* [PATCH net-next 1/6] mptcp: add sysctl allow_join_initial_addr_port
2021-06-22 19:25 [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes Mat Martineau
@ 2021-06-22 19:25 ` Mat Martineau
2021-06-22 19:25 ` [PATCH net-next 2/6] mptcp: add allow_join_id0 in mptcp_out_options Mat Martineau
` (5 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-06-22 19:25 UTC (permalink / raw)
To: netdev
Cc: Geliang Tang, davem, kuba, matthieu.baerts, mptcp,
Florian Westphal, Paolo Abeni, Mat Martineau
From: Geliang Tang <geliangtang@gmail.com>
This patch added a new sysctl, named allow_join_initial_addr_port, to
control whether allow peers to send join requests to the IP address and
port number used by the initial subflow.
Suggested-by: Florian Westphal <fw@strlen.de>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Geliang Tang <geliangtang@gmail.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
Documentation/networking/mptcp-sysctl.rst | 13 +++++++++++++
net/mptcp/ctrl.c | 16 ++++++++++++++++
net/mptcp/protocol.h | 1 +
3 files changed, 30 insertions(+)
diff --git a/Documentation/networking/mptcp-sysctl.rst b/Documentation/networking/mptcp-sysctl.rst
index ee06fd782465..76d939e688b8 100644
--- a/Documentation/networking/mptcp-sysctl.rst
+++ b/Documentation/networking/mptcp-sysctl.rst
@@ -32,3 +32,16 @@ checksum_enabled - BOOLEAN
per-namespace sysctl.
Default: 0
+
+allow_join_initial_addr_port - BOOLEAN
+ Allow peers to send join requests to the IP address and port number used
+ by the initial subflow if the value is 1. This controls a flag that is
+ sent to the peer at connection time, and whether such join requests are
+ accepted or denied.
+
+ Joins to addresses advertised with ADD_ADDR are not affected by this
+ value.
+
+ This is a per-namespace sysctl.
+
+ Default: 1
diff --git a/net/mptcp/ctrl.c b/net/mptcp/ctrl.c
index 6c2639bb9c19..7d738bd06f2c 100644
--- a/net/mptcp/ctrl.c
+++ b/net/mptcp/ctrl.c
@@ -24,6 +24,7 @@ struct mptcp_pernet {
u8 mptcp_enabled;
unsigned int add_addr_timeout;
u8 checksum_enabled;
+ u8 allow_join_initial_addr_port;
};
static struct mptcp_pernet *mptcp_get_pernet(struct net *net)
@@ -46,11 +47,17 @@ int mptcp_is_checksum_enabled(struct net *net)
return mptcp_get_pernet(net)->checksum_enabled;
}
+int mptcp_allow_join_id0(struct net *net)
+{
+ return mptcp_get_pernet(net)->allow_join_initial_addr_port;
+}
+
static void mptcp_pernet_set_defaults(struct mptcp_pernet *pernet)
{
pernet->mptcp_enabled = 1;
pernet->add_addr_timeout = TCP_RTO_MAX;
pernet->checksum_enabled = 0;
+ pernet->allow_join_initial_addr_port = 1;
}
#ifdef CONFIG_SYSCTL
@@ -80,6 +87,14 @@ static struct ctl_table mptcp_sysctl_table[] = {
.extra1 = SYSCTL_ZERO,
.extra2 = SYSCTL_ONE
},
+ {
+ .procname = "allow_join_initial_addr_port",
+ .maxlen = sizeof(u8),
+ .mode = 0644,
+ .proc_handler = proc_dou8vec_minmax,
+ .extra1 = SYSCTL_ZERO,
+ .extra2 = SYSCTL_ONE
+ },
{}
};
@@ -98,6 +113,7 @@ static int mptcp_pernet_new_table(struct net *net, struct mptcp_pernet *pernet)
table[0].data = &pernet->mptcp_enabled;
table[1].data = &pernet->add_addr_timeout;
