The sockmap feature allows bpf syscall from userspace, or based
on bpf sockops, replacing the sk_prot of sockets during protocol stack
processing with sockmap's custom read/write interfaces.
'''
tcp_rcv_state_process()
syn_recv_sock()/subflow_syn_recv_sock()
tcp_init_transfer(BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB)
bpf_skops_established <== sockops
bpf_sock_map_update(sk) <== call bpf helper
tcp_bpf_update_proto() <== update sk_prot
'''
When the server has MPTCP enabled but the client sends a TCP SYN
without MPTCP, subflow_syn_recv_sock() performs a fallback on the
subflow, replacing the subflow sk's sk_prot with the native sk_prot.
'''
subflow_syn_recv_sock()
subflow_ulp_fallback()
subflow_drop_ctx()
mptcp_subflow_ops_undo_override()
'''
Then, this subflow can be normally used by sockmap, which replaces the
native sk_prot with sockmap's custom sk_prot. The issue occurs when the
user executes accept::mptcp_stream_accept::mptcp_fallback_tcp_ops().
Here, it uses sk->sk_prot to compare with the native sk_prot, but this
is incorrect when sockmap is used, as we may incorrectly set
sk->sk_socket->ops.
This fix uses the more generic sk_family for the comparison instead.
Additionally, this also prevents a WARNING from occurring:
result from ./scripts/decode_stacktrace.sh:
------------[ cut here ]------------
WARNING: CPU: 0 PID: 337 at net/mptcp/protocol.c:68 mptcp_stream_accept \
(net/mptcp/protocol.c:4005)
Modules linked in:
...
PKRU: 55555554
Call Trace:
<TASK>
do_accept (net/socket.c:1989)
__sys_accept4 (net/socket.c:2028 net/socket.c:2057)
__x64_sys_accept (net/socket.c:2067)
x64_sys_call (arch/x86/entry/syscall_64.c:41)
do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94)
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
RIP: 0033:0x7f87ac92b83d
---[ end trace 0000000000000000 ]---
Fixes: 0b4f33def7 ("mptcp: fix tcp fallback crash")
Signed-off-by: Jiayuan Chen <jiayuan.chen@linux.dev>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Cc: <stable@vger.kernel.org>
Link: https://patch.msgid.link/20251111060307.194196-3-jiayuan.chen@linux.dev
The sockmap feature allows bpf syscall from userspace, or based on bpf
sockops, replacing the sk_prot of sockets during protocol stack processing
with sockmap's custom read/write interfaces.
'''
tcp_rcv_state_process()
subflow_syn_recv_sock()
tcp_init_transfer(BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB)
bpf_skops_established <== sockops
bpf_sock_map_update(sk) <== call bpf helper
tcp_bpf_update_proto() <== update sk_prot
'''
Consider two scenarios:
1. When the server has MPTCP enabled and the client also requests MPTCP,
the sk passed to the BPF program is a subflow sk. Since subflows only
handle partial data, replacing their sk_prot is meaningless and will
cause traffic disruption.
2. When the server has MPTCP enabled but the client sends a TCP SYN
without MPTCP, subflow_syn_recv_sock() performs a fallback on the
subflow, replacing the subflow sk's sk_prot with the native sk_prot.
'''
subflow_ulp_fallback()
subflow_drop_ctx()
mptcp_subflow_ops_undo_override()
'''
Subsequently, accept::mptcp_stream_accept::mptcp_fallback_tcp_ops()
converts the subflow to plain TCP.
For the first case, we should prevent it from being combined with sockmap
by setting sk_prot->psock_update_sk_prot to NULL, which will be blocked by
sockmap's own flow.
For the second case, since subflow_syn_recv_sock() has already restored
sk_prot to native tcp_prot/tcpv6_prot, no further action is needed.
Fixes: cec37a6e41 ("mptcp: Handle MP_CAPABLE options for outgoing connections")
Signed-off-by: Jiayuan Chen <jiayuan.chen@linux.dev>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Cc: <stable@vger.kernel.org>
Link: https://patch.msgid.link/20251111060307.194196-2-jiayuan.chen@linux.dev
Explicitly account for MPTCP-level zero windows probe, to catch
hopefully earlier issues alike the one addressed by the previous
patch.
