[PATCH mptcp-next v10 00/23] dump for userspace pm

[PATCH mptcp-next v10 00/23] dump for userspace pm

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

v10:
 - fix a checkpatch.sh warning reported by CI.
 - fix build warnings reported by kernel test robot <lkp@intel.com>.

v9:
 - add missing "sock_put(sk)" in patch 7, 17 and 21 to fix the "kmemleak"
   errors reported by CI.
 - rename the helper to mptcp_pm_has_addr_attr_id.
 - use my new mail address.

v8:
 - Address Mat's comments in v7

v7:
 - fix checkpatch.sh warning reported by CI.
 - fix build errors reported by CI.

v6:
 - patches 1-5, fixes for -net
 - patches 6, 15, selftests for "fixes for -net".
 - patches 7-14, 16-21, update "dump for userspace pm v5"
 - patches 22, 23, cleanups for set_flags
 - patch 24, flush selftest.

v5:
 - patch 2, save bitmap instead of id.
 - patch 3, drop token in reply.
 - update patch 4 and patch 12, add mptcp_pm_dump_addr and
   mptcp_pm_get_addr wrappers in pm.c

v4:
 - update patch 2, patch 3 and patch 12.
 - fix the warnings reported by CI in v3:

Unable to validate one commit: 7633a101a mptcp: get addr in userspace pm list
Unable to compile mptcp source code with make W=1 net/mptcp/pm_netlink.o
Unable to compile mptcp source code with make C=1 net/mptcp/pm_netlink.o: net/mptcp/pm_netlink.c:1656:5: warning: symbol 'mptcp_pm_nl_get_addr' was not declared. Should it be static?

v3:
 - fix the errors reported by CI in v2:

[10:55:51.288] [  897.523894][ T5664] BUG: sleeping function called from invalid context at include/linux/sched/mm.h:306
[10:55:51.288] [  897.532157][ T5664] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 5664, name: pm_nl_ctl
[10:55:51.289] [  897.539503][ T5664] preempt_count: 201, expected: 0

 - drop mptcp_pm_nl_put_entry_msg and mptcp_pm_nl_put_entry_info
   helpers.

v2:
Address Mat's comments in v1:
 - patch 2: pass token to dump_addr too.
 - patch 2: add back id in mptcp_userspace_pm_dump_addr().
 - patch 5: update mptcp.yaml

v1:
Address Mat's comments in "userspace pm enhancements" v14.

Geliang Tang (23):
  mptcp: add pm_has_addr_attr_id helper
  mptcp: add needs_id for userspace appending addr
  mptcp: add needs_id for netlink appending addr
  mptcp: map v4 address to v6 when destroying subflow
  selftests: mptcp: rm subflow with v4/v4mapped addr
  mptcp: export mptcp_genl_family & mptcp_nl_fill_addr
  mptcp: implement mptcp_userspace_pm_dump_addr
  mptcp: add token for get-addr in yaml
  mptcp: dump addrs in userspace pm list
  mptcp: check userspace pm subflow flag
  selftests: mptcp: add userspace pm subflow flag
  selftests: mptcp: add token for dump_addr
  selftests: mptcp: add check_output helper
  selftests: mptcp: dump after creating id 0 subflow
  selftests: mptcp: dump userspace addrs list
  mptcp: add userspace_pm_lookup_addr_by_id helper
  mptcp: implement mptcp_userspace_pm_get_addr
  mptcp: get addr in userspace pm list
  selftests: mptcp: add token for get_addr
  selftests: mptcp: userspace pm get addr tests
  mptcp: update set_flags interfaces
  mptcp: set error messages for set_flags
  selftests: mptcp: flush userspace addrs list

 Documentation/netlink/specs/mptcp_pm.yaml     |   3 +-
 net/mptcp/mptcp_pm_gen.c                      |   7 +-
 net/mptcp/mptcp_pm_gen.h                      |   2 +-
 net/mptcp/pm.c                                |  38 ++-
 net/mptcp/pm_netlink.c                        | 100 ++++----
 net/mptcp/pm_userspace.c                      | 241 +++++++++++++++---
 net/mptcp/protocol.h                          |  25 +-
 .../testing/selftests/net/mptcp/mptcp_join.sh | 150 ++++++++++-
 tools/testing/selftests/net/mptcp/pm_nl_ctl.c |  39 ++-
 9 files changed, 505 insertions(+), 100 deletions(-)

-- 
2.40.1

[PATCH mptcp-next v10 01/23] mptcp: add pm_has_addr_attr_id helper

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch adds a new helper mptcp_pm_has_addr_attr_id() to check whether
an address ID is set from PM or not. It will be used in the next two
commits.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm.c       | 12 ++++++++++++
 net/mptcp/protocol.h |  2 ++
 2 files changed, 14 insertions(+)

diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c
index 4ae19113b8eb..6856d10368be 100644
--- a/net/mptcp/pm.c
+++ b/net/mptcp/pm.c
@@ -533,6 +533,18 @@ void mptcp_pm_data_reset(struct mptcp_sock *msk)
 	bitmap_fill(msk->pm.id_avail_bitmap, MPTCP_PM_MAX_ADDR_ID + 1);
 }
 
+bool mptcp_pm_has_addr_attr_id(const struct nlattr *attr,
+			       struct genl_info *info)
+{
+	struct nlattr *tb[MPTCP_PM_ADDR_ATTR_MAX + 1];
+
+	if (!nla_parse_nested_deprecated(tb, MPTCP_PM_ADDR_ATTR_MAX, attr,
+					 mptcp_pm_address_nl_policy, info->extack) &&
+	    tb[MPTCP_PM_ADDR_ATTR_ID])
+		return true;
+	return false;
+}
+
 void mptcp_pm_data_init(struct mptcp_sock *msk)
 {
 	spin_lock_init(&msk->pm.lock);
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index d442d876f465..0d634e01c47d 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -887,6 +887,8 @@ int mptcp_pm_parse_addr(struct nlattr *attr, struct genl_info *info,
 int mptcp_pm_parse_entry(struct nlattr *attr, struct genl_info *info,
 			 bool require_family,
 			 struct mptcp_pm_addr_entry *entry);
+bool mptcp_pm_has_addr_attr_id(const struct nlattr *attr,
+			       struct genl_info *info);
 bool mptcp_pm_addr_families_match(const struct sock *sk,
 				  const struct mptcp_addr_info *loc,
 				  const struct mptcp_addr_info *rem);
-- 
2.40.1

[PATCH mptcp-next v10 02/23] mptcp: add needs_id for userspace appending addr

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

When userspace PM requires to create an ID 0 subflow in "userspace pm
create id 0 subflow" test like this:

        userspace_pm_add_sf $ns2 10.0.3.2 0

An ID 1 subflow, in fact, is created.

Since in mptcp_pm_nl_append_new_local_addr(), 'id 0' will be treated as
no ID is set by userspace, and will allocate a new ID immediately:

     if (!e->addr.id)
             e->addr.id = find_next_zero_bit(pernet->id_bitmap,
                                             MPTCP_PM_MAX_ADDR_ID + 1,
                                             1);

To solve this issue, a new parameter needs_id is added for
mptcp_userspace_pm_append_new_local_addr() to distinguish between
whether userspace PM has set an ID 0 or whether userspace PM has
not set any address.

In mptcp_userspace_pm_get_local_id(), needs_id is always true, but in
mptcp_pm_nl_announce_doit() and mptcp_pm_nl_subflow_create_doit(),
pass mptcp_pm_has_addr_attr_id() to needs_it.

Fixes: e5ed101a6028 ("mptcp: userspace pm allow creating id 0 subflow")
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm_userspace.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index 4f3901d5b8ef..8ca6e28a121e 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -26,7 +26,8 @@ void mptcp_free_local_addr_list(struct mptcp_sock *msk)
 }
 
 static int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
-						    struct mptcp_pm_addr_entry *entry)
+						    struct mptcp_pm_addr_entry *entry,
+						    bool needs_id)
 {
 	DECLARE_BITMAP(id_bitmap, MPTCP_PM_MAX_ADDR_ID + 1);
 	struct mptcp_pm_addr_entry *match = NULL;
@@ -41,7 +42,7 @@ static int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
 	spin_lock_bh(&msk->pm.lock);
 	list_for_each_entry(e, &msk->pm.userspace_pm_local_addr_list, list) {
 		addr_match = mptcp_addresses_equal(&e->addr, &entry->addr, true);
-		if (addr_match && entry->addr.id == 0)
+		if (addr_match && entry->addr.id == 0 && needs_id)
 			entry->addr.id = e->addr.id;
 		id_match = (e->addr.id == entry->addr.id);
 		if (addr_match && id_match) {
@@ -64,7 +65,7 @@ static int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
 		}
 
 		*e = *entry;
-		if (!e->addr.id)
+		if (!e->addr.id && needs_id)
 			e->addr.id = find_next_zero_bit(id_bitmap,
 							MPTCP_PM_MAX_ADDR_ID + 1,
 							1);
@@ -153,7 +154,7 @@ int mptcp_userspace_pm_get_local_id(struct mptcp_sock *msk,
 	if (new_entry.addr.port == msk_sport)
 		new_entry.addr.port = 0;
 
-	return mptcp_userspace_pm_append_new_local_addr(msk, &new_entry);
+	return mptcp_userspace_pm_append_new_local_addr(msk, &new_entry, true);
 }
 
 int mptcp_pm_nl_announce_doit(struct sk_buff *skb, struct genl_info *info)
@@ -198,7 +199,8 @@ int mptcp_pm_nl_announce_doit(struct sk_buff *skb, struct genl_info *info)
 		goto announce_err;
 	}
 
-	err = mptcp_userspace_pm_append_new_local_addr(msk, &addr_val);
+	err = mptcp_userspace_pm_append_new_local_addr(msk, &addr_val,
+						       !mptcp_pm_has_addr_attr_id(addr, info));
 	if (err < 0) {
 		GENL_SET_ERR_MSG(info, "did not match address and id");
 		goto announce_err;
@@ -378,7 +380,8 @@ int mptcp_pm_nl_subflow_create_doit(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	local.addr = addr_l;
-	err = mptcp_userspace_pm_append_new_local_addr(msk, &local);
+	err = mptcp_userspace_pm_append_new_local_addr(msk, &local,
+						       !mptcp_pm_has_addr_attr_id(laddr, info));
 	if (err < 0) {
 		GENL_SET_ERR_MSG(info, "did not match address and id");
 		goto create_err;
-- 
2.40.1

[PATCH mptcp-next v10 03/23] mptcp: add needs_id for netlink appending addr

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Just the same as userspace PM, a new parameter needs_id is added for
in-kernel PM mptcp_pm_nl_append_new_local_addr() too.

