If mtu probing is enabled tcp_mtu_probing() could very well end up
with a too small MSS.

Use the new sysctl tcp_min_snd_mss to make sure MSS search
is performed in an acceptable range.

CVE-2019-11479 -- tcp mss hardcoded to 48
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reported-by: NJonathan Lemon <jonathan.lemon@gmail.com>
Cc: Jonathan Looney <jtl@netflix.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: Tyler Hicks <tyhicks@canonical.com>
Cc: Bruce Curtis <brucec@netflix.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add SPDX license identifiers to all files which:

 - Have no license information of any form

 - Have EXPORT_.*_SYMBOL_GPL inside which was used in the
   initial scan/conversion to ignore the file

These files fall under the project license, GPL v2 only. The resulting SPDX
license identifier is:

  GPL-2.0-only
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

TCP sender would use congestion window of 1 packet on the second SYN
and SYNACK timeout except passive TCP Fast Open. This makes passive
TFO too aggressive and unfair during congestion at handshake. This
patch fixes this issue so TCP (fast open or not, passive or active)
always conforms to the RFC6298.

Note that tcp_enter_loss() is called only once during recurring
timeouts.  This is because during handshake, high_seq and snd_una
are the same so tcp_enter_loss() would incorrect set the undo state
variables multiple times.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Instead of using pingpong as a single bit information, we refactor the
code to treat it as a counter. When interactive session is detected,
we set pingpong count to TCP_PINGPONG_THRESH. And when pingpong count
is >= TCP_PINGPONG_THRESH, we consider the session in pingpong mode.

This patch is a pure refactor and sets foundation for the next patch.
This patch itself does not change any pingpong logic.
Signed-off-by: NWei Wang <weiwan@google.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Previously when the sender fails to retransmit a data packet on
timeout due to congestion in the local host (e.g. throttling in
qdisc), it'll retry within an RTO up to 500ms.

In low-RTT networks such as data-centers, RTO is often far
below the default minimum 200ms (and the cap 500ms). Then local
host congestion could trigger a retry storm pouring gas to the
fire. Worse yet, the retry counter (icsk_retransmits) is not
properly updated so the aggressive retry may exceed the system
limit (15 rounds) until the packet finally slips through.

On such rare events, it's wise to retry more conservatively (500ms)
and update the stats properly to reflect these incidents and follow
the system limit. Note that this is consistent with the behavior
when a keep-alive probe is dropped due to local congestion.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NNeal Cardwell <ncardwell@google.com>
Reviewed-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Previously we use the next unsent skb's timestamp to determine
when to abort a socket stalling on window probes. This no longer
works as skb timestamp reflects the last instead of the first
transmission.

Instead we can estimate how long the socket has been stalling
with the probe count and the exponential backoff behavior.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NNeal Cardwell <ncardwell@google.com>
Reviewed-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Create a helper to model TCP exponential backoff for the next patch.
This is pure refactor w no behavior change.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NNeal Cardwell <ncardwell@google.com>
Reviewed-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch addresses a corner issue on timeout behavior of a
passive Fast Open socket.  A passive Fast Open server may write
and close the socket when it is re-trying SYN-ACK to complete
the handshake. After the handshake is completely, the server does
not properly stamp the recovery start time (tp->retrans_stamp is
0), and the socket may abort immediately on the very first FIN
timeout, instead of retying until it passes the system or user
specified limit.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NNeal Cardwell <ncardwell@google.com>
Reviewed-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Previously TCP socket's retrans_stamp is not set if the
retransmission has failed to send. As a result if a socket is
experiencing local issues to retransmit packets, determining when
to abort a socket is complicated w/o knowning the starting time of
the recovery since retrans_stamp may remain zero.

This complication causes sub-optimal behavior that TCP may use the
latest, instead of the first, retransmission time to compute the
elapsed time of a stalling connection due to local issues. Then TCP
may disrecard TCP retries settings and keep retrying until it finally
succeed: not a good idea when the local host is already strained.

