_TCLASS.

changed the return value type of in6_addr2zoneid(), from u_int32_t to
int64_t, so that the caller can tell an error from valid "4+28" ID values.

make net.inet6.ip6.v6only a read-only variable on bsdi3.

allow user to modify net.inet[6].ip6.v6only

do not chase intermediate IPv6 header in IPv6-over-IPv6 case.  from sumikawa

process IPV6CTL_PREFER_TEMPADDR for [no]bsd
(tested on nbsd only)

netbsd don't need opt_hoge.h. remove it from sources.
freebsd3 need opt_hoge.h, add opt_mip6.h to sys/conf/files.

 Modified Files:
 	freebsd3/sys/conf/options kame/sys/netinet6/dest6.c
 	kame/sys/netinet6/icmp6.c kame/sys/netinet6/in6.c
 	kame/sys/netinet6/in6_proto.c kame/sys/netinet6/in6_src.c
 	kame/sys/netinet6/ip6_forward.c kame/sys/netinet6/ip6_input.c
 	kame/sys/netinet6/ip6_output.c kame/sys/netinet6/mip6.c
 	kame/sys/netinet6/mip6_binding.c kame/sys/netinet6/mip6_ha.c
 	kame/sys/netinet6/mip6_icmp6.c kame/sys/netinet6/mip6_prefix.c
 	kame/sys/netinet6/mip6_subnet.c kame/sys/netinet6/nd6_rtr.c

add mobileip source code.

currently, many parts of spec are not implemented yet.  even
uncompilable under other than freebsd4 now.  also very unstable.
of course, without MIP6 kernel option, there is no problem. all mip6
related code is separated by ifdef MIP6.

this integration is just for KAME mip6 developper's convinience.
the purpose of this early integration is to decrease the maintenance cost
to synchronize KAME main tree and developping mip6 tree.  after this merge,
mip6 code modifications are reflected directly to KAME code.

i will make these codes compilable on the other OSes as soon as possible.

 Modified Files:
	kame/sys/netinet6/dest6.c kame/sys/netinet6/icmp6.c
	kame/sys/netinet6/in6.c kame/sys/netinet6/in6.h
	kame/sys/netinet6/in6_ifattach.c
	kame/sys/netinet6/in6_ifattach.h kame/sys/netinet6/in6_proto.c
	kame/sys/netinet6/in6_src.c kame/sys/netinet6/in6_var.h
	kame/sys/netinet6/ip6_forward.c kame/sys/netinet6/ip6_input.c
	kame/sys/netinet6/ip6_output.c kame/sys/netinet6/mip6.c
	kame/sys/netinet6/mip6.h kame/sys/netinet6/mip6_ha.c
 	kame/sys/netinet6/nd6.h kame/sys/netinet6/nd6_rtr.c
 	freebsd4/sys/net/if_types.h freebsd4/sys/conf/files
 	freebsd4/sys/conf/options
 Added Files:
 	kame/sys/net/if_hif.c kame/sys/net/if_hif.h
 	kame/sys/netinet6/mip6_binding.c
 	kame/sys/netinet6/mip6_icmp6.c kame/sys/netinet6/mip6_prefix.c
 	kame/sys/netinet6/mip6_subnet.c kame/sys/netinet6/mip6_var.h

3rd argument 'port' of ip6_fw_chk is unnecessary.

declare ipsec_getnhist().  in preparation to ipsec m->m_pkthdr.aux change.

added several improvements on IPv6 output routine
(I should've done some of them by separate commits...sorry):

- added a seprate function in6_selectroute() to centralize the route and
  outgoing interface selection algorithm
- call in6_selectroute from in6_selecsrc() and ip6_output(), and use same
  codebase for both unicast and multicast packets as much as possible.
- added stricter (and thorough) scope checks for outgoing packets
  (e.g. this check would prevent a packet generated by
   'ping6 -S ::1 fe80::1%ne0')
- renamed in6_addr2scopeid to in6_addr2zoneid to be more intuitive in terms
  of the scope architecture
- made in6_addr2scopeid take care of the loopback address
- set the tentative flag before joinging multicast addresses, so that the
  corresponding MLD packet would not have a tentative source address

