summaryrefslogtreecommitdiff
path: root/sys/net
AgeCommit message (Expand)Author
2004-02-20Make pfsync deal with clearing states bound to a group or interface (egRyan Thomas McBride
2004-02-19Makes pfctl -Fs and pfctl -w works with the optional -i specifier.Cedric Berger
2004-02-19the 2nd round of the qid assignment change.Kenjiro Cho
2004-02-17Tighten pfi_skip_if() up, and a bit of KNF. ok mcbride@Cedric Berger
2004-02-1518 -> ETHER_ADDR_LEN*3 in ether_sprintf() for clarityAnil Madhavapeddy
2004-02-13Do an explicit pf_update_anchor_rules() after an anchor gets removed.Marco Pfatschbacher
2004-02-12from camield:Henning Brauer
2004-02-10KNFDaniel Hartmeier
2004-02-10plug mbuf leak (ip_fragment() always free mbuf on error). tested by cedric,Jun-ichiro itojun Hagino
2004-02-10KNFHenning Brauer
2004-02-10Make pfsync work correctly with IP options on 64-bit alignmentRyan Thomas McBride
2004-02-09Repair "set loginterface". Don't flush stats on pfctl -e. pf_status.sinceCedric Berger
2004-02-08if_detach_rtdelete(): abort and restart rn_walktree() if a cloning routeMarkus Friedl
2004-02-08Fix kernel panic which occurs under very high load:Ryan Thomas McBride
2004-02-07Use the offset provided to us by m_pulldown(), rather than using size ofRyan Thomas McBride
2004-02-06as seen in netbsd. crank bpf sizes to adapt to faster networks.Ted Unangst
2004-02-04Fix a number of bugs with setting pool limits which I introduced withRyan Thomas McBride
2004-02-02missing #if NPF > 0. ok henning@Cedric Berger
2004-02-02Do not evaluate pfi_index2kif[ifp->if_index] if PF is disabled.Cedric Berger
2004-01-27drop packet if kif == NULL; ok henning deraadtMarkus Friedl
2004-01-27don't convert tcpmd5 to ip-over-ip in SADB_X_GETSPROTO; from hshoexerMarkus Friedl
2004-01-26- use SIOC[GS]WAVELAN.Federico G. Schwindt
2004-01-22- Include the value of pf_state.timeout in pfsync messagesRyan Thomas McBride
2004-01-20the pfsync interface does not have a baudrate, so don't claim 100 MBit/sHenning Brauer
2004-01-20Ignore pfsync packets if pf is not running.Ryan Thomas McBride
2004-01-19Update comment; handling PFSYNC_ACT_UPD in pfsync_input() is no longerRyan Thomas McBride
2004-01-19Clean up creation and expiry timestamp calculations.Ryan Thomas McBride
2004-01-18Port is already stored in network byte order, no need to convert.Ryan Thomas McBride
2004-01-16Fix IPv6 stateful tcp scrubbing by not dereferencing a null pointer.Ryan Thomas McBride
2004-01-15add a RTM_IFANNOUNCE message; from netbsd; ok itojun, henningMarkus Friedl
2004-01-12use klist_invalidate to permit destroy while kqueued. ok mpf@Ted Unangst
2004-01-09fix leak ether_deatch(): if if_free_sadl() is called before if_detach()Markus Friedl
2004-01-07PFI_MTYPE leak; ok cedric@Markus Friedl
2004-01-07ieee80211 framework from NetBSD; ok'd by several people some time ago.Federico G. Schwindt
2004-01-06Drop UDP packets with destination port 0, or zero or oversized payloadDaniel Hartmeier
2004-01-05stop ifc_destroy() if there are still knotes registered.Marco Pfatschbacher
2004-01-050 -> (void *)NULL for last argument of icmp_error(), which is of typeDaniel Hartmeier
2004-01-05Repair my merging error, simplify DIOCCLRSTATUS code. ok dhartmei@Cedric Berger
2004-01-05Repair merge errors. Thanks Pyun YongHyeon, Sorry Henning :)Cedric Berger
2004-01-04oops... string.h ended up being included twice; pointed out by espiePeter Valchev
2004-01-04better macro name (IF_LOCKED -> BOUND_IFACE). from markus.Cedric Berger
2004-01-04include proper protos for userland; deraadtPeter Valchev
2004-01-03make sure userland sees memcmp and friends (gcc3)Marc Espie
2004-01-03put an mi wrapper around stdarg.h/varargs.h. gcc3 moved stdarg/varargs macrosMarc Espie
2003-12-31spacing. note this, cedricTheo de Raadt
2003-12-31delay interfaces attach until "self" has been created; ok cedric@Markus Friedl
2003-12-31Many improvements to the handling of interfaces in PF.Cedric Berger
2003-12-28Add a new PFSYNC_ACT_UREQ message type.Ryan Thomas McBride
2003-12-22pasto in pf_status.src_nodes backup, from 'kirash'Daniel Hartmeier
2003-12-19more const-correctness, ok mcbride@Daniel Hartmeier