Here is a small optimisation for the in-kernel PM, joined by a small behavioural change to avoid confusions, and followed by a few more tests.
- Patch 1: record fullmesh endpoints numbers, not to iterate over all endpoints to check if one is marked as fullmesh.
- Patch 2: when at least one endpoint is marked as fullmesh, only use these endpoints when reacting to an ADD_ADDR, even if there are no endpoints for this IP family: this is less confusing.
- Patch 3: reduce duplicated code to prepare the next patch.
- Patch 4: extra "bind" cases: the listen socket restrict the bind to one IP address, not allowing MP_JOIN to extra IP addresses, except if another listening socket accepts them.
Signed-off-by: Matthieu Baerts (NGI0) matttbe@kernel.org --- Matthieu Baerts (NGI0) (4): mptcp: pm: in-kernel: record fullmesh endp nb mptcp: pm: in kernel: only use fullmesh endp if any selftests: mptcp: join: do_transfer: reduce code dup selftests: mptcp: join: validate extra bind cases
include/uapi/linux/mptcp.h | 3 +- net/mptcp/pm_kernel.c | 36 ++++- net/mptcp/protocol.h | 1 + net/mptcp/sockopt.c | 2 + tools/testing/selftests/net/mptcp/mptcp_connect.c | 10 +- tools/testing/selftests/net/mptcp/mptcp_join.sh | 187 +++++++++++++++++++--- 6 files changed, 213 insertions(+), 26 deletions(-) --- base-commit: 01cc760632b875c4ad0d8fec0b0c01896b8a36d4 change-id: 20251101-net-next-mptcp-fm-endp-nb-bind-cf7ab688d9f1
Best regards,
Instead of iterating over all endpoints, under RCU read lock, just to check if one of them as the fullmesh flag, we can keep a counter of fullmesh endpoint, similar to what is done with the other flags.
This counter is now checked, before iterating over all endpoints.
Similar to the other counters, this new one is also exposed. A userspace app can then know when it is being used in a fullmesh mode, with potentially (too) many subflows.
Reviewed-by: Geliang Tang geliang@kernel.org Signed-off-by: Matthieu Baerts (NGI0) matttbe@kernel.org --- include/uapi/linux/mptcp.h | 3 ++- net/mptcp/pm_kernel.c | 38 +++++++++++++++++++++++++++++++++++--- net/mptcp/protocol.h | 1 + net/mptcp/sockopt.c | 2 ++ 4 files changed, 40 insertions(+), 4 deletions(-)
diff --git a/include/uapi/linux/mptcp.h b/include/uapi/linux/mptcp.h index 87cfab874e24..04eea6d1d0a9 100644 --- a/include/uapi/linux/mptcp.h +++ b/include/uapi/linux/mptcp.h @@ -70,7 +70,8 @@ struct mptcp_info { __u64 mptcpi_bytes_acked; __u8 mptcpi_subflows_total; __u8 mptcpi_endp_laminar_max; - __u8 reserved[2]; + __u8 mptcpi_endp_fullmesh_max; + __u8 reserved; __u32 mptcpi_last_data_sent; __u32 mptcpi_last_data_recv; __u32 mptcpi_last_ack_recv; diff --git a/net/mptcp/pm_kernel.c b/net/mptcp/pm_kernel.c index 2ae95476dba3..e2918c68ff02 100644 --- a/net/mptcp/pm_kernel.c +++ b/net/mptcp/pm_kernel.c @@ -22,6 +22,7 @@ struct pm_nl_pernet { u8 endp_signal_max; u8 endp_subflow_max; u8 endp_laminar_max; + u8 endp_fullmesh_max; u8 limit_add_addr_accepted; u8 limit_extra_subflows; u8 next_id; @@ -70,6 +71,14 @@ u8 mptcp_pm_get_endp_laminar_max(const struct mptcp_sock *msk) } EXPORT_SYMBOL_GPL(mptcp_pm_get_endp_laminar_max);
+u8 mptcp_pm_get_endp_fullmesh_max(const struct mptcp_sock *msk) +{ + struct pm_nl_pernet *pernet = pm_nl_get_pernet_from_msk(msk); + + return READ_ONCE(pernet->endp_fullmesh_max); +} +EXPORT_SYMBOL_GPL(mptcp_pm_get_endp_fullmesh_max); + u8 mptcp_pm_get_limit_add_addr_accepted(const struct mptcp_sock *msk) { struct pm_nl_pernet *pernet = pm_nl_get_pernet_from_msk(msk); @@ -603,9 +612,12 @@ fill_local_addresses_vec(struct mptcp_sock *msk, struct mptcp_addr_info *remote, int i;
