Commit Graph

54 Commits

Author SHA1 Message Date
itojun e127ba2f0b fix well-known CPI handling bug. (sync with KAME code) 1999-11-05 14:57:00 +00:00
itojun 37ecccca6b decrease amount of history buffer to use for IPcomp.
the default setting of zlib allocates too much memory and of no use for
network packets (which are like < 2k).

From: Laine Stump <lainestump@rcn.com>
1999-11-05 14:56:26 +00:00
itojun 23c246780c consistent logging for icmp6 redirects
XXX should make logs 1-liner so that duplicated logs can be compressed
by syslog(8)?
1999-10-01 10:16:16 +00:00
itojun 97999826c8 sanity check against truncated extension headers. 1999-10-01 10:15:16 +00:00
is 1eb4d34bba Add missing "case IFT_ARCNET". 1999-09-26 20:08:15 +00:00
is 3a43ac11b5 Don't even pretend we can create a nonglobal EUI64 out of an ARCNET link
level address. Instead, create the link-local address directly.
1999-09-26 20:04:08 +00:00
is 1f417b8ca9 invert u bit to convert EUI64 to RFC2373 interface ID for ARCnet 1999-09-25 22:29:21 +00:00
itojun 62c568fac4 tiny fix to ARCnet IPv6 support.
- in in6_ifattach_getifid(), we can grab interface id source iff the source
  is universally (worldwide) unique.  ARCnet hardware address is of 8bit and
  does not satisfy the condition.
  (in6_ifattach_getifid() is for getting interface id usable for pseudo
  interfaces like gif*)
- xx_to_eui64() should return EUI64 format, not IPv6 interface id format.
  this may seem awkward so I wish to clean these things up.
- in nd6.c, change if clause into case clause to allow future addition
  of IFT_xxx easier.
1999-09-20 02:35:44 +00:00
is 072a96b17d fix mergo 1999-09-19 21:42:23 +00:00
is 6a793d8a47 Zeroth version of IPv6 support for ARCnet. Correct MTU handling still needs
to be done.
1999-09-19 21:31:33 +00:00
itojun b3d9b0130c eliminate unnecessary splnet(). 1999-09-17 12:26:04 +00:00
itojun 6b9bfc0691 - Call in{,6}_pcbdetach if ipsec initialization is failed during PRU_ATTACH.
This situation happens on severe memory shortage.  We may need more
  improvements here and there.
- Grab IEEE802 address from IFT_ETHER card, even if the card is
  inserted after bootup time.  Is there any other card that can be
  inserted afterwards?  pcmcia fddi card? :-P
- RFC2373 u bit handling suggests that we SHOULD NOT copy interface id from
  ethernet card to pseudo interface, when ethernet card has IEEE802/EUI64
  with u bit != 0 (this means that IEEE802/EUI64 is not universally unique).
  Do not use such address as, for example, interface id for gif interface.
  (I have such an ethernet card myself)
  This may change interface id for your gif interface.  be careful upgrading
  rc files.

(sync with recent KAME)
1999-09-13 12:15:54 +00:00
itojun 770a224d1e fix ipcomp behavior against -R to meet documentation.
From: Laine Stump <lainestump@rcn.com>
1999-09-10 02:05:24 +00:00
itojun a00c3d5ea1 fix u bit in interface identifier for ether and p2p-802 interfacde. 1999-09-08 00:50:15 +00:00
itojun 45538437d9 - invert u bit on interface id for pseudo interfaces, as suggested in RFC2373.
- do not perform IPv6 initialization for faith* interface, as they become
  mistakingly IFF_UP.  we are wondering if we should nuke in6_ifattach_p2p().
(sync with recent kame)
1999-09-05 01:57:10 +00:00
itojun 5a219342f2 sync with kame; typo in comment. 1999-08-26 11:11:51 +00:00
itojun 0647902fc8 fix IPv6 fragment ID initialization - random() does not return
random value when frag6_init() is called, so use microtime() to stir
the value better.
1999-08-26 11:10:49 +00:00
itojun 4e69e908fe fix AH computation for HbB options. 1999-08-25 17:47:47 +00:00
itojun ad853adad7 sync with recent kame: fix source address selection on IPv6 tunnel ipsec. 1999-08-25 12:56:38 +00:00
itojun d48c55f4f0 ctlinput handling must look at ip6_src, not ip6_dst.
(this makes path mtu handling wrong)
1999-08-25 12:38:14 +00:00
itojun 00cf8638cb do not capture packets by gif, when gif interface is down. 1999-08-20 10:07:40 +00:00
itojun b0ad0eaf51 typo fix (from koji@dti.ad.jp).
remove things that are already done.
1999-08-14 19:17:19 +00:00
itojun 8dba61e664 fix multicast demux.
From: Matthias Drochner <drochner@zel459.zel.kfa-juelich.de>
1999-08-13 10:57:05 +00:00
itojun a9b7fe4621 return with doing nothing from xx_ctlinput(), when sa->sa_family
is not the expected one.

