<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel/git/stable/linux-stable.git/include/net/tcp.h, branch linux-2.6.37.y</title>
<subtitle>Linux kernel stable tree</subtitle>
<id>https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/atom?h=linux-2.6.37.y</id>
<link rel='self' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/atom?h=linux-2.6.37.y'/>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/'/>
<updated>2010-11-10T20:12:00+00:00</updated>
<entry>
<title>net: avoid limits overflow</title>
<updated>2010-11-10T20:12:00+00:00</updated>
<author>
<name>Eric Dumazet</name>
<email>eric.dumazet@gmail.com</email>
</author>
<published>2010-11-09T23:24:26+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=8d987e5c75107ca7515fa19e857cfa24aab6ec8f'/>
<id>urn:sha1:8d987e5c75107ca7515fa19e857cfa24aab6ec8f</id>
<content type='text'>
Robin Holt tried to boot a 16TB machine and found some limits were
reached : sysctl_tcp_mem[2], sysctl_udp_mem[2]

We can switch infrastructure to use long "instead" of "int", now
atomic_long_t primitives are available for free.

Signed-off-by: Eric Dumazet &lt;eric.dumazet@gmail.com&gt;
Reported-by: Robin Holt &lt;holt@sgi.com&gt;
Reviewed-by: Robin Holt &lt;holt@sgi.com&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tcp: tcp_enter_quickack_mode can be static</title>
<updated>2010-09-30T02:45:36+00:00</updated>
<author>
<name>stephen hemminger</name>
<email>shemminger@vyatta.com</email>
</author>
<published>2010-09-28T19:30:14+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=1b9f409293529da4630bfc5d6d8e7d7451a6ccb5'/>
<id>urn:sha1:1b9f409293529da4630bfc5d6d8e7d7451a6ccb5</id>
<content type='text'>
Function only used in tcp_input.c

Signed-off-by: Stephen Hemminger &lt;shemminger@vyatta.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6</title>
<updated>2010-09-27T08:03:03+00:00</updated>
<author>
<name>David S. Miller</name>
<email>davem@davemloft.net</email>
</author>
<published>2010-09-27T08:03:03+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=e40051d134f7ee95c8c1f7a3471e84eafc9ab326'/>
<id>urn:sha1:e40051d134f7ee95c8c1f7a3471e84eafc9ab326</id>
<content type='text'>
Conflicts:
	drivers/net/qlcnic/qlcnic_init.c
	net/ipv4/ip_output.c
</content>
</entry>
<entry>
<title>tcp: Prevent overzealous packetization by SWS logic.</title>
<updated>2010-09-15T19:01:44+00:00</updated>
<author>
<name>Alexey Kuznetsov</name>
<email>kuznet@ms2.inr.ac.ru</email>
</author>
<published>2010-09-15T17:27:52+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=01f83d69844d307be2aa6fea88b0e8fe5cbdb2f4'/>
<id>urn:sha1:01f83d69844d307be2aa6fea88b0e8fe5cbdb2f4</id>
<content type='text'>
If peer uses tiny MSS (say, 75 bytes) and similarly tiny advertised
window, the SWS logic will packetize to half the MSS unnecessarily.

This causes problems with some embedded devices.

However for large MSS devices we do want to half-MSS packetize
otherwise we never get enough packets into the pipe for things
like fast retransmit and recovery to work.

Be careful also to handle the case where MSS &gt; window, otherwise
we'll never send until the probe timer.

Reported-by: ツ Leandro Melo de Sales &lt;leandroal@gmail.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6</title>
<updated>2010-09-10T05:27:33+00:00</updated>
<author>
<name>David S. Miller</name>
<email>davem@davemloft.net</email>
</author>
<published>2010-09-10T05:27:33+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=e548833df83c3554229eff0672900bfe958b45fd'/>
<id>urn:sha1:e548833df83c3554229eff0672900bfe958b45fd</id>
<content type='text'>
Conflicts:
	net/mac80211/main.c
</content>
</entry>
<entry>
<title>TCP: update initial windows according to RFC 5681</title>
<updated>2010-08-30T20:50:44+00:00</updated>
<author>
<name>Gerrit Renker</name>
<email>gerrit@erg.abdn.ac.uk</email>
</author>
<published>2010-08-29T19:27:34+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=3d5b99ae82f8742e3bb1f8634fd11ac36ea19ee1'/>
<id>urn:sha1:3d5b99ae82f8742e3bb1f8634fd11ac36ea19ee1</id>
<content type='text'>
This updates the use of larger initial windows, as originally specified in
RFC 3390, to use the newer IW values specified in RFC 5681, section 3.1.

