userspace pm: subflows counter: not updated when a subflows is removed from the other end

[matttbe]

It looks like the subflow counter is not updated when the other end initiates the removal of a subflow, e.g. after having received a RM_ADDR:

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index 2b66c5fa71eb..424745490b78 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -3338,17 +3335,19 @@ userspace_tests()
        if reset_with_events "userspace pm add & remove address" &&
           continue_if mptcp_lib_has_file '/proc/sys/net/mptcp/pm_type'; then
                set_userspace_pm $ns1
-               pm_nl_set_limits $ns2 2 2
+               pm_nl_set_limits $ns2 2 5
+               pm_nl_add_endpoint $ns2 10.0.1.2 flags subflow,fullmesh
+               pm_nl_add_endpoint $ns2 10.0.2.2 flags subflow,fullmesh
                speed=5 \
                        run_tests $ns1 $ns2 10.0.1.1 &
                local tests_pid=$!
                wait_mpj $ns1
                userspace_pm_add_addr $ns1 10.0.2.1 10
                userspace_pm_add_addr $ns1 10.0.3.1 20
-               chk_join_nr 2 2 2
+               chk_join_nr 5 5 5
                chk_add_nr 2 2
-               chk_mptcp_info subflows 2 subflows 2
-               chk_subflows_total 3 3
+               chk_mptcp_info subflows 5 subflows 5
+               chk_subflows_total 6 6
                chk_mptcp_info add_addr_signal 2 add_addr_accepted 2
                userspace_pm_chk_dump_addr "${ns1}" \
                        $'id 10 flags signal 10.0.2.1\nid 20 flags signal 10.0.3.1' \
@@ -3356,15 +3355,13 @@ userspace_tests()
                userspace_pm_chk_get_addr "${ns1}" "10" "id 10 flags signal 10.0.2.1"
                userspace_pm_chk_get_addr "${ns1}" "20" "id 20 flags signal 10.0.3.1"
                userspace_pm_rm_addr $ns1 10
-               userspace_pm_rm_sf $ns1 "::ffff:10.0.2.1" $MPTCP_LIB_EVENT_SUB_ESTABLISHED
                userspace_pm_chk_dump_addr "${ns1}" \
                        "id 20 flags signal 10.0.3.1" "after rm_addr 10"
                userspace_pm_rm_addr $ns1 20
-               userspace_pm_rm_sf $ns1 10.0.3.1 $MPTCP_LIB_EVENT_SUB_ESTABLISHED
                userspace_pm_chk_dump_addr "${ns1}" "" "after rm_addr 20"
-               chk_rm_nr 2 2 invert
-               chk_mptcp_info subflows 0 subflows 0
-               chk_subflows_total 1 1
+               chk_rm_nr 2 0 invert
+               chk_mptcp_info subflows 1 subflows 1
+               chk_subflows_total 2 2
                kill_events_pids
                mptcp_lib_kill_wait $tests_pid
        fi

We get the following errors:

mptcp_info subflows=1:1             [FAIL] got 5:1 subflows:subflows expected 1:1
mptcp_info subflows_total=2:2       [FAIL] got 6:2 subflows_total:subflows_total expected 2:2

The counter should be the same on each side.

I don't know why this test sends 2 remove address, each time followed by the removal of the subflow: this hid the bug. I suggest removing only one subflow, not the two of them. @geliangtang: WDYT?

We are certainly missing some actions when a subflow is removed, e.g. retry if we are under the limit, etc.

(Note: I modified the test to validate https://patchwork.kernel.org/project/mptcp/patch/[email protected]/ : this patch is needed not to have errors with chk_rm_nr 2 0 invert)