Its not enough to check for available tcp send space.

We also hold on to transmitted data for mptcp-level retransmits.
Right now we will send more and more data if the peer can ack data
at the tcp level fast enough, since that frees up tcp send buffer space.

But we also need to check that data was acked and reclaimed at the mptcp
level.

Therefore add needed check in mptcp_sendmsg, flush tcp data and
wait until more mptcp snd space becomes available if we are over the
limit.  Before we wait for more data, also make sure we start the
retransmit timer if we ran out of sndbuf space.

Otherwise there is a very small chance that we wait forever:

 * receiver is waiting for data
 * sender is blocked because mptcp socket buffer is full
 * at tcp level, all data was acked
 * mptcp-level snd_una was not updated, because last ack
   that acknowledged the last data packet carried an older
   MPTCP-ack.

Restarting the retransmit timer avoids this problem: if TCP
subflow is idle, data is retransmitted from the RTX queue.

New data will make the peer send a new, updated MPTCP-Ack.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Paolo noticed that ssk_check_wmem() has same pattern, so add/use
common helper for both places.
Suggested-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

RFC8684 allows to send 32-bit DATA_ACKs as long as the peer is not
sending 64-bit data-sequence numbers. The 64-bit DSN is only there for
extreme scenarios when a very high throughput subflow is combined with a
long-RTT subflow such that the high-throughput subflow wraps around the
32-bit sequence number space within an RTT of the high-RTT subflow.

It is thus a rare scenario and we should try to use the 32-bit DATA_ACK
instead as long as possible. It allows to reduce the TCP-option overhead
by 4 bytes, thus makes space for an additional SACK-block. It also makes
tcpdumps much easier to read when the DSN and DATA_ACK are both either
32 or 64-bit.
Signed-off-by: NChristoph Paasch <cpaasch@apple.com>
Reviewed-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, on MP_JOIN failure we reset the child
socket, but leave the request socket untouched.

tcp_check_req will deal with it according to the
'tcp_abort_on_overflow' sysctl value - by default the
req socket will stay alive.

The above leads to inconsistent behavior on MP JOIN
failure, and bad listener overflow accounting.

This patch addresses the issue leveraging the infrastructure
just introduced to ask the TCP stack to drop the req on
failure.

The child socket is not freed anymore by subflow_syn_recv_sock(),
instead it's moved to a dead state and will be disposed by the
next sock_put done by the TCP stack, so that listener overflow
accounting is not affected by MP JOIN failure.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Reviewed-by: NChristoph Paasch <cpaasch@apple.com>
Reviewed-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

MP_JOIN subflows must not land into the accept queue.
Currently tcp_check_req() calls an mptcp specific helper
to detect such scenario.

Such helper leverages the subflow context to check for
MP_JOIN subflows. We need to deal also with MP JOIN
failures, even when the subflow context is not available
due allocation failure.

A possible solution would be changing the syn_recv_sock()
signature to allow returning a more descriptive action/
error code and deal with that in tcp_check_req().

Since the above need is MPTCP specific, this patch instead
uses a TCP request socket hole to add a MPTCP specific flag.
Such flag is used by the MPTCP syn_recv_sock() to tell
tcp_check_req() how to deal with the request socket.

This change is a no-op for !MPTCP build, and makes the
MPTCP code simpler. It allows also the next patch to deal
correctly with MP JOIN failure.

v1 -> v2:
 - be more conservative on drop_req initialization (Mat)

RFC -> v1:
 - move the drop_req bit inside tcp_request_sock (Eric)
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Reviewed-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Reviewed-by: NChristoph Paasch <cpaasch@apple.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the other MPTCP-peer uses 32-bit data-sequence numbers, we rely on
map_seq to indicate how to expand to a 64-bit data-sequence number in
expand_seq() when receiving data.

For new subflows, this field is not initialized, thus results in an
"invalid" mapping being discarded.

Fix this by initializing map_seq upon subflow establishment time.

Fixes: f296234c ("mptcp: Add handling of incoming MP_JOIN requests")
Signed-off-by: NChristoph Paasch <cpaasch@apple.com>
Reviewed-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a subflow is created via mptcp_subflow_create_socket(),
a new 'struct socket' is allocated, with a new i_ino value.