The simple fix is to always timestamp the start of a recovery.
It's worth noting that retrans_stamp is also used to compare echo
timestamp values to detect spurious recovery. This patch does
not break that because retrans_stamp is still later than when the
original packet was sent.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NNeal Cardwell <ncardwell@google.com>
Reviewed-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Previously TCP only warns if its RTO timer fires and the
retransmission queue is empty, but it'll cause null pointer
reference later on. It's better to avoid such catastrophic failure
and simply exit with a warning.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NNeal Cardwell <ncardwell@google.com>
Reviewed-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Previously upon SYN timeouts the sender recomputes the txhash to
try a different path. However this does not apply on the initial
timeout of SYN-data (active Fast Open). Therefore an active IPv6
Fast Open connection may incur one second RTO penalty to take on
a new path after the second SYN retransmission uses a new flow label.

This patch removes this undesirable behavior so Fast Open changes
the flow label just like the regular connections. This also helps
avoid falsely disabling Fast Open on the sender which triggers
after two consecutive SYN timeouts on Fast Open.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Reviewed-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Previously the SNMP TCPTIMEOUTS counter has inconsistent accounting:
1. It counts all SYN and SYN-ACK timeouts
2. It counts timeouts in other states except recurring timeouts and
   timeouts after fast recovery or disorder state.

Such selective accounting makes analysis difficult and complicated. For
example the monitoring system needs to collect many other SNMP counters
to infer the total amount of timeout events. This patch makes TCPTIMEOUTS
counter simply counts all the retransmit timeout (SYN or data or FIN).
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Previously there is an off-by-one bug on determining when to abort
a stalled window-probing socket. This patch fixes that so it is
consistent with tcp_write_timeout().
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a qdisc setup including pacing FQ is dismantled and recreated,
some TCP packets are sent earlier than instructed by TCP stack.

TCP can be fooled when ACK comes back, because the following
operation can return a negative value.

    tcp_time_stamp(tp) - tp->rx_opt.rcv_tsecr;

Some paths in TCP stack were not dealing properly with this,
this patch addresses four of them.

Fixes: ab408b6d ("tcp: switch tcp and sch_fq to new earliest departure time model")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Jean-Louis reported a TCP regression and bisected to recent SACK
compression.

After a loss episode (receiver not able to keep up and dropping
packets because its backlog is full), linux TCP stack is sending
a single SACK (DUPACK).

Sender waits a full RTO timer before recovering losses.

While RFC 6675 says in section 5, "Algorithm Details",

   (2) If DupAcks < DupThresh but IsLost (HighACK + 1) returns true --
       indicating at least three segments have arrived above the current
       cumulative acknowledgment point, which is taken to indicate loss
       -- go to step (4).
...
   (4) Invoke fast retransmit and enter loss recovery as follows:

there are old TCP stacks not implementing this strategy, and
still counting the dupacks before starting fast retransmit.

While these stacks probably perform poorly when receivers implement
LRO/GRO, we should be a little more gentle to them.

This patch makes sure we do not enable SACK compression unless
3 dupacks have been sent since last rcv_nxt update.

Ideally we should even rearm the timer to send one or two
more DUPACK if no more packets are coming, but that will
be work aiming for linux-4.21.

Many thanks to Jean-Louis for bisecting the issue, providing
packet captures and testing this patch.

Fixes: 5d9f4262 ("tcp: add SACK compression")
Reported-by: NJean-Louis Dupond <jean-louis@dupond.be>
Tested-by: NJean-Louis Dupond <jean-louis@dupond.be>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In EDT design, I made the mistake of using tcp_wstamp_ns
to store the last tcp_clock_ns() sample and to store the
pacing virtual timer.

This causes major regressions at high speed flows.

Introduce tcp_clock_cache to store last tcp_clock_ns().
This is needed because some arches have slow high-resolution
kernel time service.

tcp_wstamp_ns is only updated when a packet is sent.

Note that we can remove tcp_mstamp in the future since
tcp_mstamp is essentially tcp_clock_cache/1000, so the
apparent socket size increase is temporary.