In mptcp_pm_nl_get_local_id(), needs_id is always true, but in
mptcp_pm_nl_add_addr_doit(), pass mptcp_pm_has_addr_attr_id() to
needs_it.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm_netlink.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index 287a60381eae..d158bfab1abc 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -901,7 +901,8 @@ static void __mptcp_pm_release_addr_entry(struct mptcp_pm_addr_entry *entry)
 }
 
 static int mptcp_pm_nl_append_new_local_addr(struct pm_nl_pernet *pernet,
-					     struct mptcp_pm_addr_entry *entry)
+					     struct mptcp_pm_addr_entry *entry,
+					     bool needs_id)
 {
 	struct mptcp_pm_addr_entry *cur, *del_entry = NULL;
 	unsigned int addr_max;
@@ -949,7 +950,7 @@ static int mptcp_pm_nl_append_new_local_addr(struct pm_nl_pernet *pernet,
 		}
 	}
 
-	if (!entry->addr.id) {
+	if (!entry->addr.id && needs_id) {
 find_next:
 		entry->addr.id = find_next_zero_bit(pernet->id_bitmap,
 						    MPTCP_PM_MAX_ADDR_ID + 1,
@@ -960,7 +961,7 @@ static int mptcp_pm_nl_append_new_local_addr(struct pm_nl_pernet *pernet,
 		}
 	}
 
-	if (!entry->addr.id)
+	if (!entry->addr.id && needs_id)
 		goto out;
 
 	__set_bit(entry->addr.id, pernet->id_bitmap);
@@ -1092,7 +1093,7 @@ int mptcp_pm_nl_get_local_id(struct mptcp_sock *msk, struct mptcp_addr_info *skc
 	entry->ifindex = 0;
 	entry->flags = MPTCP_PM_ADDR_FLAG_IMPLICIT;
 	entry->lsk = NULL;
-	ret = mptcp_pm_nl_append_new_local_addr(pernet, entry);
+	ret = mptcp_pm_nl_append_new_local_addr(pernet, entry, true);
 	if (ret < 0)
 		kfree(entry);
 
@@ -1326,7 +1327,8 @@ int mptcp_pm_nl_add_addr_doit(struct sk_buff *skb, struct genl_info *info)
 			goto out_free;
 		}
 	}
-	ret = mptcp_pm_nl_append_new_local_addr(pernet, entry);
+	ret = mptcp_pm_nl_append_new_local_addr(pernet, entry,
+						!mptcp_pm_has_addr_attr_id(attr, info));
 	if (ret < 0) {
 		GENL_SET_ERR_MSG_FMT(info, "too many addresses or duplicate one: %d", ret);
 		goto out_free;
-- 
2.40.1

[PATCH mptcp-next v10 04/23] mptcp: map v4 address to v6 when destroying subflow

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Address family of server side mismatches with that of client side, like
in "userspace pm add & remove address" test:

    userspace_pm_add_addr $ns1 10.0.2.1 10
    userspace_pm_rm_sf $ns1 "::ffff:10.0.2.1" $SUB_ESTABLISHED

That's because on the server side, the family is set to AF_INET6 and the
v4 address is mapped in a v6 one.

This patch fixes this issue. In mptcp_pm_nl_subflow_destroy_doit(), before
checking local address family with remote address family, map an IPv4
address to an IPv6 address if the pair is a v4-mapped address.

Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/387
Fixes: 702c2f646d42 ("mptcp: netlink: allow userspace-driven subflow establishment")
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm_userspace.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index 8ca6e28a121e..7bb3574cc65a 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -497,6 +497,16 @@ int mptcp_pm_nl_subflow_destroy_doit(struct sk_buff *skb, struct genl_info *info
 		goto destroy_err;
 	}
 
+#if IS_ENABLED(CONFIG_MPTCP_IPV6)
+	if (addr_l.family == AF_INET && ipv6_addr_v4mapped(&addr_r.addr6)) {
+		ipv6_addr_set_v4mapped(addr_l.addr.s_addr, &addr_l.addr6);
+		addr_l.family = AF_INET6;
+	}
+	if (addr_r.family == AF_INET && ipv6_addr_v4mapped(&addr_l.addr6)) {
+		ipv6_addr_set_v4mapped(addr_r.addr.s_addr, &addr_r.addr6);
+		addr_r.family = AF_INET6;
+	}
+#endif
 	if (addr_l.family != addr_r.family) {
 		GENL_SET_ERR_MSG(info, "address families do not match");
 		err = -EINVAL;
-- 
2.40.1

[PATCH mptcp-next v10 05/23] selftests: mptcp: rm subflow with v4/v4mapped addr

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Now both a v4 address and a v4-mapped address are supported when
destroying a userspace pm subflow, this patch adds random tests for both
addresses.

Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/387
Fixes: 48d73f609dcc ("selftests: mptcp: update userspace pm addr tests")
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/mptcp_join.sh | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index 3a5b63026191..915ec79f5030 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -3340,12 +3340,13 @@ userspace_pm_rm_sf()
 {
 	local evts=$evts_ns1
 	local t=${3:-1}
-	local ip=4
+	local ip
 	local tk da dp sp
 	local cnt
 
 	[ "$1" == "$ns2" ] && evts=$evts_ns2
-	if mptcp_lib_is_v6 $2; then ip=6; fi
+	[ -n "$(mptcp_lib_evts_get_info "saddr4" "$evts" $t)" ] && ip=4
+	[ -n "$(mptcp_lib_evts_get_info "saddr6" "$evts" $t)" ] && ip=6
 	tk=$(mptcp_lib_evts_get_info token "$evts")
 	da=$(mptcp_lib_evts_get_info "daddr$ip" "$evts" $t)
 	dp=$(mptcp_lib_evts_get_info dport "$evts" $t)
@@ -3448,7 +3449,11 @@ userspace_tests()
 		chk_subflows_total 2 2
 		chk_mptcp_info add_addr_signal 1 add_addr_accepted 1
 		userspace_pm_rm_addr $ns1 10
-		userspace_pm_rm_sf $ns1 "::ffff:10.0.2.1" $SUB_ESTABLISHED
+		if [ $((RANDOM%2)) -eq 0 ]; then
+			userspace_pm_rm_sf $ns1 ::ffff:10.0.2.1 $SUB_ESTABLISHED
+		else
+			userspace_pm_rm_sf $ns1 10.0.2.1 $SUB_ESTABLISHED
+		fi
 		chk_rm_nr 1 1 invert
 		chk_mptcp_info subflows 0 subflows 0
 		chk_subflows_total 1 1
-- 
2.40.1

[PATCH mptcp-next v10 06/23] mptcp: export mptcp_genl_family & mptcp_nl_fill_addr

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch exports struct mptcp_genl_family and mptcp_nl_fill_addr() helper
to allow them can be used in pm_userspace.c.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm_netlink.c | 9 +++------
 net/mptcp/protocol.h   | 4 ++++
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index d158bfab1abc..1c3091fc4e1d 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -18,9 +18,6 @@
 #include "protocol.h"
 #include "mib.h"
 
-/* forward declaration */
-static struct genl_family mptcp_genl_family;
-
 static int pm_nl_pernet_id;
 
 struct mptcp_pm_add_entry {
@@ -1621,8 +1618,8 @@ int mptcp_pm_nl_flush_addrs_doit(struct sk_buff *skb, struct genl_info *info)
 	return 0;
 }
 
-static int mptcp_nl_fill_addr(struct sk_buff *skb,
-			      struct mptcp_pm_addr_entry *entry)
+int mptcp_nl_fill_addr(struct sk_buff *skb,
+		       struct mptcp_pm_addr_entry *entry)
 {
 	struct mptcp_addr_info *addr = &entry->addr;
 	struct nlattr *attr;
@@ -2266,7 +2263,7 @@ void mptcp_event(enum mptcp_event_type type, const struct mptcp_sock *msk,
 	nlmsg_free(skb);
 }
 
-static struct genl_family mptcp_genl_family __ro_after_init = {
+struct genl_family mptcp_genl_family __ro_after_init = {
 	.name		= MPTCP_PM_NAME,
 	.version	= MPTCP_PM_VER,
 	.netnsok	= true,
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 0d634e01c47d..bfb1bfea83c3 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -341,6 +341,8 @@ struct mptcp_sock {
 #define mptcp_for_each_subflow_safe(__msk, __subflow, __tmp)			\
 	list_for_each_entry_safe(__subflow, __tmp, &((__msk)->conn_list), node)
 
+extern struct genl_family mptcp_genl_family;
+
 static inline void msk_owned_by_me(const struct mptcp_sock *msk)
 {
 	sock_owned_by_me((const struct sock *)msk);
@@ -964,6 +966,8 @@ void __mptcp_fastopen_gen_msk_ackseq(struct mptcp_sock *msk, struct mptcp_subflo
 				     const struct mptcp_options_received *mp_opt);
 void mptcp_fastopen_subflow_synack_set_params(struct mptcp_subflow_context *subflow,
 					      struct request_sock *req);
+int mptcp_nl_fill_addr(struct sk_buff *skb,
+		       struct mptcp_pm_addr_entry *entry);
 
 static inline bool mptcp_pm_should_add_signal(struct mptcp_sock *msk)
 {
-- 
2.40.1

[PATCH mptcp-next v10 07/23] mptcp: implement mptcp_userspace_pm_dump_addr

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch implements mptcp_userspace_pm_dump_addr() to dump addresses
from userspace pm address list. Use mptcp_token_get_sock() to get the
msk from the given token, if userspace PM is enabled in it, traverse
each address entry in address list, put every entry to userspace using
mptcp_pm_nl_put_entry_msg().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm_userspace.c | 60 ++++++++++++++++++++++++++++++++++++++++
 net/mptcp/protocol.h     |  2 ++
 2 files changed, 62 insertions(+)

diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index 7bb3574cc65a..f1582f40f70e 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -574,3 +574,63 @@ int mptcp_userspace_pm_set_flags(struct net *net, struct nlattr *token,
 	sock_put(sk);
 	return ret;
 }
+
+int mptcp_userspace_pm_dump_addr(struct sk_buff *msg,
+				 struct netlink_callback *cb)
+{
+	struct id_bitmap {
+		DECLARE_BITMAP(map, MPTCP_PM_MAX_ADDR_ID + 1);
+	} *bitmap;
+	const struct genl_info *info = genl_info_dump(cb);
+	struct net *net = sock_net(msg->sk);
+	struct mptcp_pm_addr_entry *entry;
+	struct mptcp_sock *msk;
+	struct nlattr *token;
+	int ret = -EINVAL;
+	struct sock *sk;
+	void *hdr;
+
+	bitmap = (struct id_bitmap *)cb->ctx;
+	token = info->attrs[MPTCP_PM_ATTR_TOKEN];
+
+	msk = mptcp_token_get_sock(net, nla_get_u32(token));
+	if (!msk) {
+		NL_SET_ERR_MSG_ATTR(info->extack, token, "invalid token");
+		return ret;
+	}
+
+	sk = (struct sock *)msk;
+
+	if (!mptcp_pm_is_userspace(msk)) {
+		GENL_SET_ERR_MSG(info, "invalid request; userspace PM not selected");
+		goto out;
+	}
+
+	lock_sock(sk);
+	spin_lock_bh(&msk->pm.lock);
+	list_for_each_entry(entry, &msk->pm.userspace_pm_local_addr_list, list) {
+		if (test_bit(entry->addr.id, bitmap->map))
+			continue;
+
+		hdr = genlmsg_put(msg, NETLINK_CB(cb->skb).portid,
+				  cb->nlh->nlmsg_seq, &mptcp_genl_family,
+				  NLM_F_MULTI, MPTCP_PM_CMD_GET_ADDR);
+		if (!hdr)
+			break;
+
+		if (mptcp_nl_fill_addr(msg, entry) < 0) {
+			genlmsg_cancel(msg, hdr);
+			break;
+		}
+
+		__set_bit(entry->addr.id, bitmap->map);
+		genlmsg_end(msg, hdr);
+	}
+	spin_unlock_bh(&msk->pm.lock);
+	release_sock(sk);
+	ret = msg->len;
+
+out:
+	sock_put(sk);
+	return ret;
+}
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index bfb1bfea83c3..6d21223ec65d 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -1032,6 +1032,8 @@ bool mptcp_pm_rm_addr_signal(struct mptcp_sock *msk, unsigned int remaining,
 int mptcp_pm_get_local_id(struct mptcp_sock *msk, struct sock_common *skc);
 int mptcp_pm_nl_get_local_id(struct mptcp_sock *msk, struct mptcp_addr_info *skc);
 int mptcp_userspace_pm_get_local_id(struct mptcp_sock *msk, struct mptcp_addr_info *skc);
+int mptcp_userspace_pm_dump_addr(struct sk_buff *msg,
+				 struct netlink_callback *cb);
 
 void __init mptcp_pm_nl_init(void);
 void mptcp_pm_nl_work(struct mptcp_sock *msk);
-- 
2.40.1

[PATCH mptcp-next v10 08/23] mptcp: add token for get-addr in yaml

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch adds token paramenter together with addr in get-addr section in
mptcp_pm.yaml, then use the following commands to update mptcp_pm_gen.c
and mptcp_pm_gen.h:

./tools/net/ynl/ynl-gen-c.py --mode kernel \
        --spec Documentation/netlink/specs/mptcp_pm.yaml --source \
        -o net/mptcp/mptcp_pm_gen.c
./tools/net/ynl/ynl-gen-c.py --mode kernel \
        --spec Documentation/netlink/specs/mptcp_pm.yaml --header \
        -o net/mptcp/mptcp_pm_gen.h