table[2].data = &pernet->checksum_enabled;
+ table[3].data = &pernet->allow_join_initial_addr_port;
hdr = register_net_sysctl(net, MPTCP_SYSCTL_PATH, table);
if (!hdr)
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 160c2ab09f19..9aab5fb54716 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -540,6 +540,7 @@ static inline void mptcp_subflow_delegated_done(struct mptcp_subflow_context *su
int mptcp_is_enabled(struct net *net);
unsigned int mptcp_get_add_addr_timeout(struct net *net);
int mptcp_is_checksum_enabled(struct net *net);
+int mptcp_allow_join_id0(struct net *net);
void mptcp_subflow_fully_established(struct mptcp_subflow_context *subflow,
struct mptcp_options_received *mp_opt);
bool mptcp_subflow_data_available(struct sock *sk);
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 2/6] mptcp: add allow_join_id0 in mptcp_out_options
2021-06-22 19:25 [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes Mat Martineau
2021-06-22 19:25 ` [PATCH net-next 1/6] mptcp: add sysctl allow_join_initial_addr_port Mat Martineau
@ 2021-06-22 19:25 ` Mat Martineau
2021-06-22 19:25 ` [PATCH net-next 3/6] mptcp: add deny_join_id0 in mptcp_options_received Mat Martineau
` (4 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-06-22 19:25 UTC (permalink / raw)
To: netdev
Cc: Geliang Tang, davem, kuba, matthieu.baerts, mptcp, Paolo Abeni,
Mat Martineau
From: Geliang Tang <geliangtang@gmail.com>
This patch defined a new flag MPTCP_CAP_DENY_JOIN_ID0 for the third bit,
labeled "C" of the MP_CAPABLE option.
Add a new flag allow_join_id0 in struct mptcp_out_options. If this flag is
set, send out the MP_CAPABLE option with the flag MPTCP_CAP_DENY_JOIN_ID0.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Geliang Tang <geliangtang@gmail.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
include/net/mptcp.h | 3 ++-
net/mptcp/options.c | 6 ++++++
net/mptcp/protocol.h | 6 ++++--
net/mptcp/subflow.c | 1 +
4 files changed, 13 insertions(+), 3 deletions(-)
diff --git a/include/net/mptcp.h b/include/net/mptcp.h
index d61bbbf11979..cb580b06152f 100644
--- a/include/net/mptcp.h
+++ b/include/net/mptcp.h
@@ -67,7 +67,8 @@ struct mptcp_out_options {
u8 backup;
u8 reset_reason:4,
reset_transient:1,
- csum_reqd:1;
+ csum_reqd:1,
+ allow_join_id0:1;
u32 nonce;
u64 thmac;
u32 token;
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 25189595ed1d..7a4b6d0bf3f6 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -402,6 +402,7 @@ bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb,
if (subflow->request_mptcp) {
opts->suboptions = OPTION_MPTCP_MPC_SYN;
opts->csum_reqd = mptcp_is_checksum_enabled(sock_net(sk));
+ opts->allow_join_id0 = mptcp_allow_join_id0(sock_net(sk));
*size = TCPOLEN_MPTCP_MPC_SYN;
return true;
} else if (subflow->request_join) {
@@ -490,6 +491,7 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
opts->sndr_key = subflow->local_key;
opts->rcvr_key = subflow->remote_key;
opts->csum_reqd = READ_ONCE(msk->csum_enabled);
+ opts->allow_join_id0 = mptcp_allow_join_id0(sock_net(sk));
/* Section 3.1.