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Tested-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20251028-net-mptcp-send-timeout-v1-4-38ffff5a9ec8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since commit 72377ab2d6 ("mptcp: more conservative check for zero
probes") the MPTCP-level zero window probe check is always disabled, as
the TCP-level write queue always contains at least the newly allocated
skb.
Refine the relevant check tacking in account that the above condition
and that such skb can have zero length.
Fixes: 72377ab2d6 ("mptcp: more conservative check for zero probes")
Cc: stable@vger.kernel.org
Reported-by: Geliang Tang <geliang@kernel.org>
Closes: https://lore.kernel.org/d0a814c364e744ca6b836ccd5b6e9146882e8d42.camel@kernel.org
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Tested-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20251028-net-mptcp-send-timeout-v1-3-38ffff5a9ec8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
If a MSG_PEEK | MSG_WAITALL read operation consumes all the bytes in the
receive queue and recvmsg() need to waits for more data - i.e. it's a
blocking one - upon arrival of the next packet the MPTCP protocol will
start again copying the oldest data present in the receive queue,
corrupting the data stream.
Address the issue explicitly tracking the peeked sequence number,
restarting from the last peeked byte.
Fixes: ca4fb89257 ("mptcp: add MSG_PEEK support")
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Geliang Tang <geliang@kernel.org>
Tested-by: Geliang Tang <geliang@kernel.org>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20251028-net-mptcp-send-timeout-v1-2-38ffff5a9ec8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Accessing the transmit queue without owning the msk socket lock is
inherently racy, hence __mptcp_check_push() could actually quit early
even when there is pending data.
That in turn could cause unexpected tx lock and timeout.
Dropping the early check avoids the race, implicitly relaying on later
tests under the relevant lock. With such change, all the other
mptcp_send_head() call sites are now under the msk socket lock and we
can additionally drop the now unneeded annotation on the transmit head
pointer accesses.
Fixes: 6e628cd3a8 ("mptcp: use mptcp release_cb for delayed tasks")
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Geliang Tang <geliang@kernel.org>
Tested-by: Geliang Tang <geliang@kernel.org>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20251028-net-mptcp-send-timeout-v1-1-38ffff5a9ec8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch has no functional change, and prepares the following one.
tcp_rcvbuf_grow() will need to have access to tp->rcvq_space.space
old and new values.
Change mptcp_rcvbuf_grow() in a similar way.
Signed-off-by: Eric Dumazet <edumazet@google.com>
[ Moved 'oldval' declaration to the next patch to avoid warnings at
build time. ]
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Link: https://patch.msgid.link/20251028-net-tcp-recv-autotune-v3-3-74b43ba4c84c@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The mptcp PM can add subflow to the conn_list before tcp_init_transfer().
Calling tcp_rcvbuf_grow() on such subflow is not correct as later
init will overwrite the update.
Fix the issue calling tcp_rcvbuf_grow() only after init buffer
initialization.
Fixes: e118cdc34d ("mptcp: rcvbuf auto-tuning improvement")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20251028-net-tcp-recv-autotune-v3-1-74b43ba4c84c@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The special C-flag case expects the ADD_ADDR to be received when
switching to 'fully-established'. But for various reasons, the ADD_ADDR
could be sent after the "4th ACK", and the special case doesn't work.
On NIPA, the new test validating this special case for the C-flag failed
a few times, e.g.
102 default limits, server deny join id 0
syn rx [FAIL] got 0 JOIN[s] syn rx expected 2
Server ns stats
(...)
MPTcpExtAddAddrTx 1
MPTcpExtEchoAdd 1
Client ns stats
(...)
MPTcpExtAddAddr 1
MPTcpExtEchoAddTx 1
synack rx [FAIL] got 0 JOIN[s] synack rx expected 2
ack rx [FAIL] got 0 JOIN[s] ack rx expected 2
join Rx [FAIL] see above
syn tx [FAIL] got 0 JOIN[s] syn tx expected 2
join Tx [FAIL] see above
I had a suspicion about what the issue could be: the ADD_ADDR might have
been received after the switch to the 'fully-established' state. The
issue was not easy to reproduce. The packet capture shown that the
ADD_ADDR can indeed be sent with a delay, and the client would not try
to establish subflows to it as expected.