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 Documentation/netlink/specs/mptcp_pm.yaml | 3 ++-
 net/mptcp/mptcp_pm_gen.c                  | 7 ++++---
 net/mptcp/mptcp_pm_gen.h                  | 2 +-
 3 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/Documentation/netlink/specs/mptcp_pm.yaml b/Documentation/netlink/specs/mptcp_pm.yaml
index 49f90cfb4698..af525ed29792 100644
--- a/Documentation/netlink/specs/mptcp_pm.yaml
+++ b/Documentation/netlink/specs/mptcp_pm.yaml
@@ -292,13 +292,14 @@ operations:
     -
       name: get-addr
       doc: Get endpoint information
-      attribute-set: endpoint
+      attribute-set: attr
       dont-validate: [ strict ]
       flags: [ uns-admin-perm ]
       do: &get-addr-attrs
         request:
           attributes:
            - addr
+           - token
         reply:
           attributes:
            - addr
diff --git a/net/mptcp/mptcp_pm_gen.c b/net/mptcp/mptcp_pm_gen.c
index 670da7822e6c..c30a2a90a192 100644
--- a/net/mptcp/mptcp_pm_gen.c
+++ b/net/mptcp/mptcp_pm_gen.c
@@ -32,8 +32,9 @@ const struct nla_policy mptcp_pm_del_addr_nl_policy[MPTCP_PM_ENDPOINT_ADDR + 1]
 };
 
 /* MPTCP_PM_CMD_GET_ADDR - do */
-const struct nla_policy mptcp_pm_get_addr_nl_policy[MPTCP_PM_ENDPOINT_ADDR + 1] = {
-	[MPTCP_PM_ENDPOINT_ADDR] = NLA_POLICY_NESTED(mptcp_pm_address_nl_policy),
+const struct nla_policy mptcp_pm_get_addr_nl_policy[MPTCP_PM_ATTR_TOKEN + 1] = {
+	[MPTCP_PM_ATTR_ADDR] = NLA_POLICY_NESTED(mptcp_pm_address_nl_policy),
+	[MPTCP_PM_ATTR_TOKEN] = { .type = NLA_U32, },
 };
 
 /* MPTCP_PM_CMD_FLUSH_ADDRS - do */
@@ -110,7 +111,7 @@ const struct genl_ops mptcp_pm_nl_ops[11] = {
 		.doit		= mptcp_pm_nl_get_addr_doit,
 		.dumpit		= mptcp_pm_nl_get_addr_dumpit,
 		.policy		= mptcp_pm_get_addr_nl_policy,
-		.maxattr	= MPTCP_PM_ENDPOINT_ADDR,
+		.maxattr	= MPTCP_PM_ATTR_TOKEN,
 		.flags		= GENL_UNS_ADMIN_PERM,
 	},
 	{
diff --git a/net/mptcp/mptcp_pm_gen.h b/net/mptcp/mptcp_pm_gen.h
index ac9fc7225b6a..e24258f6f819 100644
--- a/net/mptcp/mptcp_pm_gen.h
+++ b/net/mptcp/mptcp_pm_gen.h
@@ -18,7 +18,7 @@ extern const struct nla_policy mptcp_pm_add_addr_nl_policy[MPTCP_PM_ENDPOINT_ADD
 
 extern const struct nla_policy mptcp_pm_del_addr_nl_policy[MPTCP_PM_ENDPOINT_ADDR + 1];
 
-extern const struct nla_policy mptcp_pm_get_addr_nl_policy[MPTCP_PM_ENDPOINT_ADDR + 1];
+extern const struct nla_policy mptcp_pm_get_addr_nl_policy[MPTCP_PM_ATTR_TOKEN + 1];
 
 extern const struct nla_policy mptcp_pm_flush_addrs_nl_policy[MPTCP_PM_ENDPOINT_ADDR + 1];
 
-- 
2.40.1

[PATCH mptcp-next v10 09/23] mptcp: dump addrs in userspace pm list

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch renames mptcp_pm_nl_get_addr_dumpit() as a dedicated in-kernel
netlink PM dump addrs function mptcp_pm_nl_dump_addr(), and invoke a newly
added wrapper mptcp_pm_dump_addr() in mptcp_pm_nl_get_addr_dumpit().

Invoke in-kernel PM dump addrs function mptcp_pm_nl_dump_addr() or
userspace PM dump addrs function mptcp_userspace_pm_dump_addr() based on
whether the token parameter is passed in or not in the wrapper.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm.c         |  9 +++++++++
 net/mptcp/pm_netlink.c | 10 ++++++++--
 net/mptcp/protocol.h   |  3 +++
 3 files changed, 20 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c
index 6856d10368be..481f8becc3b4 100644
--- a/net/mptcp/pm.c
+++ b/net/mptcp/pm.c
@@ -441,6 +441,15 @@ int mptcp_pm_get_flags_and_ifindex_by_id(struct mptcp_sock *msk, unsigned int id
 	return mptcp_pm_nl_get_flags_and_ifindex_by_id(msk, id, flags, ifindex);
 }
 
+int mptcp_pm_dump_addr(struct sk_buff *msg, struct netlink_callback *cb)
+{
+	const struct genl_info *info = genl_info_dump(cb);
+
+	if (info->attrs[MPTCP_PM_ATTR_TOKEN])
+		return mptcp_userspace_pm_dump_addr(msg, cb);
+	return mptcp_pm_nl_dump_addr(msg, cb);
+}
+
 int mptcp_pm_set_flags(struct net *net, struct nlattr *token,
 		       struct mptcp_pm_addr_entry *loc,
 		       struct mptcp_pm_addr_entry *rem, u8 bkup)
diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index 1c3091fc4e1d..6fedf271e26d 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -1707,8 +1707,8 @@ int mptcp_pm_nl_get_addr_doit(struct sk_buff *skb, struct genl_info *info)
 	return ret;
 }
 
-int mptcp_pm_nl_get_addr_dumpit(struct sk_buff *msg,
-				struct netlink_callback *cb)
+int mptcp_pm_nl_dump_addr(struct sk_buff *msg,
+			  struct netlink_callback *cb)
 {
 	struct net *net = sock_net(msg->sk);
 	struct mptcp_pm_addr_entry *entry;
@@ -1750,6 +1750,12 @@ int mptcp_pm_nl_get_addr_dumpit(struct sk_buff *msg,
 	return msg->len;
 }
 
+int mptcp_pm_nl_get_addr_dumpit(struct sk_buff *msg,
+				struct netlink_callback *cb)
+{
+	return mptcp_pm_dump_addr(msg, cb);
+}
+
 static int parse_limit(struct genl_info *info, int id, unsigned int *limit)
 {
 	struct nlattr *attr = info->attrs[id];
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 6d21223ec65d..0dc3891a45d4 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -1032,6 +1032,9 @@ bool mptcp_pm_rm_addr_signal(struct mptcp_sock *msk, unsigned int remaining,
 int mptcp_pm_get_local_id(struct mptcp_sock *msk, struct sock_common *skc);
 int mptcp_pm_nl_get_local_id(struct mptcp_sock *msk, struct mptcp_addr_info *skc);
 int mptcp_userspace_pm_get_local_id(struct mptcp_sock *msk, struct mptcp_addr_info *skc);
+int mptcp_pm_dump_addr(struct sk_buff *msg, struct netlink_callback *cb);
+int mptcp_pm_nl_dump_addr(struct sk_buff *msg,
+			  struct netlink_callback *cb);
 int mptcp_userspace_pm_dump_addr(struct sk_buff *msg,
 				 struct netlink_callback *cb);
 
-- 
2.40.1

[PATCH mptcp-next v10 10/23] mptcp: check userspace pm subflow flag

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Just like MPTCP_PM_ADDR_FLAG_SIGNAL flag is checked in userspace PM
announce mptcp_pm_nl_announce_doit(), MPTCP_PM_ADDR_FLAG_SUBFLOW flag
should be checked in mptcp_pm_nl_subflow_create_doit() too.

If MPTCP_PM_ADDR_FLAG_SUBFLOW flag is not set, there's no flags field
in the output of dump_addr. This looks a bit strange:

	id 10 flags  10.0.3.2

This patch uses mptcp_pm_parse_entry() instead of mptcp_pm_parse_addr()
to get the flags of the entry. Add MPTCP_PM_ADDR_FLAG_SUBFLOW flag check
in mptcp_pm_nl_subflow_create_doit().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm_userspace.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index f1582f40f70e..ca0d6e1dfade 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -361,11 +361,18 @@ int mptcp_pm_nl_subflow_create_doit(struct sk_buff *skb, struct genl_info *info)
 		goto create_err;
 	}
 
-	err = mptcp_pm_parse_addr(laddr, info, &addr_l);
+	err = mptcp_pm_parse_entry(laddr, info, true, &local);
 	if (err < 0) {
 		NL_SET_ERR_MSG_ATTR(info->extack, laddr, "error parsing local addr");
 		goto create_err;
 	}
+	addr_l = local.addr;
+
+	if (!(local.flags & MPTCP_PM_ADDR_FLAG_SUBFLOW)) {
+		GENL_SET_ERR_MSG(info, "invalid addr flags");
+		err = -EINVAL;
+		goto create_err;
+	}
 
 	err = mptcp_pm_parse_addr(raddr, info, &addr_r);
 	if (err < 0) {
@@ -379,7 +386,6 @@ int mptcp_pm_nl_subflow_create_doit(struct sk_buff *skb, struct genl_info *info)
 		goto create_err;
 	}
 
-	local.addr = addr_l;
 	err = mptcp_userspace_pm_append_new_local_addr(msk, &local,
 						       !mptcp_pm_has_addr_attr_id(laddr, info));
 	if (err < 0) {
-- 
2.40.1

[PATCH mptcp-next v10 11/23] selftests: mptcp: add userspace pm subflow flag

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch adds the address flag MPTCP_PM_ADDR_FLAG_SUBFLOW in csf() in
pm_nl_ctl.c when subflow is created by a userspace PM.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/pm_nl_ctl.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/tools/testing/selftests/net/mptcp/pm_nl_ctl.c b/tools/testing/selftests/net/mptcp/pm_nl_ctl.c
index 49369c4a5f26..e97856323ec3 100644
--- a/tools/testing/selftests/net/mptcp/pm_nl_ctl.c
+++ b/tools/testing/selftests/net/mptcp/pm_nl_ctl.c
@@ -453,6 +453,7 @@ int csf(int fd, int pm_family, int argc, char *argv[])
 	char data[NLMSG_ALIGN(sizeof(struct nlmsghdr)) +
 		  NLMSG_ALIGN(sizeof(struct genlmsghdr)) +
 		  1024];
+	u_int32_t flags = MPTCP_PM_ADDR_FLAG_SUBFLOW;
 	const char *params[5];
 	struct nlmsghdr *nh;
 	struct rtattr *addr;
@@ -558,6 +559,13 @@ int csf(int fd, int pm_family, int argc, char *argv[])
 			off += NLMSG_ALIGN(rta->rta_len);
 		}
 
+		/* addr flags */
+		rta = (void *)(data + off);
+		rta->rta_type = MPTCP_PM_ADDR_ATTR_FLAGS;
+		rta->rta_len = RTA_LENGTH(4);
+		memcpy(RTA_DATA(rta), &flags, 4);
+		off += NLMSG_ALIGN(rta->rta_len);
+
 		addr->rta_len = off - addr_start;
 	}
 
-- 
2.40.1

[PATCH mptcp-next v10 12/23] selftests: mptcp: add token for dump_addr

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

The command dump_addr() of pm_nl_ctl can be used like this in in-kernel PM:

        pm_nl_ctl dump

This patch adds token argument for it to support userspace PM:

        pm_nl_ctl dump token $token

If 'token $token' is passed to dump_addr(), copy it into the kernel
netlink.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/pm_nl_ctl.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/tools/testing/selftests/net/mptcp/pm_nl_ctl.c b/tools/testing/selftests/net/mptcp/pm_nl_ctl.c
index e97856323ec3..8d7d1b4ed28e 100644
--- a/tools/testing/selftests/net/mptcp/pm_nl_ctl.c
+++ b/tools/testing/selftests/net/mptcp/pm_nl_ctl.c
@@ -1127,8 +1127,16 @@ int dump_addrs(int fd, int pm_family, int argc, char *argv[])
 		  1024];
 	pid_t pid = getpid();
 	struct nlmsghdr *nh;
+	u_int32_t token = 0;
+	struct rtattr *rta;
 	int off = 0;
 
