| Commit message (Expand) | Author | Age | Files | Lines |
* | Rename netinet/tcp_cc.h to netinet/cc/cc.h. | glebius | 2016-01-27 | 1 | -1/+1 |
* | Persist timers TCPTV_PERSMIN and TCPTV_PERSMAX are hardcoded with 5 seconds and | hiren | 2016-01-26 | 1 | -0/+8 |
* | - Rename cc.h to more meaningful tcp_cc.h. | glebius | 2016-01-21 | 1 | -1/+2 |
* | Historically we have two fields in tcpcb to describe sender MSS: t_maxopd, | glebius | 2016-01-07 | 1 | -13/+8 |
* | Implementation of server-side TCP Fast Open (TFO) [RFC7413]. | pkelsey | 2015-12-24 | 1 | -1/+2 |
* | First cut of the modularization of our TCP stack. Still | rrs | 2015-12-16 | 1 | -3/+18 |
* | This fixes several places where callout_stops return is examined. The | rrs | 2015-11-13 | 1 | -2/+2 |
* | Fix an unnecessarily aggressive behavior where mtu clamping begins on first | hiren | 2015-10-14 | 1 | -2/+10 |
* | dd DTrace probe points, translators and a corresponding script | gnn | 2015-09-13 | 1 | -0/+7 |
* | Put r284245 back in place: If at first this fix was seen as a temporary | jch | 2015-08-30 | 1 | -15/+39 |
* | Revert r284245: "Fix a callout race condition introduced in TCP | jch | 2015-08-24 | 1 | -39/+15 |
* | Make clear that TIME_WAIT timeout expiration is managed solely by | jch | 2015-08-18 | 1 | -4/+12 |
* | Decompose TCP INP_INFO lock to increase short-lived TCP connections scalability: | jch | 2015-08-03 | 1 | -30/+14 |
* | Fix a callout race condition introduced in TCP timers callouts with r281599. | jch | 2015-06-10 | 1 | -16/+39 |
* | Fix an old and well-documented use-after-free race condition in | jch | 2015-04-16 | 1 | -75/+93 |
* | Provide better debugging information in tcp_timer_activate() and | jch | 2015-04-02 | 1 | -2/+2 |
* | Use appropriate timeout_t* instead of void* in tcp_timer_activate() | jch | 2015-03-31 | 1 | -1/+1 |
* | Refactor / restructure the RSS code into generic, IPv4 and IPv6 specific | adrian | 2015-01-18 | 1 | -0/+1 |
* | Fix a race condition in TCP timewait between tcp_tw_2msl_reuse() and | jch | 2014-10-30 | 1 | -1/+1 |
* | Fix multiple incorrect SYSCTL arguments in the kernel: | hselasky | 2014-10-21 | 1 | -6/+6 |
* | Handle small file case with regards to plpmtud blackhole detection. | sbruno | 2014-10-13 | 1 | -1/+9 |
* | Implement PLPMTUD blackhole detection (RFC 4821), inspired by code | sbruno | 2014-10-07 | 1 | -0/+148 |
* | If we're doing RSS then ensure the TCP timer selection uses the multi-CPU | adrian | 2014-06-30 | 1 | -0/+4 |
* | When RSS is enabled and per cpu TCP timers are enabled, do an RSS | adrian | 2014-05-18 | 1 | -4/+49 |
* | Currently, the TCP slow timer can starve TCP input processing while it | jhb | 2014-04-10 | 1 | -3/+1 |
* | - Make callout(9) tickless, relying on eventtimers(4) as backend for | davide | 2013-03-04 | 1 | -7/+11 |
* | Don't drop options from the third retransmitted SYN by default. If the | jhb | 2013-01-09 | 1 | -1/+7 |
* | Make sure that tcp_timer_activate() correctly sees TCP_OFFLOAD (or not). | np | 2012-11-27 | 1 | -0/+1 |
* | If the user has closed the socket then drop a persisting connection | andre | 2012-10-28 | 1 | -0/+10 |
* | Update comment to reflect the change made in r242263. | andre | 2012-10-28 | 1 | -1/+1 |
* | Add SACK_PERMIT to the list of TCP options that are switched off after | andre | 2012-10-28 | 1 | -1/+1 |
* | When retransmitting SYN in TCPS_SYN_SENT state use TCPTV_RTOBASE, | andre | 2012-10-28 | 1 | -1/+1 |
* | Remove bogus 'else' in #ifdef that prevented the rttvar from being reset | andre | 2012-10-28 | 1 | -1/+0 |
* | When SYN or SYN/ACK had to be retransmitted RFC5681 requires us to | andre | 2012-10-28 | 1 | -1/+7 |
* | In tcp timers, check INP_DROPPED flag a little later, after | trociny | 2012-08-05 | 1 | -9/+38 |
* | - Updated TOE support in the kernel. | np | 2012-06-19 | 1 | -0/+5 |
* | Add new socket options: TCP_KEEPINIT, TCP_KEEPIDLE, TCP_KEEPINTVL and | glebius | 2012-02-05 | 1 | -10/+11 |
* | Make sure the inp wasn't dropped when rexmt let go of the inp and | np | 2011-10-12 | 1 | -0/+7 |
* | Decompose the current single inpcbinfo lock into two locks: | rwatson | 2011-05-30 | 1 | -1/+1 |
* | TCP reuses t_rxtshift to determine the backoff timer used for both the | jhb | 2011-04-29 | 1 | -1/+3 |
* | Trim extra spaces before tabs. | jhb | 2011-01-07 | 1 | -9/+9 |
* | Fix a typo in a comment. | jhb | 2010-12-21 | 1 | -1/+1 |
* | Pass NULL instead of 0 for the th pointer value. NULL != 0 on all platforms. | lstewart | 2010-12-02 | 1 | -1/+1 |
* | This commit marks the first formal contribution of the "Five New TCP Congestion | lstewart | 2010-11-12 | 1 | -38/+11 |
* | - spread tcp timer callout load evenly across cpus if net.inet.tcp.per_cpu_ti... | kmacy | 2010-03-20 | 1 | -10/+30 |
* | Locking the tcbinfo structure should not be necessary in tcp_timer_delack(), | rwatson | 2010-03-07 | 1 | -3/+0 |
* | Add the ability to see TCP timers via netstat -x. This can be a useful | silby | 2009-09-16 | 1 | -0/+21 |
* | Merge the remainder of kern_vimage.c and vimage.h into vnet.c and | rwatson | 2009-08-01 | 1 | -1/+1 |
* | Reimplement and/or implement vnet list locking by replacing a mostly | rwatson | 2009-07-19 | 1 | -2/+2 |
* | Build on Jeff Roberson's linker-set based dynamic per-CPU allocator | rwatson | 2009-07-14 | 1 | -7/+0 |