tcp: fix tcp_trim_head() to adjust segment count with skb MSS
authorNeal Cardwell <ncardwell@google.com>
Sat, 28 Jan 2012 17:29:46 +0000 (17:29 +0000)
committerSimone Willett <swillett@nvidia.com>
Wed, 8 Feb 2012 22:31:34 +0000 (14:31 -0800)
commitc6d519412d7ef7ab7fb351b64421876b3b95d88d
tree6f7a3dd47b3a19f25f89d87aa4d3a0287dfb64d3
parentfd0ee4525f9cd149b960a1b33e86129eaede864c
tcp: fix tcp_trim_head() to adjust segment count with skb MSS

[ Upstream commit 5b35e1e6e9ca651e6b291c96d1106043c9af314a ]

This commit fixes tcp_trim_head() to recalculate the number of
segments in the skb with the skb's existing MSS, so trimming the head
causes the skb segment count to be monotonically non-increasing - it
should stay the same or go down, but not increase.

Previously tcp_trim_head() used the current MSS of the connection. But
if there was a decrease in MSS between original transmission and ACK
(e.g. due to PMTUD), this could cause tcp_trim_head() to
counter-intuitively increase the segment count when trimming bytes off
the head of an skb. This violated assumptions in tcp_tso_acked() that
tcp_trim_head() only decreases the packet count, so that packets_acked
in tcp_tso_acked() could underflow, leading tcp_clean_rtx_queue() to
pass u32 pkts_acked values as large as 0xffffffff to
ca_ops->pkts_acked().

As an aside, if tcp_trim_head() had really wanted the skb to reflect
the current MSS, it should have called tcp_set_skb_tso_segs()
unconditionally, since a decrease in MSS would mean that a
single-packet skb should now be sliced into multiple segments.

Signed-off-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Nandita Dukkipati <nanditad@google.com>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Change-Id: I8e7359bbd561e78d0d9be4c3f871b22114504c8b
Reviewed-on: http://git-master/r/79696
Reviewed-by: Automatic_Commit_Validation_User
net/ipv4/tcp_output.c