A simple fix is not to mark the endpoints as 'used' in the C-flag case,
when looking at creating subflows to the remote initial IP address and
port. In this case, there is no need to try.
Note: newly added fullmesh endpoints will still continue to be used as
expected, thanks to the conditions behind mptcp_pm_add_addr_c_flag_case.
Fixes: 4b1ff850e0 ("mptcp: pm: in-kernel: usable client side with C-flag")
Cc: stable@vger.kernel.org
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20251020-net-mptcp-c-flag-late-add-addr-v1-1-8207030cb0e8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Such function is called only by __mptcp_data_ready(), which in turn
is always invoked when msk is not owned by the user: we can drop the
redundant, related check.
Additionally mptcp needs to propagate the socket error only for
current subflow.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Tested-by: Geliang Tang <geliang@kernel.org>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250927-net-next-mptcp-rcv-path-imp-v1-7-5da266aa9c1a@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The upcoming patch will introduced backlog processing for MPTCP
socket, and we want to leverage coalescing in such data path.
Factor out the relevant bits not touching memory accounting to
deal with such use-case.
Co-developed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Geliang Tang <geliang@kernel.org>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250927-net-next-mptcp-rcv-path-imp-v1-6-5da266aa9c1a@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since commit b7535cfed2 ("mptcp: drop legacy code around RX EOF"),
sk_shutdown can't change during the main recvmsg loop, we can drop
the related race breaker.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Tested-by: Geliang Tang <geliang@kernel.org>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250927-net-next-mptcp-rcv-path-imp-v1-5-5da266aa9c1a@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Factor out all the skb initialization step in a new helper and
use it. Note that this change moves the MPTCP CB initialization
earlier: we can do such step as soon as the skb leaves the
subflow socket receive queues.
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Geliang Tang <geliang@kernel.org>
Tested-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250927-net-next-mptcp-rcv-path-imp-v1-4-5da266aa9c1a@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Apply to the MPTCP auto-tuning the same improvements introduced for the
TCP protocol by the merge commit 2da35e4b4d ("Merge branch
'tcp-receive-side-improvements'").
The main difference is that TCP subflow and the main MPTCP socket need
to account separately for OoO: MPTCP does not care for TCP-level OoO
and vice versa, as a consequence do not reflect MPTCP-level rcvbuf
increase due to OoO packets at the subflow level.
This refeactor additionally allow dropping the msk receive buffer update
at receive time, as the latter only intended to cope with subflow receive
buffer increase due to OoO packets.
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/487
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/559
Reviewed-by: Geliang Tang <geliang@kernel.org>
Tested-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250927-net-next-mptcp-rcv-path-imp-v1-3-5da266aa9c1a@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Usage of the skb deferral API is straight-forward; with multiple
subflows actives this allow moving part of the received application
load into multiple CPUs.
Also fix a typo in the related comment.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Tested-by: Geliang Tang <geliang@kernel.org>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250927-net-next-mptcp-rcv-path-imp-v1-1-5da266aa9c1a@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently, upon the reception of an ADD_ADDR (and when the fullmesh flag
is not used), the in-kernel PM will create new subflows using the local
address the routing configuration will pick.
It would be easier to pick local addresses from a selected list of
endpoints, and use it only once, than relying on routing rules.
Use case: both the client (C) and the server (S) have two addresses (a
and b). The client establishes the connection between C(a) and S(a).
Once established, the server announces its additional address S(b). Once
received, the client connects to it using its second address C(b).
Compared to a situation without the 'laminar' endpoint for C(b), the
client didn't use this address C(b) to establish a subflow to the
server's primary address S(a). So at the end, we have:
C S
C(a) --- S(a)
C(b) --- S(b)
In case of a 3rd address on each side (C(c) and S(c)), upon the
reception of an ADD_ADDR with S(c), the client should not pick C(b)
because it has already been used. C(c) should then be used.
