summaryrefslogtreecommitdiffstats
path: root/sys/netinet
Commit message (Expand)AuthorAgeFilesLines
* Fixe some whitespace nits that were introduced in r216758.lstewart2011-01-111-7/+7
* Reset the last_sack_ack SACK hint for TCP input processing to ensure that thelstewart2011-01-101-0/+1
* Make sure to always do source address selection ondeischen2011-01-081-8/+9
* Trim extra spaces before tabs.jhb2011-01-0710-21/+21
* Fix a memory leak in ARP queues.gnn2011-01-071-2/+1
* Adjust ARP hold queue locking.gnn2011-01-071-2/+3
* Use a regular taskqueue for dummynet rather than a "fast" taskqueue.jhb2011-01-071-1/+1
* Bugfix: Make sure that the COMM_UP notificatin is delivered first alsotuexen2011-01-021-5/+18
* Fix a typo.tuexen2011-01-011-1/+1
* Try to catch a possible divide-by-zero as early as possible if "mtu" is 0bz2010-12-311-0/+3
* Define and use SCTP_SSN_GE, SCTP_SSN_GT, SCTP_TSN_GE, SCTP_TSN_GT macrostuexen2010-12-307-210/+125
* Code cleanup: Use LIST_FOREACH, LIST_FOREACH_SAFE, TAILQ_FOREACH,tuexen2010-12-309-738/+439
* Fix three bugs related to the sequence number wrap-around affectingtuexen2010-12-303-3/+3
* Add a comment for the ccv member of struct tcpcb.lstewart2010-12-281-1/+1
* - Add some helper hook points to the TCP stack. The hooks allow Khelp modules tolstewart2010-12-284-0/+97
* Add a new sack hint to track the most recent and highest sacked sequence number.lstewart2010-12-282-0/+2
* Fix a whitespace nit introduced in r215166.lstewart2010-12-281-1/+1
* Remove comment bemoaning the lack of an INP_INHASHLIST above in_pcbdrop();rwatson2010-12-271-6/+0
* Provide a possibility to configure the inital congestion window to thetuexen2010-12-222-9/+13
* Improve plausibility check in sctp_handle_sack().tuexen2010-12-227-79/+68
* Fix a typo in a comment.jhb2010-12-211-1/+1
* Fix a flightsize bug related to the processing of PKTDRP reports.tuexen2010-12-171-1/+1
* Bugfix: Take also the nr-mapping array into account when detectingtuexen2010-12-162-10/+14
* Add a missing cast. Reported by blade_ly at yahoo.com.cn.tuexen2010-12-161-1/+1
* Bring back (most of) NATM to avoid further bitrot after r186119.bz2010-12-151-0/+6
* Bugfix: Do correct accounting using the MIB counters when antuexen2010-12-121-0/+5
* Use correct field to track statistics counting error as bad header length.bz2010-12-051-1/+1
* Fix a bug where also the number of non-renegable gap reportstuexen2010-12-043-47/+25
* Import a clean-room implementation of the experimental H-TCP congestion controllstewart2010-12-021-0/+521
* Import a clean-room implementation of the experimental CUBIC congestion controllstewart2010-12-022-0/+625
* General cleanup of the NewReno CC module (no functional changes):lstewart2010-12-021-52/+40
* - Reinstantiate the after_idle hook call in tcp_output(), which got lostlstewart2010-12-022-36/+20
* Set ssthresh appropriately on RTO. This change was accidentally not ported fromlstewart2010-12-021-0/+2
* Pass NULL instead of 0 for the th pointer value. NULL != 0 on all platforms.lstewart2010-12-021-1/+1
* Use time_uptime instead of non-monotonic time_second to drive ARPglebius2010-11-302-6/+6
* Fix more continuous/contiguous typos (cf. r215955)brucec2010-11-271-1/+1
* Adds new dtrace for cwnd functions and lay'srrs2010-11-255-3/+332
* Redo r166423. It is important not only skip freeing multicastglebius2010-11-241-14/+17
* After some off-list discussion, revert a number of changes to thedim2010-11-2222-110/+110
* Remove an apparently redundant CURVNET_SET() / CURVNET_RESTORE() pair.zec2010-11-221-2/+0
* Fix a minor code redundancy nit.lstewart2010-11-201-3/+1
* When enabling or disabling SIFTR with a VIMAGE kernel, ensure we add or removelstewart2010-11-201-12/+24
* Add new, per connection, statistics for TCP, including:gnn2010-11-175-3/+15
* Add an SCTP socket option to retrieve the number of timeoutstuexen2010-11-163-0/+35
* Make the CC framework more VIMAGE friendly by adding the machinery to allowlstewart2010-11-162-32/+34
* - Querying the default CC algo is more common than setting it and the functionlstewart2010-11-161-3/+2
* Move protocol specific implementation detail out of the core CC framework.lstewart2010-11-163-48/+70
* On CC algorithm module unload, we walk the list of active TCP control blocks.lstewart2010-11-161-24/+35
* cc_init() should only be run once on system boot, but with VIMAGE kernels itlstewart2010-11-163-5/+4
* Apply the STATIC_VNET_DEFINE and STATIC_DPCPU_DEFINE macros throughoutdim2010-11-1422-110/+110
OpenPOWER on IntegriCloud