Note that sysctl_tcp_thin_dupack was not used, I deleted it.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The SMC protocol [1] relies on the use of a new TCP experimental
option [2, 3]. With this option, SMC capabilities are exchanged
between peers during the TCP three way handshake. This patch adds
support for this experimental option to TCP.
References:
[1] SMC-R Informational RFC: http://www.rfc-editor.org/info/rfc7609
[2] Shared Use of TCP Experimental Options RFC 6994:
https://tools.ietf.org/rfc/rfc6994.txt
[3] IANA ExID SMCR:
http://www.iana.org/assignments/tcp-parameters/tcp-parameters.xhtml#tcp-exids
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Current implementation calls tcp_rate_skb_sent() when tcp_transmit_skb()
is called when it clones skb only. Not calling tcp_rate_skb_sent() is OK
for all such code paths except from __tcp_retransmit_skb() which happens
when skb->data address is not aligned. This may rarely happen e.g. when
small amount of data is sent initially and the receiver partially acks
odd number of bytes for some reason, possibly malicious.
Signed-off-by: Yousuk Seung <ysseung@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In my first attempt to fix the lockdep splat, I forgot we could
enter inet_csk_route_req() with a freshly allocated request socket,
for which refcount has not yet been elevated, due to complex
SLAB_TYPESAFE_BY_RCU rules.
We either are in rcu_read_lock() section _or_ we own a refcount on the
request.
Correct RCU verb to use here is rcu_dereference_check(), although it is
not possible to prove we actually own a reference on a shared
refcount :/
In v2, I added ireq_opt_deref() helper and use in three places, to fix other
possible splats.
[ 49.844590] lockdep_rcu_suspicious+0xea/0xf3
[ 49.846487] inet_csk_route_req+0x53/0x14d
[ 49.848334] tcp_v4_route_req+0xe/0x10
[ 49.850174] tcp_conn_request+0x31c/0x6a0
[ 49.851992] ? __lock_acquire+0x614/0x822
[ 49.854015] tcp_v4_conn_request+0x5a/0x79
[ 49.855957] ? tcp_v4_conn_request+0x5a/0x79
[ 49.858052] tcp_rcv_state_process+0x98/0xdcc
[ 49.859990] ? sk_filter_trim_cap+0x2f6/0x307
[ 49.862085] tcp_v4_do_rcv+0xfc/0x145
[ 49.864055] ? tcp_v4_do_rcv+0xfc/0x145
[ 49.866173] tcp_v4_rcv+0x5ab/0xaf9
[ 49.868029] ip_local_deliver_finish+0x1af/0x2e7
[ 49.870064] ip_local_deliver+0x1b2/0x1c5
[ 49.871775] ? inet_del_offload+0x45/0x45
[ 49.873916] ip_rcv_finish+0x3f7/0x471
[ 49.875476] ip_rcv+0x3f1/0x42f
[ 49.876991] ? ip_local_deliver_finish+0x2e7/0x2e7
[ 49.878791] __netif_receive_skb_core+0x6d3/0x950
[ 49.880701] ? process_backlog+0x7e/0x216
[ 49.882589] __netif_receive_skb+0x1d/0x5e
[ 49.884122] process_backlog+0x10c/0x216
[ 49.885812] net_rx_action+0x147/0x3df
Fixes: a6ca7abe53 ("tcp/dccp: fix lockdep splat in inet_csk_route_req()")
Fixes: c92e8c02fe ("tcp/dccp: fix ireq->opt races")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: kernel test robot <fengguang.wu@intel.com>
Reported-by: Maciej Żenczykowski <maze@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
after previous commit xt_replace_table will wait until all cpus
had even seqcount (i.e., no cpu is accessing old ruleset).
Add a 'old' counter retrival version that doesn't synchronize counters.
Its not needed, the old counters are not in use anymore at this point.
This speeds up table replacement on busy systems with large tables
(and many cores).
Cc: Dan Williams <dcbw@redhat.com>
Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
We currently pass down the l4 protocol to the conntrack ->packet()
function, but the only user of this is the debug info decision.
Same information can be derived from struct nf_conn.
As a first step, add and use a new log function for this, similar to
nf_ct_helper_log().
Add __cold annotation -- invalid packets should be infrequent so
gcc can consider all call paths that lead to such a function as
unlikely.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
We already allow to enable TFO without a cookie by using the
fastopen-sysctl and setting it to TFO_SERVER_COOKIE_NOT_REQD (or
TFO_CLIENT_NO_COOKIE).
This is safe to do in certain environments where we know that there
isn't a malicous host (aka., data-centers) or when the
application-protocol already provides an authentication mechanism in the
first flight of data.
A server however might be providing multiple services or talking to both
sides (public Internet and data-center). So, this server would want to
enable cookie-less TFO for certain services and/or for connections that
go to the data-center.
This patch exposes a socket-option and a per-route attribute to enable such
fine-grained configurations.
Signed-off-by: Christoph Paasch <cpaasch@apple.com>
Reviewed-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use BUG_ON instead of if condition followed by BUG in tcp_time_wait.
This issue was detected with the help of Coccinelle.
Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use BUG_ON instead of if condition followed by BUG in icmp_timestamp.
This issue was detected with the help of Coccinelle.
Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds tracepoint trace_tcp_set_state. Besides usual fields
(s/d ports, IP addresses), old and new state of the socket is also
printed with TP_printk, with __print_symbolic().
Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds trace event trace_tcp_destroy_sock.
Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
New tracepoint trace_tcp_receive_reset is added and called from
tcp_reset(). This tracepoint is define with a new class tcp_event_sk.
Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
New tracepoint trace_tcp_send_reset is added and called from
tcp_v4_send_reset(), tcp_v6_send_reset() and tcp_send_active_reset().
Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes the following lockdep splat in inet_csk_route_req()
lockdep_rcu_suspicious
inet_csk_route_req
tcp_v4_send_synack
tcp_rtx_synack
inet_rtx_syn_ack
tcp_fastopen_synack_time
tcp_retransmit_timer
tcp_write_timer_handler
tcp_write_timer
call_timer_fn
Thread running inet_csk_route_req() owns a reference on the request
socket, so we have the guarantee ireq->ireq_opt wont be changed or
freed.
lockdep can enforce this invariant for us.
Fixes: c92e8c02fe ("tcp/dccp: fix ireq->opt races")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When retransmission on TSQ handler was introduced in the commit
f9616c35a0 ("tcp: implement TSQ for retransmits"), the retransmitted
skbs' timestamps were updated on the actual transmission. In the later
commit 385e20706f ("tcp: use tp->tcp_mstamp in output path"), it stops
being done so. In the commit, the comment says "We try to refresh
tp->tcp_mstamp only when necessary", and at present tcp_tsq_handler and
tcp_v4_mtu_reduced applies to this. About the latter, it's okay since
it's rare enough.
About the former, even though possible retransmissions on the tasklet
comes just after the destructor run in NET_RX softirq handling, the time
between them could be nonnegligibly large to the extent that
tcp_rack_advance or rto rearming be affected if other (remaining) RX,
BLOCK and (preceding) TASKLET sofirq handlings are unexpectedly heavy.
So in the same way as tcp_write_timer_handler does, doing tcp_mstamp_refresh
ensures the accuracy of algorithms relying on it.
Fixes: 385e20706f ("tcp: use tp->tcp_mstamp in output path")
Signed-off-by: Koichiro Den <den@klaipeden.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There were quite a few overlapping sets of changes here.
Daniel's bug fix for off-by-ones in the new BPF branch instructions,
along with the added allowances for "data_end > ptr + x" forms
collided with the metadata additions.
Along with those three changes came veritifer test cases, which in
their final form I tried to group together properly. If I had just
trimmed GIT's conflict tags as-is, this would have split up the
meta tests unnecessarily.
In the socketmap code, a set of preemption disabling changes
overlapped with the rename of bpf_compute_data_end() to
bpf_compute_data_pointers().
Changes were made to the mv88e6060.c driver set addr method
which got removed in net-next.
The hyperv transport socket layer had a locking change in 'net'
which overlapped with a change of socket state macro usage
in 'net-next'.
Signed-off-by: David S. Miller <davem@davemloft.net>
TCP_NV will try to get the base RTT from a socket_ops BPF program if one
is loaded. NV will then use the base RTT to bound its min RTT (its
notion of the base RTT). It uses the base RTT as an upper bound and 80%
of the base RTT as its lower bound.
In other words, NV will consider filtered RTTs larger than base RTT as a
sign of congestion. As a result, there is no minRTT inflation when there
is a lot of congestion. For example, in a DC where the RTTs are less
than 40us when there is no congestion, a base RTT value of 80us improves
the performance of NV. The difference between the uncongested RTT and
the base RTT provided represents how much queueing we are willing to
have (in practice it can be higher).
NV has been tunned to reduce congestion when there are many flows at the
cost of one flow not achieving full bandwith utilization. When a
reasonable base RTT is provided, one NV flow can now fully utilize the
full bandwidth. In addition, the performance is also improved when there
are many flows.
In the following examples the NV results are using a kernel with this
patch set (i.e. both NV results are using the new nv_loss_dec_factor).
With one host sending to another host and only one flow the
goodputs are:
Cubic: 9.3 Gbps, NV: 5.5 Gbps, NV (baseRTT=80us): 9.2 Gbps
With 2 hosts sending to one host (1 flow per host, the goodput per flow
is:
Cubic: 4.6 Gbps, NV: 4.5 Gbps, NV (baseRTT=80us)L 4.6 Gbps
But the RTTs seen by a ping process in the sender is:
Cubic: 3.3ms NV: 97us, NV (baseRTT=80us): 146us
With a lot of flows things look even better for NV with baseRTT. Here we
have 3 hosts sending to one host. Each sending host has 6 flows: 1
stream, 4x1MB RPC, 1x10KB RPC. Cubic, NV and NV with baseRTT all fully
utilize the full available bandwidth. However, the distribution of
bandwidth among the flows is very different. For the 10KB RPC flow:
Cubic: 27Mbps, NV: 111Mbps, NV (baseRTT=80us): 222Mbps
The 99% latencies for the 10KB flows are:
Cubic: 26ms, NV: 1ms, NV (baseRTT=80us): 500us
The RTT seen by a ping process at the senders:
Cubic: 3.2ms NV: 720us, NV (baseRTT=80us): 330us
Signed-off-by: Lawrence Brakmo <brakmo@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Syzkaller stumbled upon a way to trigger
WARNING: CPU: 1 PID: 13881 at net/core/sock_reuseport.c:41
reuseport_alloc+0x306/0x3b0 net/core/sock_reuseport.c:39
There are two initialization paths for the sock_reuseport structure in a
socket: Through the udp/tcp bind paths of SO_REUSEPORT sockets or through
SO_ATTACH_REUSEPORT_[CE]BPF before bind. The existing implementation
assumedthat the socket lock protected both of these paths when it actually
only protects the SO_ATTACH_REUSEPORT path. Syzkaller triggered this
double allocation by running these paths concurrently.
This patch moves the check for double allocation into the reuseport_alloc
function which is protected by a global spin lock.