Fixes: 9799ccb0 ("tcp: add tcp_wstamp_ns socket field")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In the recent TCP/EDT patch series, I switched TCP and sch_fq
clocks from MONOTONIC to TAI, in order to meet the choice done
earlier for sch_etf packet scheduler.

But sure enough, this broke some setups were the TAI clock
jumps forward (by almost 50 year...), as reported
by Leonard Crestez.

If we want to converge later, we'll probably need to add
an skb field to differentiate the clock bases, or a socket option.

In the meantime, an UDP application will need to use CLOCK_MONOTONIC
base for its SCM_TXTIME timestamps if using fq packet scheduler.

Fixes: 72b0094f ("tcp: switch tcp_clock_ns() to CLOCK_TAI base")
Fixes: 142537e4 ("net_sched: sch_fq: switch to CLOCK_TAI")
Fixes: fd2bca2a ("tcp: switch internal pacing timer to CLOCK_TAI")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reported-by: NLeonard Crestez <leonard.crestez@nxp.com>
Tested-by: NLeonard Crestez <leonard.crestez@nxp.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Next patch will use tcp_wstamp_ns to feed internal
TCP pacing timer, so switch to CLOCK_TAI to share same base.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Switch internal TCP skb->skb_mstamp to skb->skb_mstamp_ns,
from usec units to nsec units.

Do not clear skb->tstamp before entering IP stacks in TX,
so that qdisc or devices can implement pacing based on the
earliest departure time instead of socket sk->sk_pacing_rate

Packets are fed with tcp_wstamp_ns, and following patch
will update tcp_wstamp_ns when both TCP and sch_fq switch to
the earliest departure time mechanism.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fixes the following sparse warnings:

net/ipv4/tcp_timer.c:25:5: warning:
 symbol 'tcp_retransmit_stamp' was not declared. Should it be static?
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Create the tcp_clamp_rto_to_user_timeout() helper routine. To calculate
the correct rto, so that the TCP_USER_TIMEOUT socket option is more
accurate. Taking suggestions and feedback into account from
Eric Dumazet, Neal Cardwell and David Laight. Due to the 1st commit we
can avoid the msecs_to_jiffies() and jiffies_to_msecs() dance.
Signed-off-by: NJon Maxwell <jmaxwell37@gmail.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Create a seperate helper routine as per Neal Cardwells suggestion. To
be used by the final commit in this series and retransmits_timed_out().
Signed-off-by: NJon Maxwell <jmaxwell37@gmail.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is a preparatory commit. Part of this series that improves the
socket TCP_USER_TIMEOUT option accuracy. Implement Eric Dumazets idea
to convert icsk->icsk_user_timeout from jiffies to msecs. To eliminate
the msecs_to_jiffies() and jiffies_to_msecs() dance in future.
Signed-off-by: NJon Maxwell <jmaxwell37@gmail.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When TCP receives an out-of-order packet, it immediately sends
a SACK packet, generating network load but also forcing the
receiver to send 1-MSS pathological packets, increasing its
RTX queue length/depth, and thus processing time.

Wifi networks suffer from this aggressive behavior, but generally
speaking, all these SACK packets add fuel to the fire when networks
are under congestion.

This patch adds a high resolution timer and tp->compressed_ack counter.

Instead of sending a SACK, we program this timer with a small delay,
based on RTT and capped to 1 ms :

	delay = min ( 5 % of RTT, 1 ms)

If subsequent SACKs need to be sent while the timer has not yet
expired, we simply increment tp->compressed_ack.

When timer expires, a SACK is sent with the latest information.
Whenever an ACK is sent (if data is sent, or if in-order
data is received) timer is canceled.

Note that tcp_sack_new_ofo_skb() is able to force a SACK to be sent
if the sack blocks need to be shuffled, even if the timer has not
expired.

A new SNMP counter is added in the following patch.

