net: clear skb->tstamp in forwarding paths
authorEric Dumazet <edumazet@google.com>
Fri, 14 Dec 2018 14:46:49 +0000 (06:46 -0800)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 9 Jan 2019 16:38:31 +0000 (17:38 +0100)
commit281731c8170accbec0f56e31b94bee62c0fcf2a4
treebc922b303864f699283ba43586ea59cf782d3918
parentec820972c5436e45f398904af1bc715b969c3d5d
net: clear skb->tstamp in forwarding paths

[ Upstream commit 8203e2d844d34af247a151d8ebd68553a6e91785 ]

Sergey reported that forwarding was no longer working
if fq packet scheduler was used.

This is caused by the recent switch to EDT model, since incoming
packets might have been timestamped by __net_timestamp()

__net_timestamp() uses ktime_get_real(), while fq expects packets
using CLOCK_MONOTONIC base.

The fix is to clear skb->tstamp in forwarding paths.

Fixes: 80b14dee2bea ("net: Add a new socket option for a future transmit time.")
Fixes: fb420d5d91c1 ("tcp/fq: move back to CLOCK_MONOTONIC")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Sergey Matyukevich <geomatsi@gmail.com>
Tested-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/ipv4/ip_forward.c
net/ipv6/ip6_output.c