Fixes: e32ea7e747 ("soreuseport: fast reuseport UDP socket selection")
Fixes: c125e80b88 ("soreuseport: fast reuseport TCP socket selection")
Signed-off-by: Craig Gallek <kraig@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the UDP code there are two leftover error messages with very few meaning.
Replace them with a more descriptive error message as some users
reported them as "strange network error".
Signed-off-by: Matteo Croce <mcroce@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ipv4_default_advmss() incorrectly uses the device MTU instead
of the route provided one. IPv6 has the proper behavior,
lets harmonize the two protocols.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
syn_data was allocated by sk_stream_alloc_skb(), meaning
its destructor and _skb_refdst fields are mangled.
We need to call tcp_skb_tsorted_anchor_cleanup() before
calling kfree_skb() or kernel crashes.
Bug was reported by syzkaller bot.
Fixes: e2080072ed ("tcp: new list for sent but unacked skbs for RACK recovery")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All of the notifier data (fib_info, tos, type and table id) are
contained in the fib_alias. Pass it to the notifier instead of
each data separately shortening the argument list by 3.
Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
New socket option TCP_FASTOPEN_KEY to allow different keys per
listener. The listener by default uses the global key until the
socket option is set. The key is a 16 bytes long binary data. This
option has no effect on regular non-listener TCP sockets.
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Christoph Paasch <cpaasch@apple.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add extack to in_validator_info and in6_validator_info. Update the one
user of each, ipvlan, to return an error message for failures.
Only manual configuration of an address is plumbed in the IPv6 code path.
Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I tried to hard avoiding a call to rb_first() (via tcp_rtx_queue_head)
in tcp_xmit_retransmit_queue(). But this was probably too bold.
Quoting Yuchung :
We might miss re-arming the RTO if tp->retransmit_skb_hint is not NULL.
This can happen when RACK marks the first packet lost again and resets
tp->retransmit_skb_hint for example (tcp_rack_mark_skb_lost())
Fixes: 75c119afe1 ("tcp: implement rb-tree based retransmit queue")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In preparation to enabling -Wimplicit-fallthrough, mark switch cases
where we are expecting to fall through.
Notice that in some cases I placed the "fall through" comment
on its own line, which is what GCC is expecting to find.
Addresses-Coverity-ID: 115108
Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com>
Signed-off-by: David 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: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: netdev@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David 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: Alexander Aring <alex.aring@gmail.com>
Cc: Stefan Schmidt <stefan@osg.samsung.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Pablo Neira Ayuso <pablo@netfilter.org>
Cc: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>
Cc: Florian Westphal <fw@strlen.de>
Cc: linux-wpan@vger.kernel.org
Cc: netdev@vger.kernel.org
Cc: netfilter-devel@vger.kernel.org
Cc: coreteam@netfilter.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: Stefan Schmidt <stefan@osg.samsung.com> # for ieee802154
Signed-off-by: David 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: Kees Cook <keescook@chromium.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
rtnl af_ops currently rely on rtnl mutex: unregister (called from module
exit functions) takes the rtnl mutex and all users that do af_ops lookup
also take the rtnl mutex. IOW, parallel rmmod will block until doit()
callback is done.
As none of the af_ops implementation sleep we can use rcu instead.
doit functions that need the af_ops can now use rcu instead of the
rtnl mutex provided the mutex isn't needed for other reasons.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
The structure tcp_cdg is local to the source and
does not need to be in global scope, so make it static.
Cleans up sparse warning:
symbol 'tcp_cdg' was not declared. Should it be static?
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need a real-time notification for tcp retransmission
for monitoring.
Of course we could use ftrace to dynamically instrument this
kernel function too, however we can't retrieve the connection
information at the same time, for example perf-tools [1] reads
/proc/net/tcp for socket details, which is slow when we have
a lots of connections.
Therefore, this patch adds a tracepoint for __tcp_retransmit_skb()
and exposes src/dst IP addresses and ports of the connection.
This also makes it easier to integrate into perf.
Note, I expose both IPv4 and IPv6 addresses at the same time:
for a IPv4 socket, v4 mapped address is used as IPv6 addresses,
for a IPv6 socket, LOOPBACK4_IPV6 is already filled by kernel.
Also, add sk and skb pointers as they are useful for BPF.
1. https://github.com/brendangregg/perf-tools/blob/master/net/tcpretrans
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Brendan Gregg <bgregg@netflix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ICMP implementation currently replies to an ICMP time exceeded message
(type 11) with an ICMP host unreachable message (type 3, code 1).
However, time exceeded messages can either represent "time to live exceeded
in transit" (code 0) or "fragment reassembly time exceeded" (code 1).
Unconditionally replying to "fragment reassembly time exceeded" with
host unreachable messages might cause unjustified connection resets
which are now easily triggered as UFO has been removed, because, in turn,
sending large buffers triggers IP fragmentation.
The issue can be easily reproduced by running a lot of UDP streams
which is likely to trigger IP fragmentation:
# start netserver in the test namespace
ip netns add test
ip netns exec test netserver
# create a VETH pair
ip link add name veth0 type veth peer name veth0 netns test
ip link set veth0 up
ip -n test link set veth0 up
for i in $(seq 20 29); do
# assign addresses to both ends
ip addr add dev veth0 192.168.$i.1/24
ip -n test addr add dev veth0 192.168.$i.2/24
# start the traffic
netperf -L 192.168.$i.1 -H 192.168.$i.2 -t UDP_STREAM -l 0 &
done
# wait
send_data: data send error: No route to host (errno 113)
netperf: send_omni: send_data failed: No route to host
We need to differentiate instead: if fragment reassembly time exceeded
is reported, we need to silently drop the packet,
if time to live exceeded is reported, maintain the current behaviour.
