Skip to contentSkip to conversation

only remove entry from local_addr_list when sending a REMOVE_ADDR

PATCH Superseded Geliang Tang 4 participants 7 messages Jun 26 – Jul 23 1 message not in the archive

Subsystems (3)

v5 of 6 Superseded Latest is v6 →

lore
Share & tools
lore

Geliang Tang

author reviewer
#0 lore raw

v5 cover only remove entry from local_addr_list when sending a REMOVE_ADDR

From: Geliang Tang <[email protected]>

v5:
- rebased.
- add Matt's reviewed tags.
- add a RM_ADDR from the server to the client as Sashiko suggested.
- mptcpd tests (make check) are still OK after this modification.

v4:
- rebased.
- Link: https://patchwork.kernel.org/project/mptcp/cover/[email protected]/

v3:
- update userspace selftests.
- Link: https://patchwork.kernel.org/project/mptcp/cover/[email protected]/

v2:
- squash patch 1 into patch 2 as Mat suggested.

Geliang Tang (3):
mptcp: pm: userspace: drop delete_local_addr helper
selftests: mptcp: join: update userspace dump_addr outputs
selftests: mptcp: userspace: send RM_ADDR between server and client

 net/mptcp/pm_userspace.c                      | 35 ++-----------------
 .../testing/selftests/net/mptcp/mptcp_join.sh |  6 ++--
 .../selftests/net/mptcp/userspace_pm.sh       | 16 +++++++++
 3 files changed, 21 insertions(+), 36 deletions(-)
Signature · 1 line
2.53.0
author reviewer
#1 lore raw

PATCH v5 1/3 mptcp: pm: userspace: drop delete_local_addr helper

From: Geliang Tang <[email protected]>

Address entries should not be removed from local_addr_list when a subflow is deleted by the userspace PM, should only be removed when sending a REMOVE_ADDR.

So mptcp_userspace_pm_delete_local_addr() helper shouldn't be called in mptcp_pm_nl_subflow_create_doit() and mptcp_pm_nl_subflow_destroy_doit().

Since this helper is open-coding in mptcp_pm_nl_remove_doit(), it can be dropped now.

Address entries are removed from local_addr_list when sending a REMOVE_ADDR by the userspace PM, the local_addr_used counter of PM should also be decremented accordingly.

---
 net/mptcp/pm_userspace.c | 35 ++---------------------------------
 1 file changed, 2 insertions(+), 33 deletions(-)
net/mptcp/pm_userspace.c +2 -33
index ad6ba658e5a5..45463f2f1e60 100644
@@ -90,30 +90,6 @@ static int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
 	return ret;
 }
 
-/* If the subflow is closed from the other peer (not via a
- * subflow destroy command then), we want to keep the entry
- * not to assign the same ID to another address and to be
- * able to send RM_ADDR after the removal of the subflow.
- */
-static int mptcp_userspace_pm_delete_local_addr(struct mptcp_sock *msk,
-						struct mptcp_pm_addr_entry *addr)
-{
-	struct sock *sk = (struct sock *)msk;
-	struct mptcp_pm_addr_entry *entry;
-
-	entry = mptcp_userspace_pm_lookup_addr(msk, &addr->addr);
-	if (!entry)
-		return -EINVAL;
-
-	/* TODO: a refcount is needed because the entry can
-	 * be used multiple times (e.g. fullmesh mode).
-	 */
-	list_del_rcu(&entry->list);
-	sock_kfree_s(sk, entry, sizeof(*entry));
-	msk->pm.local_addr_used--;
-	return 0;
-}
-
 static struct mptcp_pm_addr_entry *
 mptcp_userspace_pm_lookup_addr_by_id(struct mptcp_sock *msk, unsigned int id)
 {
@@ -331,6 +307,7 @@ int mptcp_pm_nl_remove_doit(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	list_del_rcu(&match->list);
+	msk->pm.local_addr_used--;
 	spin_unlock_bh(&msk->pm.lock);
 
 	mptcp_pm_remove_addr_entry(msk, match);
@@ -416,14 +393,9 @@ int mptcp_pm_nl_subflow_create_doit(struct sk_buff *skb, struct genl_info *info)
 	err = __mptcp_subflow_connect(sk, &local, &addr_r);
 	release_sock(sk);
 
-	if (err) {
+	if (err)
 		GENL_SET_ERR_MSG_FMT(info, "connect error: %d", err);
 