/* If there is at least one MPTCP endpoint with a fullmesh flag */ - i = fill_local_addresses_vec_fullmesh(msk, remote, locals, c_flag_case); - if (i) - return i; + if (mptcp_pm_get_endp_fullmesh_max(msk)) { + i = fill_local_addresses_vec_fullmesh(msk, remote, locals, + c_flag_case); + if (i) + return i; + }
/* If there is at least one MPTCP endpoint with a laminar flag */ if (mptcp_pm_get_endp_laminar_max(msk)) @@ -790,6 +802,10 @@ static int mptcp_pm_nl_append_new_local_addr(struct pm_nl_pernet *pernet, addr_max = pernet->endp_laminar_max; WRITE_ONCE(pernet->endp_laminar_max, addr_max + 1); } + if (entry->flags & MPTCP_PM_ADDR_FLAG_FULLMESH) { + addr_max = pernet->endp_fullmesh_max; + WRITE_ONCE(pernet->endp_fullmesh_max, addr_max + 1); + }
pernet->endpoints++; if (!entry->addr.port) @@ -1187,6 +1203,10 @@ int mptcp_pm_nl_del_addr_doit(struct sk_buff *skb, struct genl_info *info) addr_max = pernet->endp_laminar_max; WRITE_ONCE(pernet->endp_laminar_max, addr_max - 1); } + if (entry->flags & MPTCP_PM_ADDR_FLAG_FULLMESH) { + addr_max = pernet->endp_fullmesh_max; + WRITE_ONCE(pernet->endp_fullmesh_max, addr_max - 1); + }
pernet->endpoints--; list_del_rcu(&entry->list); @@ -1502,6 +1522,18 @@ int mptcp_pm_nl_set_flags(struct mptcp_pm_addr_entry *local, changed = (local->flags ^ entry->flags) & mask; entry->flags = (entry->flags & ~mask) | (local->flags & mask); *local = *entry; + + if (changed & MPTCP_PM_ADDR_FLAG_FULLMESH) { + u8 addr_max = pernet->endp_fullmesh_max; + + if (entry->flags & MPTCP_PM_ADDR_FLAG_FULLMESH) + addr_max++; + else + addr_max--; + + WRITE_ONCE(pernet->endp_fullmesh_max, addr_max); + } + spin_unlock_bh(&pernet->lock);
mptcp_pm_nl_set_flags_all(net, local, changed); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 379a88e14e8d..9a3429175758 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -1183,6 +1183,7 @@ void __mptcp_pm_kernel_worker(struct mptcp_sock *msk); u8 mptcp_pm_get_endp_signal_max(const struct mptcp_sock *msk); u8 mptcp_pm_get_endp_subflow_max(const struct mptcp_sock *msk); u8 mptcp_pm_get_endp_laminar_max(const struct mptcp_sock *msk); +u8 mptcp_pm_get_endp_fullmesh_max(const struct mptcp_sock *msk); u8 mptcp_pm_get_limit_add_addr_accepted(const struct mptcp_sock *msk); u8 mptcp_pm_get_limit_extra_subflows(const struct mptcp_sock *msk);
diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index a28a48385885..de90a2897d2d 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -982,6 +982,8 @@ void mptcp_diag_fill_info(struct mptcp_sock *msk, struct mptcp_info *info) mptcp_pm_get_endp_subflow_max(msk); info->mptcpi_endp_laminar_max = mptcp_pm_get_endp_laminar_max(msk); + info->mptcpi_endp_fullmesh_max = + mptcp_pm_get_endp_fullmesh_max(msk); }
if (__mptcp_check_fallback(msk))
Our documentation is saying that the in-kernel PM is only using fullmesh endpoints to establish subflows to announced addresses when at least one endpoint has a fullmesh flag. But this was not totally correct: only fullmesh endpoints were used if at least one endpoint *from the same address family as the received ADD_ADDR* has the fullmesh flag.
