Skip to contentSkip to conversation

mptcp: mp_fail related fixes

PATCH Superseded Paolo Abeni 3 participants 11 messages Jun 15, 2022 – Jun 17, 2022

Subsystems (2)

v2 of 5 Superseded Latest is v5 →

lore
Share & tools
lore

Paolo Abeni

author maintainer
#0 lore raw

v2 cover mptcp: mp_fail related fixes

While cooking a follow-up for "mptcp: invoke MP_FAIL response when needed" I stumbled upon a few other MP_FAIL related issues hit my the self-tests while mangling the above.

All the patches for for the -net tree, and should be applied in-order around to the squashed-to patch.

v1 -> v2:
- take care of UaF in mptcp_worker with new patch 6/6

Paolo Abeni (6):
mptcp: fix error mibs accounting
mptcp: introduce MAPPING_BAD_CSUM
Squash-to: "mptcp: invoke MP_FAIL response when needed"
mptcp: fix shutdown vs fallback race
mptcp: consistent map handling on failure
mptcp: fix race on unaccepted mptcp sockets

 net/mptcp/options.c  |   8 +--
 net/mptcp/pm.c       |   5 +-
 net/mptcp/protocol.c |  61 +++++++++++++++++++----
 net/mptcp/protocol.h |  25 ++++++++--
 net/mptcp/subflow.c  | 113 ++++++++++++++++++++++++++++++++++---------
 5 files changed, 168 insertions(+), 44 deletions(-)
Signature · 1 line
2.35.3
author maintainer
#1 lore raw

PATCH v2 1/6 mptcp: fix error mibs accounting

The current accounting for MP_FAIL and FASTCLOSE is not very
accurate: both can be increased even when the related option is not really sent. Move the accounting into the correct place.

Fixes eb7f33654dc1 ("mptcp: add the mibs for MP_FAIL")
Fixes 1e75629cb964 ("mptcp: add the mibs for MP_FASTCLOSE")
---
 net/mptcp/options.c | 6 +++---
 net/mptcp/pm.c      | 1 -
 net/mptcp/subflow.c | 4 +---
 3 files changed, 4 insertions(+), 7 deletions(-)
net/mptcp/options.c +3 -3
index be3b918a6d15..0bfa6662447c 100644
@@ -765,6 +765,7 @@ static noinline bool mptcp_established_options_rst(struct sock *sk, struct sk_bu
 	opts->suboptions |= OPTION_MPTCP_RST;
 	opts->reset_transient = subflow->reset_transient;
 	opts->reset_reason = subflow->reset_reason;
+	MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPRSTTX);
 
 	return true;
 }
@@ -788,6 +789,7 @@ static bool mptcp_established_options_fastclose(struct sock *sk,
 	opts->rcvr_key = msk->remote_key;
 
 	pr_debug("FASTCLOSE key=%llu", opts->rcvr_key);
+	MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPFASTCLOSETX);
 	return true;
 }
 
@@ -807,7 +809,7 @@ static bool mptcp_established_options_mp_fail(struct sock *sk,
 	*size = TCPOLEN_MPTCP_FAIL;
 	opts->suboptions |= OPTION_MPTCP_FAIL;
 	opts->fail_seq = subflow->map_seq;
-
+	MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPFAILTX);
 	pr_debug("MP_FAIL fail_seq=%llu", opts->fail_seq);
 
 	return true;
@@ -833,13 +835,11 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
 		    mptcp_established_options_mp_fail(sk, &opt_size, remaining, opts)) {
 			*size += opt_size;
 			remaining -= opt_size;
-			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPFASTCLOSETX);
 		}
 		/* MP_RST can be used with MP_FASTCLOSE and MP_FAIL if there is room */
 		if (mptcp_established_options_rst(sk, skb, &opt_size, remaining, opts)) {
 			*size += opt_size;
 			remaining -= opt_size;
-			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPRSTTX);
 		}
 		return true;
 	}
net/mptcp/pm.c +0 -1
index 2a57d95d5492..3c7f07bb124e 100644
@@ -309,7 +309,6 @@ void mptcp_pm_mp_fail_received(struct sock *sk, u64 fail_seq)
 		pr_debug("send MP_FAIL response and infinite map");
 
