summaryrefslogtreecommitdiff
path: root/usr.sbin/bgpd/rde_rib.c
AgeCommit message (Expand)Author
2013-08-14Rewrite the internals of the RDE reload logic.Claudio Jeker
2013-07-17on graceful restart, the number of prefixes could be counted wrong,Sebastian Benoit
2013-05-20D'oh, add missing break in prefix_writebuf(). Another find by blambert@Claudio Jeker
2013-03-14correct struct in sizeof(); found by llvmFlorian Obser
2012-09-12Better graceful restart support (implementing more then just the EoR record).Claudio Jeker
2012-07-01typo in comment, s/withdrawl/withdrawal/Stuart Henderson
2012-05-22Flush the right networks, the dynamic ones not the static ones.Claudio Jeker
2011-09-21Fix nexthop_modify() to reset the flags when called. Until nowClaudio Jeker
2011-09-20Move a few functions into util.c because bgpctl will need them soon.Claudio Jeker
2011-09-17Implement new mrt table dump format as specified in draft-ietf-grow-mrt.Claudio Jeker
2011-01-14plug memleak in err path; from zinovik, ok claudioHenning Brauer
2010-05-03Make it possible to load multiple routing tables at the same time and useClaudio Jeker
2010-04-20prefix_unlink() must remove the rib entry. Currently this was only doneClaudio Jeker
2010-04-07Call nexthop_delete() a bit later in nexthop_update(). The nh->state needsClaudio Jeker
2010-04-06Switch to a more address family independent nexthop imsg. Instead of passingClaudio Jeker
2010-03-29We always allocate rib ids dynamicaly so there is no need for allowingClaudio Jeker
2010-03-26Be more careful when walking the tree looking for a non-empty element,Claudio Jeker
2010-03-03Replace enum rib_state with enum reconf_action since their doing the same.Claudio Jeker
2010-01-13Add support for BGP MPLS VPN aka RFC 4364. This is only the RDE part soClaudio Jeker
2010-01-10Switch rib_dump() to use AID instead of AFs. OK henning@Claudio Jeker
2009-12-01Use an artificial address family id in struct bgpd_addr and almost everywhereClaudio Jeker
2009-10-05Load prefixes into new created RIBs at reload time by walking over theClaudio Jeker
2009-06-29Protect the other rde_send_kroute() with a F_RIB_NOFIB check.Claudio Jeker
2009-06-07First attempt at reload support for RIBs. There is some magic that I doClaudio Jeker
2009-06-04Implement rib_find and add a rib id to struct rde_peer.Claudio Jeker
2009-06-04Add "rde rib <name>" to the config and allow the rde to use these other RIBs.Claudio Jeker
2009-06-03Allocate all rib contextes for tree walks and don't use static stack memory.Claudio Jeker
2009-06-03Initial stab at rib_free() until now unused.Claudio Jeker
2009-06-03Better way to allocate new RIBs.Claudio Jeker
2009-06-03Make prefix_destroy more generic and use it in prefix_remove this is possibleClaudio Jeker
2009-06-02Move the rest of the rib dump functions into rde_rib.c where it belongs.Claudio Jeker
2009-06-01Use only one list to queue the dump contextes on. Use the list in structClaudio Jeker
2009-06-01Do not call the upcall twice on some prefixes. Move the upcall back toClaudio Jeker
2009-06-01Holy simplification batman. Use the per rib entry flags to lock entriesClaudio Jeker
2009-06-01Instead of storing a pointer to the RIB head in the RIB element use thatClaudio Jeker
2009-05-27Move update and withdraw code into own functions to simplify the necessaryClaudio Jeker
2009-05-21Make it possible to turn off the decision process per RIB. This is mainlyClaudio Jeker
2009-05-17F_LOCAL and F_ORIGINAL are gone. The Adj-Rib-In is now a distinct tree.Claudio Jeker
2009-05-17rib_dump_r needs to check the af of the prefix because there is no rib perClaudio Jeker
2009-05-17Rework most of the RDE to allow multiple RIBs. This is mostly preparationClaudio Jeker
2009-04-23Rework the way we handle announced networks. Instead of two freak rde_peersClaudio Jeker
2008-11-21Track nexthops when the underlying route is changing. Until now true nexthopsClaudio Jeker
2007-06-01Remove a stupid wrapper function that does nothing more then calling anotherClaudio Jeker
2007-05-11Various spelling fixes from Stuart Henderson.Claudio Jeker
2007-04-02Typo.Claudio Jeker
2007-02-22KNFHenning Brauer
2007-01-11Correct logic in path_update() so that moves are only done when needed.Claudio Jeker
2006-12-12Change nexthop_delete() to be more obvious what's going on. No functionalClaudio Jeker
2006-12-12Even IPv6 has a prefixlen limit. This ensures that we do not overflow theClaudio Jeker
2006-12-12In path_remove() remove only local prefixes from the pftable.Claudio Jeker