Note that this situation is currently possible if C doesn't add any
endpoint, but configure the routing in order to pick C(b) for the route
to S(b), and pick C(c) for the route to S(c). That doesn't sound very
practical because it means knowing in advance the IP addresses that
will be used and announced by the server.
'laminar', like the idea of laminar flows: the different subflows don't
mix with each other on an endpoint, unlike the "turbulent" way traffic
is mixed by 'fullmesh'.
In the code, the new endpoint type is added. Similar to the other
subflow types, an MPTCP_INFO counter is added. While at it, hole are now
commented in struct mptcp_info, to remember next time that these holes
can no longer be used.
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/503
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-15-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When receiving an ADD_ADDR right after the 3WHS, the connection will
switch to 'fully established'. It means the MPTCP worker will be called
to treat two events, in this order: ADD_ADDR_RECEIVED, PM_ESTABLISHED.
The MPTCP endpoints cannot have the ID 0, because it is reserved to the
address and port used by the initial subflow. To be able to deal with
this case in different places, msk->mpc_endpoint_id contains the
endpoint ID linked to the initial subflow. This variable was only set
when treating the first PM_ESTABLISHED event, after ADD_ADDR_RECEIVED.
That's why in fill_local_addresses_vec(), the endpoint addresses were
compared with the one of the initial subflow, instead of only comparing
the IDs.
Instead, msk->mpc_endpoint_id is now set when treating ADD_ADDR_RECEIVED
as well, if needed, then the IDs can be compared.
To be able to do so, the code doing that is now in a dedicated helper,
and called from the functions linked to the two actions.
While at it, mptcp_endp_get_local_id() has also been moved up, next to
this new helper, because they are linked, and to be able to use it in
fill_local_addresses_vec() in the next commit.
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-14-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
All the 'unsigned int' variables from the 'pm_nl_pernet' structure are
bounded to MPTCP_PM_ADDR_MAX, currently set to 8. The endpoint ID is
also bounded by the protocol to 8-bit. MPTCP_PM_ADDR_MAX, if extended
later, will never over 8-bit.
So no need to use 'unsigned int' variables, 'u8' is enough.
Note that the exposed counters in MPTCP_INFO are already limited to
8-bit, same for pm->extra_subflows, and others. So it seems even better
to limit them to 8-bit.
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-13-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
It is currently not used.
It was in fact never used since its introduction in commit ff5a0b421c
("mptcp: faster active backup recovery"). It was probably initially
added to struct pm_nl_pernet during the development of this commit,
before being added to struct mptcp_pernet in ctrl.c, but not removed
from the first place.
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-12-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A few variables linked to the in-kernel Path-Manager are confusing, and
it would help current and future developers, to clarify them.
One of them is 'addrs', which in fact represents the number of declared
endpoints, and not only the 'signal' endpoints.
No functional changes intended.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-11-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A few variables linked to the in-kernel Path-Manager are confusing, and
it would help current and future developers, to clarify them.
One of them is 'local_addr_list', which in fact represents the list of
endpoints, and not only the 'subflow' endpoints.
No functional changes intended.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-10-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A few variables linked to the in-kernel Path-Manager are confusing, and
it would help current and future developers, to clarify them.
One of them is 'local_addr_max', which in fact represents the maximum
number of 'subflow' endpoints that can be used to create new subflows,
and not the number of local addresses that have been used to create
subflows.
While at it, add an additional name for the corresponding variable in
MPTCP INFO: mptcpi_endp_subflow_max. Not to break the current uAPI, the
new name is added as a 'define' pointing to the former name. This will
then also help userspace devs.
Also move the variable and function next to the other 'endp_X_max' ones.
No functional changes intended.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-9-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A few variables linked to the in-kernel Path-Manager are confusing, and
it would help current and future developers, to clarify them.
One of them is 'add_addr_accept_max', which in fact represents the limit
of ADD_ADDR that can be accepted: the limit set via 'ip mptcp limit
add_addr_accepted X' for example. It is not linked to the maximum number
of accepted ADD_ADDR.