* The MP_CAPABLE option is carried on the SYN, SYN/ACK, and ACK
@@ -827,6 +829,7 @@ bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
opts->suboptions = OPTION_MPTCP_MPC_SYNACK;
opts->sndr_key = subflow_req->local_key;
opts->csum_reqd = subflow_req->csum_reqd;
+ opts->allow_join_id0 = subflow_req->allow_join_id0;
*size = TCPOLEN_MPTCP_MPC_SYNACK;
pr_debug("subflow_req=%p, local_key=%llu",
subflow_req, subflow_req->local_key);
@@ -1201,6 +1204,9 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
if (opts->csum_reqd)
flag |= MPTCP_CAP_CHECKSUM_REQD;
+ if (!opts->allow_join_id0)
+ flag |= MPTCP_CAP_DENY_JOIN_ID0;
+
*ptr++ = mptcp_option(MPTCPOPT_MP_CAPABLE, len,
MPTCP_SUPPORTED_VERSION,
flag);
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 9aab5fb54716..f2326f6074b9 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -79,8 +79,9 @@
#define MPTCP_VERSION_MASK (0x0F)
#define MPTCP_CAP_CHECKSUM_REQD BIT(7)
#define MPTCP_CAP_EXTENSIBILITY BIT(6)
+#define MPTCP_CAP_DENY_JOIN_ID0 BIT(5)
#define MPTCP_CAP_HMAC_SHA256 BIT(0)
-#define MPTCP_CAP_FLAG_MASK (0x3F)
+#define MPTCP_CAP_FLAG_MASK (0x1F)
/* MPTCP DSS flags */
#define MPTCP_DSS_DATA_FIN BIT(4)
@@ -350,7 +351,8 @@ struct mptcp_subflow_request_sock {
u16 mp_capable : 1,
mp_join : 1,
backup : 1,
- csum_reqd : 1;
+ csum_reqd : 1,
+ allow_join_id0 : 1;
u8 local_id;
u8 remote_id;
u64 local_key;
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 585951e7e52f..e9e8ce862218 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -109,6 +109,7 @@ static void subflow_init_req(struct request_sock *req, const struct sock *sk_lis
subflow_req->mp_capable = 0;
subflow_req->mp_join = 0;
subflow_req->csum_reqd = mptcp_is_checksum_enabled(sock_net(sk_listener));
+ subflow_req->allow_join_id0 = mptcp_allow_join_id0(sock_net(sk_listener));
subflow_req->msk = NULL;
mptcp_token_init_request(req);
}
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 3/6] mptcp: add deny_join_id0 in mptcp_options_received
2021-06-22 19:25 [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes Mat Martineau
2021-06-22 19:25 ` [PATCH net-next 1/6] mptcp: add sysctl allow_join_initial_addr_port Mat Martineau
2021-06-22 19:25 ` [PATCH net-next 2/6] mptcp: add allow_join_id0 in mptcp_out_options Mat Martineau
@ 2021-06-22 19:25 ` Mat Martineau
2021-06-22 19:25 ` [PATCH net-next 4/6] selftests: mptcp: add deny_join_id0 testcases Mat Martineau
` (3 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-06-22 19:25 UTC (permalink / raw)
To: netdev
Cc: Geliang Tang, davem, kuba, matthieu.baerts, mptcp,
Florian Westphal, Paolo Abeni, Mat Martineau
From: Geliang Tang <geliangtang@gmail.com>
This patch added a new flag named deny_join_id0 in struct
mptcp_options_received. Set it when MP_CAPABLE with the flag
MPTCP_CAP_DENYJOIN_ID0 is received.
Also add a new flag remote_deny_join_id0 in struct mptcp_pm_data. When the
flag deny_join_id0 is set, set this remote_deny_join_id0 flag.
In mptcp_pm_create_subflow_or_signal_addr, if the remote_deny_join_id0 flag
is set, and the remote address id is zero, stop this connection.
