linux-stable.git/net/ipv4, branch v4.19.166

netfilter: x_tables: Switch synchronization to RCU

2020-12-30T10:25:45+00:00

[ Upstream commit cc00bcaa589914096edef7fb87ca5cee4a166b5c ]

When running concurrent iptables rules replacement with data, the per CPU
sequence count is checked after the assignment of the new information.
The sequence count is used to synchronize with the packet path without the
use of any explicit locking. If there are any packets in the packet path using
the table information, the sequence count is incremented to an odd value and
is incremented to an even after the packet process completion.

The new table value assignment is followed by a write memory barrier so every
CPU should see the latest value. If the packet path has started with the old
table information, the sequence counter will be odd and the iptables
replacement will wait till the sequence count is even prior to freeing the
old table info.

However, this assumes that the new table information assignment and the memory
barrier is actually executed prior to the counter check in the replacement
thread. If CPU decides to execute the assignment later as there is no user of
the table information prior to the sequence check, the packet path in another
CPU may use the old table information. The replacement thread would then free
the table information under it leading to a use after free in the packet
processing context-

Unable to handle kernel NULL pointer dereference at virtual
address 000000000000008e
pc : ip6t_do_table+0x5d0/0x89c
lr : ip6t_do_table+0x5b8/0x89c
ip6t_do_table+0x5d0/0x89c
ip6table_filter_hook+0x24/0x30
nf_hook_slow+0x84/0x120
ip6_input+0x74/0xe0
ip6_rcv_finish+0x7c/0x128
ipv6_rcv+0xac/0xe4
__netif_receive_skb+0x84/0x17c
process_backlog+0x15c/0x1b8
napi_poll+0x88/0x284
net_rx_action+0xbc/0x23c
__do_softirq+0x20c/0x48c

This could be fixed by forcing instruction order after the new table
information assignment or by switching to RCU for the synchronization.

Fixes: 80055dab5de0 ("netfilter: x_tables: make xt_replace_table wait until old rules are not used anymore")
Reported-by: Sean Tranchetti 
Reported-by: kernel test robot 
Suggested-by: Florian Westphal 
Signed-off-by: Subash Abhinov Kasiviswanathan 
Signed-off-by: Pablo Neira Ayuso 
Signed-off-by: Sasha Levin

tcp: fix cwnd-limited bug for TSO deferral where we send nothing

2020-12-30T10:25:41+00:00

[ Upstream commit 299bcb55ecd1412f6df606e9dc0912d55610029e ]

When cwnd is not a multiple of the TSO skb size of N*MSS, we can get
into persistent scenarios where we have the following sequence:

(1) ACK for full-sized skb of N*MSS arrives
  -> tcp_write_xmit() transmit full-sized skb with N*MSS
  -> move pacing release time forward
  -> exit tcp_write_xmit() because pacing time is in the future

(2) TSQ callback or TCP internal pacing timer fires
  -> try to transmit next skb, but TSO deferral finds remainder of
     available cwnd is not big enough to trigger an immediate send
     now, so we defer sending until the next ACK.

(3) repeat...

So we can get into a case where we never mark ourselves as
cwnd-limited for many seconds at a time, even with
bulk/infinite-backlog senders, because:

o In case (1) above, every time in tcp_write_xmit() we have enough
cwnd to send a full-sized skb, we are not fully using the cwnd
(because cwnd is not a multiple of the TSO skb size). So every time we
send data, we are not cwnd limited, and so in the cwnd-limited
tracking code in tcp_cwnd_validate() we mark ourselves as not
cwnd-limited.

o In case (2) above, every time in tcp_write_xmit() that we try to
transmit the "remainder" of the cwnd but defer, we set the local
variable is_cwnd_limited to true, but we do not send any packets, so
sent_pkts is zero, so we don't call the cwnd-limited logic to update
tp->is_cwnd_limited.

