Commit Graph

71 Commits

Author SHA1 Message Date
thorpej 974aa74abd Use the pool allocator for ipqent structures. 1998-10-08 01:19:25 +00:00
matt bf4e491879 Fix boolean dyslexic test. Duh! 1998-10-06 00:41:13 +00:00
matt 8e8f38e0f2 Add a sysctl for newreno (default to off). 1998-10-06 00:20:44 +00:00
matt 25054b5cf7 Adapt the NEWRENO changes from the UCSB diffs of BSDI 3.0's TCP
to NetBSD.  Ignore the SACK & FACK stuff for now.
1998-10-04 21:33:52 +00:00
mycroft 4a000a54e6 Fix a typo (not mine) in a comment. 1998-09-19 04:34:34 +00:00
mycroft 04ef3bf88d If we're in LISTEN state and all of RST, SYN and ACK are clear, send a RST. 1998-09-19 04:32:51 +00:00
mouse b95116821c Create tcp.keepidle, tcp.keepintvl, tcp.keepcnt, tcp.slowhz sysctls. 1998-09-10 10:46:03 +00:00
thorpej 4dbfe05f1f Use an algorithm similar to that in tcp_notify() to determine if
syn_cache_unreach() should remove the entry, or just continue on.

Algorithm is to only remove the entry if we've had more than one unreach
error and have retransmitted 3 or more times.  This prevents the following
scenario, as noted in PR #5909 (PR from Ty Sarna, scenario from
Charles Hannum):

	* Host A sends a SYN.
	* Host A retransmits the SYN.
	* Host B gets the first SYN and sends a SYN-ACK.
	* Host B gets the second SYN and sends a SYN-ACK.
	* One of the SYN-ACK bounces with an
	  ICMP unreachable, causing the `SYN cache' entry to be
	  removed with no notification.
	* Host A receives the other SYN-ACK, sends an ACK, and goes to
	  ESTABLISHED state.

Should fix PR #5909.
1998-09-09 01:32:27 +00:00
thorpej d319e4b419 Use the pool allocator for syn_cache entries. 1998-08-02 00:35:51 +00:00
thorpej a3f4316cba Clarify that we are using the Loss Window if a retransmission occurred
during the three-way handshake.
1998-07-17 22:58:56 +00:00
thorpej b22946827d Add a comment explaining why we do _not_ ACK data that might accompany
a SYN (avoidance of a DoS attack).
1998-06-02 18:33:02 +00:00
thorpej 5596fe2614 Nuke TUBA per my note to tech-net; there's no reason to keep it around. 1998-05-11 19:57:23 +00:00
thorpej ce3d776874 Rework the syn cache code somewhat:
- Don't use home-grown queue manipulation.  Use <sys/queue.h> instead.  The
  data structures are a little larger, but we are otherwise wasting the
  memory chunk anyway (we're already a 64-byte malloc bucket).
- Fix a bug in the cache-is-full case: if the oldest element removed from
  the first non-empty bucket was the only element in the bucket, the
  bucket wouldn't be removed from the bucket cache, causing queue corruption
  later.
- Optimize the syn cache timers by using PRT timers rather than home-grown
  decrement-and-propagate timers.