Suggested-by: Florian Westphal <fw@strlen.de>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Geliang Tang <geliangtang@gmail.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
net/mptcp/options.c | 7 +++++++
net/mptcp/pm.c | 1 +
net/mptcp/pm_netlink.c | 3 ++-
net/mptcp/protocol.h | 4 +++-
net/mptcp/subflow.c | 2 ++
5 files changed, 15 insertions(+), 2 deletions(-)
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index 7a4b6d0bf3f6..a05270996613 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -83,6 +83,9 @@ static void mptcp_parse_option(const struct sk_buff *skb,
if (flags & MPTCP_CAP_CHECKSUM_REQD)
mp_opt->csum_reqd = 1;
+ if (flags & MPTCP_CAP_DENY_JOIN_ID0)
+ mp_opt->deny_join_id0 = 1;
+
mp_opt->mp_capable = 1;
if (opsize >= TCPOLEN_MPTCP_MPC_SYNACK) {
mp_opt->sndr_key = get_unaligned_be64(ptr);
@@ -360,6 +363,7 @@ void mptcp_get_options(const struct sock *sk,
mp_opt->mp_prio = 0;
mp_opt->reset = 0;
mp_opt->csum_reqd = READ_ONCE(msk->csum_enabled);
+ mp_opt->deny_join_id0 = 0;
length = (th->doff * 4) - sizeof(struct tcphdr);
ptr = (const unsigned char *)(th + 1);
@@ -908,6 +912,9 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *ssk,
return false;
}
+ if (mp_opt->deny_join_id0)
+ WRITE_ONCE(msk->pm.remote_deny_join_id0, true);
+
if (unlikely(!READ_ONCE(msk->pm.server_side)))
pr_warn_once("bogus mpc option on established client sk");
mptcp_subflow_fully_established(subflow, mp_opt);
diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c
index 9d00fa6d22e9..639271e09604 100644
--- a/net/mptcp/pm.c
+++ b/net/mptcp/pm.c
@@ -320,6 +320,7 @@ void mptcp_pm_data_init(struct mptcp_sock *msk)
WRITE_ONCE(msk->pm.addr_signal, 0);
WRITE_ONCE(msk->pm.accept_addr, false);
WRITE_ONCE(msk->pm.accept_subflow, false);
+ WRITE_ONCE(msk->pm.remote_deny_join_id0, false);
msk->pm.status = 0;
spin_lock_init(&msk->pm.lock);
diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index d4732a4f223e..d2591ebf01d9 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -451,7 +451,8 @@ static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk)
/* check if should create a new subflow */
if (msk->pm.local_addr_used < local_addr_max &&
- msk->pm.subflows < subflows_max) {
+ msk->pm.subflows < subflows_max &&
+ !READ_ONCE(msk->pm.remote_deny_join_id0)) {
local = select_local_address(pernet, msk);
if (local) {
struct mptcp_addr_info remote = { 0 };
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index f2326f6074b9..f4eaa5f57e3f 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -138,7 +138,8 @@ struct mptcp_options_received {
mp_prio : 1,
echo : 1,
csum_reqd : 1,
- backup : 1;
+ backup : 1,
+ deny_join_id0 : 1;
u32 token;
u32 nonce;
u64 thmac;
@@ -193,6 +194,7 @@ struct mptcp_pm_data {
bool work_pending;
bool accept_addr;
bool accept_subflow;
+ bool remote_deny_join_id0;
u8 add_addr_signaled;
u8 add_addr_accepted;
u8 local_addr_used;
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index e9e8ce862218..d55f4ef736a5 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -408,6 +408,8 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb)
if (mp_opt.csum_reqd)
WRITE_ONCE(mptcp_sk(parent)->csum_enabled, true);
+ if (mp_opt.deny_join_id0)
+ WRITE_ONCE(mptcp_sk(parent)->pm.remote_deny_join_id0, true);
subflow->mp_capable = 1;
subflow->can_ack = 1;
subflow->remote_key = mp_opt.sndr_key;
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 4/6] selftests: mptcp: add deny_join_id0 testcases
2021-06-22 19:25 [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes Mat Martineau
` (2 preceding siblings ...)