While at it, add an additional name for the corresponding variable in
MPTCP INFO: mptcpi_limit_add_addr_accepted. Not to break the current
uAPI, the new name is added as a 'define' pointing to the former name.
This will then also help userspace devs.
No functional changes intended.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-8-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A few variables linked to the in-kernel Path-Manager are confusing, and
it would help current and future developers, to clarify them.
One of them is 'add_addr_signal_max', which in fact represents the
maximum number of 'signal' endpoints that can be used to announced
addresses, and not the number of ADD_ADDR that can be signalled.
While at it, add an additional name for the corresponding variable in
MPTCP INFO: mptcpi_endp_signal_max. Not to break the current uAPI, the
new name is added as a 'define' pointing to the former name. This will
then also help userspace devs.
No functional changes intended.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-7-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A few variables linked to the in-kernel Path-Manager are confusing, and
it would help current and future developers, to clarify them.
One of them is 'subflows_max', which in fact represents the limit of
extra subflows: the limit set via 'ip mptcp limit subflows X' for
example. It is not linked to the maximum number of created / possible
subflows.
While at it, add an additional name for the corresponding variable in
MPTCP INFO: mptcpi_limit_extra_subflows. Not to break the current uAPI,
the new name is added as a 'define' pointing to the former name. This
will then also help userspace devs.
No functional changes intended.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-6-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A few variables linked to the Path-Managers are confusing, and it would
help current and future developers, to clarify them.
One of them is 'subflows', which in fact represents the number of extra
subflows: all the additional subflows created after the initial one, and
not the total number of subflows.
While at it, add an additional name for the corresponding variable in
MPTCP INFO: mptcpi_extra_subflows. Not to break the current uAPI, the
new name is added as a 'define' pointing to the former name. This will
then also help userspace devs.
No functional changes intended.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-5-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Before this modification, this function was quite long with many levels
of indentations.
Each case can be split in a dedicated function: fullmesh, non-fullmesh.
To remove one level of indentation, msk->pm.subflows >= subflows_max is
now checked after having added one subflow, and stops the loop if it is
no longer possible to add new subflows. This is fine to do this because
this function should only be called if msk->pm.subflows < subflows_max.
No functional changes intended.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-4-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Before this modification, this function was quite long with many levels
of indentations.
Each case can be split in a dedicated function: fullmesh, C flag, any.
No functional changes intended.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-3-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When servers set the C-flag in their MP_CAPABLE to tell clients not to
create subflows to the initial address and port, clients will likely not
use their other endpoints. That's because the in-kernel path-manager
uses the 'subflow' endpoints to create subflows only to the initial
address and port.
If the limits have not been modified to accept ADD_ADDR, the client
doesn't try to establish new subflows. If the limits accept ADD_ADDR,
the routing routes will be used to select the source IP.
The C-flag is typically set when the server is operating behind a legacy
Layer 4 load balancer, or using anycast IP address. Clients having their
different 'subflow' endpoints setup, don't end up creating multiple
subflows as expected, and causing some deployment issues.
A special case is then added here: when servers set the C-flag in the
MPC and directly sends an ADD_ADDR, this single ADD_ADDR is accepted.
The 'subflows' endpoints will then be used with this new remote IP and
port. This exception is only allowed when the ADD_ADDR is sent
immediately after the 3WHS, and makes the client switching to the 'fully
established' mode. After that, 'select_local_address()' will not be able
to find any subflows, because 'id_avail_bitmap' will be filled in
mptcp_pm_create_subflow_or_signal_addr(), when switching to 'fully
established' mode.
Fixes: df377be387 ("mptcp: add deny_join_id0 in mptcp_options_received")
Cc: stable@vger.kernel.org
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/536
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250925-net-next-mptcp-c-flag-laminar-v1-1-ad126cc47c6b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When working on a fix modifying mptcp_check_data_fin(), I noticed the
returned value was no longer used.
It looks like it was used for 3 days, between commit 7ed90803a2
("mptcp: send explicit ack on delayed ack_seq incr") and commit
ea4ca586b1 ("mptcp: refine MPTCP-level ack scheduling").
