src/sys/netinet/if_ether.c, branch release/4.8.0_cvs

This commit was manufactured by cvs2svn to create tag

2003-04-03T18:49:23Z

'RELENG_4_8_0_RELEASE'. This commit was manufactured to restore the state of the 4.8-RELEASE image. Releases prior to 5.3-RELEASE are omitting the secure/ and crypto/ subdirs.

Re-enable non pre-emptive ARP requests.

2002-12-12T23:19:02Z

Submitted by: "Diomidis Spinellis" PR: kern/46116

MFC 1.95: Register the arpintr netisr early.

2002-10-21T20:48:54Z

This fixes panics (mainly seen on alphas) which happen when dhclient is run and the machine crashes because the arpintr netisr has not yet been installed. PR: 41642 Tested by: pirzyk, others on freebsd-alpha

MFC 1.76: Use macros rather than fondling implementation

2002-10-13T07:11:46Z

details.

MFC netinet/if_ether.c:1.93-1.94,

2002-06-18T00:15:31Z

net/if_iso88025subr.c:1.21, net/iso88025.h:1.6-1.8: Move ISO88025 source routing information into sockaddr_dl's sdl_data field. This returns the sdl_data field to a variable-length field. More importantly, this prevents a easily-reproduceable data-corruption bug when the interface name plus the hardware address exceed the sdl_data field's original 12 byte limit. However, token-ring interfaces may still overflow the new sdl_data field's 46 byte limit if the interface name exceeds 6 characters (since 6 characters for interface name plus 6 for hardware address plus 34 for source routing = the size of sdl_data). Sponsored by: NTT Multimedia Communications Labs

MFC (rev 1.91): support for pre-expiry arp requests for hosts we

2002-03-27T16:40:59Z

are sending data to. PR: kern/25517 Reviewed by: luigi

unbreak binary compatibility for network drivers

2002-02-20T23:34:09Z

MFC: ARP support for variable length link level addresses

2002-02-13T21:38:56Z

MFC: prevent a panic when using bridging and one of the interface does

2001-12-26T17:36:36Z

not have an address. Also add a comment to point out what the correct solution should be, but this is enough of a fix for the time being. Hopefully we will have time to come up with a proper one for 4.5 (subject to re approval). Approved by: re

MFC:

2001-12-20T10:30:19Z

net/ if.c,v 1.124, if_disc.c,v 1.29, if_faith.c,v 1.8, if_loop.c,v 1.65, if_stf.c,v 1.14, if_var.h,v 1.40, route.c,v 1.67, route.h,v 1.41, rtsock.c,v 1.61 netinet/ if_atm.c,v 1.9, if_atm.h,v 1.4, if_ether.c,v 1.85, in_pcb.c,v 1.91 netinet6/ in6_pcb.c,v 1.21, nd6.c,v 1.14, nd6.h,v 1.9 Pull post-4.4BSD change to sys/net/route.c from BSD/OS 4.2, also found in CSRG archive. Have sys/net/route.c:rtrequest1(), which takes ``rt_addrinfo *'' as the argument. Pass rt_addrinfo all the way down to rtrequest1 and ifa->ifa_rtrequest. 3rd argument of ifa->ifa_rtrequest is now ``rt_addrinfo *'' instead of ``sockaddr *'' (almost no one is using it anyways). Benefit: the following command now works. Previously we needed two route(8) invocations, "add" then "change". # route add -inet6 default ::1 -ifp gif0 Remove unsafe typecast in rtrequest(), from ``rtentry *'' to ``sockaddr *''. It was introduced by 4.3BSD-Reno and never corrected. Also, bring in the following fixes from -CURRENT that prevented this code from being MFC'ed earlier: net/ rtsock.c,v 1.62 In RTM_CHANGE, only call rt_getifa() (to set the new ifp/ifa) if we've either been passed a gateway (and the route is marked as RTF_GATEWAY) or if we've been given an RTA_IFP or changed RTA_IFA sockaddr. net/ rtsock.c,v 1.63 Back out revision 1.7 -- it's no longer necessary to ensure that ``gate'' is set when RTF_GATEWAY is passed, as subsequent code does that check now anyway.