| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Abstract inpcb drop logic, previously just setting of INP_DROPPED in TCP, | Robert Watson | 2006-04-25 | 1 | -3/+2 |
| * | Replace isn_mtx direct use with ISN_*() lock macros so that locking | Robert Watson | 2006-04-23 | 1 | -5/+9 |
| * | Introduce a new TCP mutex, isn_mtx, which protects the initial sequence | Robert Watson | 2006-04-22 | 1 | -3/+6 |
| * | Allow for nmbclusters and maxsockets to be increased via sysctl. | Paul Saab | 2006-04-21 | 1 | -0/+11 |
| * | Add a tunable net.inet.tcp.maxtcptw, that allows to set a limit | Gleb Smirnoff | 2006-04-04 | 1 | -1/+8 |
| * | Before dereferencing intotw() when INP_TIMEWAIT, check for inp_ppcb being | Robert Watson | 2006-04-04 | 1 | -5/+15 |
| * | In TCP notify routines, check inpcb for INP_TIMEWAIT and INP_DROPPED. | Robert Watson | 2006-04-03 | 1 | -66/+81 |
| * | Change inp_ppcb from caddr_t to void *, fix/remove associated related | Robert Watson | 2006-04-03 | 1 | -7/+10 |
| * | Style tweaks: convert to ANSI from K&R function prototypes. | Robert Watson | 2006-04-03 | 1 | -59/+26 |
| * | Update comment on tcp_close() for new world order. | Robert Watson | 2006-04-03 | 1 | -4/+2 |
| * | Fix up locking surrounding tcp_drop sysctl: in the new world order, we | Robert Watson | 2006-04-03 | 1 | -8/+8 |
| * | Properly handle an edge case previously not handled correctly: a | Robert Watson | 2006-04-01 | 1 | -20/+37 |
| * | Update TCP for infrastructural changes to the socket/pcb refcount model, | Robert Watson | 2006-04-01 | 1 | -33/+123 |
| * | Have TCP Inflight disable itself if the RTT is below a certain | Andre Oppermann | 2006-02-16 | 1 | -1/+7 |
| * | Retire MT_HEADER mbuf type and change its users to use MT_DATA. | Andre Oppermann | 2005-11-02 | 1 | -2/+2 |
| * | Unbreak the net.inet6.tcp6.getcred sysctl. | Philip Paeps | 2005-10-12 | 1 | -3/+3 |
| * | o Teach sysctl_drop() how to deal with the sockets in TIME_WAIT state. | Maxim Konovalov | 2005-10-02 | 1 | -1/+5 |
| * | In tcp_ctlinput() do not swap ip->ip_len a second time. It | Andre Oppermann | 2005-09-10 | 1 | -2/+3 |
| * | Use the correct mbuf type for MGET(). | Andre Oppermann | 2005-08-30 | 1 | -1/+1 |
| * | recover the line which was wrongly disappeared during scope cleanup. | Hajimu UMEMOTO | 2005-08-01 | 1 | -0/+1 |
| * | scope cleanup. with this change | Hajimu UMEMOTO | 2005-07-25 | 1 | -12/+10 |
| * | Remove no-op spl's and most comment references to spls, as TCP locking | Robert Watson | 2005-07-19 | 1 | -16/+4 |
| * | Fix for a bug in the change that defers sack option processing until | Paul Saab | 2005-07-01 | 1 | -2/+1 |
| * | Assert tcbinfo lock in tcp_drop() due to its call of tcp_close() | Robert Watson | 2005-06-01 | 1 | -0/+6 |
| * | Fix two issues which were missed in FreeBSD-SA-05:08.kmem. | Colin Percival | 2005-05-07 | 1 | -0/+2 |
| * | If we don't get a suggested MTU during path MTU discovery | Andre Oppermann | 2005-05-04 | 1 | -9/+20 |
| * | - Make the sack scoreboard logic use the TAILQ macros. This improves | Paul Saab | 2005-04-21 | 1 | -0/+2 |
| * | Move Path MTU discovery ICMP processing from icmp_input() to | Andre Oppermann | 2005-04-21 | 1 | -7/+36 |
| * | Ignore ICMP Source Quench messages for TCP sessions. Source Quench is | Andre Oppermann | 2005-04-21 | 1 | -24/+11 |
| * | - If the reassembly queue limit was reached or if we couldn't allocate | Paul Saab | 2005-04-10 | 1 | -0/+1 |
| * | Use NET_CALLOUT_MPSAFE macro. | Gleb Smirnoff | 2005-03-01 | 1 | -7/+5 |
| * | o Add handling of an IPv4-mapped IPv6 address. | Maxim Konovalov | 2005-02-14 | 1 | -0/+98 |
| * | teach scope of IPv6 address to net.inet6.tcp6.getcred. | Hajimu UMEMOTO | 2005-02-04 | 1 | -4/+10 |
| * | Update an additional reference to the rate of ISN tick callouts that was | Robert Watson | 2005-01-31 | 1 | -1/+1 |
| * | Have tcp_isn_tick() fire 100 times a second, rather than HZ times a | Robert Watson | 2005-01-30 | 1 | -1/+1 |
| * | /* -> /*- for license, minor formatting changes | Warner Losh | 2005-01-07 | 1 | -1/+1 |
| * | Attempt to consistently use () around return values in calls to | Robert Watson | 2004-12-23 | 1 | -18/+18 |
| * | Remove an XXXRW comment relating to whether or not the TCP timers are | Robert Watson | 2004-12-23 | 1 | -6/+1 |
| * | Assert inpcb lock in: | Robert Watson | 2004-12-05 | 1 | -1/+12 |
| * | tcp_timewait() performs multiple non-atomic reads on the tcptw | Robert Watson | 2004-11-23 | 1 | -0/+2 |
| * | Assert the inpcb lock in tcp_twstart(), which does both read-modify-write | Robert Watson | 2004-11-23 | 1 | -0/+10 |
| * | Assert inpcb lock in tcp_quench(), tcp_drop_syn_sent(), tcp_mtudisc(), | Robert Watson | 2004-11-23 | 1 | -0/+4 |
| * | Assert the tcbinfo write lock in tcp_new_isn(), as the tcbinfo lock | Robert Watson | 2004-11-23 | 1 | -4/+11 |
| * | support TCP-MD5(IPv4) in KAME-IPSEC, too. | SUZUKI Shinsuke | 2004-11-08 | 1 | -0/+1 |
| * | Remove RFC1644 T/TCP support from the TCP side of the network stack. | Andre Oppermann | 2004-11-02 | 1 | -39/+2 |
| * | Push acquisition of the accept mutex out of sofree() into the caller | Robert Watson | 2004-10-18 | 1 | -0/+1 |
| * | - Estimate the amount of data in flight in sack recovery and use it | Paul Saab | 2004-10-05 | 1 | -5/+0 |
| * | fix up socket/ip layer violation... don't assume/know that | John-Mark Gurney | 2004-09-05 | 1 | -1/+2 |
| * | For IPv6 access pointer to tcpcb only after we have checked it is valid. | Andre Oppermann | 2004-08-19 | 1 | -1/+4 |
| * | White space cleanup for netinet before branch: | Robert Watson | 2004-08-16 | 1 | -68/+68 |