This is confusing, and it seems clearer not to have differences depending on the address family.
So, now, when at least one MPTCP endpoint has a fullmesh flag, the local addresses are picked from all fullmesh endpoints, which might be 0 if there are no endpoints for the correct address family.
One selftest needs to be adapted for this behaviour change.
Reviewed-by: Geliang Tang geliang@kernel.org Signed-off-by: Matthieu Baerts (NGI0) matttbe@kernel.org --- net/mptcp/pm_kernel.c | 10 +++------- tools/testing/selftests/net/mptcp/mptcp_join.sh | 6 +++++- 2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/net/mptcp/pm_kernel.c b/net/mptcp/pm_kernel.c index e2918c68ff02..e50721c670d0 100644 --- a/net/mptcp/pm_kernel.c +++ b/net/mptcp/pm_kernel.c @@ -609,15 +609,11 @@ fill_local_addresses_vec(struct mptcp_sock *msk, struct mptcp_addr_info *remote, struct mptcp_pm_local *locals) { bool c_flag_case = remote->id && mptcp_pm_add_addr_c_flag_case(msk); - int i;
/* If there is at least one MPTCP endpoint with a fullmesh flag */ - if (mptcp_pm_get_endp_fullmesh_max(msk)) { - i = fill_local_addresses_vec_fullmesh(msk, remote, locals, - c_flag_case); - if (i) - return i; - } + if (mptcp_pm_get_endp_fullmesh_max(msk)) + return fill_local_addresses_vec_fullmesh(msk, remote, locals, + c_flag_case);
/* If there is at least one MPTCP endpoint with a laminar flag */ if (mptcp_pm_get_endp_laminar_max(msk)) diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index 78a1aa4ecff2..e7a498dd5a46 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -2952,7 +2952,11 @@ mixed_tests() pm_nl_add_endpoint $ns1 10.0.1.1 flags signal speed=slow \ run_tests $ns1 $ns2 dead:beef:2::1 - chk_join_nr 1 1 1 + if mptcp_lib_kallsyms_has "mptcp_pm_get_endp_fullmesh_max$"; then + chk_join_nr 0 0 0 + else + chk_join_nr 1 1 1 + fi fi
# fullmesh still tries to create all the possibly subflows with
The same extra long commands are present twice, with small differences: the variable for the stdin file is different.
Use new dedicated variables in one command to avoid this code duplication.
Reviewed-by: Geliang Tang geliang@kernel.org Signed-off-by: Matthieu Baerts (NGI0) matttbe@kernel.org --- tools/testing/selftests/net/mptcp/mptcp_join.sh | 30 +++++++++---------------- 1 file changed, 11 insertions(+), 19 deletions(-)
diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index e7a498dd5a46..4c9ee094381e 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -951,6 +951,8 @@ do_transfer() local FAILING_LINKS=${FAILING_LINKS:-""} local fastclose=${fastclose:-""} local speed=${speed:-"fast"} + local listener_in="${sin}" + local connector_in="${cin}" port=$(get_port)
:> "$cout" @@ -999,16 +1001,12 @@ do_transfer()
extra_srv_args="$extra_args $extra_srv_args" if [ "$test_linkfail" -gt 1 ];then - timeout ${timeout_test} \ - ip netns exec ${listener_ns} \ - ./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \ - $extra_srv_args "::" < "$sinfail" > "$sout" & - else - timeout ${timeout_test} \ - ip netns exec ${listener_ns} \ - ./mptcp_connect -t ${timeout_poll} -l -p $port -s ${srv_proto} \ - $extra_srv_args "::" < "$sin" > "$sout" & + listener_in="${sinfail}" fi + timeout ${timeout_test} \ + ip netns exec ${listener_ns} \ + ./mptcp_connect -t ${timeout_poll} -l -p ${port} -s ${srv_proto} \ + ${extra_srv_args} "::" < "${listener_in}" > "${sout}" & local spid=$!