Two other patches add sysctls to allow changing the 1,000,000 and 44
values that this commit hard-coded.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Acked-by: NYuchung Cheng <ycheng@google.com>
Acked-by: NToke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

linux-4.16 got support for softirq based hrtimers.
TCP can switch its pacing hrtimer to this variant, since this
avoids going through a tasklet and some atomic operations.

pacing timer logic looks like other (jiffies based) tcp timers.

v2: use hrtimer_try_to_cancel() in tcp_clear_xmit_timers()
    to correctly release reference on socket if needed.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the connection is aborted, there is no point in
keeping the packets on the write queue until the connection
is closed.

Similar to a27fd7a8 ('tcp: purge write queue upon RST'),
this is essential for a correct MSG_ZEROCOPY implementation,
because userspace cannot call close(fd) before receiving
zerocopy signals even when the connection is aborted.

Fixes: f214f915 ("tcp: enable MSG_ZEROCOPY")
Signed-off-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Reviewed-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Adds an optional call to sock_ops BPF program based on whether the
BPF_SOCK_OPS_RTO_CB_FLAG is set in bpf_sock_ops_flags.
The BPF program is passed 2 arguments: icsk_retransmits and whether the
RTO has expired.
Signed-off-by: NLawrence Brakmo <brakmo@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>

When a tcp socket is closed, if it detects that its net namespace is
exiting, close immediately and do not wait for FIN sequence.

For normal sockets, a reference is taken to their net namespace, so it will
never exit while the socket is open.  However, kernel sockets do not take a
reference to their net namespace, so it may begin exiting while the kernel
socket is still open.  In this case if the kernel socket is a tcp socket,
it will stay open trying to complete its close sequence.  The sock's dst(s)
hold a reference to their interface, which are all transferred to the
namespace's loopback interface when the real interfaces are taken down.
When the namespace tries to take down its loopback interface, it hangs
waiting for all references to the loopback interface to release, which
results in messages like:

unregister_netdevice: waiting for lo to become free. Usage count = 1

These messages continue until the socket finally times out and closes.
Since the net namespace cleanup holds the net_mutex while calling its
registered pernet callbacks, any new net namespace initialization is
blocked until the current net namespace finishes exiting.

After this change, the tcp socket notices the exiting net namespace, and
closes immediately, releasing its dst(s) and their reference to the
loopback interface, which lets the net namespace continue exiting.

Link: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1711407
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=97811Signed-off-by: NDan Streetman <ddstreet@canonical.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Only the retransmit timer currently refreshes tcp_mstamp

We should do the same for delayed acks and keepalives.

Even if RFC 7323 does not request it, this is consistent to what linux
did in the past, when TS values were based on jiffies.

Fixes: 385e2070 ("tcp: use tp->tcp_mstamp in output path")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Soheil Hassas Yeganeh <soheil@google.com>
Cc: Mike Maloney <maloney@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Acked-by: NSoheil Hassas Yeganeh <soheil@google.com>
Acked-by: NMike Maloney <maloney@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Prior to this patch, active Fast Open is paused on a specific
destination IP address if the previous connections to the
IP address have experienced recurring timeouts . But recent
experiments by Microsoft (https://goo.gl/cykmn7) and Mozilla
browsers indicate the isssue is often caused by broken middle-boxes
sitting close to the client. Therefore it is much better user
experience if Fast Open is disabled out-right globally to avoid
experiencing further timeouts on connections toward other
destinations.

This patch changes the destination-IP disablement to global
disablement if a connection experiencing recurring timeouts
or aborts due to timeout.  Repeated incidents would still
exponentially increase the pause time, starting from an hour.
This is extremely conservative but an unfortunate compromise to
minimize bad experience due to broken middle-boxes.
Reported-by: NDragana Damjanovic <ddamjanovic@mozilla.com>
Reported-by: NPatrick McManus <mcmanus@ducksong.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Reviewed-by: NWei Wang <weiwan@google.com>
Reviewed-by: NNeal Cardwell <ncardwell@google.com>
Reviewed-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Reduce one indentation level to make code more readable.
tcp_sync_mss() can be factorized.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Note that sysctl_tcp_thin_dupack was not used, I deleted it.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In preparation for unconditionally passing the struct timer_list pointer to
all timer callbacks, switch to using the new timer_setup() and from_timer()
to pass the timer pointer explicitly.