2021-06-22 19:25 ` [PATCH net-next 3/6] mptcp: add deny_join_id0 in mptcp_options_received Mat Martineau
@ 2021-06-22 19:25 ` Mat Martineau
2021-06-22 19:25 ` [PATCH net-next 5/6] selftests: mptcp: turn rp_filter off on each NIC Mat Martineau
` (2 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-06-22 19:25 UTC (permalink / raw)
To: netdev
Cc: Geliang Tang, davem, kuba, matthieu.baerts, mptcp, Paolo Abeni,
Mat Martineau
From: Geliang Tang <geliangtang@gmail.com>
This patch added a new argument '-d' for mptcp_join.sh script, to invoke
the testcases for the MP_CAPABLE 'C' flag.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Geliang Tang <geliangtang@gmail.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
.../testing/selftests/net/mptcp/mptcp_join.sh | 75 ++++++++++++++++++-
1 file changed, 74 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh
index 523c7797f30a..9a191c1a5de8 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_join.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh
@@ -139,6 +139,17 @@ reset_with_checksum()
ip netns exec $ns2 sysctl -q net.mptcp.checksum_enabled=$ns2_enable
}
+reset_with_allow_join_id0()
+{
+ local ns1_enable=$1
+ local ns2_enable=$2
+
+ reset
+
+ ip netns exec $ns1 sysctl -q net.mptcp.allow_join_initial_addr_port=$ns1_enable
+ ip netns exec $ns2 sysctl -q net.mptcp.allow_join_initial_addr_port=$ns2_enable
+}
+
ip -Version > /dev/null 2>&1
if [ $? -ne 0 ];then
echo "SKIP: Could not run test without ip tool"
@@ -1462,6 +1473,63 @@ checksum_tests()
chk_csum_nr "checksum test 1 0"
}
+deny_join_id0_tests()
+{
+ # subflow allow join id0 ns1
+ reset_with_allow_join_id0 1 0
+ ip netns exec $ns1 ./pm_nl_ctl limits 1 1
+ ip netns exec $ns2 ./pm_nl_ctl limits 1 1
+ ip netns exec $ns2 ./pm_nl_ctl add 10.0.3.2 flags subflow
+ run_tests $ns1 $ns2 10.0.1.1
+ chk_join_nr "single subflow allow join id0 ns1" 1 1 1
+
+ # subflow allow join id0 ns2
+ reset_with_allow_join_id0 0 1
+ ip netns exec $ns1 ./pm_nl_ctl limits 1 1
+ ip netns exec $ns2 ./pm_nl_ctl limits 1 1
+ ip netns exec $ns2 ./pm_nl_ctl add 10.0.3.2 flags subflow
+ run_tests $ns1 $ns2 10.0.1.1
+ chk_join_nr "single subflow allow join id0 ns2" 0 0 0
+
+ # signal address allow join id0 ns1
+ # ADD_ADDRs are not affected by allow_join_id0 value.
+ reset_with_allow_join_id0 1 0
+ ip netns exec $ns1 ./pm_nl_ctl limits 1 1
+ ip netns exec $ns2 ./pm_nl_ctl limits 1 1
+ ip netns exec $ns1 ./pm_nl_ctl add 10.0.2.1 flags signal
+ run_tests $ns1 $ns2 10.0.1.1
+ chk_join_nr "signal address allow join id0 ns1" 1 1 1
+ chk_add_nr 1 1
+
+ # signal address allow join id0 ns2
+ # ADD_ADDRs are not affected by allow_join_id0 value.
+ reset_with_allow_join_id0 0 1
+ ip netns exec $ns1 ./pm_nl_ctl limits 1 1
+ ip netns exec $ns2 ./pm_nl_ctl limits 1 1
+ ip netns exec $ns1 ./pm_nl_ctl add 10.0.2.1 flags signal
+ run_tests $ns1 $ns2 10.0.1.1
+ chk_join_nr "signal address allow join id0 ns2" 1 1 1
+ chk_add_nr 1 1
+
+ # subflow and address allow join id0 ns1
+ reset_with_allow_join_id0 1 0
+ ip netns exec $ns1 ./pm_nl_ctl limits 2 2
+ ip netns exec $ns2 ./pm_nl_ctl limits 2 2
+ ip netns exec $ns1 ./pm_nl_ctl add 10.0.2.1 flags signal
+ ip netns exec $ns2 ./pm_nl_ctl add 10.0.3.2 flags subflow
+ run_tests $ns1 $ns2 10.0.1.1
+ chk_join_nr "subflow and address allow join id0 1" 2 2 2
+
+ # subflow and address allow join id0 ns2
+ reset_with_allow_join_id0 0 1
+ ip netns exec $ns1 ./pm_nl_ctl limits 2 2
+ ip netns exec $ns2 ./pm_nl_ctl limits 2 2
+ ip netns exec $ns1 ./pm_nl_ctl add 10.0.2.1 flags signal
+ ip netns exec $ns2 ./pm_nl_ctl add 10.0.3.2 flags subflow
+ run_tests $ns1 $ns2 10.0.1.1
+ chk_join_nr "subflow and address allow join id0 2" 1 1 1
+}
+
all_tests()
{
subflows_tests
@@ -1476,6 +1544,7 @@ all_tests()
add_addr_ports_tests
syncookies_tests
checksum_tests
+ deny_join_id0_tests
}
usage()
@@ -1493,6 +1562,7 @@ usage()
echo " -p add_addr_ports_tests"
echo " -k syncookies_tests"
echo " -S checksum_tests"
+ echo " -d deny_join_id0_tests"
echo " -c capture pcap files"
echo " -C enable data checksum"
echo " -h help"
@@ -1528,7 +1598,7 @@ if [ $do_all_tests -eq 1 ]; then
exit $ret
fi
-while getopts 'fsltra64bpkchCS' opt; do
+while getopts 'fsltra64bpkdchCS' opt; do
case $opt in
f)
subflows_tests
@@ -1566,6 +1636,9 @@ while getopts 'fsltra64bpkchCS' opt; do
S)
checksum_tests
;;
+ d)
+ deny_join_id0_tests
+ ;;
c)
;;
C)
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 5/6] selftests: mptcp: turn rp_filter off on each NIC
2021-06-22 19:25 [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes Mat Martineau
` (3 preceding siblings ...)
