aboutsummaryrefslogtreecommitdiff
path: root/sys/netinet
Commit message (Expand)AuthorAgeFilesLines
* Correctly move the packet header in ip_insertoptions().Alan Cox2005-01-231-1/+2
* Sort sections.Ruslan Ermilov2005-01-201-5/+5
* - Reduce number of arguments passed to dummynet_io(), we already have cookieGleb Smirnoff2005-01-163-8/+5
* o Clean up interface between ip_fw_chk() and its callers:Gleb Smirnoff2005-01-143-82/+109
* Fix a TCP SACK related crash resulting from incorrect computationPaul Saab2005-01-121-6/+16
* include "alias.h", not <alias.h>Brian Somers2005-01-101-1/+1
* /* -> /*- for license, minor formatting changesWarner Losh2005-01-0767-70/+70
* Add a sysctl (net.inet.tcp.insecure_rst) which allows one to specifyMike Silbersack2005-01-032-2/+14
* Port randomization leads to extremely fast port reuse at highMike Silbersack2005-01-024-4/+69
* Remove an errant blank line apparently introduced inRobert Watson2004-12-251-1/+0
* In the dropafterack case of tcp_input(), it's OK to release the TCPRobert Watson2004-12-252-2/+2
* Revert parts of tcp_input.c:1.255 associated with the header predictedRobert Watson2004-12-252-4/+14
* Attempt to consistently use () around return values in calls toRobert Watson2004-12-232-36/+36
* Remove an XXXRW comment relating to whether or not the TCP timers areRobert Watson2004-12-232-12/+2
* Remove the now unused tcp_canceltimers() function. tcpcb timers areRobert Watson2004-12-232-16/+0
* Remove an annotation of a minor race relating to the update ofRobert Watson2004-12-231-7/+0
* In certain cases ip_output() can free our route, so checkGleb Smirnoff2004-12-101-1/+2
* Revert last change.Gleb Smirnoff2004-12-101-4/+4
* This commit adds a shared locking mechanism very similar to theChristian S.J. Peron2004-12-101-29/+69
* Check that DUMMYNET_LOADED before seeking dummynet m_tag.Gleb Smirnoff2004-12-091-4/+4
* More fixing of multiple addresses in the same prefix. This time do not tryMax Laier2004-12-091-2/+20
* Time out routes created by redirect.Ruslan Ermilov2004-12-061-1/+4
* - Make route cacheing optional, configurable via IFF_LINK0 flag.Gleb Smirnoff2004-12-061-0/+6
* Assert the tcptw inpcb lock in tcp_timer_2msl_reset(), as fields inRobert Watson2004-12-051-0/+1
* Assert inpcb lock in:Robert Watson2004-12-052-2/+24
* Minor grammer fix in comment.Robert Watson2004-12-051-1/+2
* Pass the inpcb reference into ip_getmoptions() rather than just theRobert Watson2004-12-051-6/+14
* Define INP_UNLOCK_ASSERT() to assert that an inpcb is unlocked.Robert Watson2004-12-051-0/+1
* Push the inpcb argument into ip_setmoptions() when setting IP multicastRobert Watson2004-12-051-10/+8
* Start working through inpcb locking for ip_ctloutput() by cleaning upRobert Watson2004-12-051-10/+13
* Fixes a bug in SACK causing us to send data beyond the receive window.Paul Saab2004-11-291-2/+4
* Assert the inpcb lock in tcp_xmit_timer() as it performs read-modify-Robert Watson2004-11-282-0/+4
* Expand coverage of the receive socket buffer lock when handling urgentRobert Watson2004-11-282-4/+6
* Do export the advertised receive window via the tcpi_rcv_space field ofRobert Watson2004-11-272-1/+2
* Implement parts of the TCP_INFO socket option as found in Linux 2.6.Robert Watson2004-11-262-2/+120
* Fix a problem where our TCP stack would ignore RST packets if the receiveMike Silbersack2004-11-252-4/+6
* In tcp_reass(), assert the inpcb lock on the passed tcpcb, since theRobert Watson2004-11-232-24/+38
* tcp_timewait() performs multiple non-atomic reads on the tcptwRobert Watson2004-11-235-0/+21
* De-spl tcp_slowtimo; tcp_maxidle assignment is subject to possibleRobert Watson2004-11-231-15/+11
* Assert the inpcb lock in tcp_twstart(), which does both read-modify-writeRobert Watson2004-11-232-0/+20
* Assert inpcb lock in tcp_quench(), tcp_drop_syn_sent(), tcp_mtudisc(),Robert Watson2004-11-232-0/+8
* Assert the tcbinfo write lock in tcp_new_isn(), as the tcbinfo lockRobert Watson2004-11-232-8/+22
* Remove "Unlocked read" annotations associated with previously unlockedRobert Watson2004-11-222-6/+0
* s/send/sent/ in comment describing TCPS_SYN_RECEIVED.Robert Watson2004-11-211-1/+1
* - Since divert protocol is not connection oriented, remove SS_ISCONNECTED flagGleb Smirnoff2004-11-181-33/+0
* Fix host route addition for more than one address to a loopback interfaceMax Laier2004-11-171-1/+1
* Merge copyright notices.Max Laier2004-11-131-28/+1
* Fix ng_ksocket(4) operation as a divert socket, which is pretty usefulGleb Smirnoff2004-11-121-11/+12
* Change the way we automatically add prefix routes when adding a new address.Max Laier2004-11-121-27/+147
* Add missing '='Poul-Henning Kamp2004-11-111-1/+1