When inspecting TCP sockets via the procfs and or the diag
interface, the above ones are not related to the process owning
the MPTCP master socket, even if they are a logical part of it
('ss -p' shows an empty process field)

Additionally, subflows created by the path manager get
the uid/gid from the running workqueue.

Subflows are part of the owning MPTCP master socket, let's
adjust the vfs info to reflect this.

After this patch, 'ss' correctly displays subflows as belonging
to the msk socket creator.

Fixes: 2303f994 ("mptcp: Associate MPTCP context with TCP socket")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

tcp_v{4,6}_syn_recv_sock() set 'own_req' only when returning
a not NULL 'child', let's check 'own_req' only if child is
available to avoid an - unharmful - UBSAN splat.

v1 -> v2:
 - reference the correct hash

Fixes: 4c8941de ("mptcp: avoid flipping mp_capable field in syn_recv_sock()")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When parsing MPC+data packets we set the dss field, so
we must also initialize the data_fin, or we can find stray
value there.

Fixes: 9a19371b ("mptcp: fix data_fin handing in RX path")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The mentioned RX option field is initialized only for DSS
packet, we must access it only if 'dss' is set too, or
the subflow will end-up in a bad status, leading to
RFC violations.

Fixes: d22f4988 ("mptcp: process MP_CAPABLE data option")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Syzcaller has found a way to trigger the WARN_ON_ONCE condition
in check_fully_established().

The root cause is a legit fallback to TCP scenario, so replace
the WARN with a plain message on a more strict condition.

Fixes: f296234c ("mptcp: Add handling of incoming MP_JOIN requests")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The mptcp_options_received structure carries several per
packet flags (mp_capable, mp_join, etc.). Such fields must
be cleared on each packet, even on dropped ones or packet
not carrying any MPTCP options, but the current mptcp
code clears them only on TCP option reset.

On several races/corner cases we end-up with stray bits in
incoming options, leading to WARN_ON splats. e.g.:

[  171.164906] Bad mapping: ssn=32714 map_seq=1 map_data_len=32713
[  171.165006] WARNING: CPU: 1 PID: 5026 at net/mptcp/subflow.c:533 warn_bad_map (linux-mptcp/net/mptcp/subflow.c:533 linux-mptcp/net/mptcp/subflow.c:531)
[  171.167632] Modules linked in: ip6_vti ip_vti ip_gre ipip sit tunnel4 ip_tunnel geneve ip6_udp_tunnel udp_tunnel macsec macvtap tap ipvlan macvlan 8021q garp mrp xfrm_interface veth netdevsim nlmon dummy team bonding vcan bridge stp llc ip6_gre gre ip6_tunnel tunnel6 tun binfmt_misc intel_rapl_msr intel_rapl_common rfkill kvm_intel kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel joydev virtio_balloon pcspkr i2c_piix4 sunrpc ip_tables xfs libcrc32c crc32c_intel serio_raw virtio_console ata_generic virtio_blk virtio_net net_failover failover ata_piix libata
[  171.199464] CPU: 1 PID: 5026 Comm: repro Not tainted 5.7.0-rc1.mptcp_f227fdf5d388+ #95
[  171.200886] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-2.fc30 04/01/2014
[  171.202546] RIP: 0010:warn_bad_map (linux-mptcp/net/mptcp/subflow.c:533 linux-mptcp/net/mptcp/subflow.c:531)
[  171.206537] Code: c1 ea 03 0f b6 14 02 48 89 f8 83 e0 07 83 c0 03 38 d0 7c 04 84 d2 75 1d 8b 55 3c 44 89 e6 48 c7 c7 20 51 13 95 e8 37 8b 22 fe <0f> 0b 48 83 c4 08 5b 5d 41 5c c3 89 4c 24 04 e8 db d6 94 fe 8b 4c
[  171.220473] RSP: 0018:ffffc90000150560 EFLAGS: 00010282
[  171.221639] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
[  171.223108] RDX: 0000000000000000 RSI: 0000000000000008 RDI: fffff5200002a09e
[  171.224388] RBP: ffff8880aa6e3c00 R08: 0000000000000001 R09: fffffbfff2ec9955
[  171.225706] R10: ffffffff9764caa7 R11: fffffbfff2ec9954 R12: 0000000000007fca
[  171.227211] R13: ffff8881066f4a7f R14: ffff8880aa6e3c00 R15: 0000000000000020
[  171.228460] FS:  00007f8623719740(0000) GS:ffff88810be00000(0000) knlGS:0000000000000000
[  171.230065] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  171.231303] CR2: 00007ffdab190a50 CR3: 00000001038ea006 CR4: 0000000000160ee0
[  171.232586] Call Trace:
[  171.233109]  <IRQ>
[  171.233531] get_mapping_status (linux-mptcp/net/mptcp/subflow.c:691)
[  171.234371] mptcp_subflow_data_available (linux-mptcp/net/mptcp/subflow.c:736 linux-mptcp/net/mptcp/subflow.c:832)
[  171.238181] subflow_state_change (linux-mptcp/net/mptcp/subflow.c:1085 (discriminator 1))
[  171.239066] tcp_fin (linux-mptcp/net/ipv4/tcp_input.c:4217)
[  171.240123] tcp_data_queue (linux-mptcp/./include/linux/compiler.h:199 linux-mptcp/net/ipv4/tcp_input.c:4822)
[  171.245083] tcp_rcv_established (linux-mptcp/./include/linux/skbuff.h:1785 linux-mptcp/./include/net/tcp.h:1774 linux-mptcp/./include/net/tcp.h:1847 linux-mptcp/net/ipv4/tcp_input.c:5238 linux-mptcp/net/ipv4/tcp_input.c:5730)
[  171.254089] tcp_v4_rcv (linux-mptcp/./include/linux/spinlock.h:393 linux-mptcp/net/ipv4/tcp_ipv4.c:2009)
[  171.258969] ip_protocol_deliver_rcu (linux-mptcp/net/ipv4/ip_input.c:204 (discriminator 1))
[  171.260214] ip_local_deliver_finish (linux-mptcp/./include/linux/rcupdate.h:651 linux-mptcp/net/ipv4/ip_input.c:232)
[  171.261389] ip_local_deliver (linux-mptcp/./include/linux/netfilter.h:307 linux-mptcp/./include/linux/netfilter.h:301 linux-mptcp/net/ipv4/ip_input.c:252)
[  171.265884] ip_rcv (linux-mptcp/./include/linux/netfilter.h:307 linux-mptcp/./include/linux/netfilter.h:301 linux-mptcp/net/ipv4/ip_input.c:539)
[  171.273666] process_backlog (linux-mptcp/./include/linux/rcupdate.h:651 linux-mptcp/net/core/dev.c:6135)
[  171.275328] net_rx_action (linux-mptcp/net/core/dev.c:6572 linux-mptcp/net/core/dev.c:6640)
[  171.280472] __do_softirq (linux-mptcp/./arch/x86/include/asm/jump_label.h:25 linux-mptcp/./include/linux/jump_label.h:200 linux-mptcp/./include/trace/events/irq.h:142 linux-mptcp/kernel/softirq.c:293)
[  171.281379] do_softirq_own_stack (linux-mptcp/arch/x86/entry/entry_64.S:1083)
[  171.282358]  </IRQ>

We could address the issue clearing explicitly the relevant fields
in several places - tcp_parse_option, tcp_fast_parse_options,
possibly others.

Instead we move the MPTCP option parsing into the already existing
mptcp ingress hook, so that we need to clear the fields in a single
place.

This allows us dropping an MPTCP hook from the TCP code and
removing the quite large mptcp_options_received from the tcp_sock
struct. On the flip side, the MPTCP sockets will traverse the
option space twice (in tcp_parse_option() and in
mptcp_incoming_options(). That looks acceptable: we already
do that for syn and 3rd ack packets, plain TCP socket will
benefit from it, and even MPTCP sockets will experience better
code locality, reducing the jumps between TCP and MPTCP code.

v1 -> v2:
 - rebased on current '-net' tree

Fixes: 648ef4b8 ("mptcp: Implement MPTCP receive path")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently the MPTCP code uses 2 hooks to process syn-ack
packets, mptcp_rcv_synsent() and the sk_rx_dst_set()
callback.

We can drop the first, moving the relevant code into the
latter, reducing the hooking into the TCP code. This is
also needed by the next patch.

v1 -> v2:
 - use local tcp sock ptr instead of casting the sk variable
   several times - DaveM
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Paolo points out that mptcp_disconnect is bogus:
"lock_sock(sk);
looks suspicious (lock should be already held by the caller)
And call to: tcp_disconnect(sk, flags); too, sk is not a tcp
socket".