+	if (argc != 2 && argc != 4)
+		syntax(argv);
+
+	if (argc == 4 && !strcmp(argv[2], "token"))
+		token = strtoul(argv[3], NULL, 10);
+
 	memset(data, 0, sizeof(data));
 	nh = (void *)data;
 	off = init_genl_req(data, pm_family, MPTCP_PM_CMD_GET_ADDR,
@@ -1138,6 +1146,15 @@ int dump_addrs(int fd, int pm_family, int argc, char *argv[])
 	nh->nlmsg_pid = pid;
 	nh->nlmsg_len = off;
 
+	/* token */
+	if (token) {
+		rta = (void *)(data + off);
+		rta->rta_type = MPTCP_PM_ATTR_TOKEN;
+		rta->rta_len = RTA_LENGTH(4);
+		memcpy(RTA_DATA(rta), &token, 4);
+		off += NLMSG_ALIGN(rta->rta_len);
+	}
+
 	print_addrs(nh, pm_family, do_nl_req(fd, nh, off, sizeof(data)));
 	return 0;
 }
-- 
2.40.1

[PATCH mptcp-next v10 13/23] selftests: mptcp: add check_output helper

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Similar to check() in pm_netlink.sh, add a new helper check_output()
in mptcp_join.sh to check the output of the given commands.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../testing/selftests/net/mptcp/mptcp_join.sh | 27 +++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index 915ec79f5030..f4043baa59c6 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -21,6 +21,7 @@ cinfail=""
 cinsent=""
 tmpfile=""
 cout=""
+check_output_err=""
 capout=""
 ns1=""
 ns2=""
@@ -186,6 +187,7 @@ init() {
 	cout=$(mktemp)
 	evts_ns1=$(mktemp)
 	evts_ns2=$(mktemp)
+	check_output_err=$(mktemp)
 
 	trap cleanup EXIT
 
@@ -199,6 +201,7 @@ cleanup()
 	rm -f "$sin" "$sout" "$cinsent" "$cinfail"
 	rm -f "$tmpfile"
 	rm -rf $evts_ns1 $evts_ns2
+	rm -f $check_output_err
 	cleanup_partial
 }
 
@@ -3358,6 +3361,30 @@ userspace_pm_rm_sf()
 	wait_rm_sf $1 "${cnt}"
 }
 
+check_output()
+{
+	local cmd="$1"
+	local expected="$2"
+	local msg="$3"
+	local out=`$cmd 2>$check_output_err`
+	local cmd_ret=$?
+
+	printf "%-42s" "$msg"
+	if [ $cmd_ret -ne 0 ]; then
+		mptcp_lib_print_err "[ FAIL ] command execution '$cmd' stderr "
+		cat $check_output_err
+		ret=${KSFT_FAIL}
+		return $cmd_ret
+	elif [ "$out" = "$expected" ]; then
+		mptcp_lib_print_ok "[ OK ]"
+		return 0
+	else
+		mptcp_lib_print_err "[ FAIL ] expected '$expected' got '$out'"
+		ret=${KSFT_FAIL}
+		return 1
+	fi
+}
+
 userspace_tests()
 {
 	# userspace pm type prevents add_addr
-- 
2.40.1

[PATCH mptcp-next v10 14/23] selftests: mptcp: dump after creating id 0 subflow

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch adds a new helper userspace_pm_dump() to dump addresses
for the userspace PM. Use this helper to check whether an ID 0 subflow
is listed in the output of dump command after creating an ID 0 subflow
in "userspace pm create id 0 subflow" test.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/mptcp_join.sh | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index f4043baa59c6..a92fb5469621 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -3361,6 +3361,18 @@ userspace_pm_rm_sf()
 	wait_rm_sf $1 "${cnt}"
 }
 
+# $1: ns
+userspace_pm_dump()
+{
+	local evts=$evts_ns1
+	local tk
+
+	[ "$1" == "$ns2" ] && evts=$evts_ns2
+	tk=$(mptcp_lib_evts_get_info token "$evts")
+
+	ip netns exec $1 ./pm_nl_ctl dump token $tk
+}
+
 check_output()
 {
 	local cmd="$1"
@@ -3522,6 +3534,9 @@ userspace_tests()
 		chk_mptcp_info subflows 0 subflows 0
 		chk_subflows_total 1 1
 		userspace_pm_add_sf $ns2 10.0.3.2 0
+		check_output "userspace_pm_dump $ns2" \
+			     "id 0 flags subflow 10.0.3.2" \
+			     "      dump addrs id 0 subflow"
 		chk_join_nr 1 1 1
 		chk_mptcp_info subflows 1 subflows 1
 		chk_subflows_total 2 2
-- 
2.40.1

[PATCH mptcp-next v10 15/23] selftests: mptcp: dump userspace addrs list

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch adds two new tests for userspace pm dump address and subflow.
Use userspace_pm_add_addr() and userspace_pm_add_sf() to add an address
and a suflow.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../testing/selftests/net/mptcp/mptcp_join.sh | 44 +++++++++++++++++++
 1 file changed, 44 insertions(+)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index a92fb5469621..cec4bfc5193b 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -3593,6 +3593,50 @@ userspace_tests()
 		kill_events_pids
 		wait $tests_pid
 	fi
+
+	# userspace pm dump address
+	if reset_with_events "userspace pm dump address" &&
+	   continue_if mptcp_lib_has_file '/proc/sys/net/mptcp/pm_type'; then
+		set_userspace_pm $ns1
+		pm_nl_set_limits $ns2 1 1
+		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
+		chk_join_nr 1 1 1
+		chk_add_nr 1 1
+		chk_mptcp_info subflows 1 subflows 1
+		chk_subflows_total 2 2
+		chk_mptcp_info add_addr_signal 1 add_addr_accepted 1
+		local dump="id 10 flags signal 10.0.2.1"
+		check_output "userspace_pm_dump $ns1" \
+			     "$dump" "      dump addrs signal"
+		kill_events_pids
+		wait $tests_pid
+	fi
+
+	# userspace pm dump subflow
+	if reset_with_events "userspace pm dump subflow" &&
+	   continue_if mptcp_lib_has_file '/proc/sys/net/mptcp/pm_type'; then
+		set_userspace_pm $ns2
+		pm_nl_set_limits $ns1 0 1
+		speed=5 \
+			run_tests $ns1 $ns2 10.0.1.1 &
+		local tests_pid=$!
+		wait_mpj $ns2
+		chk_mptcp_info subflows 0 subflows 0
+		chk_subflows_total 1 1
+		userspace_pm_add_sf $ns2 10.0.3.2 20
+		chk_join_nr 1 1 1
+		chk_mptcp_info subflows 1 subflows 1
+		chk_subflows_total 2 2
+		local dump="id 20 flags subflow 10.0.3.2"
+		check_output "userspace_pm_dump $ns2" \
+			     "$dump" "      dump addrs subflow"
+		kill_events_pids
+		wait $tests_pid
+	fi
 }
 
 endpoint_tests()
-- 
2.40.1

[PATCH mptcp-next v10 16/23] mptcp: add userspace_pm_lookup_addr_by_id helper

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

Corresponding __lookup_addr_by_id() helper in the in-kernel netlink PM,
this patch adds a new helper mptcp_userspace_pm_lookup_addr_by_id() to
lookup the address entry with the given id on the userspace pm local
address list.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm_userspace.c | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index ca0d6e1dfade..d1457a443b78 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -106,19 +106,26 @@ static int mptcp_userspace_pm_delete_local_addr(struct mptcp_sock *msk,
 	return -EINVAL;
 }
 
+static struct mptcp_pm_addr_entry *
+mptcp_userspace_pm_lookup_addr_by_id(struct mptcp_sock *msk, unsigned int id)
+{
+	struct mptcp_pm_addr_entry *entry;
+
+	list_for_each_entry(entry, &msk->pm.userspace_pm_local_addr_list, list) {
+		if (entry->addr.id == id)
+			return entry;
+	}
+	return NULL;
+}
+
 int mptcp_userspace_pm_get_flags_and_ifindex_by_id(struct mptcp_sock *msk,
 						   unsigned int id,
 						   u8 *flags, int *ifindex)
 {
-	struct mptcp_pm_addr_entry *entry, *match = NULL;
+	struct mptcp_pm_addr_entry *match;
 
 	spin_lock_bh(&msk->pm.lock);
-	list_for_each_entry(entry, &msk->pm.userspace_pm_local_addr_list, list) {
-		if (id == entry->addr.id) {
-			match = entry;
-			break;
-		}
-	}
+	match = mptcp_userspace_pm_lookup_addr_by_id(msk, id);
 	spin_unlock_bh(&msk->pm.lock);
 	if (match) {
 		*flags = match->flags;
@@ -262,7 +269,7 @@ int mptcp_pm_nl_remove_doit(struct sk_buff *skb, struct genl_info *info)
 {
 	struct nlattr *token = info->attrs[MPTCP_PM_ATTR_TOKEN];
 	struct nlattr *id = info->attrs[MPTCP_PM_ATTR_LOC_ID];
-	struct mptcp_pm_addr_entry *match = NULL;
+	struct mptcp_pm_addr_entry *match;
 	struct mptcp_pm_addr_entry *entry;
 	struct mptcp_sock *msk;
 	LIST_HEAD(free_list);
@@ -299,13 +306,7 @@ int mptcp_pm_nl_remove_doit(struct sk_buff *skb, struct genl_info *info)
 
 	lock_sock(sk);
 
-	list_for_each_entry(entry, &msk->pm.userspace_pm_local_addr_list, list) {
-		if (entry->addr.id == id_val) {
-			match = entry;
-			break;
-		}
-	}
-
+	match = mptcp_userspace_pm_lookup_addr_by_id(msk, id_val);
 	if (!match) {
 		GENL_SET_ERR_MSG(info, "address with specified id not found");
 		release_sock(sk);
-- 
2.40.1