Fixes: ca8a22634381 ("tcp: make cwnd-limited checks measurement-based, and gentler")
Reported-by: Ingemar Johansson 
Signed-off-by: Neal Cardwell 
Signed-off-by: Yuchung Cheng 
Acked-by: Soheil Hassas Yeganeh 
Signed-off-by: Eric Dumazet 
Link: https://lore.kernel.org/r/20201209035759.1225145-1-ncardwell.kernel@gmail.com
Signed-off-by: Jakub Kicinski 
Signed-off-by: Greg Kroah-Hartman

tcp: select sane initial rcvq_space.space for big MSS

2020-12-30T10:25:40+00:00

[ Upstream commit 72d05c00d7ecda85df29abd046da7e41cc071c17 ]

Before commit a337531b942b ("tcp: up initial rmem to 128KB and SYN rwin to around 64KB")
small tcp_rmem[1] values were overridden by tcp_fixup_rcvbuf() to accommodate various MSS.

This is no longer the case, and Hazem Mohamed Abuelfotoh reported
that DRS would not work for MTU 9000 endpoints receiving regular (1500 bytes) frames.

Root cause is that tcp_init_buffer_space() uses tp->rcv_wnd for upper limit
of rcvq_space.space computation, while it can select later a smaller
value for tp->rcv_ssthresh and tp->window_clamp.

ss -temoi on receiver would show :

skmem:(r0,rb131072,t0,tb46080,f0,w0,o0,bl0,d0) rcv_space:62496 rcv_ssthresh:56596

This means that TCP can not increase its window in tcp_grow_window(),
and that DRS can never kick.

Fix this by making sure that rcvq_space.space is not bigger than number of bytes
that can be held in TCP receive queue.

People unable/unwilling to change their kernel can work around this issue by
selecting a bigger tcp_rmem[1] value as in :

echo "4096 196608 6291456" >/proc/sys/net/ipv4/tcp_rmem

Based on an initial report and patch from Hazem Mohamed Abuelfotoh
 https://lore.kernel.org/netdev/20201204180622.14285-1-abuehaze@amazon.com/

Fixes: a337531b942b ("tcp: up initial rmem to 128KB and SYN rwin to around 64KB")
Fixes: 041a14d26715 ("tcp: start receiver buffer autotuning sooner")
Reported-by: Hazem Mohamed Abuelfotoh 
Signed-off-by: Eric Dumazet 
Acked-by: Soheil Hassas Yeganeh 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

ipv4: Fix tos mask in inet_rtm_getroute()

2020-12-08T09:18:53+00:00

[ Upstream commit 1ebf179037cb46c19da3a9c1e2ca16e7a754b75e ]

When inet_rtm_getroute() was converted to use the RCU variants of
ip_route_input() and ip_route_output_key(), the TOS parameters
stopped being masked with IPTOS_RT_MASK before doing the route lookup.

As a result, "ip route get" can return a different route than what
would be used when sending real packets.

For example:

    $ ip route add 192.0.2.11/32 dev eth0
    $ ip route add unreachable 192.0.2.11/32 tos 2
    $ ip route get 192.0.2.11 tos 2
    RTNETLINK answers: No route to host

But, packets with TOS 2 (ECT(0) if interpreted as an ECN bit) would
actually be routed using the first route:

    $ ping -c 1 -Q 2 192.0.2.11
    PING 192.0.2.11 (192.0.2.11) 56(84) bytes of data.
    64 bytes from 192.0.2.11: icmp_seq=1 ttl=64 time=0.173 ms

    --- 192.0.2.11 ping statistics ---
    1 packets transmitted, 1 received, 0% packet loss, time 0ms
    rtt min/avg/max/mdev = 0.173/0.173/0.173/0.000 ms

This patch re-applies IPTOS_RT_MASK in inet_rtm_getroute(), to
return results consistent with real route lookups.