 		subflow->send_mp_fail = 1;
-		MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPFAILTX);
 		subflow->send_infinite_map = 1;
 	} else {
 		pr_debug("MP_FAIL response received");
net/mptcp/subflow.c +1 -3
index 5351d54e514a..57d2d8d933d0 100644
@@ -958,10 +958,8 @@ static enum mapping_status validate_data_csum(struct sock *ssk, struct sk_buff *
 				 subflow->map_data_csum);
 	if (unlikely(csum)) {
 		MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_DATACSUMERR);
-		if (subflow->mp_join || subflow->valid_csum_seen) {
+		if (subflow->mp_join || subflow->valid_csum_seen)
 			subflow->send_mp_fail = 1;
-			MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_MPFAILTX);
-		}
 		return subflow->mp_join ? MAPPING_INVALID : MAPPING_DUMMY;
 	}
 
Signature · 1 line
2.35.3
author maintainer
#2 lore raw

PATCH v2 2/6 mptcp: introduce MAPPING_BAD_CSUM

In reply to Paolo Abeni

This allow moving a couple of conditional out of the fast path, making the code more easy to follow and will simplify the next patch.

---
 net/mptcp/subflow.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)
net/mptcp/subflow.c +9 -9
index 57d2d8d933d0..98b12a9c4eb5 100644
@@ -843,7 +843,8 @@ enum mapping_status {
 	MAPPING_INVALID,
 	MAPPING_EMPTY,
 	MAPPING_DATA_FIN,
-	MAPPING_DUMMY
+	MAPPING_DUMMY,
+	MAPPING_BAD_CSUM
 };
 
 static void dbg_bad_map(struct mptcp_subflow_context *subflow, u32 ssn)
@@ -958,9 +959,7 @@ static enum mapping_status validate_data_csum(struct sock *ssk, struct sk_buff *
 				 subflow->map_data_csum);
 	if (unlikely(csum)) {
 		MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_DATACSUMERR);
-		if (subflow->mp_join || subflow->valid_csum_seen)
-			subflow->send_mp_fail = 1;
-		return subflow->mp_join ? MAPPING_INVALID : MAPPING_DUMMY;
+		return MAPPING_BAD_CSUM;
 	}
 
 	subflow->valid_csum_seen = 1;
@@ -1178,10 +1177,8 @@ static bool subflow_check_data_avail(struct sock *ssk)
 
 		status = get_mapping_status(ssk, msk);
 		trace_subflow_check_data_avail(status, skb_peek(&ssk->sk_receive_queue));
-		if (unlikely(status == MAPPING_INVALID))
-			goto fallback;
-
-		if (unlikely(status == MAPPING_DUMMY))
+		if (unlikely(status == MAPPING_INVALID || status == MAPPING_DUMMY ||
+			     status == MAPPING_BAD_CSUM))
 			goto fallback;
 
 		if (status != MAPPING_OK)
@@ -1223,7 +1220,10 @@ static bool subflow_check_data_avail(struct sock *ssk)
 fallback:
 	if (!__mptcp_check_fallback(msk)) {
 		/* RFC 8684 section 3.7. */
-		if (subflow->send_mp_fail) {
+		if (status == MAPPING_BAD_CSUM &&
+		    ((subflow->mp_join || subflow->valid_csum_seen))) {
+			subflow->send_mp_fail = 1;
+
 			if (!READ_ONCE(msk->allow_infinite_fallback)) {
 				ssk->sk_err = EBADMSG;
 				tcp_set_state(ssk, TCP_CLOSE);
Signature · 1 line
2.35.3
author maintainer
#3 lore raw

PATCH v2 3/6 Squash-to: "mptcp: invoke MP_FAIL response when needed"