Although the change is quite large, I just compiled the code on bsdi4.
Please check it on other platforms, thanks.

style: added spaces after /* and before */
(this might be a matter-of-taste issue, though)

reject incoming packets to an interface-local multicast address from the wire.

interface-local multicast support;
- install ff01::%ifname/32 in in6.c.
- change source address selection for ff01::xxx.
- change outgoing interface selection for ff01::xxxx.
- prohibit packets to ff01::xxx from being sent on the wire.

retire sys/netinet6/in6_prefix.c altogether.  i'm sure it's now safe
to do this, as there was no operation that added entries into rr_prefix chain.

fill scopeid field on interface local address.  ping6 ff01::1%lo0 now works
(but not ff01::1%ne2... why)

removed white spaces after EOL.

remove 'by kazu'.

remove redundant if (m) around ipsec mbuf flags.

revert the previous change (on which I was wrong, sorry)

simplified the logic; do validation first, and skip further validations
beyond the point.

netbsd: recover INET6_BINDV6ONLY by protecting ip6_v6only variable.
it is part of netbsd releases, so we need to keep it.
(no #ifdef in in6_pcb.c - which is good)

get rid of INET6_BINDV6ONLY (netbsd).
XXX IPv4 mapped address support cases are way too complicated!
we really really need to clean them up and check if we have made any mistakes
(it looks to me that there there are quite a few).

sync with openbsd 2.9.  GENERIC.KAME compiles, but not tested.
ALTQ portion needs a serious checking.

MIP6 removed.
Ericsson's code is provided as a patch to the plain (without MIP6)
KAME source after this point.

sysctl access to rip6stat (net.inet6.ip6.rip6stats).  for bsdi/freebsd.
not tested.

M_COPY_PKTHDR has to be done before MCLGET.

do not accept packet if the corresponding route has the cloned bit.
this should solve the problem that some OSes mistakenly accept a packet
to fe80::2%lo0.

stats per address for freebsd4

make sure to clear m->m_pkthdr.rcvif after MGETHDR.  it is not ensured in
sys/sys/mbuf.h.

correct mbuf cluster allocation code in natpt*.

clarified goto-ours logic:

1. separated checks against spoofed ::1 src/dst from the goto-ours check.
   this also fixed a bug that the kernel accepted a packet with
   src=::1, dst=invalid, rcvif=lo0
   (you can test it by 'ping6 -S ::1 fe80::xxxx%lo0", where xxxx is not an
    interface ID of lo0)
2. (experimentally) omitted a specical case for link-local destinations at a
   loopback interface.  I believe this is correct, because
   - we now have a host route for fe80::1%lo0, so we can accept a packet to
     the address using the generic logic.
   - we can reject packets to fe80::xxxx%lo0 (xxxx != 1) by the check for
     the RTF_GATEWAY bit for rt_flags (ip6_input.c line 872).
   *** NOTE to developers:***
   this is the case for bsdi4, but please check it on other platforms.
   after the confirmation, I'll completely remove the part (currently, it's
   just escaped by '#ifdef 0')

integrate new ericsson mobile-ip6 snapshot (based on 13 draft, after
Connectathon 2001).

changes from original ericsson shipment:
- avoid using "time" as variable name, we can't do that in BSD kernel.
- on function signature change for MIP6 (like static -> non-static) use #ifdef
  MIP6.  it has been making problems for *BSD-current merge.
- add __attribute__((__packed__)).
- make the tree compile without MIP6.
- make the tree compile on openbsd (should be okay on netbsd too).
	WARNING: two #warning line were suppressed, or replaced with panic().
- avoid unnecessary cosmetic change.
- avoid $Id$ (previous commit to mip6{config,stat}.
- fix nd6.c neighbor state management breakage, due to one of additional MIP6
  debugging messages.