[PATCH mptcp-next v10 17/23] mptcp: implement mptcp_userspace_pm_get_addr

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch implements mptcp_userspace_pm_get_addr() to get an address
from userspace pm address list according the given 'token' and 'id'.
Use nla_get_u32() to get the u32 value of 'token', then pass it to
mptcp_token_get_sock() to get the msk. Pass 'msk' and 'id' to the helper
mptcp_userspace_pm_lookup_addr_by_id() to get the address entry. Put
this entry to userspace using mptcp_pm_nl_put_entry_info().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm_userspace.c | 74 ++++++++++++++++++++++++++++++++++++++++
 net/mptcp/protocol.h     |  2 ++
 2 files changed, 76 insertions(+)

diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index d1457a443b78..96620038a4e9 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -641,3 +641,77 @@ int mptcp_userspace_pm_dump_addr(struct sk_buff *msg,
 	sock_put(sk);
 	return ret;
 }
+
+int mptcp_userspace_pm_get_addr(struct sk_buff *skb,
+				struct genl_info *info)
+{
+	struct nlattr *attr = info->attrs[MPTCP_PM_ENDPOINT_ADDR];
+	struct nlattr *token = info->attrs[MPTCP_PM_ATTR_TOKEN];
+	struct mptcp_pm_addr_entry addr, *entry;
+	struct net *net = sock_net(skb->sk);
+	struct mptcp_sock *msk;
+	struct sk_buff *msg;
+	int ret = -EINVAL;
+	struct sock *sk;
+	void *reply;
+
+	msk = mptcp_token_get_sock(net, nla_get_u32(token));
+	if (!msk) {
+		NL_SET_ERR_MSG_ATTR(info->extack, token, "invalid token");
+		return ret;
+	}
+
+	sk = (struct sock *)msk;
+
+	if (!mptcp_pm_is_userspace(msk)) {
+		GENL_SET_ERR_MSG(info, "invalid request; userspace PM not selected");
+		goto out;
+	}
+
+	ret = mptcp_pm_parse_entry(attr, info, false, &addr);
+	if (ret < 0)
+		goto out;
+
+	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	reply = genlmsg_put_reply(msg, info, &mptcp_genl_family, 0,
+				  info->genlhdr->cmd);
+	if (!reply) {
+		GENL_SET_ERR_MSG(info, "not enough space in Netlink message");
+		ret = -EMSGSIZE;
+		goto fail;
+	}
+
+	lock_sock(sk);
+	spin_lock_bh(&msk->pm.lock);
+	entry = mptcp_userspace_pm_lookup_addr_by_id(msk, addr.addr.id);
+	if (!entry) {
+		GENL_SET_ERR_MSG(info, "address not found");
+		ret = -EINVAL;
+		goto unlock_fail;
+	}
+
+	ret = mptcp_nl_fill_addr(msg, entry);
+	if (ret)
+		goto unlock_fail;
+
+	genlmsg_end(msg, reply);
+	ret = genlmsg_reply(msg, info);
+	spin_unlock_bh(&msk->pm.lock);
+	release_sock(sk);
+	sock_put(sk);
+	return ret;
+
+unlock_fail:
+	spin_unlock_bh(&msk->pm.lock);
+	release_sock(sk);
+fail:
+	nlmsg_free(msg);
+out:
+	sock_put(sk);
+	return ret;
+}
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 0dc3891a45d4..e2a8833bca05 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -1037,6 +1037,8 @@ int mptcp_pm_nl_dump_addr(struct sk_buff *msg,
 			  struct netlink_callback *cb);
 int mptcp_userspace_pm_dump_addr(struct sk_buff *msg,
 				 struct netlink_callback *cb);
+int mptcp_userspace_pm_get_addr(struct sk_buff *skb,
+				struct genl_info *info);
 
 void __init mptcp_pm_nl_init(void);
 void mptcp_pm_nl_work(struct mptcp_sock *msk);
-- 
2.40.1

[PATCH mptcp-next v10 18/23] mptcp: get addr in userspace pm list

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch renames mptcp_pm_nl_get_addr_doit() as a dedicated in-kernel
netlink PM get addr function mptcp_pm_nl_get_addr(). and invoke a new
wrapper mptcp_pm_get_addr() in mptcp_pm_nl_get_addr_doit.

If a token is gotten in the wrapper, that means a userspace PM is used.
So invoke mptcp_userspace_pm_get_addr() to get addr in userspace PM list.
Otherwise, invoke mptcp_pm_nl_get_addr().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm.c         | 7 +++++++
 net/mptcp/pm_netlink.c | 7 ++++++-
 net/mptcp/protocol.h   | 2 ++
 3 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c
index 481f8becc3b4..d05c2c221932 100644
--- a/net/mptcp/pm.c
+++ b/net/mptcp/pm.c
@@ -441,6 +441,13 @@ int mptcp_pm_get_flags_and_ifindex_by_id(struct mptcp_sock *msk, unsigned int id
 	return mptcp_pm_nl_get_flags_and_ifindex_by_id(msk, id, flags, ifindex);
 }
 
+int mptcp_pm_get_addr(struct sk_buff *skb, struct genl_info *info)
+{
+	if (info->attrs[MPTCP_PM_ATTR_TOKEN])
+		return mptcp_userspace_pm_get_addr(skb, info);
+	return mptcp_pm_nl_get_addr(skb, info);
+}
+
 int mptcp_pm_dump_addr(struct sk_buff *msg, struct netlink_callback *cb)
 {
 	const struct genl_info *info = genl_info_dump(cb);
diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index 6fedf271e26d..c227c8909cd3 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -1657,7 +1657,7 @@ int mptcp_nl_fill_addr(struct sk_buff *skb,
 	return -EMSGSIZE;
 }
 
-int mptcp_pm_nl_get_addr_doit(struct sk_buff *skb, struct genl_info *info)
+int mptcp_pm_nl_get_addr(struct sk_buff *skb, struct genl_info *info)
 {
 	struct nlattr *attr = info->attrs[MPTCP_PM_ENDPOINT_ADDR];
 	struct pm_nl_pernet *pernet = genl_info_pm_nl(info);
@@ -1707,6 +1707,11 @@ int mptcp_pm_nl_get_addr_doit(struct sk_buff *skb, struct genl_info *info)
 	return ret;
 }
 
+int mptcp_pm_nl_get_addr_doit(struct sk_buff *skb, struct genl_info *info)
+{
+	return mptcp_pm_get_addr(skb, info);
+}
+
 int mptcp_pm_nl_dump_addr(struct sk_buff *msg,
 			  struct netlink_callback *cb)
 {
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index e2a8833bca05..a11d1a98beca 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -1037,6 +1037,8 @@ int mptcp_pm_nl_dump_addr(struct sk_buff *msg,
 			  struct netlink_callback *cb);
 int mptcp_userspace_pm_dump_addr(struct sk_buff *msg,
 				 struct netlink_callback *cb);
+int mptcp_pm_get_addr(struct sk_buff *skb, struct genl_info *info);
+int mptcp_pm_nl_get_addr(struct sk_buff *skb, struct genl_info *info);
 int mptcp_userspace_pm_get_addr(struct sk_buff *skb,
 				struct genl_info *info);
 
-- 
2.40.1

[PATCH mptcp-next v10 19/23] selftests: mptcp: add token for get_addr

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

The command get_addr() of pm_nl_ctl can be used like this in in-kernel PM:

	pm_nl_ctl get $id

This patch adds token argument for it to support userspace PM:

	pm_nl_ctl get $id token $token

If 'token $token' is passed to get_addr(), copy it into the kernel netlink.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/pm_nl_ctl.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/mptcp/pm_nl_ctl.c b/tools/testing/selftests/net/mptcp/pm_nl_ctl.c
index 8d7d1b4ed28e..7426a2cbd4a0 100644
--- a/tools/testing/selftests/net/mptcp/pm_nl_ctl.c
+++ b/tools/testing/selftests/net/mptcp/pm_nl_ctl.c
@@ -1087,6 +1087,7 @@ int get_addr(int fd, int pm_family, int argc, char *argv[])
 		  1024];
 	struct rtattr *rta, *nest;
 	struct nlmsghdr *nh;
+	u_int32_t token = 0;
 	int nest_start;
 	u_int8_t id;
 	int off = 0;
@@ -1097,10 +1098,12 @@ int get_addr(int fd, int pm_family, int argc, char *argv[])
 			    MPTCP_PM_VER);
 
 	/* the only argument is the address id */
-	if (argc != 3)
+	if (argc != 3 && argc != 5)
 		syntax(argv);
 
 	id = atoi(argv[2]);
+	if (argc == 5 && !strcmp(argv[3], "token"))
+		token = strtoul(argv[4], NULL, 10);
 
 	nest_start = off;
 	nest = (void *)(data + off);
@@ -1116,6 +1119,15 @@ int get_addr(int fd, int pm_family, int argc, char *argv[])
 	off += NLMSG_ALIGN(rta->rta_len);
 	nest->rta_len = off - nest_start;
 
+	/* token */
+	if (token) {
+		rta = (void *)(data + off);
+		rta->rta_type = MPTCP_PM_ATTR_TOKEN;
+		rta->rta_len = RTA_LENGTH(4);
+		memcpy(RTA_DATA(rta), &token, 4);
+		off += NLMSG_ALIGN(rta->rta_len);
+	}
+
 	print_addrs(nh, pm_family, do_nl_req(fd, nh, off, sizeof(data)));
 	return 0;
 }
-- 
2.40.1

[PATCH mptcp-next v10 20/23] selftests: mptcp: userspace pm get addr tests

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch adds a new helper userspace_pm_get_addr() in mptcp_join.sh.
In it, parse the token value from the output of 'pm_nl_ctl events', then
pass it to pm_nl_ctl get_addr command. Use this helper in userspace pm
dump tests.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 tools/testing/selftests/net/mptcp/mptcp_join.sh | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index cec4bfc5193b..0b325b5efccf 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -3373,6 +3373,18 @@ userspace_pm_dump()
 	ip netns exec $1 ./pm_nl_ctl dump token $tk
 }
 
+# $1: ns ; $2: id
+userspace_pm_get_addr()
+{
+	local evts=$evts_ns1
+	local tk
+
+	[ "$1" == "$ns2" ] && evts=$evts_ns2
+	tk=$(mptcp_lib_evts_get_info token "$evts")
+
+	ip netns exec $1 ./pm_nl_ctl get $2 token $tk
+}
+
 check_output()
 {
 	local cmd="$1"
@@ -3610,6 +3622,8 @@ userspace_tests()
 		chk_subflows_total 2 2
 		chk_mptcp_info add_addr_signal 1 add_addr_accepted 1
 		local dump="id 10 flags signal 10.0.2.1"
+		check_output "userspace_pm_get_addr $ns1 10" \
+			     "$dump" "      get id 10 addr"
 		check_output "userspace_pm_dump $ns1" \
 			     "$dump" "      dump addrs signal"
 		kill_events_pids
@@ -3632,6 +3646,8 @@ userspace_tests()
 		chk_mptcp_info subflows 1 subflows 1
 		chk_subflows_total 2 2
 		local dump="id 20 flags subflow 10.0.3.2"
+		check_output "userspace_pm_get_addr $ns2 20" \
+			     "$dump" "      get id 20 addr"
 		check_output "userspace_pm_dump $ns2" \
 			     "$dump" "      dump addrs subflow"
 		kill_events_pids
-- 
2.40.1

[PATCH mptcp-next v10 21/23] mptcp: update set_flags interfaces

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch updates set_flags interfaces, make it more similar to the
interfaces of dump_addr and get_addr:

 mptcp_pm_set_flags(struct sk_buff *skb, struct genl_info *info)
 mptcp_pm_nl_set_flags(struct sk_buff *skb, struct genl_info *info)
 mptcp_userspace_pm_set_flags(struct sk_buff *skb, struct genl_info *info)

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm.c           | 10 +++----
 net/mptcp/pm_netlink.c   | 58 +++++++++++++++++-----------------------
 net/mptcp/pm_userspace.c | 32 +++++++++++++++++-----
 net/mptcp/protocol.h     | 10 +++----
 4 files changed, 58 insertions(+), 52 deletions(-)

diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c
index d05c2c221932..30cedafb009e 100644
--- a/net/mptcp/pm.c
+++ b/net/mptcp/pm.c
@@ -457,13 +457,11 @@ int mptcp_pm_dump_addr(struct sk_buff *msg, struct netlink_callback *cb)
 	return mptcp_pm_nl_dump_addr(msg, cb);
 }
 
-int mptcp_pm_set_flags(struct net *net, struct nlattr *token,
-		       struct mptcp_pm_addr_entry *loc,
-		       struct mptcp_pm_addr_entry *rem, u8 bkup)
+int mptcp_pm_set_flags(struct sk_buff *skb, struct genl_info *info)
 {
-	if (token)
-		return mptcp_userspace_pm_set_flags(net, token, loc, rem, bkup);
-	return mptcp_pm_nl_set_flags(net, loc, bkup);
+	if (info->attrs[MPTCP_PM_ATTR_TOKEN])
+		return mptcp_userspace_pm_set_flags(skb, info);
+	return mptcp_pm_nl_set_flags(skb, info);
 }
 
 void mptcp_pm_subflow_chk_stale(const struct mptcp_sock *msk, struct sock *ssk)
diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index c227c8909cd3..6e913498ff79 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -1875,66 +1875,58 @@ static int mptcp_nl_set_flags(struct net *net,
 	return ret;
 }
 