This returned value can be safely removed.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250919-net-next-mptcp-server-side-flag-v1-6-a97a5d561a8b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Now that such info is in the 'flags' attribute, it is time to deprecate
the dedicated 'server-side' attribute.
It will be removed in a few versions.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250919-net-next-mptcp-server-side-flag-v1-3-a97a5d561a8b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Now that the 'flags' attribute is used, it seems interesting to add one
flag for 'server-side', a boolean value.
This is duplicating the info from the dedicated 'server-side' attribute,
but it will be deprecated in the next commit, and removed in a few
versions.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250919-net-next-mptcp-server-side-flag-v1-2-a97a5d561a8b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This attribute is a boolean. No need to add it to set it to 'false'.
Indeed, the default value when this attribute is not set is naturally
'false'. A few bytes can then be saved by not adding this attribute if
the connection is not on the server side.
This prepares the future deprecation of its attribute, in favour of a
new flag.
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250919-net-next-mptcp-server-side-flag-v1-1-a97a5d561a8b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When a first MPTCP connection gets successfully established after a
blackhole period, 'active_disable_times' was supposed to be reset when
this connection was done via any non-loopback interfaces.
Unfortunately, the opposite condition was checked: only reset when the
connection was established via a loopback interface. Fixing this by
simply looking at the opposite.
This is similar to what is done with TCP FastOpen, see
tcp_fastopen_active_disable_ofo_check().
This patch is a follow-up of a previous discussion linked to commit
893c49a78d ("mptcp: Use __sk_dst_get() and dst_dev_rcu() in
mptcp_active_enable()."), see [1].
Fixes: 27069e7cb3 ("mptcp: disable active MPTCP in case of blackhole")
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/4209a283-8822-47bd-95b7-87e96d9b7ea3@kernel.org [1]
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20250918-net-next-mptcp-blackhole-reset-loopback-v1-1-bf5818326639@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cross-merge networking fixes after downstream PR (net-6.17-rc7).
No conflicts.
Adjacent changes:
drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
9536fbe10c ("net/mlx5e: Add PSP steering in local NIC RX")
7601a0a462 ("net/mlx5e: Add a miss level for ipsec crypto offload")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
mptcp_active_enable() is called from subflow_finish_connect(),
which is icsk->icsk_af_ops->sk_rx_dst_set() and it's not always
under RCU.
Using sk_dst_get(sk)->dev could trigger UAF.
Let's use __sk_dst_get() and dst_dev_rcu().
Fixes: 27069e7cb3 ("mptcp: disable active MPTCP in case of blackhole")
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250916214758.650211-8-kuniyu@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
mptcp_active_enable() calls sk_dst_get(), which returns dst with its
refcount bumped, but forgot dst_release().
Let's add missing dst_release().
Cc: stable@vger.kernel.org
Fixes: 27069e7cb3 ("mptcp: disable active MPTCP in case of blackhole")
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20250916214758.650211-7-kuniyu@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
As pointed out by Donald, when parsing an entry, the wrong type was set
for the temp value: this value is signed.
There are no real issues here, because the intermediate variable was
only wrong for the sign, not for the size, and the final variable had
the right sign. But this feels wrong, and is confusing, so fixing this
small typo introduced by commit ef0da3b8a2 ("mptcp: move address
attribute into mptcp_addr_info").
Reported-by: Donald Hunter <donald.hunter@gmail.com>
Closes: https://lore.kernel.org/m2plc0ui9z.fsf@gmail.com
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250912-net-next-mptcp-minor-fixes-6-18-v1-3-99d179b483ad@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When TFO is used, the check to see if the 'C' flag (deny join id0) was
set was bypassed.
This flag can be set when TFO is used, so the check should also be done
when TFO is used.
Note that the set_fully_established label is also used when a 4th ACK is
received. In this case, deny_join_id0 will not be set.
Fixes: dfc8d06030 ("mptcp: implement delayed seq generation for passive fastopen")
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250912-net-mptcp-pm-uspace-deny_join_id0-v1-4-40171884ade8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
During the connection establishment, a peer can tell the other one that
it cannot establish new subflows to the initial IP address and port by
setting the 'C' flag [1]. Doing so makes sense when the sender is behind
a strict NAT, operating behind a legacy Layer 4 load balancer, or using
anycast IP address for example.