while we are here, made a couple of changes:
- add __attribute__((__packed__)) here and there.
- remove KAME-only (pre-2292bis) MIP6 header declaratations.

do not inject packet to ipfilter, if the packet went through ipsec tunnel mode.
netbsd only.

drop linklocal packets with s6_addr16[1] already filled.
From: Markku Savela <msa@burp.tkv.asdf.org>

byebye M_ANYCAST6

correct openbsd compilation (no PR_LASTHDR)

no netinet6/ipsec.h for openbsd

enforce IPsec policy checking for transport mode input.
fixes *bsd ip6 (modulo openbsd).

slightly reduce #ifdef in in6_proto.c.

on bsdi[34], set net.inet6.ip6.v6only readonly

supply net.inet6.ip6.v6only on all platforms, to indicate AF_INET6 wildcard
bind behavior.

openbsd/bsdi3/freebsd2: 1, cannot change
netbsd: 1, can change unless INET6_BINDV6ONLY is defined at compilation time
bsdi4: 0, can change (no effect - should be made readonly)
freebsd3/4: 0, cannot change
	(should be made modifyable after conflict with ip6_mapped_addr gets
	fixed)

label must follow statement.  from netbsd-current

avoid non-intuitive hardcoded value for ip6stat.ip6s_m2m[]

- include "opt_natpt.h" when FreeBSD version is more than 3.

cope with freebsd4 bridge code.

inden

- renamed in6_linklocal_auto to ip6_auto_linklocal to be compatible with
  other sysctl names.
- added sysctl knob for ip6_auto_linklocal
  work on bsdi4
  code is available for bsdi3, nbsd, and openbsd (not compiled)

use callout_xx on freebsd[34].  less ifdef!  notified by thorpej

style

use timeout_foo for openbsd.  mip6 is not touched, maybe later
(the goal here is to sync with openbsd-current and use timeout_foo there)

do not log() on packetes to tentative address.

revisit comment in pfctlinput()

revisit safety check for pfctlinput2().

goodbye OLD_LOOPBACK_IF

fix comment for special consideration against fe80::%lo0/10

if rtalert or jumbo payload option has unexpected value in length field,
return parameter problem (pointing option length byte).

document assuption of ip6_process_hopopts

undo part of previous style change - did s/\t /\t/ too much for multi-line
function call (personally prefer openbsd KNF, but anyway, i don't think
i should enforce that).

style.

if rtalert/jumboopt has invalid length, drop the packet on the floor.
old packet just logs the invalid length and continued on - if we have
shorter option length, we may touch out-of-bounds memory.

topmost bit in option type (RFC2460 p9) talks about behavior when we received
unknown option type, not "known option with invalid length".  so p9 does not
apply to the change here.

indent

local variables in ip6_sysctl for netbsd and openbsd are now
necessary on both OSes.

renamed ip6_anon_regen_advance to ip6_temp_regen_advance

made it compile on netbsd

- sysctl knobs for addrconf privacy on BSDI
  (tested on bsdi4 only)
- changed the type of addrconv privacy lifetimes to u_int32_t

added sysctl knobs for autoconf privacy extension:
  net.inet6.ip6.use_tempaddr(bool): whether to use the extension
  net.inet6.ip6.temppltime(int): default preferred lifetime for temporary addrs
  net.inet6.ip6.tempvltime(int): default valid lifetime for temporary addrs
(currently for netbsd only)

changed naming convention of variables/constants related to addrconf
privacy, according to changed from the latest I-D to RFC 3041.

made them compilable on netbsd

several bug fixes on addrconf privacy extension.
now it seems to work well.