In both cases increment the related error count "icmpInTimeExcds".
While at it, fix a typo in a comment, and convert the if statement
into a switch to mate it more readable.
Signed-off-by: Matteo Croce <mcroce@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The kernel config help for policy routing was still pointing at
an ancient document from 2000 that refers to Linux 2.1. Update it
to point to something that is at least occasionally updated.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This reverts commit fbb1fb4ad4.
This was not the proper fix, lets cleanly revert it, so that
following patch can be carried to stable versions.
sock_cgroup_ptr() callers do not expect a NULL return value.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
sk_clone_lock() might run while TCP/DCCP listener already vanished.
In order to prevent use after free, it is better to defer cgroup_sk_alloc()
to the point we know both parent and child exist, and from process context.
Fixes: e994b2f0fb ("tcp: do not lock listener to process SYN packets")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of calling mem_cgroup_sk_alloc() from BH context,
it is better to call it from inet_csk_accept() in process context.
Not only this removes code in mem_cgroup_sk_alloc(), but it also
fixes a bug since listener might have been dismantled and css_get()
might cause a use-after-free.
Fixes: e994b2f0fb ("tcp: do not lock listener to process SYN packets")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso says:
====================
Netfilter/IPVS fixes for net
The following patchset contains Netfilter/IPVS fixes for your net tree,
they are:
1) Fix packet drops due to incorrect ECN handling in IPVS, from Vadim
Fedorenko.
2) Fix splat with mark restoration in xt_socket with non-full-sock,
patch from Subash Abhinov Kasiviswanathan.
3) ipset bogusly bails out when adding IPv4 range containing more than
2^31 addresses, from Jozsef Kadlecsik.
4) Incorrect pernet unregistration order in ipset, from Florian Westphal.
5) Races between dump and swap in ipset results in BUG_ON splats, from
Ross Lagerwall.
6) Fix chain renames in nf_tables, from JingPiao Chen.
7) Fix race in pernet codepath with ebtables table registration, from
Artem Savkov.
8) Memory leak in error path in set name allocation in nf_tables, patch
from Arvind Yadav.
9) Don't dump chain counters if they are not available, this fixes a
crash when listing the ruleset.
10) Fix out of bound memory read in strlcpy() in x_tables compat code,
from Eric Dumazet.
11) Make sure we only process TCP packets in SYNPROXY hooks, patch from
Lin Zhang.
12) Cannot load rules incrementally anymore after xt_bpf with pinned
objects, added in revision 1. From Shmulik Ladkani.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The commit bc044e8db7 ("udp: perform source validation for
mcast early demux") does not take into account that broadcast packets
lands in the same code path and they need different checks for the
source address - notably, zero source address are valid for bcast
and invalid for mcast.
As a result, 2nd and later broadcast packets with 0 source address
landing to the same socket are dropped. This breaks dhcp servers.
Since we don't have stringent performance requirements for ingress
broadcast traffic, fix it by disabling UDP early demux such traffic.
Reported-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Fixes: bc044e8db7 ("udp: perform source validation for mcast early demux")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A recent patch removed the dst_free() on the allocated
dst_entry in ipv4_blackhole_route(). The dst_free() marked the
dst_entry as dead and added it to the gc list. I.e. it was setup
for a one time usage. As a result we may now have a blackhole
route cached at a socket on some IPsec scenarios. This makes the
connection unusable.
Fix this by marking the dst_entry directly at allocation time
as 'dead', so it is used only once.
Fixes: b838d5e1c5 ("ipv4: mark DST_NOGC and remove the operation of dst_free()")
Reported-by: Tobias Brunner <tobias@strongswan.org>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In function {ipv4,ipv6}_synproxy_hook we expect a normal tcp packet, but
the real server maybe reply an icmp error packet related to the exist
tcp conntrack, so we will access wrong tcp data.
Fix it by checking for the protocol field and only process tcp traffic.
Signed-off-by: Lin Zhang <xiaolou4617@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
When gso_size reset to zero for the tail segment in skb_segment(), later
in ipv6_gso_segment(), __skb_udp_tunnel_segment() and gre_gso_segment()
we will get incorrect results (payload length, pcsum) for that segment.
inet_gso_segment() already has a check for gso_size before calculating
payload.
The issue was found with LTP vxlan & gre tests over ixgbe NIC.
Fixes: 07b26c9454 ("gso: Support partial splitting at the frag_list pointer")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Acked-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Similarly to early patch for erspan_xmit(), the ARPHDR_ETHER device
is the length of the whole ether packet. So skb->len should subtract
the dev->hard_header_len.
Fixes: 1a66a836da ("gre: add collect_md mode to ERSPAN tunnel")
Fixes: 84e54fe0a5 ("gre: introduce native tunnel support for ERSPAN")
Signed-off-by: William Tu <u9012063@gmail.com>
Cc: Xin Long <lucien.xin@gmail.com>
Cc: David Laight <David.Laight@aculab.com>
Reviewed-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David 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: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
No need to recompute previous skb, as it will be a bit more
expensive when rtx queue is converted to RB tree.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
With upcoming rb-tree implementation, the checks will trigger
more often, and this is expected.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It will be a bit more expensive to get the head of rtx queue
once rtx queue is converted to an rb-tree.
We can avoid this extra cost in case tp->lost_skb_hint is set.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since the upcoming rtx rbtree will add some extra code,
it is time to not inline this fat function anymore.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Geeralize private netem_rb_to_skb()
TCP rtx queue will soon be converted to rb-tree,
so we will need skb_rbtree_walk() helpers.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Refactor the RACK loop to improve readability and speed up the checks.