-int mptcp_pm_nl_set_flags(struct net *net, struct mptcp_pm_addr_entry *addr, u8 bkup)
+int mptcp_pm_nl_set_flags(struct sk_buff *skb, struct genl_info *info)
 {
-	struct pm_nl_pernet *pernet = pm_nl_get_pernet(net);
+	struct mptcp_pm_addr_entry addr = { .addr = { .family = AF_UNSPEC }, };
+	struct nlattr *attr = info->attrs[MPTCP_PM_ATTR_ADDR];
 	u8 changed, mask = MPTCP_PM_ADDR_FLAG_BACKUP |
 			   MPTCP_PM_ADDR_FLAG_FULLMESH;
+	struct net *net = sock_net(skb->sk);
 	struct mptcp_pm_addr_entry *entry;
+	struct pm_nl_pernet *pernet;
 	u8 lookup_by_id = 0;
+	u8 bkup = 0;
+	int ret;
+
+	pernet = pm_nl_get_pernet(net);
+
+	ret = mptcp_pm_parse_entry(attr, info, false, &addr);
+	if (ret < 0)
+		return ret;
 
-	if (addr->addr.family == AF_UNSPEC) {
+	if (addr.addr.family == AF_UNSPEC) {
 		lookup_by_id = 1;
-		if (!addr->addr.id)
+		if (!addr.addr.id)
 			return -EOPNOTSUPP;
 	}
 
+	if (addr.flags & MPTCP_PM_ADDR_FLAG_BACKUP)
+		bkup = 1;
+
 	spin_lock_bh(&pernet->lock);
-	entry = __lookup_addr(pernet, &addr->addr, lookup_by_id);
+	entry = __lookup_addr(pernet, &addr.addr, lookup_by_id);
 	if (!entry) {
 		spin_unlock_bh(&pernet->lock);
 		return -EINVAL;
 	}
-	if ((addr->flags & MPTCP_PM_ADDR_FLAG_FULLMESH) &&
+	if ((addr.flags & MPTCP_PM_ADDR_FLAG_FULLMESH) &&
 	    (entry->flags & MPTCP_PM_ADDR_FLAG_SIGNAL)) {
 		spin_unlock_bh(&pernet->lock);
 		return -EINVAL;
 	}
 
-	changed = (addr->flags ^ entry->flags) & mask;
-	entry->flags = (entry->flags & ~mask) | (addr->flags & mask);
-	*addr = *entry;
+	changed = (addr.flags ^ entry->flags) & mask;
+	entry->flags = (entry->flags & ~mask) | (addr.flags & mask);
+	addr = *entry;
 	spin_unlock_bh(&pernet->lock);
 
-	mptcp_nl_set_flags(net, &addr->addr, bkup, changed);
+	mptcp_nl_set_flags(net, &addr.addr, bkup, changed);
 	return 0;
 }
 
 int mptcp_pm_nl_set_flags_doit(struct sk_buff *skb, struct genl_info *info)
 {
-	struct mptcp_pm_addr_entry remote = { .addr = { .family = AF_UNSPEC }, };
-	struct mptcp_pm_addr_entry addr = { .addr = { .family = AF_UNSPEC }, };
-	struct nlattr *attr_rem = info->attrs[MPTCP_PM_ATTR_ADDR_REMOTE];
-	struct nlattr *token = info->attrs[MPTCP_PM_ATTR_TOKEN];
-	struct nlattr *attr = info->attrs[MPTCP_PM_ATTR_ADDR];
-	struct net *net = sock_net(skb->sk);
-	u8 bkup = 0;
-	int ret;
-
-	ret = mptcp_pm_parse_entry(attr, info, false, &addr);
-	if (ret < 0)
-		return ret;
-
-	if (attr_rem) {
-		ret = mptcp_pm_parse_entry(attr_rem, info, false, &remote);
-		if (ret < 0)
-			return ret;
-	}
-
-	if (addr.flags & MPTCP_PM_ADDR_FLAG_BACKUP)
-		bkup = 1;
-
-	return mptcp_pm_set_flags(net, token, &addr, &remote, bkup);
+	return mptcp_pm_set_flags(skb, info);
 }
 
 static void mptcp_nl_mcast_send(struct net *net, struct sk_buff *nlskb, gfp_t gfp)
diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index 96620038a4e9..bb5672a4332e 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -549,14 +549,19 @@ int mptcp_pm_nl_subflow_destroy_doit(struct sk_buff *skb, struct genl_info *info
 	return err;
 }
 
-int mptcp_userspace_pm_set_flags(struct net *net, struct nlattr *token,
-				 struct mptcp_pm_addr_entry *loc,
-				 struct mptcp_pm_addr_entry *rem, u8 bkup)
+int mptcp_userspace_pm_set_flags(struct sk_buff *skb, struct genl_info *info)
 {
+	struct mptcp_pm_addr_entry loc = { .addr = { .family = AF_UNSPEC }, };
+	struct mptcp_pm_addr_entry rem = { .addr = { .family = AF_UNSPEC }, };
+	struct nlattr *attr_rem = info->attrs[MPTCP_PM_ATTR_ADDR_REMOTE];
+	struct nlattr *token = info->attrs[MPTCP_PM_ATTR_TOKEN];
+	struct nlattr *attr = info->attrs[MPTCP_PM_ATTR_ADDR];
+	struct net *net = sock_net(skb->sk);
 	struct mptcp_sock *msk;
 	int ret = -EINVAL;
 	struct sock *sk;
 	u32 token_val;
+	u8 bkup = 0;
 
 	token_val = nla_get_u32(token);
 
@@ -569,12 +574,27 @@ int mptcp_userspace_pm_set_flags(struct net *net, struct nlattr *token,
 	if (!mptcp_pm_is_userspace(msk))
 		goto set_flags_err;
 
-	if (loc->addr.family == AF_UNSPEC ||
-	    rem->addr.family == AF_UNSPEC)
+	ret = mptcp_pm_parse_entry(attr, info, false, &loc);
+	if (ret < 0)
+		goto set_flags_err;
+
+	if (attr_rem) {
+		ret = mptcp_pm_parse_entry(attr_rem, info, false, &rem);
+		if (ret < 0)
+			goto set_flags_err;
+	}
+
+	if (loc.addr.family == AF_UNSPEC ||
+	    rem.addr.family == AF_UNSPEC) {
+		ret = -EINVAL;
 		goto set_flags_err;
+	}
+
+	if (loc.flags & MPTCP_PM_ADDR_FLAG_BACKUP)
+		bkup = 1;
 
 	lock_sock(sk);
-	ret = mptcp_pm_nl_mp_prio_send_ack(msk, &loc->addr, &rem->addr, bkup);
+	ret = mptcp_pm_nl_mp_prio_send_ack(msk, &loc.addr, &rem.addr, bkup);
 	release_sock(sk);
 
 set_flags_err:
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index a11d1a98beca..cd193737b369 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -936,13 +936,9 @@ int mptcp_pm_nl_get_flags_and_ifindex_by_id(struct mptcp_sock *msk, unsigned int
 int mptcp_userspace_pm_get_flags_and_ifindex_by_id(struct mptcp_sock *msk,
 						   unsigned int id,
 						   u8 *flags, int *ifindex);
-int mptcp_pm_set_flags(struct net *net, struct nlattr *token,
-		       struct mptcp_pm_addr_entry *loc,
-		       struct mptcp_pm_addr_entry *rem, u8 bkup);
-int mptcp_pm_nl_set_flags(struct net *net, struct mptcp_pm_addr_entry *addr, u8 bkup);
-int mptcp_userspace_pm_set_flags(struct net *net, struct nlattr *token,
-				 struct mptcp_pm_addr_entry *loc,
-				 struct mptcp_pm_addr_entry *rem, u8 bkup);
+int mptcp_pm_set_flags(struct sk_buff *skb, struct genl_info *info);
+int mptcp_pm_nl_set_flags(struct sk_buff *skb, struct genl_info *info);
+int mptcp_userspace_pm_set_flags(struct sk_buff *skb, struct genl_info *info);
 int mptcp_pm_announce_addr(struct mptcp_sock *msk,
 			   const struct mptcp_addr_info *addr,
 			   bool echo);
-- 
2.40.1

[PATCH mptcp-next v10 22/23] mptcp: set error messages for set_flags

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

In addition to returning the error value, this patch also sets an error
messages with GENL_SET_ERR_MSG or NL_SET_ERR_MSG_ATTR both for pm_netlink.c
and pm_userspace.c.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/pm_netlink.c   | 6 +++++-
 net/mptcp/pm_userspace.c | 9 +++++++--
 2 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index 6e913498ff79..4ab68f83ddd5 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -1896,8 +1896,10 @@ int mptcp_pm_nl_set_flags(struct sk_buff *skb, struct genl_info *info)
 
 	if (addr.addr.family == AF_UNSPEC) {
 		lookup_by_id = 1;
-		if (!addr.addr.id)
+		if (!addr.addr.id) {
+			GENL_SET_ERR_MSG(info, "missing required inputs");
 			return -EOPNOTSUPP;
+		}
 	}
 
 	if (addr.flags & MPTCP_PM_ADDR_FLAG_BACKUP)
@@ -1907,11 +1909,13 @@ int mptcp_pm_nl_set_flags(struct sk_buff *skb, struct genl_info *info)
 	entry = __lookup_addr(pernet, &addr.addr, lookup_by_id);
 	if (!entry) {
 		spin_unlock_bh(&pernet->lock);
+		GENL_SET_ERR_MSG(info, "address not found");
 		return -EINVAL;
 	}
 	if ((addr.flags & MPTCP_PM_ADDR_FLAG_FULLMESH) &&
 	    (entry->flags & MPTCP_PM_ADDR_FLAG_SIGNAL)) {
 		spin_unlock_bh(&pernet->lock);
+		GENL_SET_ERR_MSG(info, "invalid addr flags");
 		return -EINVAL;
 	}
 
diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
index bb5672a4332e..ea9e8612da68 100644
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -566,13 +566,17 @@ int mptcp_userspace_pm_set_flags(struct sk_buff *skb, struct genl_info *info)
 	token_val = nla_get_u32(token);
 
 	msk = mptcp_token_get_sock(net, token_val);
-	if (!msk)
+	if (!msk) {
+		NL_SET_ERR_MSG_ATTR(info->extack, token, "invalid token");
 		return ret;
+	}
 
 	sk = (struct sock *)msk;
 
-	if (!mptcp_pm_is_userspace(msk))
+	if (!mptcp_pm_is_userspace(msk)) {
+		GENL_SET_ERR_MSG(info, "invalid request; userspace PM not selected");
 		goto set_flags_err;
+	}
 
 	ret = mptcp_pm_parse_entry(attr, info, false, &loc);
 	if (ret < 0)
@@ -586,6 +590,7 @@ int mptcp_userspace_pm_set_flags(struct sk_buff *skb, struct genl_info *info)
 
 	if (loc.addr.family == AF_UNSPEC ||
 	    rem.addr.family == AF_UNSPEC) {
+		GENL_SET_ERR_MSG(info, "address families do not match");
 		ret = -EINVAL;
 		goto set_flags_err;
 	}
-- 
2.40.1

[PATCH mptcp-next v10 23/23] selftests: mptcp: flush userspace addrs list

[Geliang Tang]

From: Geliang Tang <tanggeliang@kylinos.cn>

This patch adds a new helper userspace_pm_flush() to flush all addresses
for the userspace PM. Invoke it in userspace pm dump address and subflow
tests. And use dump commands to check if the userspace pm local address
list is empty after addresses flushing.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 .../testing/selftests/net/mptcp/mptcp_join.sh | 45 +++++++++++++++++--
 1 file changed, 41 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index 0b325b5efccf..65e8d9d61ac9 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -3385,6 +3385,31 @@ userspace_pm_get_addr()
 	ip netns exec $1 ./pm_nl_ctl get $2 token $tk
 }
 
+# $1: ns ; $2: addr
+userspace_pm_flush()
+{
+	local ns=$1
+	local line
+
+	userspace_pm_dump $ns | while read -r line; do
+		local arr=($line)
+		local nr=0
+		local id
+		local addr
+		local i
+		for i in "${arr[@]}"; do
+			if [ $i = "id" ]; then
+				id=${arr[$nr+1]}
+			fi
+			nr=$((nr + 1))
+		done
+		addr=${arr[$nr-1]}
+		[ $ip_mptcp -eq 1 ] && addr=${arr[0]}
+		userspace_pm_rm_addr $ns $id
+		userspace_pm_rm_sf $ns "$addr" $MPTCP_LIB_SUB_ESTABLISHED
+	done
+}
+
 check_output()
 {
 	local cmd="$1"
@@ -3606,8 +3631,8 @@ userspace_tests()
 		wait $tests_pid
 	fi
 