->disconnect() gets called from e.g. inet_stream_connect when
one tries to disassociate a connected socket again (to re-connect
without closing the socket first).
MPTCP however uses mptcp_stream_connect, not inet_stream_connect,
for the mptcp-socket connect call.

inet_stream_connect only gets called indirectly, for the tcp socket,
so any ->disconnect() calls end up calling tcp_disconnect for that
tcp subflow sk.

This also explains why syzkaller has not yet reported a problem
here.  So for now replace this with a stub that doesn't do anything.

Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/14Acked-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently subflow_finish_connect() changes unconditionally
any msk socket status other than TCP_ESTABLISHED.

If an unblocking connect() races with close(), we can end-up
triggering:

IPv4: Attempt to release TCP socket in state 1 00000000e32b8b7e

when the msk socket is disposed.

Be sure to enter the established status only from SYN_SENT.

Fixes: c3c123d1 ("net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Reviewed-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In MPTCP, the receive window is shared across all subflows, because it
refers to the mptcp-level sequence space.

MPTCP receivers already place incoming packets on the mptcp socket
receive queue and will charge it to the mptcp socket rcvbuf until
userspace consumes the data.

Update __tcp_select_window to use the occupancy of the parent/mptcp
socket instead of the subflow socket in case the tcp socket is part
of a logical mptcp connection.

This commit doesn't change choice of initial window for passive or active
connections.
While it would be possible to change those as well, this adds complexity
(especially when handling MP_JOIN requests).  Furthermore, the MPTCP RFC
specifically says that a MPTCP sender 'MUST NOT use the RCV.WND field
of a TCP segment at the connection level if it does not also carry a DSS
option with a Data ACK field.'

SYN/SYNACK packets do not carry a DSS option with a Data ACK field.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Normal there should be checked for nla_put_in6_addr like other
usage in net.

Detected by CoverityScan, CID# 1461639

Fixes: 01cacb00 ("mptcp: add netlink-based PM")
Signed-off-by: NBo YU <tsu.yubo@gmail.com>
Acked-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The data fin flag is set only via a DSS option, but
mptcp_incoming_options() copies it unconditionally from the
provided RX options.

Since we do not clear all the mptcp sock RX options in a
socket free/alloc cycle, we can end-up with a stray data_fin
value while parsing e.g. MPC packets.

That would lead to mapping data corruption and will trigger
a few WARN_ON() in the RX path.

Instead of adding a costly memset(), fetch the data_fin flag
only for DSS packets - when we always explicitly initialize
such bit at option parsing time.

Fixes: 648ef4b8 ("mptcp: Implement MPTCP receive path")
Reviewed-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We don't need them, as we can use the current ingress opt
data instead. Setting them in syn_recv_sock() may causes
inconsistent mptcp socket status, as per previous commit.

Fixes: cc7972ea ("mptcp: parse and emit MP_CAPABLE option according to v1 spec")
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If multiple CPUs races on the same req_sock in syn_recv_sock(),
flipping such field can cause inconsistent child socket status.

When racing, the CPU losing the req ownership may still change
the mptcp request socket mp_capable flag while the CPU owning
the request is cloning the socket, leaving the child socket with
'is_mptcp' set but no 'mp_capable' flag.

Such socket will stay with 'conn' field cleared, heading to oops
in later mptcp callback.

Address the issue tracking the fallback status in a local variable.

Fixes: 58b09919 ("mptcp: create msk early")
Co-developed-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Following splat can occur during self test:

 BUG: KASAN: use-after-free in subflow_data_ready+0x156/0x160
 Read of size 8 at addr ffff888100c35c28 by task mptcp_connect/4808

  subflow_data_ready+0x156/0x160
  tcp_child_process+0x6a3/0xb30
  tcp_v4_rcv+0x2231/0x3730
  ip_protocol_deliver_rcu+0x5c/0x860
  ip_local_deliver_finish+0x220/0x360
  ip_local_deliver+0x1c8/0x4e0
  ip_rcv_finish+0x1da/0x2f0
  ip_rcv+0xd0/0x3c0
  __netif_receive_skb_one_core+0xf5/0x160
  __netif_receive_skb+0x27/0x1c0
  process_backlog+0x21e/0x780
  net_rx_action+0x35f/0xe90
  do_softirq+0x4c/0x50
  [..]

