summaryrefslogtreecommitdiffstats
path: root/sys/netinet/tcp_usrreq.c
Commit message (Expand)AuthorAgeFilesLines
* tcp_usrreq: Free allocated buffer in relock casecem2016-04-261-2/+5
* Redo r294869. The array of counters for TCP states doesn't belong toglebius2016-03-151-1/+1
* Fix dtrace probes (introduced in 287759): debug__input was usedgnn2016-03-031-0/+1
* Rename netinet/tcp_cc.h to netinet/cc/cc.h.glebius2016-01-271-1/+1
* Fix issues with TCP_CONGESTION handling after r294540:glebius2016-01-271-16/+15
* Augment struct tcpstat with tcps_states[], which is used for book-keepingglebius2016-01-271-0/+1
* Provide new socket option TCP_CCALGOOPT, which stands for TCP congestionglebius2016-01-221-1/+27
* Refactor TCP_CONGESTION setsockopt handling:glebius2016-01-211-43/+39
* - Rename cc.h to more meaningful tcp_cc.h.glebius2016-01-211-1/+2
* Historically we have two fields in tcpcb to describe sender MSS: t_maxopd,glebius2016-01-071-6/+4
* Implementation of server-side TCP Fast Open (TFO) [RFC7413].pkelsey2015-12-241-0/+58
* First cut of the modularization of our TCP stack. Stillrrs2015-12-161-15/+90
* There are times when it would be really nice to have a record of the last fewhiren2015-10-141-0/+31
* Remove unnecessary tcp state transition call.hiren2015-09-151-1/+1
* dd DTrace probe points, translators and a corresponding scriptgnn2015-09-131-0/+17
* Fix a kernel assertion issue introduced with r286227:jch2015-08-081-3/+8
* Decompose TCP INP_INFO lock to increase short-lived TCP connections scalability:jch2015-08-031-22/+22
* Revert r265338, r271089 and r271123 as those changes do not handlepkelsey2015-07-291-1/+1
* In TCP, connect() can return incorrect error code EINVALjch2015-03-091-4/+12
* Merge from projects/sendfile:glebius2014-11-301-5/+41
* Merge from projects/sendfile: extend protocols API to supportglebius2014-11-301-2/+2
* Missed in r274421: use sbavail() instead of bare access to sb_cc.glebius2014-11-301-1/+1
* Fix a race condition in TCP timewait between tcp_tw_2msl_reuse() andjch2014-10-301-0/+15
* A connection in TIME_WAIT state before calling close() actually did notjch2014-10-121-1/+3
* Make in6_pcblookup_hash_locked and in6_pcbladdr static.ae2014-09-101-37/+4
* The FreeBSD-SA-14:08.tcp was a lesson on not doing acrobatics withglebius2014-05-041-1/+1
* Decrease lock contention within the TCP accept case by removinggnn2014-01-281-9/+0
* Make TCP_KEEP* socket options readable. At least PostgreSQL wantsglebius2013-11-081-0/+21
* The r48589 promised to remove implicit inclusion of if_var.h soon. Prepareglebius2013-10-261-0/+1
* Implement the ip, tcp, and udp DTrace providers. The probe definitions usemarkj2013-08-251-7/+7
* Add checks for SO_NO_OFFLOAD in a couple of places that I missed earliernp2013-01-261-0/+2
* There is no need to call into the TOE driver twice in pru_rcvd (tod_rcvdnp2013-01-251-0/+1
* Heed SO_NO_OFFLOAD.np2013-01-251-2/+5
* Fix bug in TCP_KEEPCNT setting, which slipped in in the last roundglebius2012-09-271-8/+14
* - Updated TOE support in the kernel.np2012-06-191-22/+53
* Add new socket options: TCP_KEEPINIT, TCP_KEEPIDLE, TCP_KEEPINTVL andglebius2012-02-051-3/+58
* Always release the inp lock before returning from tcp_detach.np2012-01-061-1/+3
* Move the tcp_sendspace and tcp_recvspace sysctl's fromandre2011-10-161-14/+0
* VNET virtualize tcp_sendspace/tcp_recvspace and change theandre2011-10-161-7/+10
* Update the comment and description of tcp_sendspace and tcp_recvspaceandre2011-10-161-5/+4
* Do not leak the pcbinfohash lock in the case where in6_pcbladdr() returnsrwatson2011-06-021-1/+1
* Decompose the current single inpcbinfo lock into two locks:rwatson2011-05-301-45/+51
* Make the TCP code compile without INET. Sort #includes and add #ifdef INETs.bz2011-04-301-13/+39
* When turning off TCP_NOPUSH, only call tcp_output() to immediately flushjhb2011-02-041-2/+3
* Remove duplicate printing of TF_NOPUSH in db_print_tflags().bz2011-01-291-4/+0
* Trim extra spaces before tabs.jhb2011-01-071-1/+1
* Add new, per connection, statistics for TCP, including:gnn2010-11-171-0/+3
* This commit marks the first formal contribution of the "Five New TCP Congestionlstewart2010-11-121-1/+61
* Remove the TCP inflight bandwidth limiter as announced in r211315andre2010-09-161-13/+9
* Add a comment to tcp_usr_accept() to indicate why it is we acquire therwatson2010-03-061-3/+9
OpenPOWER on IntegriCloud