- more work on addrconf privacy extension.
  + generate a temporary address after configuring a public address
  + added random ID regeneration timer
  + prefer temporary address to public address
  + introduced random delay for regeneration/lifetimes
  + make sure not to increment lifetimes of temporary addresses
(it now works quite well, but still has some problems. I'll soon fix them.)
- more clarifications on (normal) addrconf
  + do not update prefix parameters if the on-link bit is zero
  + do not update address parameters if the autonomous bit is zero

comment out onion peeling history portion, it looks slow and too restrictive
(about header chain length)

onion peeling history.
fix paren match

style

candidate correspondent code (outside of MIP6).
wrapped into #ifdef ITOJUNMIP6CN so shouldn't affect others too much.

style (to avoid unnecessarily deep {})

do not accept a packet to link-local address on loopback, if the address
is not assigned to loopback interface (like fe80::2%lo0).
fixes KAME PR 250.

bring in OpenBSD 2.8 IPv6 ipsec support.  #ifdef hell...

s/register//

netbsd: bring in pfil_hooks

oops, string termination

make sure sizeof(protosw) == sizeof(ip6protosw).

sync with netbsd 1.5.  GENERIC.KAME compiles, not tested.

made ip6_notify_pmtu more scope-aware (for non-SCOPEDROUTING kernels).

- when fragmentation occurs, notify applications the MTU of the out IF via
  the IPV6_RECVPATHMTU mechanism.
- implemented a new ctlinput launch routine "pfctlinput2", which calls each
  pfctlinput fuction with an additional argument. This could/should be
  domain independent, but is currently INET6-specific at this momemnt
  for safety.
- adjusted xxxctlinput functions to be

IP6FW kernel loadable modulization and change SYSCTL_HANDLER_ARGS expression for fbsd 4.x-RELEASE

changed the defintion of ip6_mtuinfo{} according to the Erik's latest memo;
- changed the member names
- use the host byte order for ip6m_mtu
- swap the position of the two members

make it compile on openbsd/netbsd (type check )

supported IPV6_RECVPATHMTU socket option
(TODO: transport layer code for bsdi3 [fno]bsd)

error case consideration for ip6_setdstifaddr (comment update only)

improve comment for calls to ip6_setdstifaddr.

make sure to attach in6_ifaddr pointer to m_aux.  we won't have it
if ip6_dst is not in our interface addresses.  examples would be multicast
case, FAITH case and some others.

unused var

wrap in6_ifaddr attachment to m_aux into a function.
freebsd2: get rid of M_ANYCAST6 (other OSes should follow).

on packet input, attach pointer to in6_ifaddr into m_aux chain.
(to be used as alternative to M_ANYCAST and other flag bits)

drop TCP SYN if we forbid the use of deprecated address.

avoid confusing indentation

implement draft-itojun-ipv6-tclass-api-01.txt.

make IFA_STATS really work (netbsd)

in ip6_savecontrol(), check if a routing header is contained, in order to
set destination option headers properly.

hardcode NEW_STRUCT_ROUTE to 1 for freebsd4.
I'll do the same thing for freebsd3 soon.

add missing \n to log().

(for performance measurement only) record destination addresses as well.

for netbsd, supply sysctl MIB anonport{min,max} and lowport{min,max}.

avoid unsafe typecast.

hardcode NEW_STRUCT_ROUTE to 1 (openbsd)

reset ip6_forward_cache_miss when changing 'goto ours' algorithm.

#error if NEW_STRUCT_ROUTE is undefined on netbsd
XXX may not be the right place to do this, but i could find no other place

added the OURS_CHECK_ALG_LARGEHASH case in the algorithm selection
for 'goto ours'

removed garbage

more accurate measurement

clear log2[] buf when changig ours algorithm

- added a statistics buffer
- reset ro_dst after cached route mismach

calculate performance more accurately when forwarding

measure performance from input to forward

made the "goto ours" algorithm dynamically configurable via sysctl
notes:
  - freebsd only
  - code is adhoc and ugly. sorry, I'll eventually remove all the stuff...

fixed a typo of a compile option

added a sanity check for retunred in6hash entries (just for safety)