This occurs when accessing subflow_ctx->conn.

Problem is that tcp_child_process() calls listen sockets'
sk_data_ready() notification, but it doesn't hold the listener
lock.  Another cpu calling close() on the listener will then cause
transition of refcount to 0.

Fixes: 58b09919 ("mptcp: create msk early")
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We need to set sk_state to CLOSED, else we will get following:

IPv4: Attempt to release TCP socket in state 3 00000000b95f109e
IPv4: Attempt to release TCP socket in state 10 00000000b95f109e

First one is from inet_sock_destruct(), second one from
mptcp_sk_clone failure handling.  Setting sk_state to CLOSED isn't
enough, we also need to orphan sk so it has DEAD flag set.
Otherwise, a very similar warning is printed from inet_sock_destruct().
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Following snippet (replicated from syzkaller reproducer) generates
warning: "IPv4: Attempt to release TCP socket in state 1".

int main(void) {
 struct sockaddr_in sin1 = { .sin_family = 2, .sin_port = 0x4e20,
                             .sin_addr.s_addr = 0x010000e0, };
 struct sockaddr_in sin2 = { .sin_family = 2,
	                     .sin_addr.s_addr = 0x0100007f, };
 struct sockaddr_in sin3 = { .sin_family = 2, .sin_port = 0x4e20,
	                     .sin_addr.s_addr = 0x0100007f, };
 int r0 = socket(0x2, 0x1, 0x106);
 int r1 = socket(0x2, 0x1, 0x106);

 bind(r1, (void *)&sin1, sizeof(sin1));
 connect(r1, (void *)&sin2, sizeof(sin2));
 listen(r1, 3);
 return connect(r0, (void *)&sin3, 0x4d);
}

Reason is that the newly generated mptcp socket is closed via the ulp
release of the tcp listener socket when its accept backlog gets purged.

To fix this, delay setting the ESTABLISHED state until after userspace
calls accept and via mptcp specific destructor.

Fixes: 58b09919 ("mptcp: create msk early")
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/9Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

mptcp_connect/28740 is trying to release lock (sk_lock-AF_INET) at:
[<ffffffff82c15869>] mptcp_poll+0xb9/0x550
but there are no more locks to release!
Call Trace:
 lock_release+0x50f/0x750
 release_sock+0x171/0x1b0
 mptcp_poll+0xb9/0x550
 sock_poll+0x157/0x470
 ? get_net_ns+0xb0/0xb0
 do_sys_poll+0x63c/0xdd0

Problem is that __mptcp_tcp_fallback() releases the mptcp socket lock,
but after recent change it doesn't do this in all of its return paths.

To fix this, remove the unlock from __mptcp_tcp_fallback() and
always do the unlock in the caller.

Also add a small comment as to why we have this
__mptcp_needs_tcp_fallback().

Fixes: 0b4f33de ("mptcp: fix tcp fallback crash")
Reported-by: syzbot+e56606435b7bfeea8cf5@syzkaller.appspotmail.com
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some of the mptcp logs didn't print out the format string:

[  185.651493] DSS
[  185.651494] data_fin=0 dsn64=0 use_map=0 ack64=1 use_ack=1
[  185.651494] data_ack=13792750332298763796
[  185.651495] MPTCP: msk=00000000c4b81cfc ssk=000000009743af53 data_avail=0 skb=0000000063dc595d
[  185.651495] MPTCP: msk=00000000c4b81cfc ssk=000000009743af53 status=0
[  185.651495] MPTCP: msk ack_seq=9bbc894565aa2f9a subflow ack_seq=9bbc894565aa2f9a
[  185.651496] MPTCP: msk=00000000c4b81cfc ssk=000000009743af53 data_avail=1 skb=0000000012e809e1

So this patch added these missing pr_fmt defines. Then we can get the same
format string "MPTCP" in all mptcp logs like this:

[  142.795829] MPTCP: DSS
[  142.795829] MPTCP: data_fin=0 dsn64=0 use_map=0 ack64=1 use_ack=1
[  142.795829] MPTCP: data_ack=8089704603109242421
[  142.795830] MPTCP: msk=00000000133a24e0 ssk=000000002e508c64 data_avail=0 skb=00000000d5f230df
[  142.795830] MPTCP: msk=00000000133a24e0 ssk=000000002e508c64 status=0
[  142.795831] MPTCP: msk ack_seq=66790290f1199d9b subflow ack_seq=66790290f1199d9b
[  142.795831] MPTCP: msk=00000000133a24e0 ssk=000000002e508c64 data_avail=1 skb=00000000de5aca2e
Signed-off-by: NGeliang Tang <geliangtang@gmail.com>
Reviewed-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Obtained with:

  $ make W=1 net/mptcp/token.o
  net/mptcp/token.c:53: warning: Function parameter or member 'req' not described in 'mptcp_token_new_request'
  net/mptcp/token.c:98: warning: Function parameter or member 'sk' not described in 'mptcp_token_new_connect'
  net/mptcp/token.c:133: warning: Function parameter or member 'conn' not described in 'mptcp_token_new_accept'
  net/mptcp/token.c:178: warning: Function parameter or member 'token' not described in 'mptcp_token_destroy_request'
  net/mptcp/token.c:191: warning: Function parameter or member 'token' not described in 'mptcp_token_destroy'

Fixes: 79c0949e (mptcp: Add key generation and token tree)
Fixes: 58b09919 (mptcp: create msk early)
Signed-off-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

mptcp_subflow_data_available() is commonly called via
ssk->sk_data_ready(), in this case the mptcp socket lock
cannot be acquired.

Therefore, while we can safely discard subflow data that
was already received up to msk->ack_seq, we cannot be sure
that 'subflow->data_avail' will still be valid at the time
userspace wants to read the data -- a previous read on a
different subflow might have carried this data already.

In that (unlikely) event, msk->ack_seq will have been updated
and will be ahead of the subflow dsn.

We can check for this condition and skip/resync to the expected
sequence number.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is needed at least until proper MPTCP-Level fin/reset
signalling gets added:

We wake parent when a subflow changes, but we should do this only
when all subflows have closed, not just one.

Schedule the mptcp worker and tell it to check eof state on all
subflows.

Only flag mptcp socket as closed and wake userspace processes blocking
in poll if all subflows have closed.
Co-developed-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Christoph Paasch reports following crash:

general protection fault [..]
CPU: 0 PID: 2874 Comm: syz-executor072 Not tainted 5.6.0-rc5 #62
RIP: 0010:__pv_queued_spin_lock_slowpath kernel/locking/qspinlock.c:471
[..]
 queued_spin_lock_slowpath arch/x86/include/asm/qspinlock.h:50 [inline]
 do_raw_spin_lock include/linux/spinlock.h:181 [inline]
 spin_lock_bh include/linux/spinlock.h:343 [inline]
 __mptcp_flush_join_list+0x44/0xb0 net/mptcp/protocol.c:278
 mptcp_shutdown+0xb3/0x230 net/mptcp/protocol.c:1882
[..]

Problem is that mptcp_shutdown() socket isn't an mptcp socket,
its a plain tcp_sk.  Thus, trying to access mptcp_sk specific
members accesses garbage.

Root cause is that accept() returns a fallback (tcp) socket, not an mptcp
one.  There is code in getpeername to detect this and override the sockets
stream_ops.  But this will only run when accept() caller provided a
sockaddr struct.  "accept(fd, NULL, 0)" will therefore result in
mptcp stream ops, but with sock->sk pointing at a tcp_sk.

Update the existing fallback handling to detect this as well.

Moreover, mptcp_shutdown did not have fallback handling, and
mptcp_poll did it too late so add that there as well.
Reported-by: NChristoph Paasch <cpaasch@apple.com>
Tested-by: NChristoph Paasch <cpaasch@apple.com>
Reviewed-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Expose a new netlink family to userspace to control the PM, setting:

 - list of local addresses to be signalled.
 - list of local addresses used to created subflows.
 - maximum number of add_addr option to react

When the msk is fully established, the PM netlink attempts to
announce the 'signal' list via the ADD_ADDR option. Since we
currently lack the ADD_ADDR echo (and related event) only the
first addr is sent.

After exhausting the 'announce' list, the PM tries to create
subflow for each addr in 'local' list, waiting for each
connection to be completed before attempting the next one.

Idea is to add an additional PM hook for ADD_ADDR echo, to allow
the PM netlink announcing multiple addresses, in sequence.
Co-developed-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Exported via same /proc file as the Linux TCP MIB counters, so "netstat -s"
or "nstat" will show them automatically.