-	# userspace pm dump address
-	if reset_with_events "userspace pm dump address" &&
+	# userspace pm dump & flush address
+	if reset_with_events "userspace pm dump & flush address" &&
 	   continue_if mptcp_lib_has_file '/proc/sys/net/mptcp/pm_type'; then
 		set_userspace_pm $ns1
 		pm_nl_set_limits $ns2 1 1
@@ -3626,12 +3651,18 @@ userspace_tests()
 			     "$dump" "      get id 10 addr"
 		check_output "userspace_pm_dump $ns1" \
 			     "$dump" "      dump addrs signal"
+		userspace_pm_flush $ns1
+		check_output "userspace_pm_dump $ns1" \
+			     "" "      dump addrs after flush"
+		chk_rm_nr 1 1 invert
+		chk_mptcp_info subflows 0 subflows 0
+		chk_subflows_total 1 1
 		kill_events_pids
 		wait $tests_pid
 	fi
 
-	# userspace pm dump subflow
-	if reset_with_events "userspace pm dump subflow" &&
+	# userspace pm dump & flush subflow
+	if reset_with_events "userspace pm dump & flush subflow" &&
 	   continue_if mptcp_lib_has_file '/proc/sys/net/mptcp/pm_type'; then
 		set_userspace_pm $ns2
 		pm_nl_set_limits $ns1 0 1
@@ -3650,6 +3681,12 @@ userspace_tests()
 			     "$dump" "      get id 20 addr"
 		check_output "userspace_pm_dump $ns2" \
 			     "$dump" "      dump addrs subflow"
+		userspace_pm_flush $ns2
+		check_output "userspace_pm_dump $ns2" \
+			     "" "      dump addrs after flush"
+		chk_rm_nr 1 1
+		chk_mptcp_info subflows 0 subflows 0
+		chk_subflows_total 1 1
 		kill_events_pids
 		wait $tests_pid
 	fi
-- 
2.40.1

Re: selftests: mptcp: flush userspace addrs list: Tests Results

[MPTCP CI]

Hi Geliang,

Thank you for your modifications, that's great!

Our CI (GitHub Action) did some validations and here is its report:

- KVM Validation: normal:
  - Unstable: 3 failed test(s): packetdrill_fastopen selftest_mptcp_join selftest_simult_flows 🔴:
  - Task: https://github.com/multipath-tcp/mptcp_net-next/actions/runs/7566234667

Initiator: Patchew Applier
Commits: https://github.com/multipath-tcp/mptcp_net-next/commits/08b2c98e8cef


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)

Re: selftests: mptcp: flush userspace addrs list: Tests Results

[MPTCP CI]

Hi Geliang,

Thank you for your modifications, that's great!

Our CI (Cirrus) did some validations with a debug kernel and here is its report:

- KVM Validation: debug (except selftest_mptcp_join):
  - Success! ✅:
  - Task: https://cirrus-ci.com/task/6537673670656000
  - Summary: https://api.cirrus-ci.com/v1/artifact/task/6537673670656000/summary/summary.txt

- KVM Validation: debug (only selftest_mptcp_join):
  - Unstable: 1 failed test(s): selftest_mptcp_join 🔴:
  - Task: https://cirrus-ci.com/task/6455025212325888
  - Summary: https://api.cirrus-ci.com/v1/artifact/task/6455025212325888/summary/summary.txt

Initiator: Patchew Applier
Commits: https://github.com/multipath-tcp/mptcp_net-next/commits/08b2c98e8cef


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-debug

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)

Re: [PATCH mptcp-next v10 00/23] dump for userspace pm

[Mat Martineau]

On Thu, 18 Jan 2024, Geliang Tang wrote:

> From: Geliang Tang <tanggeliang@kylinos.cn>
>
> v10:
> - fix a checkpatch.sh warning reported by CI.
> - fix build warnings reported by kernel test robot <lkp@intel.com>.
>
> v9:
> - add missing "sock_put(sk)" in patch 7, 17 and 21 to fix the "kmemleak"
>   errors reported by CI.
> - rename the helper to mptcp_pm_has_addr_attr_id.
> - use my new mail address.
>

Hi Geliang -

Please also make sure you update MAINTAINERS and .mailmap with your 
geliang@kernel.org and tanggeliang@kylinos.cn addresses.

Thanks,
Mat

Re: [PATCH mptcp-next v10 01/23] mptcp: add pm_has_addr_attr_id helper

[Mat Martineau]

On Thu, 18 Jan 2024, Geliang Tang wrote:

> From: Geliang Tang <tanggeliang@kylinos.cn>
>
> This patch adds a new helper mptcp_pm_has_addr_attr_id() to check whether
> an address ID is set from PM or not. It will be used in the next two
> commits.
>
> Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>

Hi Geliang -

In cases like this where a new helper is added (and does not replace 
existing open coded sections), I think it's better to have a single patch 
that both adds the helper and the call to it.

Please squash patches 1 & 2.

- Mat


> ---
> net/mptcp/pm.c       | 12 ++++++++++++
> net/mptcp/protocol.h |  2 ++
> 2 files changed, 14 insertions(+)
>
> diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c
> index 4ae19113b8eb..6856d10368be 100644
> --- a/net/mptcp/pm.c
> +++ b/net/mptcp/pm.c
> @@ -533,6 +533,18 @@ void mptcp_pm_data_reset(struct mptcp_sock *msk)
> 	bitmap_fill(msk->pm.id_avail_bitmap, MPTCP_PM_MAX_ADDR_ID + 1);
> }
>
> +bool mptcp_pm_has_addr_attr_id(const struct nlattr *attr,
> +			       struct genl_info *info)
> +{
> +	struct nlattr *tb[MPTCP_PM_ADDR_ATTR_MAX + 1];
> +
> +	if (!nla_parse_nested_deprecated(tb, MPTCP_PM_ADDR_ATTR_MAX, attr,
> +					 mptcp_pm_address_nl_policy, info->extack) &&
> +	    tb[MPTCP_PM_ADDR_ATTR_ID])
> +		return true;
> +	return false;
> +}
> +
> void mptcp_pm_data_init(struct mptcp_sock *msk)
> {
> 	spin_lock_init(&msk->pm.lock);
> diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
> index d442d876f465..0d634e01c47d 100644
> --- a/net/mptcp/protocol.h
> +++ b/net/mptcp/protocol.h
> @@ -887,6 +887,8 @@ int mptcp_pm_parse_addr(struct nlattr *attr, struct genl_info *info,
> int mptcp_pm_parse_entry(struct nlattr *attr, struct genl_info *info,
> 			 bool require_family,
> 			 struct mptcp_pm_addr_entry *entry);
> +bool mptcp_pm_has_addr_attr_id(const struct nlattr *attr,
> +			       struct genl_info *info);
> bool mptcp_pm_addr_families_match(const struct sock *sk,
> 				  const struct mptcp_addr_info *loc,
> 				  const struct mptcp_addr_info *rem);
> -- 
> 2.40.1
>
>
>

Re: [PATCH mptcp-next v10 02/23] mptcp: add needs_id for userspace appending addr

[Mat Martineau]

On Thu, 18 Jan 2024, Geliang Tang wrote:

> From: Geliang Tang <tanggeliang@kylinos.cn>
>
> When userspace PM requires to create an ID 0 subflow in "userspace pm
> create id 0 subflow" test like this:
>
>        userspace_pm_add_sf $ns2 10.0.3.2 0
>
> An ID 1 subflow, in fact, is created.
>
> Since in mptcp_pm_nl_append_new_local_addr(), 'id 0' will be treated as
> no ID is set by userspace, and will allocate a new ID immediately:
>
>     if (!e->addr.id)
>             e->addr.id = find_next_zero_bit(pernet->id_bitmap,
>                                             MPTCP_PM_MAX_ADDR_ID + 1,
>                                             1);
>
> To solve this issue, a new parameter needs_id is added for
> mptcp_userspace_pm_append_new_local_addr() to distinguish between
> whether userspace PM has set an ID 0 or whether userspace PM has
> not set any address.
>
> In mptcp_userspace_pm_get_local_id(), needs_id is always true, but in
> mptcp_pm_nl_announce_doit() and mptcp_pm_nl_subflow_create_doit(),
> pass mptcp_pm_has_addr_attr_id() to needs_it.
>
> Fixes: e5ed101a6028 ("mptcp: userspace pm allow creating id 0 subflow")
> Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> ---
> net/mptcp/pm_userspace.c | 15 +++++++++------
> 1 file changed, 9 insertions(+), 6 deletions(-)
>
> diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
> index 4f3901d5b8ef..8ca6e28a121e 100644
> --- a/net/mptcp/pm_userspace.c
> +++ b/net/mptcp/pm_userspace.c
> @@ -26,7 +26,8 @@ void mptcp_free_local_addr_list(struct mptcp_sock *msk)
> }
>
> static int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
> -						    struct mptcp_pm_addr_entry *entry)
> +						    struct mptcp_pm_addr_entry *entry,
> +						    bool needs_id)
> {
> 	DECLARE_BITMAP(id_bitmap, MPTCP_PM_MAX_ADDR_ID + 1);
> 	struct mptcp_pm_addr_entry *match = NULL;
> @@ -41,7 +42,7 @@ static int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
> 	spin_lock_bh(&msk->pm.lock);
> 	list_for_each_entry(e, &msk->pm.userspace_pm_local_addr_list, list) {
> 		addr_match = mptcp_addresses_equal(&e->addr, &entry->addr, true);
> -		if (addr_match && entry->addr.id == 0)
> +		if (addr_match && entry->addr.id == 0 && needs_id)
> 			entry->addr.id = e->addr.id;
> 		id_match = (e->addr.id == entry->addr.id);
> 		if (addr_match && id_match) {
> @@ -64,7 +65,7 @@ static int mptcp_userspace_pm_append_new_local_addr(struct mptcp_sock *msk,
> 		}
>
> 		*e = *entry;
> -		if (!e->addr.id)
> +		if (!e->addr.id && needs_id)
> 			e->addr.id = find_next_zero_bit(id_bitmap,
> 							MPTCP_PM_MAX_ADDR_ID + 1,
> 							1);
> @@ -153,7 +154,7 @@ int mptcp_userspace_pm_get_local_id(struct mptcp_sock *msk,
> 	if (new_entry.addr.port == msk_sport)
> 		new_entry.addr.port = 0;
>
> -	return mptcp_userspace_pm_append_new_local_addr(msk, &new_entry);
> +	return mptcp_userspace_pm_append_new_local_addr(msk, &new_entry, true);
> }
>
> int mptcp_pm_nl_announce_doit(struct sk_buff *skb, struct genl_info *info)
> @@ -198,7 +199,8 @@ int mptcp_pm_nl_announce_doit(struct sk_buff *skb, struct genl_info *info)
> 		goto announce_err;
> 	}
>
> -	err = mptcp_userspace_pm_append_new_local_addr(msk, &addr_val);
> +	err = mptcp_userspace_pm_append_new_local_addr(msk, &addr_val,
> +						       !mptcp_pm_has_addr_attr_id(addr, info));

Why would we want the kernel to assign an ID when the userspace PM should 
be reponsible for that? Seems to me that it is a bug that the kernel was 
assigning a subflow ID at all for connections managed by a userspace PM.

> 	if (err < 0) {
> 		GENL_SET_ERR_MSG(info, "did not match address and id");
> 		goto announce_err;
> @@ -378,7 +380,8 @@ int mptcp_pm_nl_subflow_create_doit(struct sk_buff *skb, struct genl_info *info)
> 	}
>
> 	local.addr = addr_l;
> -	err = mptcp_userspace_pm_append_new_local_addr(msk, &local);
> +	err = mptcp_userspace_pm_append_new_local_addr(msk, &local,
> +						       !mptcp_pm_has_addr_attr_id(laddr, info));

Same issue here, the userspace PM should be required to assign the ID. If 
there's a problem with breaking userspace please let me know and we can 
discuss!