The changes made in RFC 5681 are:
 a) the setting now is more clearly specified in units of segments (as the
    comments  by John Heffner emphasized, this was not very clear in RFC 3390);
 b) for connections with 1095 &lt; SMSS &lt;= 2190 there is now a change:
    - RFC 3390 says that IW &lt;= 4380,
    - RFC 5681 says that IW = 3 * SMSS &lt;= 6570.

Since RFC 3390 is older and "only" proposed standard, whereas the newer RFC 5681
is already draft standard, it seems preferable to use the newer IW variant.

Signed-off-by: Gerrit Renker &lt;gerrit@erg.abdn.ac.uk&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tcp/dccp: Consolidate common code for RFC 3390 conversion</title>
<updated>2010-08-30T20:45:26+00:00</updated>
<author>
<name>Gerrit Renker</name>
<email>gerrit@erg.abdn.ac.uk</email>
</author>
<published>2010-08-29T19:23:12+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=22b71c8f4f3db8df92f5e7b081c265bc56c0bd2f'/>
<id>urn:sha1:22b71c8f4f3db8df92f5e7b081c265bc56c0bd2f</id>
<content type='text'>
This patch consolidates initial-window code common to TCP and CCID-2:
 * TCP uses RFC 3390 in a packet-oriented manner (tcp_input.c) and
 * CCID-2 uses RFC 3390 in packet-oriented manner (RFC 4341).

Signed-off-by: Gerrit Renker &lt;gerrit@erg.abdn.ac.uk&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tcp: Combat per-cpu skew in orphan tests.</title>
<updated>2010-08-25T09:27:49+00:00</updated>
<author>
<name>David S. Miller</name>
<email>davem@davemloft.net</email>
</author>
<published>2010-08-25T09:27:49+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=ad1af0fedba14f82b240a03fe20eb9b2fdbd0357'/>
<id>urn:sha1:ad1af0fedba14f82b240a03fe20eb9b2fdbd0357</id>
<content type='text'>
As reported by Anton Blanchard when we use
percpu_counter_read_positive() to make our orphan socket limit checks,
the check can be off by up to num_cpus_online() * batch (which is 32
by default) which on a 128 cpu machine can be as large as the default
orphan limit itself.

Fix this by doing the full expensive sum check if the optimized check
triggers.

Reported-by: Anton Blanchard &lt;anton@samba.org&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
Acked-by: Eric Dumazet &lt;eric.dumazet@gmail.com&gt;
</content>
</entry>
<entry>
<title>tcp: sizeof struct tcp_skb_cb is 44</title>
<updated>2010-07-16T04:41:00+00:00</updated>
<author>
<name>Eric Dumazet</name>
<email>eric.dumazet@gmail.com</email>
</author>
<published>2010-07-16T04:41:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=f86586fa48a94c7a9c7f36650a2e0e4a947881c9'/>
<id>urn:sha1:f86586fa48a94c7a9c7f36650a2e0e4a947881c9</id>
<content type='text'>
Correct comment stating sizeof(struct tcp_skb_cb) is 36 or 40, since its
44 bytes, since commit 951dbc8ac714b04 ([IPV6]: Move nextheader offset
to the IP6CB).

Signed-off-by: Eric Dumazet &lt;eric.dumazet@gmail.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>inet, inet6: make tcp_sendmsg() and tcp_sendpage() through inet_sendmsg() and inet_sendpage()</title>
<updated>2010-07-13T03:21:46+00:00</updated>
<author>
<name>Changli Gao</name>
<email>xiaosuo@gmail.com</email>
</author>
<published>2010-07-10T20:41:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux-stable.git/commit/?id=7ba42910073f8432934d61a6c08b1023c408fb62'/>
<id>urn:sha1:7ba42910073f8432934d61a6c08b1023c408fb62</id>
<content type='text'>
a new boolean flag no_autobind is added to structure proto to avoid the autobind
calls when the protocol is TCP. Then sock_rps_record_flow() is called int the
TCP's sendmsg() and sendpage() pathes.

Signed-off-by: Changli Gao &lt;xiaosuo@gmail.com&gt;
----
 include/net/inet_common.h |    4 ++++
 include/net/sock.h        |    1 +
 include/net/tcp.h         |    8 ++++----
 net/ipv4/af_inet.c        |   15 +++++++++------
 net/ipv4/tcp.c            |   11 +++++------
 net/ipv4/tcp_ipv4.c       |    3 +++
 net/ipv6/af_inet6.c       |    8 ++++----
 net/ipv6/tcp_ipv6.c       |    3 +++
 8 files changed, 33 insertions(+), 20 deletions(-)
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
</feed>