In reply to Paolo Abeni

This tries to address a few issues outstanding in the mentioned patch:
- we explicitly need to reset the timeout timer for mp_fail's sake
- we need to explicitly generate a tcp ack for mp_fail, otherwise there are no guarantees for suck option being sent out
- the timeout timer needs handling need some caring, as it's still shared between mp_fail and msk socket timeout.
- we can re-use msk->first for msk->fail_ssk, as only the first/mpc subflow can fail without reset. That additionally avoid the need to clear fail_ssk on the relevant subflow close.
- fail_tout would need some additional annotation. Just to be on the safe side move its manipulaiton under the ssk socket lock.

Last 2 paragraph of the squash to commit should be replaced with:

"""
It leverages the fact that only the MPC/first subflow can gracefully fail to avoid unneeded subflows traversal: the failing subflow can be only msk->first.

A new 'fail_tout' field is added to the subflow context to record the MP_FAIL response timeout and use such field to reliably share the timeout timer between the MP_FAIL event and the MPTCP socket close timeout.

Finally, a new ack is generated to send out MP_FAIL notification as soon as we hit the relevant condition, instead of waiting a possibly unbound time for the next data packet.
"""

---
 net/mptcp/pm.c       |  4 +++-
 net/mptcp/protocol.c | 54 ++++++++++++++++++++++++++++++++++++--------
 net/mptcp/protocol.h |  4 ++--
 net/mptcp/subflow.c  | 24 ++++++++++++++++++--
 4 files changed, 72 insertions(+), 14 deletions(-)
net/mptcp/pm.c +3 -1
index 3c7f07bb124e..45e2a48397b9 100644
@@ -305,13 +305,15 @@ void mptcp_pm_mp_fail_received(struct sock *sk, u64 fail_seq)
 	if (!READ_ONCE(msk->allow_infinite_fallback))
 		return;
 
-	if (!msk->fail_ssk) {
+	if (!subflow->fail_tout) {
 		pr_debug("send MP_FAIL response and infinite map");
 
 		subflow->send_mp_fail = 1;
 		subflow->send_infinite_map = 1;
+		tcp_send_ack(sk);
 	} else {
 		pr_debug("MP_FAIL response received");
+		WRITE_ONCE(subflow->fail_tout, 0);
 	}
 }
 
net/mptcp/protocol.c +45 -9

121 lines not on this page · raw message

net/mptcp/protocol.h +2 -2
index bef7dea9f358..077a717799a0 100644
@@ -306,8 +306,6 @@ struct mptcp_sock {
 
 	u32 setsockopt_seq;
 	char		ca_name[TCP_CA_NAME_MAX];
-	struct sock	*fail_ssk;
-	unsigned long	fail_tout;
 };
 
 #define mptcp_data_lock(sk) spin_lock_bh(&(sk)->sk_lock.slock)
@@ -484,6 +482,7 @@ struct mptcp_subflow_context {
 	u8	stale_count;
 
 	long	delegated_status;
+	unsigned long	fail_tout;
 
 	);
 
@@ -677,6 +676,7 @@ void mptcp_get_options(const struct sk_buff *skb,
 
 void mptcp_finish_connect(struct sock *sk);
 void __mptcp_set_connected(struct sock *sk);
+void mptcp_reset_timeout(struct mptcp_sock *msk, unsigned long fail_tout);
 static inline bool mptcp_is_fully_established(struct sock *sk)
 {
 	return inet_sk_state_load(sk) == TCP_ESTABLISHED &&
net/mptcp/subflow.c +22 -2
index 98b12a9c4eb5..040901c1f40c 100644
@@ -1158,6 +1158,27 @@ static bool subflow_can_fallback(struct mptcp_subflow_context *subflow)
 		return !subflow->fully_established;
 }
 