The MPTCP MIB counters are allocated in a distinct pcpu area in order to
avoid bloating/wasting TCP pcpu memory.

Counters are allocated once the first MPTCP socket is created in a
network namespace and free'd on exit.

If no sockets have been allocated, all-zero mptcp counters are shown.

The MIB counter list is taken from the multipath-tcp.org kernel, but
only a few counters have been picked up so far.  The counter list can
be increased at any time later on.

v2 -> v3:
 - remove 'inline' in foo.c files (David S. Miller)
Co-developed-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

add ulp-specific diagnostic functions, so that subflow information can be
dumped to userspace programs like 'ss'.

v2 -> v3:
- uapi: use bit macros appropriate for userspace
Co-developed-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: NMatthieu Baerts <matthieu.baerts@tessares.net>
Co-developed-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavide Caratti <dcaratti@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

On timeout event, schedule a work queue to do the retransmission.
Retransmission code closely resembles the sendmsg() implementation and
re-uses mptcp_sendmsg_frag, providing a dummy msghdr - for flags'
sake - and peeking the relevant dfrag from the rtx head.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This will simplify mptcp-level retransmission implementation
in the next patch. If dfrag is provided by the caller, skip
kernel space memory allocation and use data and metadata
provided by the dfrag itself.

Because a peer could ack data at TCP level but refrain from
sending mptcp-level ACKs, we could grow the mptcp socket
backlog indefinitely.

We should thus block mptcp_sendmsg until the peer has acked some of the
sent data.

In order to be able to do so, increment the mptcp socket wmem_queued
counter on memory allocation and decrement it when releasing the memory
on mptcp-level ack reception.

Because TCP performns sndbuf auto-tuning up to tcp_wmem_max[2], make
this the mptcp sk_sndbuf limit.

In the future we could add experiment with autotuning as TCP does in
tcp_sndbuf_expand().

v2 -> v3:
 - remove 'inline' in foo.c files (David S. Miller)
Co-developed-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

After adding wmem accounting for the mptcp socket we could get
into a situation where the mptcp socket can't transmit more data,
and mptcp_clean_una doesn't reduce wmem even if snd_una has advanced
because it currently will only remove entire dfrags.

Allow advancing the dfrag head sequence and reduce wmem,
even though this isn't correct (as we can't release the page).

Because we will soon block on mptcp sk in case wmem is too large,
call sk_stream_write_space() in case we reduced the backlog so
userspace task blocked in sendmsg or poll will be woken up.

This isn't an issue if the send buffer is large, but it is when
SO_SNDBUF is used to reduce it to a lower value.

Note we can still get a deadlock for low SO_SNDBUF values in
case both sides of the connection write to the socket: both could
be blocked due to wmem being too small -- and current mptcp stack
will only increment mptcp ack_seq on recv.

This doesn't happen with the selftest as it uses poll() and
will always call recv if there is data to read.
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Charge the data on the rtx queue to the master MPTCP socket, too.
Such memory in uncharged when the data is acked/dequeued.

Also account mptcp sockets inuse via a protocol specific pcpu
counter.
Co-developed-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The timer will be used to schedule retransmission. It's
frequency is based on the current subflow RTO estimation and
is reset on every una_seq update

The timer is clearer for good by __mptcp_clear_xmit()

Also clean MPTCP rtx queue before each transmission.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Keep the send page fragment on an MPTCP level retransmission queue.
The queue entries are allocated inside the page frag allocator,
acquiring an additional reference to the page for each list entry.

Also switch to a custom page frag refill function, to ensure that
the current page fragment can always host an MPTCP rtx queue entry.

The MPTCP rtx queue is flushed at disconnect() and close() time

Note that now we need to call __mptcp_init_sock() regardless of mptcp
enable status, as the destructor will try to walk the rtx_queue.

v2 -> v3:
 - remove 'inline' in foo.c files (David S. Miller)
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

So that we keep per unacked sequence number consistent; since
we update per msk data, use an atomic64 cmpxchg() to protect
against concurrent updates from multiple subflows.

Initialize the snd_una at connect()/accept() time.
Co-developed-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fill in more path manager functionality by adding a worker function and
modifying the related stub functions to schedule the worker.
Co-developed-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Co-developed-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NPeter Krystad <peter.krystad@linux.intel.com>
Signed-off-by: NMat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>