I see PRC_REDIRECT_HOST with sa->sa_family == AF_UNIX coming to
{tcp,udp}_ctlinput() when I use dhclient, and I feel like adding
more sanity checks, without logging - if we log it it is too noisy.
1999-08-09 10:55:29 +00:00
itojun 7a60aa28c2 log() needs "\n" at the end. 1999-08-09 02:24:52 +00:00
itojun e5d7b92a94 remove invalid initialization if in6_iflladdr. 1999-08-07 12:33:04 +00:00
itojun 6d8996ba49 import recent kAME fixes.
- initialize hoplimit for raw6 socket properly.
- respect SO_TIMESTAMP on udp6.
- more sanity checks.
1999-08-05 16:01:07 +00:00
itojun 6f3fbf1ce6 remove sys/netinet6/in6_systm.h, as it is very empty.
crypto-us IPSEC build will be broken.
could someone please update?
1999-08-02 11:04:38 +00:00
itojun 70ada0957e sync with recent KAME.
- loosen ipsec restriction on packet diredtion.
- revise icmp6 redirect handling on IsRouter bit.
- tcp/udp notification processing (link-local address case)
- cosmetic fixes (better code share across *BSD).
1999-07-31 18:41:15 +00:00
itojun 06c350054d remove reference to in6_systm.h (file itself will be removed afterwords) 1999-07-30 10:35:34 +00:00
explorer ed3b6d5791 Fix a problem where tcp_slowtimo was called twice, once for ipv4 tcp and
once for ipv6.  This patch makes the ipv6 case pass NULLs in for fast
and slow timeouts iff defined(INET) and passes in the right function
if !defined(INET).

Reveiwed by itojun@iijlab.net.
1999-07-27 06:23:57 +00:00
itojun 86ef461e35 avoid u_long and hardcoded numbers. 1999-07-22 15:46:13 +00:00
itojun 7fee35f579 - implement IPv6 pmtud, which is necessary for TCP6.
- fix memory leak on SO_DEBUG over TCP.
1999-07-22 12:56:56 +00:00
itojun 76fc599a8d change unnecessary u_long/long into u_int32_t or something relevant.
more fixes should follow.
1999-07-22 03:59:42 +00:00
itojun 869f140395 fix IPV6_CHECKSUM socket option (length computation was wrong). 1999-07-19 18:37:19 +00:00
itojun c74f79d16f fix faith interface support. need testing.
(i understand this is a dirty hack, of course)
1999-07-17 07:07:08 +00:00
itojun 4502752019 fix compilation/runtime problem on alpha.
PR: 7952, 7953
From: Dave Huang <khym@bga.com>
1999-07-11 17:45:11 +00:00
thorpej 011b02e348 Clean up some printfs(), and mark a few for possible later nuking,
since they appear to be for debugging purposes only.
1999-07-10 19:46:10 +00:00
thorpej 267920eb1a defopt INET6, and put it in opt_inet.h (most places already include this
file, which is why the file list is so short).
1999-07-09 23:41:16 +00:00
thorpej f9a7668b3f defopt IPSEC and IPSEC_ESP (both into opt_ipsec.h). 1999-07-09 22:57:15 +00:00
itojun dcc13cdd33 sync with KAME/NetBSD 1.4, SNAP kit 19990705.
key changes are:
- icmp6 redirect fix (dst check)
- revised ip6 multicast check for loopback i/f
- several RCS ID cleanups
1999-07-06 12:23:19 +00:00
itojun 6ea6ccf105 checked build on alpha and i386, with GENERIC.v6.
fixed several sizeof(void *) and sizeof(size_t) issues on alpha.

Thanks to: Dave Huang and Tim Rightnour
1999-07-06 08:55:56 +00:00
itojun f3c72d8503 fix IPSEC (but not INET6) build.
PR: 7921, 7922, 7924
From: rafal@mediaone.net
1999-07-06 05:09:22 +00:00
itojun 9b74747370 s/splnet/splsoftnet/ in IPv6/IPsec part.
hope I made no mistake (the kernel works fine but I need a regress test)

Suggested by: thorpej
1999-07-04 02:01:15 +00:00
thorpej cd3a345ea0 RCS ID police. 1999-07-03 21:24:45 +00:00
kleink dd24133620 Take a stab at namespace protection. For now, only the obvious problems are
addressed, the culprit being the lack of a namespace definition for an IPv6-
extended <netinet/in.h> in XNS5.2 D2.0; I'll try to work something out and
submit it to the review WG.
1999-07-03 14:04:57 +00:00
itojun af58d5709d try to get a non-conflicting port # when bind(2) to port number 0
is called.
1999-07-02 19:57:05 +00:00
itojun 243585320c expand insque/remque (quick hack). fundamental fix should be done
while clarifying relationship between inpcb and in6pcb.

PR: 7891
1999-07-02 12:43:42 +00:00
itojun f8346292af move ipsec sysctl index to IPPROTO_AH (instead of IPPROTO_ESP),
so that you can perform sysctl operation when ESP is not compiled in.
1999-07-02 08:46:47 +00:00
itojun 932ab72aee remove TIME_WAIT issue, it was false. 1999-07-02 07:36:33 +00:00