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the new time-ordered list to speed up RACK. The detection
logic is identical. But since the list is chronologically ordered
by skb_mstamp and contains only skbs not yet acked or sacked,
RACK can abort the loop upon hitting skbs that were sent more
recently. On YouTube servers this patch reduces the iterations on
write queue by 40x. The improvement is even bigger with large
BDP networks.
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a new queue (list) that tracks the sent but not yet
acked or SACKed skbs for a TCP connection. The list is chronologically
ordered by skb->skb_mstamp (the head is the oldest sent skb).
This list will be used to optimize TCP Rack recovery, which checks
an skb's timestamp to judge if it has been lost and needs to be
retransmitted. Since TCP write queue is ordered by sequence instead
of sent time, RACK has to scan over the write queue to catch all
eligible packets to detect lost retransmission, and iterates through
SACKed skbs repeatedly.
Special cares for rare events:
1. TCP repair fakes skb transmission so the send queue needs adjusted
2. SACK reneging would require re-inserting SACKed skbs into the
send queue. For now I believe it's not worth the complexity to
make RACK work perfectly on SACK reneging, so we do nothing here.
3. Fast Open: currently for non-TFO, send-queue correctly queues
the pure SYN packet. For TFO which queues a pure SYN and
then a data packet, send-queue only queues the data packet but
not the pure SYN due to the structure of TFO code. This is okay
because the SYN receiver would never respond with a SACK on a
missing SYN (i.e. SYN is never fast-retransmitted by SACK/RACK).
In order to not grow sk_buff, we use an union for the new list and
_skb_refdst/destructor fields. This is a bit complicated because
we need to make sure _skb_refdst and destructor are properly zeroed
before skb is cloned/copied at transmit, and before being freed.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
int rc is unmodified after initalization in net/ipv4/route.c, this patch simply cleans up that variable and returns 0.
This was found with coccicheck M=net/ipv4/ on linus' tree.
Signed-off-by: Tim Hansen <devtimhansen@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit does a cleanup and moves tcp_rearm_rto() call in the TFO
server case into a previous spot in tcp_rcv_state_process() to make
it more compact.
This is only a cosmetic change.
Suggested-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Wei Wang <weiwan@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently in the TCP code, the initialization sequence for cached
metrics, congestion control, BPF, etc, after successful connection
is very inconsistent. This introduces inconsistent bevhavior and is
prone to bugs. The current call sequence is as follows:
(1) for active case (tcp_finish_connect() case):
tcp_mtup_init(sk);
icsk->icsk_af_ops->rebuild_header(sk);
tcp_init_metrics(sk);
tcp_call_bpf(sk, BPF_SOCK_OPS_ACTIVE_ESTABLISHED_CB);
tcp_init_congestion_control(sk);
tcp_init_buffer_space(sk);
(2) for passive case (tcp_rcv_state_process() TCP_SYN_RECV case):
icsk->icsk_af_ops->rebuild_header(sk);
tcp_call_bpf(sk, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB);
tcp_init_congestion_control(sk);
tcp_mtup_init(sk);
tcp_init_buffer_space(sk);
tcp_init_metrics(sk);
(3) for TFO passive case (tcp_fastopen_create_child()):
inet_csk(child)->icsk_af_ops->rebuild_header(child);
tcp_init_congestion_control(child);
tcp_mtup_init(child);
tcp_init_metrics(child);
tcp_call_bpf(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB);
tcp_init_buffer_space(child);
This commit uniforms the above functions to have the following sequence:
tcp_mtup_init(sk);
icsk->icsk_af_ops->rebuild_header(sk);
tcp_init_metrics(sk);
tcp_call_bpf(sk, BPF_SOCK_OPS_ACTIVE/PASSIVE_ESTABLISHED_CB);
tcp_init_congestion_control(sk);
tcp_init_buffer_space(sk);
This sequence is the same as the (1) active case. We pick this sequence
because this order correctly allows BPF to override the settings
including congestion control module and initial cwnd, etc from
the route, and then allows the CC module to see those settings.
Suggested-by: Neal Cardwell <ncardwell@google.com>
Tested-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Wei Wang <weiwan@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change the ipmr module to not forward packets if:
- The packet is marked with the offload_mr_fwd_mark, and
- Both input interface and output interface share the same parent ID.
This way, a packet can go through partial multicast forwarding in the
hardware, where it will be forwarded only to the devices that share the
same parent ID (AKA, reside inside the same hardware). The kernel will
forward the packet to all other interfaces.
To do this, add the ipmr_offload_forward helper, which per skb, ingress VIF
and egress VIF, returns whether the forwarding was offloaded to hardware.
The ipmr_queue_xmit frees the skb and does not forward it if the result is
a true value.
All the forwarding path code compiles out when the CONFIG_NET_SWITCHDEV is
not set.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to allow the ipmr module to do partial multicast forwarding
according to the device parent ID, add the device parent ID field to the
VIF struct. This way, the forwarding path can use the parent ID field
without invoking switchdev calls, which requires the RTNL lock.
When a new VIF is added, set the device parent ID field in it by invoking
the switchdev_port_attr_get call.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The patch 'ip_gre: ipgre_tap device should keep dst' fixed
the issue ipgre_tap dev mtu couldn't be updated in tx path.
The same fix is needed for erspan as well.
Fixes: 84e54fe0a5 ("gre: introduce native tunnel support for ERSPAN")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
According to __gre_tunnel_init, tunnel->hlen should be set as the
headers' length between inner packet and outer iphdr.
