From: "Matthieu Baerts (NGI0)" <matttbe@xxxxxxxxxx>
commit 1c1f721375989579e46741f59523e39ec9b2a9bd upstream.
Adding the following warning ...
WARN_ON_ONCE(msk->pm.add_addr_accepted == 0)
... before decrementing the add_addr_accepted counter helped to find a
bug when running the "remove single subflow" subtest from the
mptcp_join.sh selftest.
Removing a 'subflow' endpoint will first trigger a RM_ADDR, then the
subflow closure. Before this patch, and upon the reception of the
RM_ADDR, the other peer will then try to decrement this
add_addr_accepted. That's not correct because the attached subflows have
not been created upon the reception of an ADD_ADDR.
A way to solve that is to decrement the counter only if the attached
subflow was an MP_JOIN to a remote id that was not 0, and initiated by
the host receiving the RM_ADDR.
Fixes: d0876b2284cf ("mptcp: add the incoming RM_ADDR support")
Cc: stable@xxxxxxxxxxxxxxx
Reviewed-by: Mat Martineau <martineau@xxxxxxxxxx>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@xxxxxxxxxx>
Link: https://patch.msgid.link/20240819-net-mptcp-pm-reusing-id-v1-9-38035d40de5b@xxxxxxxxxx
Signed-off-by: Jakub Kicinski <kuba@xxxxxxxxxx>
[ Conflicts in pm_netlink.c, because the context is different, but the
same lines can still be modified. The conflicts are due to commit
4d25247d3ae4 ("mptcp: bypass in-kernel PM restrictions for non-kernel
PMs") and commit a88c9e496937 ("mptcp: do not block subflows creation
on errors"), adding new features and not present in this version.
Note that because some features to better track subflows are missing
in this version, it is required to remove the WARN_ON, because the
counter could be 0 in some cases. ]
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
net/mptcp/pm_netlink.c | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -757,7 +757,7 @@ static void mptcp_pm_nl_rm_addr_or_subfl
mptcp_close_ssk(sk, ssk, subflow);
spin_lock_bh(&msk->pm.lock);
- removed = true;
+ removed |= subflow->request_join;
msk->pm.subflows--;
if (rm_type == MPTCP_MIB_RMSUBFLOW)
__MPTCP_INC_STATS(sock_net(sk), rm_type);
@@ -767,7 +767,11 @@ static void mptcp_pm_nl_rm_addr_or_subfl
if (!removed)
continue;
- if (rm_type == MPTCP_MIB_RMADDR) {
+ if (rm_type == MPTCP_MIB_RMADDR && rm_list->ids[i] &&
+ msk->pm.add_addr_accepted != 0) {
+ /* Note: if the subflow has been closed before, this
+ * add_addr_accepted counter will not be decremented.
+ */
msk->pm.add_addr_accepted--;
WRITE_ONCE(msk->pm.accept_addr, true);
} else if (rm_type == MPTCP_MIB_RMSUBFLOW) {
Patches currently in stable-queue which might be from matttbe@xxxxxxxxxx are
queue-5.15/mptcp-pm-avoid-possible-uaf-when-selecting-endp.patch
queue-5.15/mptcp-pm-only-decrement-add_addr_accepted-for-mpj-req.patch
queue-5.15/mptcp-pm-fullmesh-select-the-right-id-later.patch
queue-5.15/mptcp-pm-skip-connecting-to-already-established-sf.patch
queue-5.15/mptcp-pm-re-using-id-of-unused-flushed-subflows.patch
queue-5.15/mptcp-pm-add_addr-0-is-not-a-new-address.patch
queue-5.15/mptcp-constify-a-bunch-of-of-helpers.patch
queue-5.15/mptcp-pm-do-not-remove-already-closed-subflows.patch
queue-5.15/mptcp-pr_debug-add-missing-n-at-the-end.patch
queue-5.15/mptcp-pm-check-add_addr_accept_max-before-accepting-new-add_addr.patch
queue-5.15/mptcp-close-subflow-when-receiving-tcp-fin.patch
queue-5.15/mptcp-avoid-duplicated-sub_closed-events.patch
queue-5.15/mptcp-pm-send-ack-on-an-active-subflow.patch