-		spin_lock_bh(&msk->pm.lock);
-		mptcp_userspace_pm_delete_local_addr(msk, &entry);
-		spin_unlock_bh(&msk->pm.lock);
-	}
-
  create_err:
 	sock_put(sk);
 	return err;
@@ -540,9 +512,6 @@ int mptcp_pm_nl_subflow_destroy_doit(struct sk_buff *skb, struct genl_info *info
 		goto release_sock;
 	}
 
-	spin_lock_bh(&msk->pm.lock);
-	mptcp_userspace_pm_delete_local_addr(msk, &addr_l);
-	spin_unlock_bh(&msk->pm.lock);
 	mptcp_subflow_shutdown(sk, ssk, RCV_SHUTDOWN | SEND_SHUTDOWN);
 	mptcp_close_ssk(sk, ssk, mptcp_subflow_ctx(ssk));
 	MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_RMSUBFLOW);
Signature · 1 line
2.53.0
author reviewer
#2 lore raw

PATCH v5 2/3 selftests: mptcp: join: update userspace dump_addr outputs

In reply to Geliang Tang

From: Geliang Tang <[email protected]>

Now address entries only be removed from local_addr_list when sending a REMOVE_ADDR by the userspace PM, they're no longer removed when a subflow is deleted.

To make userspace PM selftests in mptcp_join.sh pass, this patch updates the outputs of dump_addr.

---
 tools/testing/selftests/net/mptcp/mptcp_join.sh | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
tools/testing/selftests/net/mptcp/mptcp_join.sh +3 -3
index 550a6b6117a9..e1cdde5ba722 100755
@@ -4089,9 +4089,9 @@ userspace_tests()
 		userspace_pm_chk_get_addr "${ns1}" "20" "id 20 flags signal 10.0.3.1"
 		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_sf 10"
+			$'id 10 flags signal 10.0.2.1\nid 20 flags signal 10.0.3.1' "after rm_sf 10"
 		userspace_pm_rm_addr $ns1 20
-		userspace_pm_chk_dump_addr "${ns1}" "" "after rm_addr 20"
+		userspace_pm_chk_dump_addr "${ns1}" "id 10 flags signal 10.0.2.1" "after rm_addr 20"
 		chk_rm_nr 1 1 invert
 		chk_mptcp_info subflows 0 subflows 0
 		chk_subflows_total 1 1
@@ -4119,7 +4119,7 @@ userspace_tests()
 		userspace_pm_chk_get_addr "${ns2}" "20" "id 20 flags subflow 10.0.3.2"
 		userspace_pm_rm_sf $ns2 10.0.3.2 $MPTCP_LIB_EVENT_SUB_ESTABLISHED
 		userspace_pm_chk_dump_addr "${ns2}" \
-			"" \
+			"id 20 flags subflow 10.0.3.2" \
 			"after rm_sf 20"
 		chk_rm_nr 0 1
 		chk_mptcp_info subflows 0 subflows 0
Signature · 1 line
2.53.0
author reviewer
#3 lore raw

PATCH v5 3/3 selftests: mptcp: userspace: send RM_ADDR between server and client

In reply to Geliang Tang

From: Geliang Tang <[email protected]>

Now address entries only be removed from local_addr_list when sending a REMOVE_ADDR by the userspace PM, they're no longer removed when a subflow is deleted.

To make the original userspace PM selftests pass, this patch sends RM_ADDR from server to client machine at the end of each test, and always sends RM_ADDR from client to server machine to cleanup address IDs for the next new test.

---
 .../testing/selftests/net/mptcp/userspace_pm.sh  | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
tools/testing/selftests/net/mptcp/userspace_pm.sh +16 -0
index e9ae1806ab07..3e7e2a633448 100755
@@ -608,6 +608,10 @@ test_subflows()
 	   "$client4_token"
 	sleep 0.5
 
+	# RM_ADDR from server to client machine
+	ip netns exec "$ns1" ./pm_nl_ctl rem id 23 token "$server4_token"
+	sleep 0.5
+
 	# Attempt to add a listener at dead:beef:2::2:<subflow-port>
 	ip netns exec "$ns2" ./pm_nl_ctl listen dead:beef:2::2\
 	   "$client6_port" &
@@ -642,6 +646,8 @@ test_subflows()
 			      "dead:beef:2::1" "dead:beef:2::2" "$client6_port" "23"\
 			      "$client_addr_id" "ns1" "ns2"
 