Fixes: 3765d35ed8b9 ("net: ipv4: Convert inet_rtm_getroute to rcu versions of route lookup")
Signed-off-by: Guillaume Nault 
Reviewed-by: David Ahern 
Link: https://lore.kernel.org/r/b2d237d08317ca55926add9654a48409ac1b8f5b.1606412894.git.gnault@redhat.com
Signed-off-by: Jakub Kicinski 
Signed-off-by: Greg Kroah-Hartman

tcp: Set INET_ECN_xmit configuration in tcp_reinit_congestion_control

2020-12-08T09:18:52+00:00

[ Upstream commit 55472017a4219ca965a957584affdb17549ae4a4 ]

When setting congestion control via a BPF program it is seen that the
SYN/ACK for packets within a given flow will not include the ECT0 flag. A
bit of simple printk debugging shows that when this is configured without
BPF we will see the value INET_ECN_xmit value initialized in
tcp_assign_congestion_control however when we configure this via BPF the
socket is in the closed state and as such it isn't configured, and I do not
see it being initialized when we transition the socket into the listen
state. The result of this is that the ECT0 bit is configured based on
whatever the default state is for the socket.

Any easy way to reproduce this is to monitor the following with tcpdump:
tools/testing/selftests/bpf/test_progs -t bpf_tcp_ca

Without this patch the SYN/ACK will follow whatever the default is. If dctcp
all SYN/ACK packets will have the ECT0 bit set, and if it is not then ECT0
will be cleared on all SYN/ACK packets. With this patch applied the SYN/ACK
bit matches the value seen on the other packets in the given stream.

Fixes: 91b5b21c7c16 ("bpf: Add support for changing congestion control")
Signed-off-by: Alexander Duyck 
Signed-off-by: Jakub Kicinski 
Signed-off-by: Greg Kroah-Hartman

tcp: only postpone PROBE_RTT if RTT is < current min_rtt estimate

2020-11-24T12:27:18+00:00

[ Upstream commit 1b9e2a8c99a5c021041bfb2d512dc3ed92a94ffd ]

During loss recovery, retransmitted packets are forced to use TCP
timestamps to calculate the RTT samples, which have a millisecond
granularity. BBR is designed using a microsecond granularity. As a
result, multiple RTT samples could be truncated to the same RTT value
during loss recovery. This is problematic, as BBR will not enter
PROBE_RTT if the RTT sample is <= the current min_rtt sample, meaning
that if there are persistent losses, PROBE_RTT will constantly be
pushed off and potentially never re-entered. This patch makes sure
that BBR enters PROBE_RTT by checking if RTT sample is < the current
min_rtt sample, rather than <=.

The Netflix transport/TCP team discovered this bug in the Linux TCP
BBR code during lab tests.

Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control")
Signed-off-by: Ryan Sharpelletti 
Signed-off-by: Neal Cardwell 
Signed-off-by: Soheil Hassas Yeganeh 
Signed-off-by: Yuchung Cheng 
Link: https://lore.kernel.org/r/20201116174412.1433277-1-sharpelletti.kdev@gmail.com
Signed-off-by: Jakub Kicinski 
Signed-off-by: Greg Kroah-Hartman

inet_diag: Fix error path to cancel the meseage in inet_req_diag_fill()

2020-11-24T12:27:16+00:00

[ Upstream commit e33de7c5317e2827b2ba6fd120a505e9eb727b05 ]

nlmsg_cancel() needs to be called in the error path of
inet_req_diag_fill to cancel the message.

Fixes: d545caca827b ("net: inet: diag: expose the socket mark to privileged processes.")
Reported-by: Hulk Robot 
Signed-off-by: Wang Hai 
Link: https://lore.kernel.org/r/20201116082018.16496-1-wanghai38@huawei.com
Signed-off-by: Jakub Kicinski 
Signed-off-by: Greg Kroah-Hartman

net: Update window_clamp if SOCK_RCVBUF is set

2020-11-18T18:18:51+00:00

[ Upstream commit 909172a149749242990a6e64cb55d55460d4e417 ]

When net.ipv4.tcp_syncookies=1 and syn flood is happened,
cookie_v4_check or cookie_v6_check tries to redo what
tcp_v4_send_synack or tcp_v6_send_synack did,
rsk_window_clamp will be changed if SOCK_RCVBUF is set,
which will make rcv_wscale is different, the client
still operates with initial window scale and can overshot
granted window, the client use the initial scale but local
server use new scale to advertise window value, and session
work abnormally.