+static void mptcp_subflow_fail(struct mptcp_sock *msk, struct sock *ssk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
+	unsigned long fail_tout;
+
+	/* grecefull failure can happen only on the MPC subflow */
+	if (WARN_ON_ONCE(ssk != READ_ONCE(msk->first)))
+		return;
+
+	/* we don't need extreme accuracy here, use a zero fail_tout as special
+	 * value meaning no fail timeout at all;
+	 */
+	fail_tout = jiffies + TCP_RTO_MAX;
+	if (!fail_tout)
+		fail_tout = 1;
+	WRITE_ONCE(subflow->fail_tout, fail_tout);
+	tcp_send_ack(ssk);
+
+	mptcp_reset_timeout(msk, subflow->fail_tout);
+}
+
 static bool subflow_check_data_avail(struct sock *ssk)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
@@ -1233,8 +1254,7 @@ static bool subflow_check_data_avail(struct sock *ssk)
 				while ((skb = skb_peek(&ssk->sk_receive_queue)))
 					sk_eat_skb(ssk, skb);
 			} else {
-				msk->fail_ssk = ssk;
-				msk->fail_tout = jiffies + TCP_RTO_MAX;
+				mptcp_subflow_fail(msk, ssk);
 			}
 			WRITE_ONCE(subflow->data_avail, MPTCP_SUBFLOW_NODATA);
 			return true;
Signature · 1 line
2.35.3
author maintainer
#7 lore raw

PATCH v2 4/6 mptcp: fix shutdown vs fallback race

In reply to Paolo Abeni

If the MPTCP socket shutdown happens before a fallback to TCP, and all the pending data have been already spooled, we never close the TCP connection.

Address the issue explicitly checking for critical condition at fallback time.

---
 net/mptcp/options.c  |  2 +-
 net/mptcp/protocol.c |  2 +-
 net/mptcp/protocol.h | 19 ++++++++++++++++---
 net/mptcp/subflow.c  |  2 +-
 4 files changed, 19 insertions(+), 6 deletions(-)
net/mptcp/options.c +1 -1
index 0bfa6662447c..57eab237c837 100644
@@ -966,7 +966,7 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *ssk,
 			goto reset;
 		subflow->mp_capable = 0;
 		pr_fallback(msk);
-		__mptcp_do_fallback(msk);
+		mptcp_do_fallback(ssk);
 		return false;
 	}
 
net/mptcp/protocol.c +1 -1
index 50026b8da625..4f4d6c66e190 100644
@@ -1245,7 +1245,7 @@ static void mptcp_update_infinite_map(struct mptcp_sock *msk,
 	MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_INFINITEMAPTX);
 	mptcp_subflow_ctx(ssk)->send_infinite_map = 0;
 	pr_fallback(msk);
-	__mptcp_do_fallback(msk);
+	mptcp_do_fallback(ssk);
 }
 
 static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
net/mptcp/protocol.h +16 -3
index 077a717799a0..ad9b02b1b3e6 100644
@@ -941,12 +941,25 @@ static inline void __mptcp_do_fallback(struct mptcp_sock *msk)
 	set_bit(MPTCP_FALLBACK_DONE, &msk->flags);
 }
 
-static inline void mptcp_do_fallback(struct sock *sk)
+static inline void mptcp_do_fallback(struct sock *ssk)
 {
-	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
-	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
+	struct sock *sk = subflow->conn;
+	struct mptcp_sock *msk;
 
+	msk = mptcp_sk(sk);
 	__mptcp_do_fallback(msk);
+	if (READ_ONCE(msk->snd_data_fin_enable) && !(ssk->sk_shutdown & SEND_SHUTDOWN)) {
+		gfp_t saved_allocation = ssk->sk_allocation;
+
+		/* we are in a atomic (BH) scope, override ssk default for data
+		 * fin allocation
+		 */
+		ssk->sk_allocation = GFP_ATOMIC;
+		ssk->sk_shutdown |= SEND_SHUTDOWN;
+		tcp_shutdown(ssk, SEND_SHUTDOWN);
+		ssk->sk_allocation = saved_allocation;
+	}
 }
 