2021-06-22 19:25 ` [PATCH net-next 4/6] selftests: mptcp: add deny_join_id0 testcases Mat Martineau
@ 2021-06-22 19:25 ` Mat Martineau
2021-06-22 19:25 ` [PATCH net-next 6/6] mptcp: refine mptcp_cleanup_rbuf Mat Martineau
2021-06-22 21:50 ` [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes patchwork-bot+netdevbpf
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-06-22 19:25 UTC (permalink / raw)
To: netdev
Cc: Yonglong Li, davem, kuba, matthieu.baerts, mptcp, Paolo Abeni,
Mat Martineau
From: Yonglong Li <liyonglong@chinatelecom.cn>
To turn rp_filter off we should:
echo 0 > /proc/sys/net/ipv4/conf/default/rp_filter
and
echo 0 > /proc/sys/net/ipv4/conf/all/rp_filter
before NIC created.
Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Yonglong Li <liyonglong@chinatelecom.cn>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
tools/testing/selftests/net/mptcp/simult_flows.sh | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/mptcp/simult_flows.sh b/tools/testing/selftests/net/mptcp/simult_flows.sh
index 3aeef3bcb101..fd63ebfe9a2b 100755
--- a/tools/testing/selftests/net/mptcp/simult_flows.sh
+++ b/tools/testing/selftests/net/mptcp/simult_flows.sh
@@ -60,6 +60,8 @@ setup()
for i in "$ns1" "$ns2" "$ns3";do
ip netns add $i || exit $ksft_skip
ip -net $i link set lo up
+ ip netns exec $i sysctl -q net.ipv4.conf.all.rp_filter=0
+ ip netns exec $i sysctl -q net.ipv4.conf.default.rp_filter=0
done
ip link add ns1eth1 netns "$ns1" type veth peer name ns2eth1 netns "$ns2"
@@ -80,7 +82,6 @@ setup()
ip netns exec "$ns1" ./pm_nl_ctl limits 1 1
ip netns exec "$ns1" ./pm_nl_ctl add 10.0.2.1 dev ns1eth2 flags subflow
- ip netns exec "$ns1" sysctl -q net.ipv4.conf.all.rp_filter=0
ip -net "$ns2" addr add 10.0.1.2/24 dev ns2eth1
ip -net "$ns2" addr add dead:beef:1::2/64 dev ns2eth1 nodad
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 6/6] mptcp: refine mptcp_cleanup_rbuf
2021-06-22 19:25 [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes Mat Martineau
` (4 preceding siblings ...)
