tcp: fix TSO FACK loss marking in tcp_mark_head_lost
authorYuchung Cheng <ycheng@google.com>
Fri, 24 Sep 2010 13:22:06 +0000 (13:22 +0000)
committerDavid S. Miller <davem@davemloft.net>
Mon, 27 Sep 2010 21:55:57 +0000 (14:55 -0700)
commitb3de7559afbb7a8a35b4be975a6adf6c5e3cdca0
treeb387ee28cbb0c577943e87ddc7d87c3d7c65b99c
parent3fd6c88ef875a14740801ebfc6b6e4e064a1cdd4
tcp: fix TSO FACK loss marking in tcp_mark_head_lost

When TCP uses FACK algorithm to mark lost packets in
tcp_mark_head_lost(), if the number of packets in the (TSO) skb is
greater than the number of packets that should be marked lost, TCP
incorrectly exits the loop and marks no packets lost in the skb. This
underestimates tp->lost_out and affects the recovery/retransmission.
This patch fargments the skb and marks the correct amount of packets
lost.

Signed-off-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv4/tcp_input.c