src-test2/sys/netinet/ip_var.h, branch stable/4

MFC of port randomization rate-based disabling:

2005-01-02T01:53:44Z

in_pcb.c rev 1.157 in_pcb.h rev 1.79 ip_input.c rev 1.293 ip_var.h rev 1.93 Approved by: re

MFC: Add the IP_ONESBCAST option, to enable undirected IP broadcasts to be

2003-09-09T19:09:23Z

sent on specific interfaces. This is required by aodvd, and may in future help us in getting rid of the requirement for BPF from our import of isc-dhcp. PR: kern/37486 Obtained from: BSD/OS Approved by: re, jake (mentor)

Merge from -current support for Protocol Independent Multicast.

2003-08-24T08:24:39Z

Submitted by: Pavlin Radoslavov

A correct MFC:

2003-02-27T04:50:02Z

ip_input.c revs 1.225, 1.229 ip_var.h rev 1.71 Approved by: re (bmah)

Backout:

2003-02-27T04:12:01Z

ip_input 1.130.2.49 ip_var.h 1.50.2.10 This is not even close to a straight mfc, my mistake.

IP frag per packet limits

2003-02-27T03:55:46Z

MFC ip_input.c rev 1.225 ip_var.h rev 1.71 Approved by: re (bmah)

MFC: m_tag support

2003-01-23T21:06:48Z

Replace aux mbufs with packet tags: o instead of a list of mbufs use a list of m_tag structures a la openbsd o for netgraph et. al. extend the stock openbsd m_tag to include a 32-bit ABI/module number cookie o for openbsd compatibility define a well-known cookie MTAG_ABI_COMPAT and use this in defining openbsd-compatible m_tag_find and m_tag_get routines o rewrite KAME use of aux mbufs in terms of packet tags o eliminate the most heavily used aux mbufs by adding an additional struct inpcb parameter to ip_output and ip6_output to allow the IPsec code to locate the security policy to apply to outbound packets o bump __FreeBSD_version so code can be conditionalized o fixup ipfilter's call to ip_output based on __FreeBSD_version

MFC: fix multicast routing KLD support, including some minor bug fixes

2002-11-20T23:20:55Z

(see 1.83 of ip_mroute.c). In detail: conf/files ip_mroute optional mrouting net/route.c remove useless #include, MFC mrouting KLD support netinet/ip_encap.c remove useless #include, access ipip_input through a function pointer (only a partial MFC -- -current has a better solution). fix interface to encap4_input() to avoid varargs netinet/ip_encap.h remove __P, fix interface to encap4_input netinet/ip_input.c MFC: mrouting KLD support, and move here rsvp_input() together with the rest of the rsvp code. netinet/ip_mroute.[ch] MFC 1.83 mrouting cleanup and KLD support netinet/ip_output.c MFC mrouting KLD support netinet/ip_var.h MFC support for mrouting KLD netinet/raw_ip.c MFC support for mrouting KLD Also some innocuous style cleanup (fix function headers, remove register, etc.) which have slipped in and i am too fed up to remove yet another time (and rerun all the kernel builds with all possible combinations of relevant options to make sure their removal does not introduce bugs).

MFC: in_rmx.c,v 1.39, ip_input.c,v 1.165, and ip_var.h,v 1.54:

2002-08-09T14:49:23Z

Invalidate cached forwarding route (ipforward_rt) whenever a new route is added to the routing table, otherwise we may end up using the wrong route when forwarding. PR: kern/10778 Spotted by: Sergey Starosek , Andrew Rukavishnikov

The purpose of this commit is to bring the net/ and netinet/ sources

2002-07-09T09:11:43Z

more in sync with what is in -current, so that mainteinance and bugfix of the two trees is not a nightmare. THERE IS NO FUNCTIONAL OR EXTERNAL API CHANGE IN THIS COMMIT You should not need to recompile any userland code. In (some) more detail, this commit does the following: * remove a number of static variables from the ip stack that were used to support DIVERT, IPFIREWALL_FORWARD and stateful rules/ dynamic pipes. These are replaced with packet annotations, much in the same (ugly for someone, but it is extremely fast and easy to extend) way used for ages to support dummynet annotations. On passing, fix a bug in the handling of divert for fragmented packets. * as part of the removal of static variables, change the (internal) interface of ip_fw_chk() to use a single structure to hold arguments. Adapt clients of the above (ip_input, ip_output, bridge, ether_output, ether_demux) to use the new interface. * remove some unused variables. * remove some of the __P() macros from some of the files involved Because of the NO FUNCTIONAL CHANGE you don't get the following features which are in -current: * ipfw on layer-2 packets. All the hooks and the code are there, but the controlling variable net.link.ether.ipfw: 0 is readonly because i am only 99% confident on how the old ipfw handles these frames. Just edit if_ethersubr.c to make the variable RW if you want this feature. I might commit this in due time if there is interest. these frames. Just edit if_ethersubr.c to make it RW if you want this feature. I might commit this in due time if there is interest. * ipfw2, the new, faster and more flexible firewall code. The code has hooks to make use of ipfw2, and I will make patches available to use it (it is basically 3 files, netinet/ip_fw2.[ch], sbin/ipfw/ipfw2.c, plus one-line changes in conf/options, conf/files and sbin/ipfw/Makefile, plus libalias patches).