mptcp: pm: prio: skip closed subflows

[ Upstream commit 166b78344031bf7ac9f55cb5282776cfd85f220e ]

When sending an MP_PRIO, closed subflows need to be skipped.

This fixes the case where the initial subflow got closed, re-opened
later, then an MP_PRIO is needed for the same local address.

Note that explicit MP_PRIO cannot be sent during the 3WHS, so it is fine
to use __mptcp_subflow_active().

Fixes: 067065422fcd ("mptcp: add the outgoing MP_PRIO support")
Cc: stable@vger.kernel.org
Fixes: b29fcfb54cd7 ("mptcp: full disconnect implementation")
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260505-net-mptcp-pm-fixes-7-1-rc3-v1-9-fca8091060a4@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
[ applied to renamed function `mptcp_pm_nl_mp_prio_send_ack()` in `pm_netlink.c` ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index 28f8d7fb7bdf391937205329e1daa16361e55f04..655fb90465f750773c44b209f6088b97d6608b0a 100644 (file)
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -923,6 +923,9 @@ int mptcp_pm_nl_mp_prio_send_ack(struct mptcp_sock *msk,
                struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
                struct mptcp_addr_info local, remote;
 
+               if (!__mptcp_subflow_active(subflow))
+                       continue;
+
                mptcp_local_address((struct sock_common *)ssk, &local);
                if (!mptcp_addresses_equal(&local, addr, addr->port))
                        continue;