| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Modulate tcp_now by a random amount on a per-connection basis. |   mcbride | 2004-10-28 | 1 | -2/+2 |
| * | set the congestion window to two segments (instead of only one), this matches |   markus | 2004-10-06 | 1 | -2/+2 |
| * | don't send partial segments if SS_ISSENDING is set, remember | markus | 2004-09-16 | 1 | -4/+12 |
| * | remove #ifdef TUBA |   itojun | 2004-06-20 | 1 | -12/+1 |
| * | factor out md5 code; ok+tests henning@, djm@, hshoexer@ | markus | 2004-06-08 | 1 | -68/+7 |
| * | set m_pkthdr.len early; ok mcbride, deraadt | markus | 2004-06-05 | 1 | -3/+2 |
| * | work around an LP64 problem where we report an excessively large window |   brad | 2004-05-31 | 1 | -3/+3 |
| * | Replace RSA-derived md5 code with code derived from Colin Plumb's PD version. |   millert | 2004-05-07 | 1 | -2/+2 |
| * | - allow the user to force the TCP mss below the fail-safe 216 with a low |   frantzen | 2004-04-26 | 1 | -2/+2 |
| * | don't allocate a cluster if the header fits into a mbuf; | markus | 2004-02-16 | 1 | -4/+4 |
| * | check TF_SIGNATURE when calculating the mss; add TCPOLEN_SIGLEN and | markus | 2004-02-10 | 1 | -3/+3 |
| * | !sack_disable -> sack_enable; ok deraadt@ | markus | 2004-01-31 | 1 | -11/+12 |
| * | add gettdbbysrcdst(), just like gettdb(), but compares tdb_src as well; ok mcbride@ | markus | 2004-01-22 | 1 | -11/+18 |
| * | es tanzt das KNF | markus | 2004-01-15 | 1 | -3/+3 |
| * | syncache+ipv6 support for TCP_SIGNATURE; with itojun; ok deraadt | markus | 2004-01-14 | 1 | -12/+28 |
| * | de-register. deraadt ok | itojun | 2003-12-10 | 1 | -6/+6 |
| * | do not flip ip_len/ip_off in netinet stack. deraadt ok. | itojun | 2003-07-09 | 1 | -2/+2 |
| * | backout following: | itojun | 2003-06-09 | 1 | -15/+4 |
| * | Remove the advertising clause in the UCB license which Berkeley | millert | 2003-06-02 | 1 | -6/+2 |
| * | use m_pulldown not m_pullup2. fix some bugs in IPv6 tcp_trace(). | itojun | 2003-05-29 | 1 | -4/+15 |
| * | don't send more than half of the send buffer space limit in | markus | 2003-01-25 | 1 | -5/+14 |
| * | Fix a problem where passing NULL as a pointer with varargs does not promote |   pefo | 2002-08-28 | 1 | -2/+2 |
| * | whitespace | itojun | 2002-06-09 | 1 | -34/+34 |
| * | avoid is_ipv6 construct. a step towards IPv4-less kernel | itojun | 2002-06-07 | 1 | -7/+7 |
| * | bring in ECN support from KAME. |   kjc | 2002-05-16 | 1 | -1/+62 |
| * | use timeout(9) to schedule TCP timers. this avoid traversing all |   provos | 2002-03-08 | 1 | -4/+5 |
| * | remove tcp_fasttimo and convert delayed acks to the timeout(9) API instead. | provos | 2002-03-01 | 1 | -2/+10 |
| * | when retransmitting a segment after FIN has been sent don't set FIN | provos | 2002-02-05 | 1 | -2/+2 |
| * | knf | provos | 2002-01-14 | 1 | -13/+13 |
| * | use macros to manage tcp timers; based on netbsd | provos | 2002-01-14 | 1 | -25/+25 |
| * | KNF |   deraadt | 2001-11-24 | 1 | -3/+3 |
| * | Always defer output TCP checksumming until ip_output() (or hardware, |   angelos | 2001-06-25 | 1 | -43/+8 |
| * | Use in_cksum_phdr() rather than in_cksum() -- from NetBSD | angelos | 2001-06-25 | 1 | -2/+8 |
| * | Likewise, only use outgoing TCP/UDP hardware checksumming if the | angelos | 2001-06-23 | 1 | -2/+3 |
| * | Keep stats on TCP/UDP hardware checksumming. | angelos | 2001-06-23 | 1 | -1/+2 |
| * | TCP/UDP hardware checksumming. Untested, since txp dies when it tries | angelos | 2001-06-23 | 1 | -1/+18 |
| * | fix up mtu for routes and ongoing tcp connection when if mtu changes | provos | 2001-06-23 | 1 | -1/+11 |
| * | Cut down on include files. | angelos | 2001-06-08 | 1 | -4/+1 |
| * | repair copyright notices for NRL & cmetz; cmetz | deraadt | 2001-06-05 | 1 | -14/+38 |
| * | Two fixes from Stevens via davidg@freebsd, bug report by | provos | 2001-05-31 | 1 | -5/+8 |
| * | on expiry of pmtu route, retry higher mtu. okay angelos@ | provos | 2000-09-25 | 1 | -3/+2 |
| * | correctly calculate mss | provos | 2000-09-20 | 1 | -2/+6 |
| * | Path MTU discovery based on NetBSD but with the decision to use the DF | provos | 2000-09-18 | 1 | -2/+4 |
| * | various fixes to SACK and FACK from adesai@cisco.com, tomh@tomh.org and | provos | 2000-09-05 | 1 | -2/+22 |
| * | TCP SACK fixes via Tom Henderson (tomh@cs.berkeley.edu): | provos | 2000-02-21 | 1 | -21/+32 |
| * | s/memset/bzero/ | itojun | 2000-01-07 | 1 | -2/+2 |
| * | option TCP_NEWRENO goes away, its the default case for TCP_SACK if | provos | 1999-12-21 | 1 | -3/+3 |
| * | never go into persist mode if there are still segments to be retransmitted. | provos | 1999-12-15 | 1 | -1/+14 |
| * | sync DIAGNOSTIC code with reality. we always get cluster mbuf so | itojun | 1999-12-10 | 1 | -2/+2 |
| * | bring in KAME IPv6 code, dated 19991208. | itojun | 1999-12-08 | 1 | -49/+40 |