It would be used especially to calculate a proper mtu when updating
mtu in tnl_update_pmtu. Now without setting it, a bigger mtu value
than expected would be updated, which hurts performance a lot.
This patch is to fix it by setting tunnel->hlen with:
tunnel->tun_hlen + tunnel->encap_hlen + sizeof(struct erspanhdr)
Fixes: 84e54fe0a5 ("gre: introduce native tunnel support for ERSPAN")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As a ARPHRD_ETHER device, skb->len in erspan_xmit is the length
of the whole ether packet. So before checking if a packet size
exceeds the mtu, skb->len should subtract dev->hard_header_len.
Otherwise, all packets with max size according to mtu would be
trimmed to be truncated packet.
Fixes: 84e54fe0a5 ("gre: introduce native tunnel support for ERSPAN")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
erspan only uses the first 10 bits of session_id as the key to look
up the tunnel. But in erspan_rcv, it missed 'session_id & ID_MASK'
when getting the key from session_id.
If any other flag is also set in session_id in a packet, it would
fail to find the tunnel due to incorrect key in erspan_rcv.
This patch is to add 'session_id & ID_MASK' there and also remove
the unnecessary variable session_id.
Fixes: 84e54fe0a5 ("gre: introduce native tunnel support for ERSPAN")
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Different namespace application might require different time period in
second to disable Fastopen on active TCP sockets.
Tested:
Simulate following similar situation that the server's data gets dropped
after 3WHS.
C ---- syn-data ---> S
C <--- syn/ack ----- S
C ---- ack --------> S
S (accept & write)
C? X <- data ------ S
[retry and timeout]
And then print netstat of TCPFastOpenBlackhole, the counter increased as
expected when the firewall blackhole issue is detected and active TFO is
disabled.
# cat /proc/net/netstat | awk '{print $91}'
TCPFastOpenBlackhole
1
Signed-off-by: Haishuang Yan <yanhaishuang@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Different namespace application might require different tcp_fastopen_key
independently of the host.
David Miller pointed out there is a leak without releasing the context
of tcp_fastopen_key during netns teardown. So add the release action in
exit_batch path.
Tested:
1. Container namespace:
# cat /proc/sys/net/ipv4/tcp_fastopen_key:
2817fff2-f803cf97-eadfd1f3-78c0992b
cookie key in tcp syn packets:
Fast Open Cookie
Kind: TCP Fast Open Cookie (34)
Length: 10
Fast Open Cookie: 1e5dd82a8c492ca9
2. Host:
# cat /proc/sys/net/ipv4/tcp_fastopen_key:
107d7c5f-68eb2ac7-02fb06e6-ed341702
cookie key in tcp syn packets:
Fast Open Cookie
Kind: TCP Fast Open Cookie (34)
Length: 10
Fast Open Cookie: e213c02bf0afbc8a
Signed-off-by: Haishuang Yan <yanhaishuang@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The 'publish' logic is not necessary after commit dfea2aa654 ("tcp:
Do not call tcp_fastopen_reset_cipher from interrupt context"), because
in tcp_fastopen_cookie_gen，it wouldn't call tcp_fastopen_init_key_once.
Signed-off-by: Haishuang Yan <yanhaishuang@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Different namespace application might require enable TCP Fast Open
feature independently of the host.
This patch series continues making more of the TCP Fast Open related
sysctl knobs be per net-namespace.
Reported-by: Luca BRUNO <lucab@debian.org>
Signed-off-by: Haishuang Yan <yanhaishuang@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The function ipmr_notifier_init is local to the source and does
not need to be in global scope, so make it static.
Cleans up sparse warning:
warning: symbol 'ipmr_notifier_init' was not declared. Should it be static?
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The UDP early demux can leverate the rx dst cache even for
multicast unconnected sockets.
In such scenario the ipv4 source address is validated only on
the first packet in the given flow. After that, when we fetch
the dst entry from the socket rx cache, we stop enforcing
the rp_filter and we even start accepting any kind of martian
addresses.
Disabling the dst cache for unconnected multicast socket will
cause large performace regression, nearly reducing by half the
max ingress tput.
Instead we factor out a route helper to completely validate an
skb source address for multicast packets and we call it from
the UDP early demux for mcast packets landing on unconnected
sockets, after successful fetching the related cached dst entry.
This still gives a measurable, but limited performance
regression:
rp_filter = 0 rp_filter = 1
edmux disabled: 1182 Kpps 1127 Kpps
edmux before: 2238 Kpps 2238 Kpps
edmux after: 2037 Kpps 2019 Kpps
The above figures are on top of current net tree.
Applying the net-next commit 6e617de84e ("net: avoid a full
fib lookup when rp_filter is disabled.") the delta with
rp_filter == 0 will decrease even more.
Fixes: 421b3885bf ("udp: ipv4: Add udp early demux")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently no error is emitted, but this infrastructure will
used by the next patch to allow source address validation
for mcast sockets.
Since early demux can do a route lookup and an ipv4 route
lookup can return an error code this is consistent with the
current ipv4 route infrastructure.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Without keeping dst, the tunnel will not update any mtu/pmtu info,
since it does not have a dst on the skb.
Reproducer:
client(ipgre_tap1 - eth1) <-----> (eth1 - ipgre_tap1)server
After reducing eth1's mtu on client, then perforamnce became 0.
This patch is to netif_keep_dst in gre_tap_init, as ipgre does.
Reported-by: Jianlin Shi <jishi@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
fib_check_nh does not use the fib_info arg; remove t.
