tcp: set timestamps for restored skb-s
authorAndrey Vagin <avagin@openvz.org>
Fri, 16 Aug 2013 15:04:36 +0000 (19:04 +0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sat, 14 Sep 2013 13:54:55 +0000 (06:54 -0700)
commitf3f905389f1aaae9e091a28d018c66e08c85eddd
tree6c32397fd5c7b1c99bdb256cd65c2ca60db7754f
parenta829a28873ec2f5daedd77ef91e430b260af1521
tcp: set timestamps for restored skb-s

[ Upstream commit 7ed5c5ae96d23da22de95e1c7a239537acd378b1 ]

When the repair mode is turned off, the write queue seqs are
updated so that the whole queue is considered to be 'already sent.

The "when" field must be set for such skb. It's used in tcp_rearm_rto
for example. If the "when" field isn't set, the retransmit timeout can
be calculated incorrectly and a tcp connected can stop for two minutes
(TCP_RTO_MAX).

Acked-by: Pavel Emelyanov <xemul@parallels.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Signed-off-by: Andrey Vagin <avagin@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/ipv4/tcp.c