Now that we've increased the size of our send / receive buffers, bursting
an entire window onto the network may cause congestion. As a result,
we will slow start beginning with a flightsize of 4 packets.
Problem reported by: Thomas Zenker <thz@Lennartz-electronic.de>
MFC after: 3 days
receiver was not sending an immediate ack with delayed acks turned on
when the input buffer is drained, preventing the transmitter from
restarting immediately.
Propogate the TCP_NODELAY option to accept()ed sockets. (Helps tbench and
is a good idea anyway).
Some cleanup. Identify additonal issues in comments.
MFC after: 1 day
to send all its data, especially when the data is less than one MSS.
This fixes an issue where the stack was delaying the sending
of data, eventhough there was enough window to send all the data and
the sending of data was emptying the socket buffer.
Problem found by Yoshihiro Tsuchiya (tsuchiya@flab.fujitsu.co.jp)
Submitted by: Jayanth Vijayaraghavan
connection. The information contained in a tcptemp can be
reconstructed from a tcpcb when needed.
Previously, tcp templates required the allocation of one
mbuf per connection. On large systems, this change should
free up a large number of mbufs.
Reviewed by: bmilekic, jlemon, ru
MFC after: 2 weeks
This work was based on kame-20010528-freebsd43-snap.tgz and some
critical problem after the snap was out were fixed.
There are many many changes since last KAME merge.
TODO:
- The definitions of SADB_* in sys/net/pfkeyv2.h are still different
from RFC2407/IANA assignment because of binary compatibility
issue. It should be fixed under 5-CURRENT.
- ip6po_m member of struct ip6_pktopts is no longer used. But, it
is still there because of binary compatibility issue. It should
be removed under 5-CURRENT.
Reviewed by: itojun
Obtained from: KAME
MFC after: 3 weeks
other "system" header files.
Also help the deprecation of lockmgr.h by making it a sub-include of
sys/lock.h and removing sys/lockmgr.h form kernel .c files.
Sort sys/*.h includes where possible in affected files.
OK'ed by: bde (with reservations)
because it only takes a struct tag which makes it impossible to
use unions, typedefs etc.
Define __offsetof() in <machine/ansi.h>
Define offsetof() in terms of __offsetof() in <stddef.h> and <sys/types.h>
Remove myriad of local offsetof() definitions.
Remove includes of <stddef.h> in kernel code.
NB: Kernelcode should *never* include from /usr/include !
Make <sys/queue.h> include <machine/ansi.h> to avoid polluting the API.
Deprecate <struct.h> with a warning. The warning turns into an error on
01-12-2000 and the file gets removed entirely on 01-01-2001.
Paritials reviews by: various.
Significant brucifications by: bde
It causes a panic when/if snd_una is incremented elsewhere (this
is a conservative change, because originally no rollback occurred
for any packets at all).
Submitted by: Vivek Sadananda Pai <vivek@imimic.com>
When this happens, we know for sure that the packet data was not
received by the peer. Therefore, back out any advancing of the
transmit sequence number so that we send the same data the next
time we transmit a packet, avoiding a guaranteed missed packet and
its resulting TCP transmit slowdown.
In most systems ip_output() probably never returns an error, and
so this problem is never seen. However, it is more likely to occur
with device drivers having short output queues (causing ENOBUFS to
be returned when they are full), not to mention low memory situations.
Moreover, because of this problem writers of slow devices were
required to make an unfortunate choice between (a) having a relatively
short output queue (with low latency but low TCP bandwidth because
of this problem) or (b) a long output queue (with high latency and
high TCP bandwidth). In my particular application (ISDN) it took
an output queue equal to ~5 seconds of transmission to avoid ENOBUFS.
A more reasonable output queue of 0.5 seconds resulted in only about
50% TCP throughput. With this patch full throughput was restored in
the latter case.
Reviewed by: freebsd-net
a mbuf, it may return without setting any timers. If no more data is
scheduled to be transmitted (this was a FIN) the system will sit in
LAST_ACK state forever.
Thus, when mbuf allocation fails, set the retransmit timer if neither
the retransmit or persist timer is already pending.
Problem discovered by: Mike Silbersack (silby@silby.com)
Pushed for a fix by: Bosko Milekic <bmilekic@dsuper.net>
Reviewed by: jayanth
better recovery for multiple packet losses in a single window.