When this 'C' flag is set, the path-managers must then not try to
establish new subflows to the other peer's initial IP address and port.
The in-kernel PM has access to this info, but the userspace PM didn't.
The RFC8684 [1] is strict about that:
(...) therefore the receiver MUST NOT try to open any additional
subflows toward this address and port.
So it is important to tell the userspace about that as it is responsible
for the respect of this flag.
When a new connection is created and established, the Netlink events
now contain the existing but not currently used 'flags' attribute. When
MPTCP_PM_EV_FLAG_DENY_JOIN_ID0 is set, it means no other subflows
to the initial IP address and port -- info that are also part of the
event -- can be established.
Link: https://datatracker.ietf.org/doc/html/rfc8684#section-3.1-20.6 [1]
Fixes: 702c2f646d ("mptcp: netlink: allow userspace-driven subflow establishment")
Reported-by: Marek Majkowski <marek@cloudflare.com>
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/532
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250912-net-mptcp-pm-uspace-deny_join_id0-v1-2-40171884ade8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When a SYN containing the 'C' flag (deny join id0) was received, this
piece of information was not propagated to the path-manager.
Even if this flag is mainly set on the server side, a client can also
tell the server it cannot try to establish new subflows to the client's
initial IP address and port. The server's PM should then record such
info when received, and before sending events about the new connection.
Fixes: df377be387 ("mptcp: add deny_join_id0 in mptcp_options_received")
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250912-net-mptcp-pm-uspace-deny_join_id0-v1-1-40171884ade8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When the MPTCP DATA FIN have been ACKed, there is no more MPTCP related
metadata to exchange, and all subflows can be safely shutdown.
Before this patch, the subflows were actually terminated at 'close()'
time. That's certainly fine most of the time, but not when the userspace
'shutdown()' a connection, without close()ing it. When doing so, the
subflows were staying in LAST_ACK state on one side -- and consequently
in FIN_WAIT2 on the other side -- until the 'close()' of the MPTCP
socket.
Now, when the DATA FIN have been ACKed, all subflows are shutdown. A
consequence of this is that the TCP 'FIN' flag can be set earlier now,
but the end result is the same. This affects the packetdrill tests
looking at the end of the MPTCP connections, but for a good reason.
Note that tcp_shutdown() will check the subflow state, so no need to do
that again before calling it.
Fixes: 3721b9b646 ("mptcp: Track received DATA_FIN sequence number and add related helpers")
Cc: stable@vger.kernel.org
Fixes: 16a9a9da17 ("mptcp: Add helper to process acks of DATA_FIN")
Reviewed-by: Mat Martineau <martineau@kernel.org>
Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250912-net-mptcp-fix-sft-connect-v1-1-d40e77cbbf02@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cross-merge networking fixes after downstream PR (net-6.17-rc6).
Conflicts:
net/netfilter/nft_set_pipapo.c
net/netfilter/nft_set_pipapo_avx2.c
c4eaca2e10 ("netfilter: nft_set_pipapo: don't check genbit from packetpath lookups")
84c1da7b38 ("netfilter: nft_set_pipapo: use avx2 algorithm for insertions too")
Only trivial adjacent changes (in a doc and a Makefile).
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently the ADD_ADDR option is retransmitted with a fixed timeout. This
patch makes the retransmission timeout adaptive by using the maximum RTO
among all the subflows, while still capping it at the configured maximum
value (add_addr_timeout_max). This improves responsiveness when
establishing new subflows.
Specifically:
1. Adds mptcp_adjust_add_addr_timeout() helper to compute the adaptive
timeout.
2. Uses maximum subflow RTO (icsk_rto) when available.
3. Applies exponential backoff based on retransmission count.
4. Maintains fallback to configured max timeout when no RTO data exists.
This slightly changes the behaviour of the MPTCP "add_addr_timeout"
sysctl knob to be used as a maximum instead of a fixed value. But this
is seen as an improvement: the ADD_ADDR might be sent quicker than
before to improve the overall MPTCP connection. Also, the default
value is set to 2 min, which was already way too long, and caused the
ADD_ADDR not to be retransmitted for connections shorter than 2 minutes.