+	# RM_ADDR from server to client machine
+	ip netns exec "$ns1" ./pm_nl_ctl rem id 23 token "$server6_token"
 	# RM_ADDR from client to server machine
 	ip netns exec "$ns2" ./pm_nl_ctl rem id $client_addr_id token\
 	   "$client6_token"
@@ -680,6 +686,8 @@ test_subflows()
 	verify_subflow_events "$server_evts" "$SUB_CLOSED" "$server4_token" "$AF_INET" "10.0.2.1"\
 			      "10.0.2.2" "$new4_port" "23" "$client_addr_id" "ns1" "ns2"
 
+	# RM_ADDR from server to client machine
+	ip netns exec "$ns1" ./pm_nl_ctl rem id 23 token "$server4_token"
 	# RM_ADDR from client to server machine
 	ip netns exec "$ns2" ./pm_nl_ctl rem id $client_addr_id token\
 	   "$client4_token"
@@ -721,6 +729,8 @@ test_subflows()
 	# RM_ADDR from server to client machine
 	ip netns exec "$ns1" ./pm_nl_ctl rem id $server_addr_id token\
 	   "$server4_token"
+	# RM_ADDR from client to server machine
+	ip netns exec "$ns2" ./pm_nl_ctl rem id 23 token "$client4_token"
 	sleep 0.5
 
 	# Attempt to add a listener at dead:beef:2::1:<subflow-port>
@@ -760,6 +770,8 @@ test_subflows()
 	# RM_ADDR6 from server to client machine
 	ip netns exec "$ns1" ./pm_nl_ctl rem id $server_addr_id token\
 	   "$server6_token"
+	# RM_ADDR6 from client to server machine
+	ip netns exec "$ns2" ./pm_nl_ctl rem id 23 token "$client6_token"
 	sleep 0.5
 
 	# Attempt to add a listener at 10.0.2.1:<new-port>
@@ -797,6 +809,8 @@ test_subflows()
 	# RM_ADDR from server to client machine
 	ip netns exec "$ns1" ./pm_nl_ctl rem id $server_addr_id token\
 	   "$server4_token"
+	# RM_ADDR from client to server machine
+	ip netns exec "$ns2" ./pm_nl_ctl rem id 23 token "$client4_token"
 }
 
 test_subflows_v4_v6_mix()
@@ -844,6 +858,8 @@ test_subflows_v4_v6_mix()
 	# RM_ADDR from server to client machine
 	ip netns exec "$ns1" ./pm_nl_ctl rem id $server_addr_id token\
 	   "$server6_token"
+	# RM_ADDR from client to server machine
+	ip netns exec "$ns2" ./pm_nl_ctl rem id 23 token "$client6_token"
 	sleep 0.5
 }
 
Signature · 1 line
2.53.0
#4 lore raw
In reply to Geliang Tang

Hi Geliang,

Thank you for your modifications, that's great!

Our CI did some validations and here is its report:

- KVM Validation: normal (except selftest_mptcp_join): Unstable: 1 failed test(s): packetdrill_dss ⚠️
- KVM Validation: normal (only selftest_mptcp_join): Success! ✅
- KVM Validation: debug (except selftest_mptcp_join): Success! ✅
- KVM Validation: debug (only selftest_mptcp_join): Success! ✅
- KVM Validation: btf-normal (only bpftest_all): Success! ✅
- KVM Validation: btf-debug (only bpftest_all): Success! ✅
- Task: https://github.com/multipath-tcp/mptcp_net-next/actions/runs/28223950581

Initiator: Patchew Applier
Commits: https://github.com/multipath-tcp/mptcp_net-next/commits/2e04a2de29b7
Patchwork: https://patchwork.kernel.org/project/mptcp/list/?series=1116901

If there are some issues, you can reproduce them using the same environment as the one used by the CI thanks to a docker image, e.g.:

    $ cd [kernel source code]
    $ docker run -v "${PWD}:${PWD}:rw" -w "${PWD}" --privileged --rm -it \
        --pull always mptcp/mptcp-upstream-virtme-docker:latest \
        auto-normal

For more details:

https://github.com/multipath-tcp/mptcp-upstream-virtme-docker

Please note that despite all the efforts that have been already done to have a stable tests suite when executed on a public CI like here, it is possible some reported issues are not due to your modifications. Still, do not hesitate to help us improve that ;-)

Cheers,
MPTCP GH Action bot
Bot operated by Matthieu Baerts (NGI0 Core)

A message that is not in the archive (sent off-list or not yet received): [email protected]

#6 lore raw
Quoted 9 lines

From: Geliang Tang <[email protected]>

v5:
- rebased.
- add Matt's reviewed tags.
- add a RM_ADDR from the server to the client as Sashiko suggested.
- mptcpd tests (make check) are still OK after this modification.