mptcp_lib_wait_local_port_listen "${listener_ns}" "${port}" @@ -1020,6 +1018,7 @@ do_transfer() ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ $extra_cl_args $connect_addr < "$cin" > "$cout" & elif [ "$test_linkfail" -eq 1 ] || [ "$test_linkfail" -eq 2 ];then + connector_in="${cinsent}" ( cat "$cinfail" ; sleep 2; link_failure $listener_ns ; cat "$cinfail" ) | \ tee "$cinsent" | \ timeout ${timeout_test} \ @@ -1027,6 +1026,7 @@ do_transfer() ./mptcp_connect -t ${timeout_poll} -p $port -s ${cl_proto} \ $extra_cl_args $connect_addr > "$cout" & else + connector_in="${cinsent}" tee "$cinsent" < "$cinfail" | \ timeout ${timeout_test} \ ip netns exec ${connector_ns} \ @@ -1057,17 +1057,9 @@ do_transfer() return 1 fi
- if [ "$test_linkfail" -gt 1 ];then - check_transfer $sinfail $cout "file received by client" $trunc_size - else - check_transfer $sin $cout "file received by client" $trunc_size - fi + check_transfer $listener_in $cout "file received by client" $trunc_size retc=$? - if [ "$test_linkfail" -eq 0 ];then - check_transfer $cin $sout "file received by server" $trunc_size - else - check_transfer $cinsent $sout "file received by server" $trunc_size - fi + check_transfer $connector_in $sout "file received by server" $trunc_size rets=$?
[ $retc -eq 0 ] && [ $rets -eq 0 ]
By design, an MPTCP connection will not accept extra subflows where no MPTCP listening sockets can accept such requests.
In other words, it means that if the 'server' listens on a specific address / device, it cannot accept MP_JOIN sent to a different address / device. Except if there is another MPTCP listening socket accepting them.
This is what the new tests are validating:
- Forcing a bind on the main v4/v6 address, and checking that MP_JOIN to announced addresses are not accepted.
- Also forcing a bind on the main v4/v6 address, but before, another listening socket is created to accept additional subflows. Note that 'mptcpize run nc -l' -- or something else only doing: socket(MPTCP), bind(<IP>), listen(0) -- would be enough, but here mptcp_connect is reused not to depend on another tool just for that.
- Same as the previous one, but using v6 link-local addresses: this is a bit particular because it is required to specify the outgoing network interface when connecting to a link-local address announced by the other peer. When using the routing rules, this doesn't work (the outgoing interface is not known) ; but it does work with a 'laminar' endpoint having a specified interface.
Note that extra small modifications are needed for these tests to work:
- mptcp_connect's check_getpeername_connect() check should strip the specified interface when comparing addresses.
- With IPv6 link-local addresses, it is required to wait for them to be ready (no longer in 'tentative' mode) before using them, otherwise the bind() will not be allowed.