Suggested-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/576
Reviewed-by: Christoph Paasch <cpaasch@openai.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250907-net-next-mptcp-add_addr-retrans-adapt-v1-1-824cc805772b@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Users reported a scenario where MPTCP connections that were configured
with SO_KEEPALIVE prior to connect would fail to enable their keepalives
if MTPCP fell back to TCP mode.
After investigating, this affects keepalives for any connection where
sync_socket_options is called on a socket that is in the closed or
listening state. Joins are handled properly. For connects,
sync_socket_options is called when the socket is still in the closed
state. The tcp_set_keepalive() function does not act on sockets that
are closed or listening, hence keepalive is not immediately enabled.
Since the SO_KEEPOPEN flag is absent, it is not enabled later in the
connect sequence via tcp_finish_connect. Setting the keepalive via
sockopt after connect does work, but would not address any subsequently
created flows.
Fortunately, the fix here is straight-forward: set SOCK_KEEPOPEN on the
subflow when calling sync_socket_options.
The fix was valdidated both by using tcpdump to observe keepalive
packets not being sent before the fix, and being sent after the fix. It
was also possible to observe via ss that the keepalive timer was not
enabled on these sockets before the fix, but was enabled afterwards.
Fixes: 1b3e7ede13 ("mptcp: setsockopt: handle SO_KEEPALIVE and SO_PRIORITY")
Cc: stable@vger.kernel.org
Signed-off-by: Krister Johansen <kjlx@templeofstupid.com>
Reviewed-by: Geliang Tang <geliang@kernel.org>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/aL8dYfPZrwedCIh9@templeofstupid.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use ARRAY_SIZE(), so that we know the limit at compile time.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Mat Martineau <martineau@kernel.org>
Cc: Geliang Tang <geliang@kernel.org>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250905165813.1470708-6-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Followup of commit c51da3f7a1 ("net: remove sock_i_uid()")
A recent syzbot report was the trigger for this change.
Over the years, we had many problems caused by the
read_lock[_bh](&sk->sk_callback_lock) in sock_i_uid().
We could fix smc_diag_dump_proto() or make a more radical move:
Instead of waiting for new syzbot reports, cache the socket
inode number in sk->sk_ino, so that we no longer
need to acquire sk->sk_callback_lock in sock_i_ino().
This makes socket dumps faster (one less cache line miss,
and two atomic ops avoided).
Prior art:
commit 25a9c8a443 ("netlink: Add __sock_i_ino() for __netlink_diag_dump().")
commit 4f9bf2a2f5 ("tcp: Don't acquire inet_listen_hashbucket::lock with disabled BH.")
commit efc3dbc374 ("rds: Make rds_sock_lock BH rather than IRQ safe.")
Fixes: d2d6422f8b ("x86: Allow to enable PREEMPT_RT.")
Reported-by: syzbot+50603c05bbdf4dfdaffa@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/netdev/68b73804.050a0220.3db4df.01d8.GAE@google.com/T/#u
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20250902183603.740428-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Accelerated Receive Flow Steering (aRFS) relies on sockets recording
their RX flow hash into the rps_sock_flow_table so that incoming packets
are steered to the CPU where the application runs.
With MPTCP, the application interacts with the parent MPTCP socket while
data is carried over per-subflow TCP sockets. Without recording these
subflows, aRFS cannot steer interrupts and RX processing for the flows
to the desired CPU.
Record all subflows in the RPS table by calling sock_rps_record_flow()
for each subflow at the start of mptcp_sendmsg(), mptcp_recvmsg() and
mptcp_stream_accept(), by using the new helper
mptcp_rps_record_subflows().
It does not by itself improve throughput, but ensures that IRQ and RX
processing are directed to the right CPU, which is a
prerequisite for effective aRFS.
Signed-off-by: Christoph Paasch <cpaasch@openai.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20250902-net-next-mptcp-misc-feat-6-18-v2-4-fa02bb3188b1@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
