| Commit message (Expand) | Author | Age | Files | Lines |
* | Add a new sack hint to track the most recent and highest sacked sequence number. | lstewart | 2010-12-28 | 1 | -0/+1 |
* | This commit marks the first formal contribution of the "Five New TCP Congestion | lstewart | 2010-11-12 | 1 | -1/+1 |
* | MFP4: @176978-176982, 176984, 176990-176994, 177441 | bz | 2010-04-29 | 1 | -9/+8 |
* | Merge the remainder of kern_vimage.c and vimage.h into vnet.c and | rwatson | 2009-08-01 | 1 | -1/+1 |
* | Remove unused VNET_SET() and related macros; only VNET_GET() is | rwatson | 2009-07-16 | 1 | -5/+5 |
* | Build on Jeff Roberson's linker-set based dynamic per-CPU allocator | rwatson | 2009-07-14 | 1 | -18/+19 |
* | Fix a race in the manipulation of the V_tcp_sack_globalholes global variable, | lstewart | 2009-07-13 | 1 | -2/+2 |
* | Update stats in struct tcpstat using two new macros, TCPSTAT_ADD() and | rwatson | 2009-04-11 | 1 | -1/+1 |
* | First pass at separating per-vnet initializer functions | zec | 2009-04-06 | 1 | -4/+3 |
* | Conditionally compile out V_ globals while instantiating the appropriate | zec | 2008-12-10 | 1 | -1/+0 |
* | Rather than using hidden includes (with cicular dependencies), | bz | 2008-12-02 | 1 | -0/+1 |
* | Change the initialization methodology for global variables scheduled | zec | 2008-11-19 | 1 | -4/+7 |
* | Remove endearing but syntactically unnecessary "return;" statements | rwatson | 2008-10-26 | 1 | -1/+0 |
* | Step 1.5 of importing the network stack virtualization infrastructure | zec | 2008-10-02 | 1 | -8/+10 |
* | Commit step 1 of the vimage project, (network stack) | bz | 2008-08-17 | 1 | -6/+7 |
* | Convert pcbinfo and inpcb mutexes to rwlocks, and modify macros to | rwatson | 2008-04-17 | 1 | -8/+8 |
* | Coalesce two identical UCB licenses into a single license instance with | rwatson | 2007-05-11 | 1 | -108/+87 |
* | Move universally to ANSI C function declarations, with relatively | rwatson | 2007-05-10 | 1 | -2/+1 |
* | Change the TCP timer system from using the callout system five times | andre | 2007-04-11 | 1 | -1/+1 |
* | Retire unused TCP_SACK_DEBUG. | andre | 2007-04-04 | 1 | -1/+0 |
* | In tcp_sack_doack() remove too tight KASSERT() added in last revision. This | andre | 2007-03-25 | 1 | -14/+16 |
* | Bring SACK option handling in tcp_dooptions() in line with all other | andre | 2007-03-23 | 1 | -0/+1 |
* | ANSIfy function declarations and remove register keywords for variables. | andre | 2007-03-21 | 1 | -4/+2 |
* | Make TCP_DROP_SYNFIN a standard part of TCP. Disabled by default it | andre | 2007-03-21 | 1 | -1/+0 |
* | Match up SYSCTL declaration style. | andre | 2007-03-19 | 1 | -3/+3 |
* | Eliminate debug code that catches bugs in the hinting of sack variables | mohans | 2006-04-06 | 1 | -13/+3 |
* | Certain (bad) values of sack blocks can end up corrupting the sack scoreboard. | mohans | 2006-04-05 | 1 | -0/+2 |
* | Remove unneeded includes and provide more accurate description | andre | 2006-02-18 | 1 | -2/+0 |
* | Fix for a bug that causes SACK scoreboard corruption when the limit | ps | 2005-11-21 | 1 | -6/+21 |
* | Remove unnecessary IPSEC includes. | andre | 2005-08-23 | 1 | -11/+0 |
* | - Postpone SACK option processing until after PAWS checks. SACK option | ps | 2005-06-27 | 1 | -72/+35 |
* | Fix for a bug in tcp_sack_option() causing crashes. | ps | 2005-06-23 | 1 | -2/+1 |
* | Fix a mis-merge. Remove a redundant call to tcp_sackhole_insert | ps | 2005-06-09 | 1 | -1/+0 |
* | Fix for a crash in tcp_sack_option() caused by hitting the limit on | ps | 2005-06-09 | 1 | -0/+5 |
* | Fix for a bug in the change that walks the scoreboard backwards from | ps | 2005-06-06 | 1 | -4/+9 |
* | Changes to tcp_sack_option() that | ps | 2005-06-04 | 1 | -90/+116 |
* | This is conform with the terminology in | ps | 2005-05-25 | 1 | -11/+11 |
* | Rewrite of tcp_sack_option(). Kentaro Kurahone (NetBSD) pointed out | ps | 2005-05-23 | 1 | -63/+108 |
* | Introduce routines to alloc/free sack holes. This cleans up the code | ps | 2005-05-16 | 1 | -56/+60 |
* | Fix for a bug where the "nexthole" sack hint is out of sync with the | ps | 2005-05-13 | 1 | -0/+2 |
* | When looking for the next hole to retransmit from the scoreboard, | ps | 2005-05-11 | 1 | -71/+111 |
* | - Make the sack scoreboard logic use the TAILQ macros. This improves | ps | 2005-04-21 | 1 | -75/+38 |
* | Rewrite of tcp_update_sack_list() to make it simpler and more readable | ps | 2005-04-18 | 1 | -74/+67 |
* | Fix for a TCP SACK bug where more than (win/2) bytes could have been | ps | 2005-04-14 | 1 | -1/+33 |
* | When the rightmost SACK block expands, rcv_lastsack should be updated. | ps | 2005-04-10 | 1 | -0/+2 |
* | Remove some unused sack fields. | ps | 2005-04-10 | 1 | -3/+0 |
* | Add limits on the number of elements in the sack scoreboard both | ps | 2005-03-09 | 1 | -3/+41 |
* | If the receiver sends an ack that is out of [snd_una, snd_max], | ps | 2005-02-27 | 1 | -0/+3 |
* | Remove 2 (SACK) fields from the tcpcb. These are only used by a | ps | 2005-02-17 | 1 | -7/+8 |
* | - Retransmit just one segment on initiation of SACK recovery. | ps | 2005-02-14 | 1 | -20/+16 |