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: netdev@vger.kernel.org
Cc: dccp@vger.kernel.org
Signed-off-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Using a linear list to store all skbs in write queue has been okay
for quite a while : O(N) is not too bad when N < 500.

Things get messy when N is the order of 100,000 : Modern TCP stacks
want 10Gbit+ of throughput even with 200 ms RTT flows.

40 ns per cache line miss means a full scan can use 4 ms,
blowing away CPU caches.

SACK processing often can use various hints to avoid parsing
whole retransmit queue. But with high packet losses and/or high
reordering, hints no longer work.

Sender has to process thousands of unfriendly SACK, accumulating
a huge socket backlog, burning a cpu and massively dropping packets.

Using an rb-tree for retransmit queue has been avoided for years
because it added complexity and overhead, but now is the time
to be more resistant and say no to quadratic behavior.

1) RTX queue is no longer part of the write queue : already sent skbs
are stored in one rb-tree.

2) Since reaching the head of write queue no longer needs
sk->sk_send_head, we added an union of sk_send_head and tcp_rtx_queue

Tested:

 On receiver :
 netem on ingress : delay 150ms 200us loss 1
 GRO disabled to force stress and SACK storms.

for f in `seq 1 10`
do
 ./netperf -H lpaa6 -l30 -- -K bbr -o THROUGHPUT|tail -1
done | awk '{print $0} {sum += $0} END {printf "%7u\n",sum}'

Before patch :

323.87
351.48
339.59
338.62
306.72
204.07
304.93
291.88
202.47
176.88
   2840

After patch:

1700.83
2207.98
2070.17
1544.26
2114.76
2124.89
1693.14
1080.91
2216.82
1299.94
  18053
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

syzkaller was able to trigger a divide by 0 in TCP stack [1]

Issue here is that keepalive timer needs to be updated to not attempt
to send a probe if the connection setup was deferred using
TCP_FASTOPEN_CONNECT socket option added in linux-4.11

[1]
 divide error: 0000 [#1] SMP
 CPU: 18 PID: 0 Comm: swapper/18 Not tainted
 task: ffff986f62f4b040 ti: ffff986f62fa2000 task.ti: ffff986f62fa2000
 RIP: 0010:[<ffffffff8409cc0d>]  [<ffffffff8409cc0d>] __tcp_select_window+0x8d/0x160
 Call Trace:
  <IRQ>
  [<ffffffff8409d951>] tcp_transmit_skb+0x11/0x20
  [<ffffffff8409da21>] tcp_xmit_probe_skb+0xc1/0xe0
  [<ffffffff840a0ee8>] tcp_write_wakeup+0x68/0x160
  [<ffffffff840a151b>] tcp_keepalive_timer+0x17b/0x230
  [<ffffffff83b3f799>] call_timer_fn+0x39/0xf0
  [<ffffffff83b40797>] run_timer_softirq+0x1d7/0x280
  [<ffffffff83a04ddb>] __do_softirq+0xcb/0x257
  [<ffffffff83ae03ac>] irq_exit+0x9c/0xb0
  [<ffffffff83a04c1a>] smp_apic_timer_interrupt+0x6a/0x80
  [<ffffffff83a03eaf>] apic_timer_interrupt+0x7f/0x90
  <EOI>
  [<ffffffff83fed2ea>] ? cpuidle_enter_state+0x13a/0x3b0
  [<ffffffff83fed2cd>] ? cpuidle_enter_state+0x11d/0x3b0

Tested:

Following packetdrill no longer crashes the kernel

`echo 0 >/proc/sys/net/ipv4/tcp_timestamps`

// Cache warmup: send a Fast Open cookie request
    0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
   +0 fcntl(3, F_SETFL, O_RDWR|O_NONBLOCK) = 0
   +0 setsockopt(3, SOL_TCP, TCP_FASTOPEN_CONNECT, [1], 4) = 0
   +0 connect(3, ..., ...) = -1 EINPROGRESS (Operation is now in progress)
   +0 > S 0:0(0) <mss 1460,nop,nop,sackOK,nop,wscale 8,FO,nop,nop>
 +.01 < S. 123:123(0) ack 1 win 14600 <mss 1460,nop,nop,sackOK,nop,wscale 6,FO abcd1234,nop,nop>
   +0 > . 1:1(0) ack 1
   +0 close(3) = 0
   +0 > F. 1:1(0) ack 1
   +0 < F. 1:1(0) ack 2 win 92
   +0 > .  2:2(0) ack 2

   +0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 4
   +0 fcntl(4, F_SETFL, O_RDWR|O_NONBLOCK) = 0
   +0 setsockopt(4, SOL_TCP, TCP_FASTOPEN_CONNECT, [1], 4) = 0
   +0 setsockopt(4, SOL_SOCKET, SO_KEEPALIVE, [1], 4) = 0
 +.01 connect(4, ..., ...) = 0
   +0 setsockopt(4, SOL_TCP, TCP_KEEPIDLE, [5], 4) = 0
   +10 close(4) = 0

`echo 1 >/proc/sys/net/ipv4/tcp_timestamps`

Fixes: 19f6d3f3 ("net/tcp-fastopen: Add new API support")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reported-by: NDmitry Vyukov <dvyukov@google.com>
Cc: Wei Wang <weiwan@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

prequeue is a tcp receive optimization that moves part of rx processing
from bh to process context.

This only works if the socket being processed belongs to a process that
is blocked in recv on that socket.

In practice, this doesn't happen anymore that often because nowadays
servers tend to use an event driven (epoll) model.

Even normal client applications (web browsers) commonly use many tcp
connections in parallel.

This has measureable impact only in netperf (which uses plain recv and
thus allows prequeue use) from host to locally running vm (~4%), however,
there were no changes when using netperf between two physical hosts with
ixgbe interfaces.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

After the mentioned commit, some of our packetdrill tests became flaky.

TCP_SYNCNT socket option can limit the number of SYN retransmits.

retransmits_timed_out() has to compare times computations based on
local_clock() while timers are based on jiffies. With NTP adjustments
and roundings we can observe 999 ms delay for 1000 ms timers.
We end up sending one extra SYN packet.

Gimmick added in commit 6fa12c85 ("Revert Backoff [v3]: Calculate
TCP's connection close threshold as a time value") makes no
real sense for TCP_SYN_SENT sockets where no RTO backoff can happen at
all.

Lets use a simpler logic for TCP_SYN_SENT sockets and remove @syn_set
parameter from retransmits_timed_out()

Fixes: 9a568de4 ("tcp: switch TCP TS option (RFC 7323) to 1ms clock")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Acked-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

TCP_USER_TIMEOUT is still converted to jiffies value in
icsk_user_timeout

So we need to make a conversion for the cases HZ != 1000

Fixes: 9a568de4 ("tcp: switch TCP TS option (RFC 7323) to 1ms clock")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

TCP Timestamps option is defined in RFC 7323

Traditionally on linux, it has been tied to the internal
'jiffies' variable, because it had been a cheap and good enough
generator.

For TCP flows on the Internet, 1 ms resolution would be much better
than 4ms or 10ms (HZ=250 or HZ=100 respectively)

For TCP flows in the DC, Google has used usec resolution for more
than two years with great success [1]

Receive size autotuning (DRS) is indeed more precise and converges
faster to optimal window size.

This patch converts tp->tcp_mstamp to a plain u64 value storing
a 1 usec TCP clock.

This choice will allow us to upstream the 1 usec TS option as
discussed in IETF 97.

[1] https://www.ietf.org/proceedings/97/slides/slides-97-tcpm-tcp-options-for-low-latency-00.pdfSigned-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use tcp_jiffies32 instead of tcp_time_stamp, since
tcp_time_stamp will soon be only used for TCP TS option.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NSoheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>