- Mat

> 	if (err < 0) {
> 		GENL_SET_ERR_MSG(info, "did not match address and id");
> 		goto create_err;
> -- 
> 2.40.1
>
>
>

Re: [PATCH mptcp-next v10 04/23] mptcp: map v4 address to v6 when destroying subflow

[Mat Martineau]

On Thu, 18 Jan 2024, Geliang Tang wrote:

> From: Geliang Tang <tanggeliang@kylinos.cn>
>
> Address family of server side mismatches with that of client side, like
> in "userspace pm add & remove address" test:
>
>    userspace_pm_add_addr $ns1 10.0.2.1 10
>    userspace_pm_rm_sf $ns1 "::ffff:10.0.2.1" $SUB_ESTABLISHED
>
> That's because on the server side, the family is set to AF_INET6 and the
> v4 address is mapped in a v6 one.
>
> This patch fixes this issue. In mptcp_pm_nl_subflow_destroy_doit(), before
> checking local address family with remote address family, map an IPv4
> address to an IPv6 address if the pair is a v4-mapped address.
>
> Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/387
> Fixes: 702c2f646d42 ("mptcp: netlink: allow userspace-driven subflow establishment")

Hi Geliang -

Are patches 4 & 5 for net or net-next?

Thanks,
Mat


> Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> ---
> net/mptcp/pm_userspace.c | 10 ++++++++++
> 1 file changed, 10 insertions(+)
>
> diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
> index 8ca6e28a121e..7bb3574cc65a 100644
> --- a/net/mptcp/pm_userspace.c
> +++ b/net/mptcp/pm_userspace.c
> @@ -497,6 +497,16 @@ int mptcp_pm_nl_subflow_destroy_doit(struct sk_buff *skb, struct genl_info *info
> 		goto destroy_err;
> 	}
>
> +#if IS_ENABLED(CONFIG_MPTCP_IPV6)
> +	if (addr_l.family == AF_INET && ipv6_addr_v4mapped(&addr_r.addr6)) {
> +		ipv6_addr_set_v4mapped(addr_l.addr.s_addr, &addr_l.addr6);
> +		addr_l.family = AF_INET6;
> +	}
> +	if (addr_r.family == AF_INET && ipv6_addr_v4mapped(&addr_l.addr6)) {
> +		ipv6_addr_set_v4mapped(addr_r.addr.s_addr, &addr_r.addr6);
> +		addr_r.family = AF_INET6;
> +	}
> +#endif
> 	if (addr_l.family != addr_r.family) {
> 		GENL_SET_ERR_MSG(info, "address families do not match");
> 		err = -EINVAL;
> -- 
> 2.40.1
>
>
>

Re: [PATCH mptcp-next v10 10/23] mptcp: check userspace pm subflow flag

[Mat Martineau]

On Thu, 18 Jan 2024, Geliang Tang wrote:

> From: Geliang Tang <tanggeliang@kylinos.cn>
>
> Just like MPTCP_PM_ADDR_FLAG_SIGNAL flag is checked in userspace PM
> announce mptcp_pm_nl_announce_doit(), MPTCP_PM_ADDR_FLAG_SUBFLOW flag
> should be checked in mptcp_pm_nl_subflow_create_doit() too.
>
> If MPTCP_PM_ADDR_FLAG_SUBFLOW flag is not set, there's no flags field
> in the output of dump_addr. This looks a bit strange:
>
> 	id 10 flags  10.0.3.2
>
> This patch uses mptcp_pm_parse_entry() instead of mptcp_pm_parse_addr()
> to get the flags of the entry. Add MPTCP_PM_ADDR_FLAG_SUBFLOW flag check
> in mptcp_pm_nl_subflow_create_doit().
>

Hi Geliang -

Have you tested this change with mptcpd?

Is the addr flag (SUBFLOW or SIGNAL) relevant for userspace PM addresses? 
It doesn't seem like a useful distinction: the userspace PM is always 
creating a new subflow in this context, since address signaling is handled 
by mptcp_pm_nl_announce_doit().

Two options I see:
  * update the userspace tools to cleanly handle missing flags for 
userspace connections
or
  * have mptcp_pm_nl_subflow_create_doit() always set 
MPTCP_PM_ADDR_FLAG_SUBFLOW for the local addr even if it's missing from 
the netlink attribute (and maybe don't allow MPTCP_PM_ADDR_FLAG_SIGNAL?)

- Mat


> Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> ---
> net/mptcp/pm_userspace.c | 10 ++++++++--
> 1 file changed, 8 insertions(+), 2 deletions(-)
>
> diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
> index f1582f40f70e..ca0d6e1dfade 100644
> --- a/net/mptcp/pm_userspace.c
> +++ b/net/mptcp/pm_userspace.c
> @@ -361,11 +361,18 @@ int mptcp_pm_nl_subflow_create_doit(struct sk_buff *skb, struct genl_info *info)
> 		goto create_err;
> 	}
>
> -	err = mptcp_pm_parse_addr(laddr, info, &addr_l);
> +	err = mptcp_pm_parse_entry(laddr, info, true, &local);
> 	if (err < 0) {
> 		NL_SET_ERR_MSG_ATTR(info->extack, laddr, "error parsing local addr");
> 		goto create_err;
> 	}
> +	addr_l = local.addr;
> +
> +	if (!(local.flags & MPTCP_PM_ADDR_FLAG_SUBFLOW)) {
> +		GENL_SET_ERR_MSG(info, "invalid addr flags");
> +		err = -EINVAL;
> +		goto create_err;
> +	}
>
> 	err = mptcp_pm_parse_addr(raddr, info, &addr_r);
> 	if (err < 0) {
> @@ -379,7 +386,6 @@ int mptcp_pm_nl_subflow_create_doit(struct sk_buff *skb, struct genl_info *info)
> 		goto create_err;
> 	}
>
> -	local.addr = addr_l;
> 	err = mptcp_userspace_pm_append_new_local_addr(msk, &local,
> 						       !mptcp_pm_has_addr_attr_id(laddr, info));
> 	if (err < 0) {
> -- 
> 2.40.1
>
>
>

Re: [PATCH mptcp-next v10 10/23] mptcp: check userspace pm subflow flag

[Geliang Tang]

Hi Mat & Matt,

On Tue, 2024-01-30 at 18:28 -0800, Mat Martineau wrote:
> On Thu, 18 Jan 2024, Geliang Tang wrote:
> 
> > From: Geliang Tang <tanggeliang@kylinos.cn>
> > 
> > Just like MPTCP_PM_ADDR_FLAG_SIGNAL flag is checked in userspace PM
> > announce mptcp_pm_nl_announce_doit(), MPTCP_PM_ADDR_FLAG_SUBFLOW
> > flag
> > should be checked in mptcp_pm_nl_subflow_create_doit() too.
> > 
> > If MPTCP_PM_ADDR_FLAG_SUBFLOW flag is not set, there's no flags
> > field
> > in the output of dump_addr. This looks a bit strange:
> > 
> > 	id 10 flags  10.0.3.2
> > 
> > This patch uses mptcp_pm_parse_entry() instead of
> > mptcp_pm_parse_addr()
> > to get the flags of the entry. Add MPTCP_PM_ADDR_FLAG_SUBFLOW flag
> > check
> > in mptcp_pm_nl_subflow_create_doit().
> > 
> 
> Hi Geliang -
> 
> Have you tested this change with mptcpd?

What do you think to test 'mptcpd' in our CI too? If you agree, I'll
try to do this, and add a new ticket about it in Github.

> 
> Is the addr flag (SUBFLOW or SIGNAL) relevant for userspace PM
> addresses? 
> It doesn't seem like a useful distinction: the userspace PM is always
> creating a new subflow in this context, since address signaling is
> handled 
> by mptcp_pm_nl_announce_doit().
> 
> Two options I see:
>   * update the userspace tools to cleanly handle missing flags for 
> userspace connections
> or
>   * have mptcp_pm_nl_subflow_create_doit() always set 
> MPTCP_PM_ADDR_FLAG_SUBFLOW for the local addr even if it's missing
> from 
> the netlink attribute (and maybe don't allow
> MPTCP_PM_ADDR_FLAG_SIGNAL?)

Thanks for your suggestions. Option 2 has been updated in v11.

-Geliang

> 
> - Mat
> 
> 
> > Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> > ---
> > net/mptcp/pm_userspace.c | 10 ++++++++--
> > 1 file changed, 8 insertions(+), 2 deletions(-)
> > 
> > diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c
> > index f1582f40f70e..ca0d6e1dfade 100644
> > --- a/net/mptcp/pm_userspace.c
> > +++ b/net/mptcp/pm_userspace.c
> > @@ -361,11 +361,18 @@ int mptcp_pm_nl_subflow_create_doit(struct
> > sk_buff *skb, struct genl_info *info)
> > 		goto create_err;
> > 	}
> > 
> > -	err = mptcp_pm_parse_addr(laddr, info, &addr_l);
> > +	err = mptcp_pm_parse_entry(laddr, info, true, &local);
> > 	if (err < 0) {
> > 		NL_SET_ERR_MSG_ATTR(info->extack, laddr, "error
> > parsing local addr");
> > 		goto create_err;
> > 	}
> > +	addr_l = local.addr;
> > +
> > +	if (!(local.flags & MPTCP_PM_ADDR_FLAG_SUBFLOW)) {
> > +		GENL_SET_ERR_MSG(info, "invalid addr flags");
> > +		err = -EINVAL;
> > +		goto create_err;
> > +	}
> > 
> > 	err = mptcp_pm_parse_addr(raddr, info, &addr_r);
> > 	if (err < 0) {
> > @@ -379,7 +386,6 @@ int mptcp_pm_nl_subflow_create_doit(struct
> > sk_buff *skb, struct genl_info *info)
> > 		goto create_err;
> > 	}
> > 
> > -	local.addr = addr_l;
> > 	err = mptcp_userspace_pm_append_new_local_addr(msk,
> > &local,
> > 						      
> > !mptcp_pm_has_addr_attr_id(laddr, info));
> > 	if (err < 0) {
> > -- 
> > 2.40.1
> > 
> > 
> > 

Re: [PATCH mptcp-next v10 10/23] mptcp: check userspace pm subflow flag

[Matthieu Baerts]

Hi Geliang,

On 06/02/2024 03:57, Geliang Tang wrote:
> Hi Mat & Matt,
> 
> On Tue, 2024-01-30 at 18:28 -0800, Mat Martineau wrote:
>> On Thu, 18 Jan 2024, Geliang Tang wrote:
>>
>>> From: Geliang Tang <tanggeliang@kylinos.cn>
>>>
>>> Just like MPTCP_PM_ADDR_FLAG_SIGNAL flag is checked in userspace PM
>>> announce mptcp_pm_nl_announce_doit(), MPTCP_PM_ADDR_FLAG_SUBFLOW
>>> flag
>>> should be checked in mptcp_pm_nl_subflow_create_doit() too.
>>>
>>> If MPTCP_PM_ADDR_FLAG_SUBFLOW flag is not set, there's no flags
>>> field
>>> in the output of dump_addr. This looks a bit strange:
>>>
>>> 	id 10 flags  10.0.3.2
>>>
>>> This patch uses mptcp_pm_parse_entry() instead of
>>> mptcp_pm_parse_addr()
>>> to get the flags of the entry. Add MPTCP_PM_ADDR_FLAG_SUBFLOW flag
>>> check
>>> in mptcp_pm_nl_subflow_create_doit().
>>>
>>
>> Hi Geliang -
>>
>> Have you tested this change with mptcpd?
> 
> What do you think to test 'mptcpd' in our CI too? If you agree, I'll
> try to do this, and add a new ticket about it in Github.

Thank you for the idea.

I don't think we should validate mptcpd when modifying the kernel. We
should not break the userspace API.

Maybe better to add this test on mptcpd side? The CI could download the
export branch, run the same
"multipath-tcp/mptcp-upstream-virtme-docker@latest" but with different
instruction in .virtme-exec-run file. Such test could run once a week.

WDYT?

Cheers,
Matt
-- 
Sponsored by the NGI0 Core fund.