2021-06-22 19:25 ` [PATCH net-next 5/6] selftests: mptcp: turn rp_filter off on each NIC Mat Martineau
@ 2021-06-22 19:25 ` Mat Martineau
2021-06-22 21:50 ` [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes patchwork-bot+netdevbpf
6 siblings, 0 replies; 8+ messages in thread
From: Mat Martineau @ 2021-06-22 19:25 UTC (permalink / raw)
To: netdev; +Cc: Paolo Abeni, davem, kuba, matthieu.baerts, mptcp, Mat Martineau
From: Paolo Abeni <pabeni@redhat.com>
The current cleanup rbuf tries a bit too hard to avoid acquiring
the subflow socket lock. We may end-up delaying the needed ack,
or skip acking a blocked subflow.
Address the above extending the conditions used to trigger the cleanup
to reflect more closely what TCP does and invoking tcp_cleanup_rbuf()
on all the active subflows.
Note that we can't replicate the exact tests implemented in
tcp_cleanup_rbuf(), as MPTCP lacks some of the required info - e.g.
ping-pong mode.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
net/mptcp/protocol.c | 56 ++++++++++++++++++--------------------------
net/mptcp/protocol.h | 1 -
2 files changed, 23 insertions(+), 34 deletions(-)
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index cf75be02eb00..ce0c45dfb79e 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -442,49 +442,46 @@ static void mptcp_send_ack(struct mptcp_sock *msk)
}
}
-static bool mptcp_subflow_cleanup_rbuf(struct sock *ssk)
+static void mptcp_subflow_cleanup_rbuf(struct sock *ssk)
{
bool slow;
- int ret;
slow = lock_sock_fast(ssk);
- ret = tcp_can_send_ack(ssk);
- if (ret)
+ if (tcp_can_send_ack(ssk))
tcp_cleanup_rbuf(ssk, 1);
unlock_sock_fast(ssk, slow);
- return ret;
+}
+
+static bool mptcp_subflow_could_cleanup(const struct sock *ssk, bool rx_empty)
+{
+ const struct inet_connection_sock *icsk = inet_csk(ssk);
+ bool ack_pending = READ_ONCE(icsk->icsk_ack.pending);
+ const struct tcp_sock *tp = tcp_sk(ssk);
+
+ return (ack_pending & ICSK_ACK_SCHED) &&
+ ((READ_ONCE(tp->rcv_nxt) - READ_ONCE(tp->rcv_wup) >
+ READ_ONCE(icsk->icsk_ack.rcv_mss)) ||
+ (rx_empty && ack_pending &
+ (ICSK_ACK_PUSHED2 | ICSK_ACK_PUSHED)));
}
static void mptcp_cleanup_rbuf(struct mptcp_sock *msk)
{
- struct sock *ack_hint = READ_ONCE(msk->ack_hint);
int old_space = READ_ONCE(msk->old_wspace);
struct mptcp_subflow_context *subflow;
struct sock *sk = (struct sock *)msk;
- bool cleanup;
+ int space = __mptcp_space(sk);
+ bool cleanup, rx_empty;
- /* this is a simple superset of what tcp_cleanup_rbuf() implements
- * so that we don't have to acquire the ssk socket lock most of the time
- * to do actually nothing
- */
- cleanup = __mptcp_space(sk) - old_space >= max(0, old_space);
- if (!cleanup)
- return;
+ cleanup = (space > 0) && (space >= (old_space << 1));
+ rx_empty = !atomic_read(&sk->sk_rmem_alloc);
- /* if the hinted ssk is still active, try to use it */
- if (likely(ack_hint)) {
- mptcp_for_each_subflow(msk, subflow) {
- struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
+ mptcp_for_each_subflow(msk, subflow) {
+ struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
- if (ack_hint == ssk && mptcp_subflow_cleanup_rbuf(ssk))
- return;
- }
+ if (cleanup || mptcp_subflow_could_cleanup(ssk, rx_empty))
+ mptcp_subflow_cleanup_rbuf(ssk);
}
-
- /* otherwise pick the first active subflow */
- mptcp_for_each_subflow(msk, subflow)
- if (mptcp_subflow_cleanup_rbuf(mptcp_subflow_tcp_sock(subflow)))
- return;
}
static bool mptcp_check_data_fin(struct sock *sk)
@@ -629,7 +626,6 @@ static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk,