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
fib_weight in fib_info is set but not used. Remove it and the
helpers for setting it.
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
With the commit 76174004a0 (tcp: do not slow start when cwnd equals
ssthresh), the comparison to the reduced cwnd in tcp_vegas_ssthresh() would
under-evaluate the ssthresh.
Signed-off-by: Hoang Tran <hoang.tran@uclouvain.be>
Signed-off-by: David S. Miller <davem@davemloft.net>
As measured in my prior patch ("sch_netem: faster rb tree removal"),
rbtree_postorder_for_each_entry_safe() is nice looking but much slower
than using rb_next() directly, except when tree is small enough
to fit in CPU caches (then the cost is the same)
From: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
My prior fix was not complete, as we were dereferencing a pointer
three times per node, not twice as I initially thought.
Fixes: 4cc5b44b29 ("inetpeer: fix RCU lookup()")
Fixes: b145425f26 ("inetpeer: remove AVL implementation in favor of RB tree")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the ipmr starts, it adds one default FIB rule that matches all packets
and sends them to the DEFAULT (multicast) FIB table. A more complex rule
can be added by user to specify that for a specific interface, a packet
should be look up at either an arbitrary table or according to the l3mdev
of the interface.
For drivers willing to offload the ipmr logic into a hardware but don't
want to offload all the FIB rules functionality, provide a function that
can indicate whether the FIB rule is the default multicast rule, thus only
one routing table is needed.
This way, a driver can register to the FIB notification chain, get
notifications about FIB rules added and trigger some kind of an internal
abort mechanism when a non default rule is added by the user.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow drivers, registered to the fib notification chain indicate whether a
multicast MFC route is offloaded or not, similarly to unicast routes. The
indication of whether a route is offloaded is done using the mfc_flags
field on an mfc_cache struct, and the information is sent to the userspace
via the RTNetlink interface only.
Currently, MFC routes are either offloaded or not, thus there is no need to
add per-VIF offload indication.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the newly introduced notification chain to send events upon VIF and MFC
addition and deletion. The MFC notifications are sent only on resolved MFC
entries, as unresolved cannot be offloaded.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make the ipmr module register as a FIB notifier. To do that, implement both
the ipmr_seq_read and ipmr_dump ops.
The ipmr_seq_read op returns a sequence counter that is incremented on
every notification related operation done by the ipmr. To implement that,
add a sequence counter in the netns_ipv4 struct and increment it whenever a
new MFC route or VIF are added or deleted. The sequence operations are
protected by the RTNL lock.
The ipmr_dump iterates the list of MFC routes and the list of VIF entries
and sends notifications about them. The entries dump is done under RCU
where the VIF dump uses the mrt_lock too, as the vif->dev field can change
under RCU.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Next commits will introduce MFC notifications through the atomic
fib_notification chain, thus allowing modules to be aware of MFC entries.
Due to the fact that modules may need to hold a reference to an MFC entry,
add reference count to MFC entries to prevent them from being freed while
these modules use them.
The reference counting is done only on resolved MFC entries currently.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When running LTP IPsec tests, KASan might report:
BUG: KASAN: use-after-free in vti_tunnel_xmit+0xeee/0xff0 [ip_vti]
Read of size 4 at addr ffff880dc6ad1980 by task swapper/0/0
...
Call Trace:
<IRQ>
dump_stack+0x63/0x89
print_address_description+0x7c/0x290
kasan_report+0x28d/0x370
? vti_tunnel_xmit+0xeee/0xff0 [ip_vti]
__asan_report_load4_noabort+0x19/0x20
vti_tunnel_xmit+0xeee/0xff0 [ip_vti]
? vti_init_net+0x190/0x190 [ip_vti]
? save_stack_trace+0x1b/0x20
? save_stack+0x46/0xd0
dev_hard_start_xmit+0x147/0x510
? icmp_echo.part.24+0x1f0/0x210
__dev_queue_xmit+0x1394/0x1c60
...
Freed by task 0:
save_stack_trace+0x1b/0x20
save_stack+0x46/0xd0
kasan_slab_free+0x70/0xc0
kmem_cache_free+0x81/0x1e0
kfree_skbmem+0xb1/0xe0
kfree_skb+0x75/0x170
kfree_skb_list+0x3e/0x60
__dev_queue_xmit+0x1298/0x1c60
dev_queue_xmit+0x10/0x20
neigh_resolve_output+0x3a8/0x740
ip_finish_output2+0x5c0/0xe70
ip_finish_output+0x4ba/0x680
ip_output+0x1c1/0x3a0
xfrm_output_resume+0xc65/0x13d0
xfrm_output+0x1e4/0x380
xfrm4_output_finish+0x5c/0x70
Can be fixed if we get skb->len before dst_output().
Fixes: b9959fd3b0 ("vti: switch to new ip tunnel code")
Fixes: 22e1b23daf ("vti6: Support inter address family tunneling.")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When doing my reuseport rework I screwed up and changed a
if (hlist_empty(&tb->owners))
to
if (!hlist_empty(&tb->owners))
This is obviously bad as all of the reuseport/reuse logic was reversed,
which caused weird problems like allowing an ipv4 bind conflict if we
opened an ipv4 only socket on a port followed by an ipv6 only socket on
the same port.
Fixes: b9470c2760 ("inet: kill smallest_size and smallest_port")
Reported-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In ipv6_rcv_saddr_equal() we need to use inet6_rcv_saddr(sk) for the
ipv6 compare with the fast socket information to make sure we're doing
the proper comparisons.
Fixes: 637bc8bbe6 ("inet: reset tb->fastreuseport when adding a reuseport sk")
Reported-and-tested-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need to set the tb->fast_sk_family properly so we can use the proper
comparison function for all subsequent reuseport bind requests.
