Files
linux/include/net
Eric Dumazet 94caec8ce7 ipv4: tcp: get rid of ugly unicast_sock
[ Upstream commit bdbbb8527b ]

In commit be9f4a44e7 ("ipv4: tcp: remove per net tcp_sock")
I tried to address contention on a socket lock, but the solution
I chose was horrible :

commit 3a7c384ffd ("ipv4: tcp: unicast_sock should not land outside
of TCP stack") addressed a selinux regression.

commit 0980e56e50 ("ipv4: tcp: set unicast_sock uc_ttl to -1")
took care of another regression.

commit b5ec8eeac4 ("ipv4: fix ip_send_skb()") fixed another regression.

commit 811230cd85 ("tcp: ipv4: initialize unicast_sock sk_pacing_rate")
was another shot in the dark.

Really, just use a proper socket per cpu, and remove the skb_orphan()
call, to re-enable flow control.

This solves a serious problem with FQ packet scheduler when used in
hostile environments, as we do not want to allocate a flow structure
for every RST packet sent in response to a spoofed packet.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2015-02-26 17:50:12 -08:00
..
2013-12-28 17:02:46 -05:00
2012-07-19 10:43:03 -07:00
2013-11-07 19:28:58 -05:00
2013-12-06 07:24:39 +01:00
2013-01-27 22:46:33 -05:00
2013-10-08 23:19:24 -04:00
2014-08-14 09:38:23 +08:00
2014-05-31 13:20:33 -07:00
2015-02-26 17:50:12 -08:00
2008-06-11 21:00:38 -07:00
2005-04-16 15:20:36 -07:00
2005-04-16 15:20:36 -07:00
2014-01-03 20:56:48 -05:00
2009-01-06 11:38:15 -07:00
2014-10-05 14:52:20 -07:00
2014-08-14 09:38:23 +08:00
2012-08-06 13:40:47 -07:00
2014-01-21 18:05:04 -08:00
2012-05-16 15:17:08 -04:00