This code is now a fair bit smaller, and significantly easier to read
and understand.
1998-05-07 01:37:27 +00:00
thorpej 1ffa60ac01 Use macros from tcp_timer.h to manipulate TCP timers, so that their
implementation can be changed easily.
1998-05-06 01:21:20 +00:00
thorpej e44c4fb7d3 Once again, move a declaration for the benefit of TUBA (grumble). 1998-05-03 19:54:56 +00:00
thorpej b9fc258065 Oops, move a variable declaration so TUBA won't lose. 1998-05-02 04:23:05 +00:00
thorpej b71e4ddf4c Reintroduce the immediate ACK-on-PUSH behavior removed in revision 1.47,
but make the decision to do this dependent on the sysctl variable
net.inet.tcp.ack_on_push, which is disabled by default.
1998-05-02 04:21:58 +00:00
matt 334f006538 New TCP reassembly code. The new code reduces the memory needed by
out-of-order packets and builds the infrastructure needed for sending
SACK blocks (to be added shortly).
1998-04-29 20:43:29 +00:00
thorpej 100bfaf39a Change RFC1323 timestamp update rule per Section 3.4 of RFC1323.bis. Old
rule was to update the timestamp if the sequence numbers are in range.  New
rule adds a check that the timestamp is advancing, thus preventing our notion
of the most recent timestamp from incorrectly moving backwards.
1998-04-29 00:43:46 +00:00
thorpej df750b93da Log the peer's IP address on received window scale factors larger than
TCP_MAX_WINSHIFT (14), as recommended in Section 2.3 of RFC1323.
1998-04-28 21:52:16 +00:00
kml fcf0227962 Fix to ensure that the correct MSS is advertised for loopback
TCP connections by using the MTU of the interface.  Also added
a knob, mss_ifmtu, to force all connections to use the MTU of
the interface to calculate the advertised MSS.
1998-04-13 21:18:19 +00:00
thorpej 47b4697587 Remember any source routes that may have accompanied a SYN. 1998-04-07 05:09:19 +00:00
thorpej 04d3f25df8 Now that we have a flags word in the syn cache entry, use a flag to indicate
"peer will do timestamps" rather than a bitfield, and give the now-unsed
bit to the hash, making it now 32 bits.
1998-04-03 08:02:45 +00:00
thorpej b7c562b21c Clean up some comments wrt. the syn cache code. 1998-04-03 07:54:01 +00:00
thorpej 1b176d9395 Back out a change made some time ago, that would cause the NetBSD TCP
to ACK immediately any packet that arrived with PSH set.  This breaks
delayed ACKs in a few specific common cases that delayed ACKs were
supposed to help, and ends up not making much (if any) difference in
the case where where this ACK-on-PSH change was supposed to help.

Per discussion with several members of the TCPIMPL and TCPSAT IETF
working groups.
1998-03-31 23:44:09 +00:00
thorpej 2da6c91259 Fix a potential-congestion case in the larger initial congestion window
code, as clarified in the TCPIMPL WG meeting at IETF #41: If the SYN
(active open) or SYN,ACK (passive open) was retransmitted, the initial
congestion window for the first slow start of that connection must be
one segment.
1998-03-31 22:49:09 +00:00
kml 123232e156 Fix a retransmission bug introduced by the Brakmo and Peterson
RTO estimation changes.  Under some circumstances it would return a value
of 0, while the old Van Jacobson RTO code would return a minimum of 3.
This would result in 12 retransmissions, each 1 second apart.
This takes care of those instances, and ensures that t_rttmin is
used everywhere as a lower bound.
1998-03-19 22:29:33 +00:00
thorpej 5837cc6b07 Update copyright (sigh, should have done this long ago). 1998-02-19 02:36:42 +00:00
mellon 5685520ac1 Always set sc->sc_timeout (it was missed in one case). This fixes a problem where SYN cache entries are sometimes timed out almost immediately. 1998-01-24 12:27:31 +00:00
mycroft 5ab55e91b7 Fix an old editing error from merging a bug fix into Lite,
that might cause us to erroneously drop a FIN.
Also, minor changes so the code looks more like Stevens vol 2 figure 28.30.
1998-01-24 05:04:27 +00:00
mellon babb710a0b Never free the mbuf that we give to tcp_respond(). The previous change corrected an inconsistency but in exactly the wrong way. 1998-01-21 01:21:22 +00:00
mellon ac489008ad In syn_cache_get(), don't free incoming packet before jumping to resetandabort, but do free it after sending the reset. 1998-01-18 05:56:15 +00:00
thorpej e5e283e02d Finishing merging 4.4BSD-Lite2 netinet. At this point, the only changes
left were SCCS IDs and Copyright dates.
1998-01-05 10:31:44 +00:00
thorpej 673fb149c6 Implement a queue for delayed ACK processing. This queue is used in
tcp_fasttimo() in lieu of scanning all open TCP connections.
1997-12-31 03:31:23 +00:00
thorpej 6c1840c05c Fix the "stretch ACK violation" bug documented in internet draft
draft-ietf-tcpimpl-prob-02.txt.  Also, fix another bug in the header
prediction case where an ACK would not be sent when it should be.
1997-12-11 06:33:29 +00:00
thorpej 9f18d18071 Slight change to the previous: just drop the packet in the self-connect
case.  Sending an RST to ourselves is a little silly, considering that
we'll just attempt to remove a non-existent compressed state entry and
then drop the packet anyway.
1997-11-21 06:41:54 +00:00
thorpej ebd40d53e1 In tcp_input(), if the PCB we lookup for an incoming packet is a listen
socket:
- If we received a SYN,ACK, send an RST.
- If we received a SYN, and the connection attempt appears to come from
  itself, send an RST, since it cannot possibly be valid.
1997-11-21 06:18:30 +00:00
kml 86275dc497 TCP MSS fixes to provide cleaner slow-start and recovery. 1997-11-08 02:35:22 +00:00
explorer 790e114732 Add hooks to use the kernel random system to generate TCP sequence numbers. 1997-10-10 01:51:07 +00:00
thorpej 4ed600dbd0 Fix several annoyances related to MSS handling in BSD TCP:
- Don't overload t_maxseg.  Previous behavior was to set it to the min
  of the peer's advertised MSS, our advertised MSS, and tcp_mssdflt
  (for non-local networks).  This breaks PMTU discovery running on
  either host.  Instead, remember the MSS we advertise, and use it
  as appropriate (in silly window avoidance).
- Per last bullet, split tcp_mss() into several functions for handling
  MSS (ours and peer's), and performing various tasks when a connection
  becomes ESTABLISHED.
- Introduce a new function, tcp_segsize(), which computes the max size
  for every segment transmitted in tcp_output().  This will eventually
  be used to hook in PMTU discovery.
1997-09-22 21:49:55 +00:00
thorpej 402ae96361 Garbage-collect some "extern"s. 1997-07-28 22:07:38 +00:00
thorpej e794123006 Fix a rather severe bug in handling of incoming SYNs for peer/port values
which happen to have a TCB in TIME_WAIT, where an mbuf which had been
advanced past the IP+TCP headers and TCP options would be reused as if
it had not been advanced.  Problem found by Juergen Hannken-Illjes, who
also suggested a work-around on which this fix is based.
1997-07-28 01:07:48 +00:00
thorpej efa8881dbe Pull SYN_cache_branch down into the main line. 1997-07-23 21:26:40 +00:00
thorpej e0acb98b94 Fix an old and obscure TCP bug, brought to my attention by Bill Fenner,
fixed in FreeBSD by John Polstra:

Fix a bug (apparently very old) that can cause a TCP connection to
be dropped when it has an unusual traffic pattern.  For full details
as well as a test case that demonstrates the failure, see the
referenced PR (FreeBSD's kern/3998).

   Under certain circumstances involving the persist state, it is
   possible for the receive side's tp->rcv_nxt to advance beyond its
   tp->rcv_adv.  This causes (tp->rcv_adv - tp->rcv_nxt) to become
   negative.  However, in the code affected by this fix, that difference
   was interpreted as an unsigned number by max().  Since it was
   negative, it was taken as a huge unsigned number.  The effect was
   to cause the receiver to believe that its receive window had negative
   size, thereby rejecting all received segments including ACKs.  As
   the test case shows, this led to fruitless retransmissions and
   eventually to a dropped connection.  Even connections using the
   loopback interface could be dropped.  The fix substitutes the signed
   imax() for the unsigned max() function.

Bill informs me that his research indicates this bug appeared in Reno.
1997-07-06 07:04:34 +00:00
mycroft 315bb1ab50 Fix RTT scaling problems introduced with Brakmo and Peterson changes. 1996-12-10 18:20:19 +00:00
mycroft 9bfa240a98 Hash unconnected PCBs. 1996-09-15 18:11:06 +00:00
mycroft da2cd7eb82 If we're in SYN-SENT or SYN-RECEIVED state, don't reset the keepalive
timer until we transition to ESTABLISHED state.  Suggested by TCP/IP
vol 3.
1996-09-10 23:26:05 +00:00
mycroft 62a6cce9ca Add in_nullhost() and in_hosteq() macros, to hide some protocol
details.  Also, fix a bug in TCP wrt SYN+URG packets.
1996-09-09 14:51:07 +00:00
christos 14d9cd33af netinet prototypes 1996-02-13 23:40:59 +00:00
mycroft 0a47ede0fe Ignore FIN if not yet connected. 1996-01-31 05:56:56 +00:00