Fixes: 637bc8bbe6 ("inet: reset tb->fastreuseport when adding a reuseport sk")
Reported-and-tested-by: Cole Robinson <crobinso@redhat.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since commit 1dced6a854 ("ipv4: Restore accept_local behaviour
in fib_validate_source()") a full fib lookup is needed even if
the rp_filter is disabled, if accept_local is false - which is
the default.
What we really need in the above scenario is just checking
that the source IP address is not local, and in most case we
can do that is a cheaper way looking up the ifaddr hash table.
This commit adds a helper for such lookup, and uses it to
validate the src address when rp_filter is disabled and no
'local' routes are created by the user space in the relevant
namespace.
A new ipv4 netns flag is added to account for such routes.
We need that to preserve the same behavior we had before this
patch.
It also drops the checks to bail early from __fib_validate_source,
added by the commit 1dced6a854 ("ipv4: Restore accept_local
behaviour in fib_validate_source()") they do not give any
measurable performance improvement: if we do the lookup with are
on a slower path.
This improves UDP performances for unconnected sockets
when rp_filter is disabled by 5% and also gives small but
measurable performance improvement for TCP flood scenarios.
v1 -> v2:
- use the ifaddr lookup helper in __ip_dev_find(), as suggested
by Eric
- fall-back to full lookup if custom local routes are present
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The commit 6b229cf77d ("udp: add batching to udp_rmem_release()")
reduced greatly the cacheline contention between the BH and the US
reader batching the rmem updates in most scenarios.
Such optimization is explicitly avoided if the US reader is faster
then BH processing.
My fault, I initially suggested this kind of behavior due to concerns
of possible regressions with small sk_rcvbuf values. Tests showed
such concerns are misplaced, so this commit relaxes the condition
for rmem bulk updates, obtaining small but measurable performance
gain in the scenario described above.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When dealing with a list of dismantling netns, we can scan
tcp_metrics once, saving cpu cycles.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Our recent change exposed a bug in TCP Fastopen Client that syzkaller
found right away [1]
When we prepare skb with SYN+DATA, we attempt to transmit it,
and we update socket state as if the transmit was a success.
In socket RTX queue we have two skbs, one with the SYN alone,
and a second one containing the DATA.
When (malicious) ACK comes in, we now complain that second one had no
skb_mstamp.
The proper fix is to make sure that if the transmit failed, we do not
pretend we sent the DATA skb, and make it our send_head.
When 3WHS completes, we can now send the DATA right away, without having
to wait for a timeout.
[1]
WARNING: CPU: 0 PID: 100189 at net/ipv4/tcp_input.c:3117 tcp_clean_rtx_queue+0x2057/0x2ab0 net/ipv4/tcp_input.c:3117()
WARN_ON_ONCE(last_ackt == 0);
Modules linked in:
CPU: 0 PID: 100189 Comm: syz-executor1 Not tainted
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
0000000000000000 ffff8800b35cb1d8 ffffffff81cad00d 0000000000000000
ffffffff828a4347 ffff88009f86c080 ffffffff8316eb20 0000000000000d7f
ffff8800b35cb220 ffffffff812c33c2 ffff8800baad2440 00000009d46575c0
Call Trace:
[<ffffffff81cad00d>] __dump_stack
[<ffffffff81cad00d>] dump_stack+0xc1/0x124
[<ffffffff812c33c2>] warn_slowpath_common+0xe2/0x150
[<ffffffff812c361e>] warn_slowpath_null+0x2e/0x40
[<ffffffff828a4347>] tcp_clean_rtx_queue+0x2057/0x2ab0 n
[<ffffffff828ae6fd>] tcp_ack+0x151d/0x3930
[<ffffffff828baa09>] tcp_rcv_state_process+0x1c69/0x4fd0
[<ffffffff828efb7f>] tcp_v4_do_rcv+0x54f/0x7c0
[<ffffffff8258aacb>] sk_backlog_rcv
[<ffffffff8258aacb>] __release_sock+0x12b/0x3a0
[<ffffffff8258ad9e>] release_sock+0x5e/0x1c0
[<ffffffff8294a785>] inet_wait_for_connect
[<ffffffff8294a785>] __inet_stream_connect+0x545/0xc50
[<ffffffff82886f08>] tcp_sendmsg_fastopen
[<ffffffff82886f08>] tcp_sendmsg+0x2298/0x35a0
[<ffffffff82952515>] inet_sendmsg+0xe5/0x520
[<ffffffff8257152f>] sock_sendmsg_nosec
[<ffffffff8257152f>] sock_sendmsg+0xcf/0x110
Fixes: 8c72c65b42 ("tcp: update skb->skb_mstamp more carefully")
Fixes: 783237e8da ("net-tcp: Fast Open client - sending SYN-data")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb->rbnode shares space with skb->next, skb->prev and skb->tstamp
Current uses (TCP receive ofo queue and netem) need to save/restore
tstamp, while skb->dev is either NULL (TCP) or a constant for a given
queue (netem).
Since we plan using an RB tree for TCP retransmit queue to speedup SACK
processing with large BDP, this patch exchanges skb->dev and
skb->tstamp.
This saves some overhead in both TCP and netem.
v2: removes the swtstamp field from struct tcp_skb_cb
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Soheil Hassas Yeganeh <soheil@google.com>
Cc: Wei Wang <weiwan@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
remove tcp_may_send_now and tcp_snd_test that are no longer used
Fixes: 840a3cbe89 ("tcp: remove forward retransmit feature")
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