The algorithm can be toggled via the sysctl net.inet.tcp.newreno,
which defaults to "on".
Submitted by: Jayanth Vijayaraghavan <jayanth@yahoo-inc.com>
at the same time.
When rfc1323 and rfc1644 option are enabled by sysctl,
and tcp over IPv6 is tried, kernel panic happens by the
following check in tcp_output(), because now hdrlen is bigger
in such case than before.
/*#ifdef DIAGNOSTIC*/
if (max_linkhdr + hdrlen > MHLEN)
panic("tcphdr too big");
/*#endif*/
So change the above check to compare with MCLBYTES in #ifdef INET6 case.
Also, allocate a mbuf cluster for the header mbuf, in that case.
Bug reported at KAME environment.
Approved by: jkh
Reviewed by: sumikawa
Obtained from: KAME project
-opt_ipsec.h was missing on some tcp files (sorry for basic mistake)
-made buildable as above fix
-also added some missing IPv4 mapped IPv6 addr consideration into
ipsec4_getpolicybysock
- eliminate the fast/slow timeout lists for TCP and instead use a
callout entry for each timer.
- increase the TCP timer granularity to HZ
- implement "bad retransmit" recovery, as presented in
"On Estimating End-to-End Network Path Properties", by Allman and Paxson.
Submitted by: jlemon, wollmann
(default 1) disables PMTUD globally. Although PMTUD can be disabled in
the standard case by locking the MTU on a static route (including the
default route), this method doesn't work in the face of dynamic routing
protocols like gated.
flag means that there is more data to be put into the socket buffer.
Use it in TCP to reduce the interaction between mbuf sizes and the
Nagle algorithm.
Based on: "Justin C. Walker" <justin@apple.com>'s description of Apple's
fix for this problem.
`len = min(so->so_snd.sb_cc, win) - off;'. min() has type u_int
and `off' has type int, so when min() is 0 and `off' is 1, the RHS
overflows to 0U - 1 = UINT_MAX. `len' has type long, so when
sizeof(long) == sizeof(int), the LHS normally overflows to to the
correct value of -1, but when sizeof(long) > sizeof(int), the LHS
is UINT_MAX.
Fixed some u_long's that should have been fixed-sized types.
of the TCP payload. See RFC1122 section 4.2.2.6 . This allows
Path MTU discovery to be used along with IP options.
PR: problem discovered by Kevin Lahey <kml@nas.nasa.gov>
represent in the TCP header. The old code did effectively:
win = min(win, MAX_ALLOWED);
win = max(win, what_i_think_i_advertised_last_time);
so if what_i_think_i_advertised_last_time is bigger than can be
represented in the header (e.g. large buffers and no window scaling)
then we stuff a too-big number into a short. This fix reverses the
order of the comparisons.
PR: kern/4712
cache lines. Removed the struct ip proto since only a couple of chars
were actually being used in it. Changed the order of compares in the
PCB hash lookup to take advantage of partial cache line fills (on PPro).
Discussed-with: wollman
This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.
Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.
gcc only inlines memcpy()'s whose count is constant and didn't inline
these. I want memcpy() in the kernel go away so that it's obvious that
it doesn't need to be optimized. Now it is only used for one struct
copy in si.c.
1) Set the persist timer to help time-out connections in the CLOSING state.
2) Honor the keep-alive timer in the CLOSING state.
This fixes problems with connections getting "stuck" due to incompletion
of the final connection shutdown which can be a BIG problem on busy WWW
servers.
have to decide whether to send a CC or CCnew option in our SYN segment
depending on the contents of our TAO cache. This decision has to be
made once when the connection starts. The earlier code delayed this
decision until the segment was assembled in tcp_output() and
retransmitted SYN segments could have different CC options.
Reviewed by: Richard Stevens, davidg, wollman
a few new wrinkles for MTU discovery which tcp_output() had better
be prepared to handle. ip_output() is also modified to do something
helpful in this case, since it has already calculated the information
we need.
and gated on `options MTUDISC' in the source. It is also practically
untested becausse (sniff!) I don't have easy access to a network with
an MTU of less than an Ethernet. If you have a small MTU network,
please try it and tell me if it works!
to be sent, just clean up and return ENOBUFS rather than silently
proceeding without sending any of the data. This makes it consistent
with the `#ifdef notyet' case immediately above.
Reviewed by: Andras Olah <olah@freebsd.org>
Obtained from: Lite-2