Fixes: e88c64f0a425 ("tcp: allow effective reduction of TCP's rcv-buffer via setsockopt")
Signed-off-by: Mao Wenan 
Signed-off-by: Eric Dumazet 
Link: https://lore.kernel.org/r/1604967391-123737-1-git-send-email-wenan.mao@linux.alibaba.com
Signed-off-by: Jakub Kicinski 
Signed-off-by: Greg Kroah-Hartman

netfilter: use actual socket sk rather than skb sk when routing harder

2020-11-18T18:18:44+00:00

commit 46d6c5ae953cc0be38efd0e469284df7c4328cf8 upstream.

If netfilter changes the packet mark when mangling, the packet is
rerouted using the route_me_harder set of functions. Prior to this
commit, there's one big difference between route_me_harder and the
ordinary initial routing functions, described in the comment above
__ip_queue_xmit():

   /* Note: skb->sk can be different from sk, in case of tunnels */
   int __ip_queue_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl,

That function goes on to correctly make use of sk->sk_bound_dev_if,
rather than skb->sk->sk_bound_dev_if. And indeed the comment is true: a
tunnel will receive a packet in ndo_start_xmit with an initial skb->sk.
It will make some transformations to that packet, and then it will send
the encapsulated packet out of a *new* socket. That new socket will
basically always have a different sk_bound_dev_if (otherwise there'd be
a routing loop). So for the purposes of routing the encapsulated packet,
the routing information as it pertains to the socket should come from
that socket's sk, rather than the packet's original skb->sk. For that
reason __ip_queue_xmit() and related functions all do the right thing.

One might argue that all tunnels should just call skb_orphan(skb) before
transmitting the encapsulated packet into the new socket. But tunnels do
*not* do this -- and this is wisely avoided in skb_scrub_packet() too --
because features like TSQ rely on skb->destructor() being called when
that buffer space is truely available again. Calling skb_orphan(skb) too
early would result in buffers filling up unnecessarily and accounting
info being all wrong. Instead, additional routing must take into account
the new sk, just as __ip_queue_xmit() notes.

So, this commit addresses the problem by fishing the correct sk out of
state->sk -- it's already set properly in the call to nf_hook() in
__ip_local_out(), which receives the sk as part of its normal
functionality. So we make sure to plumb state->sk through the various
route_me_harder functions, and then make correct use of it following the
example of __ip_queue_xmit().

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Jason A. Donenfeld 
Reviewed-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso 
Signed-off-by: Sasha Levin 
[Jason: backported to 4.19 from Sasha's 5.4 backport]
Signed-off-by: Jason A. Donenfeld 
Signed-off-by: Greg Kroah-Hartman

tcp: Prevent low rmem stalls with SO_RCVLOWAT.

2020-11-05T10:08:33+00:00

[ Upstream commit 435ccfa894e35e3d4a1799e6ac030e48a7b69ef5 ]

With SO_RCVLOWAT, under memory pressure,
it is possible to enter a state where:

1. We have not received enough bytes to satisfy SO_RCVLOWAT.
2. We have not entered buffer pressure (see tcp_rmem_pressure()).
3. But, we do not have enough buffer space to accept more packets.

In this case, we advertise 0 rwnd (due to #3) but the application does
not drain the receive queue (no wakeup because of #1 and #2) so the
flow stalls.

Modify the heuristic for SO_RCVLOWAT so that, if we are advertising
rwnd<=rcv_mss, force a wakeup to prevent a stall.

Without this patch, setting tcp_rmem to 6143 and disabling TCP
autotune causes a stalled flow. With this patch, no stall occurs. This
is with RPC-style traffic with large messages.

Fixes: 03f45c883c6f ("tcp: avoid extra wakeups for SO_RCVLOWAT users")
Signed-off-by: Arjun Roy 
Acked-by: Soheil Hassas Yeganeh 
Acked-by: Neal Cardwell 
Signed-off-by: Eric Dumazet 
Link: https://lore.kernel.org/r/20201023184709.217614-1-arjunroy.kdev@gmail.com
Signed-off-by: Jakub Kicinski 
Signed-off-by: Greg Kroah-Hartman