break;
}
} while (more_data_avail);
- WRITE_ONCE(msk->ack_hint, ssk);
*bytes += moved;
return done;
@@ -1910,7 +1906,6 @@ static bool __mptcp_move_skbs(struct mptcp_sock *msk)
__mptcp_update_rmem(sk);
done = __mptcp_move_skbs_from_subflow(msk, ssk, &moved);
mptcp_data_unlock(sk);
- tcp_cleanup_rbuf(ssk, moved);
if (unlikely(ssk->sk_err))
__mptcp_error_report(sk);
@@ -1926,7 +1921,6 @@ static bool __mptcp_move_skbs(struct mptcp_sock *msk)
ret |= __mptcp_ofo_queue(msk);
__mptcp_splice_receive_queue(sk);
mptcp_data_unlock(sk);
- mptcp_cleanup_rbuf(msk);
}
if (ret)
mptcp_check_data_fin((struct sock *)msk);
@@ -2175,9 +2169,6 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk,
if (ssk == msk->last_snd)
msk->last_snd = NULL;
- if (ssk == msk->ack_hint)
- msk->ack_hint = NULL;
-
if (ssk == msk->first)
msk->first = NULL;
@@ -2392,7 +2383,6 @@ static int __mptcp_init_sock(struct sock *sk)
msk->rmem_released = 0;
msk->tx_pending_data = 0;
- msk->ack_hint = NULL;
msk->first = NULL;
inet_csk(sk)->icsk_sync_mss = mptcp_sync_mss;
WRITE_ONCE(msk->csum_enabled, mptcp_is_checksum_enabled(sock_net(sk)));
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index f4eaa5f57e3f..d8ad3270dfab 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -243,7 +243,6 @@ struct mptcp_sock {
bool use_64bit_ack; /* Set when we received a 64-bit DSN */
bool csum_enabled;
spinlock_t join_list_lock;
- struct sock *ack_hint;
struct work_struct work;
struct sk_buff *ooo_last_skb;
struct rb_root out_of_order_queue;
--
2.32.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* Re: [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes
2021-06-22 19:25 [PATCH net-next 0/6] mptcp: Connection-time 'C' flag and two fixes Mat Martineau
` (5 preceding siblings ...)
2021-06-22 19:25 ` [PATCH net-next 6/6] mptcp: refine mptcp_cleanup_rbuf Mat Martineau
@ 2021-06-22 21:50 ` patchwork-bot+netdevbpf
6 siblings, 0 replies; 8+ messages in thread
From: patchwork-bot+netdevbpf @ 2021-06-22 21:50 UTC (permalink / raw)
To: Mat Martineau; +Cc: netdev, davem, kuba, matthieu.baerts, mptcp
Hello:
This series was applied to netdev/net-next.git (refs/heads/master):
On Tue, 22 Jun 2021 12:25:17 -0700 you wrote:
> Here are six more patches from the MPTCP tree.
>
> Most of them add support for the 'C' flag in the MPTCP connection-time
> option headers. This flag affects how the initial address and port are
> treated by each peer. Normally one peer may send MP_JOIN requests to the
> remote address and port that were used when initiating the MPTCP
> connection. The 'C' bit indicates that MP_JOINs should only be sent to
> remote addresses that have been advertised with ADD_ADDR.
>
> [...]
Here is the summary with links:
- [net-next,1/6] mptcp: add sysctl allow_join_initial_addr_port
https://git.kernel.org/netdev/net-next/c/d2f77960e5b0
- [net-next,2/6] mptcp: add allow_join_id0 in mptcp_out_options
https://git.kernel.org/netdev/net-next/c/bab6b88e0560
- [net-next,3/6] mptcp: add deny_join_id0 in mptcp_options_received
https://git.kernel.org/netdev/net-next/c/df377be38725
- [net-next,4/6] selftests: mptcp: add deny_join_id0 testcases
https://git.kernel.org/netdev/net-next/c/0cddb4a6f4e3
- [net-next,5/6] selftests: mptcp: turn rp_filter off on each NIC
https://git.kernel.org/netdev/net-next/c/d8e336f77e3b
- [net-next,6/6] mptcp: refine mptcp_cleanup_rbuf
https://git.kernel.org/netdev/net-next/c/fde56eea01f9
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 8+ messages in thread