Hi Matt,

Sorry for the off-list email before, I forget to cc the [email protected]

First, no offense intended — this is in no way a comment on this patch. The patch itself looks good to me. I'm replying here only because the thought came up while reading it.

While looking at mptcpd recently, I noticed that 'make check' doesn't seem to exercise the userspace PM logic in a meaningful way. The userspace PM cases in test-commands issue the per-connection (mptcpd_pm_*) calls with a fake token that doesn't correspond to any real MPTCP connection, so they only really verify that the command dispatches and the genl request is sent without crashing.

This made me wonder whether it would be worthwhile to add functional tests for mptcpd within the selftest framework — for example, via a dedicated script (something like selftest_mptcpd.sh) that actually starts mptcpd with specific plugins (e.g., sspi) in a namespace/veth environment, and then asserts the expected path-manager behavior end-to-end.

In other words, I'm asking: does the community think mptcpd's userspace PM functionality deserves similar test coverage within selftest?

Thanks
Gang

Quoted 27 lines

v4:
- rebased.
- Link: https://patchwork.kernel.org/project/mptcp/cover/[email protected]/

v3:
- update userspace selftests.
- Link: https://patchwork.kernel.org/project/mptcp/cover/[email protected]/

v2:
- squash patch 1 into patch 2 as Mat suggested.

Geliang Tang (3):
mptcp: pm: userspace: drop delete_local_addr helper
selftests: mptcp: join: update userspace dump_addr outputs
selftests: mptcp: userspace: send RM_ADDR between server and client

 net/mptcp/pm_userspace.c | 35 ++-----------------
 .../testing/selftests/net/mptcp/mptcp_join.sh | 6 ++--
 .../selftests/net/mptcp/userspace_pm.sh | 16 +++++++++
 3 files changed, 21 insertions(+), 36 deletions(-)

--
2.53.0

maintainer
#7 lore raw

Hi Gang,

Sorry for the delay.

Quoted 24 lines · On 26/06/2026 12:24, [email protected] wrote:
Quoted 9 lines

From: Geliang Tang <[email protected]>

v5:
- rebased.
- add Matt's reviewed tags.
- add a RM_ADDR from the server to the client as Sashiko suggested.
- mptcpd tests (make check) are still OK after this modification.

Hi Matt,

Sorry for the off-list email before, I forget to cc the [email protected]

First, no offense intended — this is in no way a comment on this patch. The patch itself looks good to me. I'm replying here only because the thought came up while reading it.

While looking at mptcpd recently, I noticed that 'make check' doesn't seem to exercise the userspace PM logic in a meaningful way. The userspace PM cases in test-commands issue the per-connection (mptcpd_pm_*) calls with a fake token that doesn't correspond to any real MPTCP connection, so they only really verify that the command dispatches and the genl request is sent without crashing.

Quoted 8 lines

This made me wonder whether it would be worthwhile to add functional tests for mptcpd within the selftest framework — for example, via a dedicated script (something like selftest_mptcpd.sh) that actually starts mptcpd with specific plugins (e.g., sspi) in a namespace/veth environment, and then asserts the expected path-manager behavior end-to-end.

In other words, I'm asking: does the community think mptcpd's userspace PM functionality deserves similar test coverage within selftest?

Good point. What about the opposite: adding such tests in mptcpd directly?

When mptcpd started to support the userspace PM, it was not easy to access to a kernel with such support: either the old MPTCP fork, or a v5.19 upstream kernel. The mptcpd project was and is still using GitHub Actions for the CI. A few years ago, I suppose it was not planned to build a kernel, start a VM and execute some tests on mptcpd side.

The situation is different now: tests are running on images using either a v6.8 or a v6.17 base kernel:

https://github.com/actions/runner-images/blob/main/images/ubuntu/Ubuntu2204-Readme.md

https://github.com/actions/runner-images/blob/main/images/ubuntu/Ubuntu2404-Readme.md

I think that would be better to add such tests there, rather than adding a dependence to mptcpd in the selftests where it sounds better to use pm_nl_ctl to validate the userspace PM.

WDYT?

Cheers,
Matt

Signature · 1 line
Sponsored by the NGI0 Core fund.