Link: https://github.com/multipath-tcp/mptcp_net-next/issues/591 Reviewed-by: Geliang Tang geliang@kernel.org Signed-off-by: Matthieu Baerts (NGI0) matttbe@kernel.org --- tools/testing/selftests/net/mptcp/mptcp_connect.c | 10 +- tools/testing/selftests/net/mptcp/mptcp_join.sh | 153 +++++++++++++++++++++- 2 files changed, 161 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.c b/tools/testing/selftests/net/mptcp/mptcp_connect.c index b148cadb96d0..c030b08a7195 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.c +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.c @@ -1064,6 +1064,8 @@ static void check_getpeername_connect(int fd) socklen_t salen = sizeof(ss); char a[INET6_ADDRSTRLEN]; char b[INET6_ADDRSTRLEN]; + const char *iface; + size_t len;
if (getpeername(fd, (struct sockaddr *)&ss, &salen) < 0) { perror("getpeername"); @@ -1073,7 +1075,13 @@ static void check_getpeername_connect(int fd) xgetnameinfo((struct sockaddr *)&ss, salen, a, sizeof(a), b, sizeof(b));
- if (strcmp(cfg_host, a) || strcmp(cfg_port, b)) + iface = strchr(cfg_host, '%'); + if (iface) + len = iface - cfg_host; + else + len = strlen(cfg_host) + 1; + + if (strncmp(cfg_host, a, len) || strcmp(cfg_port, b)) fprintf(stderr, "%s: %s vs %s, %s vs %s\n", __func__, cfg_host, a, cfg_port, b); } diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index 4c9ee094381e..4faf58fecc94 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -62,6 +62,7 @@ unset sflags unset fastclose unset fullmesh unset speed +unset bind_addr unset join_syn_rej unset join_csum_ns1 unset join_csum_ns2 @@ -645,6 +646,27 @@ wait_mpj() done }
+wait_ll_ready() +{ + local ns="${1}" + + local i + for i in $(seq 50); do + ip -n "${ns}" -6 addr show scope link | grep "inet6 fe80" | + grep -qw "tentative" || break + sleep 0.1 + done +} + +get_ll_addr() +{ + local ns="${1}" + local iface="${2}" + + ip -n "${ns}" -6 addr show dev "${iface}" scope link | + grep "inet6 fe80" | sed 's#.*(fe80::.*)/.*#\1#' +} + kill_events_pids() { mptcp_lib_kill_wait $evts_ns1_pid @@ -951,6 +973,7 @@ do_transfer() local FAILING_LINKS=${FAILING_LINKS:-""} local fastclose=${fastclose:-""} local speed=${speed:-"fast"} + local bind_addr=${bind_addr:-"::"} local listener_in="${sin}" local connector_in="${cin}" port=$(get_port) @@ -1006,7 +1029,7 @@ do_transfer() timeout ${timeout_test} \ ip netns exec ${listener_ns} \ ./mptcp_connect -t ${timeout_poll} -l -p ${port} -s ${srv_proto} \ - ${extra_srv_args} "::" < "${listener_in}" > "${sout}" & + ${extra_srv_args} "${bind_addr}" < "${listener_in}" > "${sout}" & local spid=$!
mptcp_lib_wait_local_port_listen "${listener_ns}" "${port}" @@ -3229,6 +3252,133 @@ add_addr_ports_tests() fi }
+bind_tests() +{ + # bind to one address should not allow extra subflows to other addresses + if reset "bind main address v4, no join v4"; then + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 10.0.2.1 flags signal + bind_addr="10.0.1.1" \ + run_tests $ns1 $ns2 10.0.1.1 + join_syn_tx=1 \ + chk_join_nr 0 0 0 + chk_add_nr 1 1 + fi + + # bind to one address should not allow extra subflows to other addresses + if reset "bind main address v6, no join v6"; then + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 dead:beef:2::1 flags signal + bind_addr="dead:beef:1::1" \ + run_tests $ns1 $ns2 dead:beef:1::1 + join_syn_tx=1 \ + chk_join_nr 0 0 0 + chk_add_nr 1 1 + fi + + # multiple binds to allow extra subflows to other addresses + if reset "multiple bind to allow joins v4"; then + local extra_bind + + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 10.0.2.1 flags signal + + # Launching another app listening on a different address + # Note: it could be a totally different app, e.g. nc, socat, ... + ip netns exec ${ns1} ./mptcp_connect -l -t -1 -p "$(get_port)" \ + -s MPTCP 10.0.2.1 & + extra_bind=$! + + bind_addr="10.0.1.1" \ + run_tests $ns1 $ns2 10.0.1.1 + chk_join_nr 1 1 1 + chk_add_nr 1 1 + + kill ${extra_bind} + fi + + # multiple binds to allow extra subflows to other addresses + if reset "multiple bind to allow joins v6"; then + local extra_bind + + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 dead:beef:2::1 flags signal + + # Launching another app listening on a different address + # Note: it could be a totally different app, e.g. nc, socat, ... + ip netns exec ${ns1} ./mptcp_connect -l -t -1 -p "$(get_port)" \ + -s MPTCP dead:beef:2::1 & + extra_bind=$! + + bind_addr="dead:beef:1::1" \ + run_tests $ns1 $ns2 dead:beef:1::1 + chk_join_nr 1 1 1 + chk_add_nr 1 1 + + kill ${extra_bind} + fi + + # multiple binds to allow extra subflows to other addresses: v6 LL case + if reset "multiple bind to allow joins v6 link-local routing"; then + local extra_bind ns1ll1 ns1ll2 + + ns1ll1="$(get_ll_addr $ns1 ns1eth1)" + ns1ll2="$(get_ll_addr $ns1 ns1eth2)" + + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 "${ns1ll2}" flags signal + + wait_ll_ready $ns1 # to be able to bind + wait_ll_ready $ns2 # also needed to bind on the client side + ip netns exec ${ns1} ./mptcp_connect -l -t -1 -p "$(get_port)" \ + -s MPTCP "${ns1ll2}%ns1eth2" & + extra_bind=$! + + bind_addr="${ns1ll1}%ns1eth1" \ + run_tests $ns1 $ns2 "${ns1ll1}%ns2eth1" + # it is not possible to connect to the announced LL addr without + # specifying the outgoing interface. + join_connect_err=1 \ + chk_join_nr 0 0 0 + chk_add_nr 1 1 + + kill ${extra_bind} + fi + + # multiple binds to allow extra subflows to v6 LL addresses: laminar + if reset "multiple bind to allow joins v6 link-local laminar" && + continue_if mptcp_lib_kallsyms_has "mptcp_pm_get_endp_laminar_max$"; then + local extra_bind ns1ll1 ns1ll2 ns2ll2 + + ns1ll1="$(get_ll_addr $ns1 ns1eth1)" + ns1ll2="$(get_ll_addr $ns1 ns1eth2)" + ns2ll2="$(get_ll_addr $ns2 ns2eth2)" + + pm_nl_set_limits $ns1 0 2 + pm_nl_set_limits $ns2 2 2 + pm_nl_add_endpoint $ns1 "${ns1ll2}" flags signal + pm_nl_add_endpoint $ns2 "${ns2ll2}" flags laminar dev ns2eth2 + + wait_ll_ready $ns1 # to be able to bind + wait_ll_ready $ns2 # also needed to bind on the client side + ip netns exec ${ns1} ./mptcp_connect -l -t -1 -p "$(get_port)" \ + -s MPTCP "${ns1ll2}%ns1eth2" & + extra_bind=$! + + bind_addr="${ns1ll1}%ns1eth1" \ + run_tests $ns1 $ns2 "${ns1ll1}%ns2eth1" + chk_join_nr 1 1 1 + chk_add_nr 1 1 + + kill ${extra_bind} + fi +} + syncookies_tests() { # single subflow, syncookies @@ -4183,6 +4333,7 @@ all_tests_sorted=( M@mixed_tests b@backup_tests p@add_addr_ports_tests + B@bind_tests k@syncookies_tests S@checksum_tests d@deny_join_id0_tests
Hello:
This series was applied to netdev/net-next.git (main) by Jakub Kicinski kuba@kernel.org:
On Sat, 01 Nov 2025 18:56:50 +0100 you wrote:
Here is a small optimisation for the in-kernel PM, joined by a small behavioural change to avoid confusions, and followed by a few more tests.
- Patch 1: record fullmesh endpoints numbers, not to iterate over all endpoints to check if one is marked as fullmesh.
[...]
Here is the summary with links: - [net-next,1/4] mptcp: pm: in-kernel: record fullmesh endp nb https://git.kernel.org/netdev/net-next/c/f88191c7f361 - [net-next,2/4] mptcp: pm: in kernel: only use fullmesh endp if any https://git.kernel.org/netdev/net-next/c/e461e8a799a2 - [net-next,3/4] selftests: mptcp: join: do_transfer: reduce code dup https://git.kernel.org/netdev/net-next/c/4a6220a453c8 - [net-next,4/4] selftests: mptcp: join: validate extra bind cases https://git.kernel.org/netdev/net-next/c/5c59df126bae
You are awesome, thank you!
linux-kselftest-mirror@lists.linaro.org
-
Matthieu Baerts (NGI0) -
patchwork-bot+netdevbpf＠kernel.org