 #define pr_fallback(a) pr_debug("%s:fallback to TCP (msk=%p)", __func__, a)
net/mptcp/subflow.c +1 -1
index 040901c1f40c..75fdc6474d0e 100644
@@ -1273,7 +1273,7 @@ static bool subflow_check_data_avail(struct sock *ssk)
 			return false;
 		}
 
-		__mptcp_do_fallback(msk);
+		mptcp_do_fallback(ssk);
 	}
 
 	skb = skb_peek(&ssk->sk_receive_queue);
Signature · 1 line
2.35.3
author maintainer
#8 lore raw

PATCH v2 5/6 mptcp: consistent map handling on failure

In reply to Paolo Abeni

When the MPTCP receive path reach a non fatal fall-back condition, e.g. when the MPC sockets must fall-back to TCP, the existing code is a little
self-inconsistent: it reports that new data is available - return true - but sets the MPC flag to the opposite value.

As the consequence read operations in some exceptional scenario may block unexpectedly.

Address the issue setting the correct MPC read status. Additionally avoid some code duplication in the fatal fall-back scenario.

Fixes 9c81be0dbc89 ("mptcp: add MP_FAIL response support")
---
 net/mptcp/subflow.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)
net/mptcp/subflow.c +9 -10
index 75fdc6474d0e..bec36e2bcc32 100644
@@ -1246,17 +1246,12 @@ static bool subflow_check_data_avail(struct sock *ssk)
 			subflow->send_mp_fail = 1;
 
 			if (!READ_ONCE(msk->allow_infinite_fallback)) {
-				ssk->sk_err = EBADMSG;
-				tcp_set_state(ssk, TCP_CLOSE);
 				subflow->reset_transient = 0;
 				subflow->reset_reason = MPTCP_RST_EMIDDLEBOX;
-				tcp_send_active_reset(ssk, GFP_ATOMIC);
-				while ((skb = skb_peek(&ssk->sk_receive_queue)))
-					sk_eat_skb(ssk, skb);
-			} else {
-				mptcp_subflow_fail(msk, ssk);
+				goto reset;
 			}
-			WRITE_ONCE(subflow->data_avail, MPTCP_SUBFLOW_NODATA);
+			mptcp_subflow_fail(msk, ssk);
+			WRITE_ONCE(subflow->data_avail, MPTCP_SUBFLOW_DATA_AVAIL);
 			return true;
 		}
 
@@ -1264,10 +1259,14 @@ static bool subflow_check_data_avail(struct sock *ssk)
 			/* fatal protocol error, close the socket.
 			 * subflow_error_report() will introduce the appropriate barriers
 			 */
-			ssk->sk_err = EBADMSG;
-			tcp_set_state(ssk, TCP_CLOSE);
 			subflow->reset_transient = 0;
 			subflow->reset_reason = MPTCP_RST_EMPTCP;
+
+reset:
+			ssk->sk_err = EBADMSG;
+			tcp_set_state(ssk, TCP_CLOSE);
+			while ((skb = skb_peek(&ssk->sk_receive_queue)))
+				sk_eat_skb(ssk, skb);
 			tcp_send_active_reset(ssk, GFP_ATOMIC);
 			WRITE_ONCE(subflow->data_avail, MPTCP_SUBFLOW_NODATA);
 			return false;
Signature · 1 line
2.35.3
author maintainer
#9 lore raw

PATCH v2 6/6 mptcp: fix race on unaccepted mptcp sockets

In reply to Paolo Abeni

When the listener socket owning the relevant request is closed, it frees the unaccepted subflows and that causes later deletion of the paired MPTCP sockets.

The mptcp socket's worker can run in the time interval between such delete operations. When that happens, any access to msk->first will cause an UaF access, as the subflow cleanup did not cleared such field in the mptcp socket.

Address the issue explictly traversing the listener socket accept queue at close time and performing the needed cleanup on the pending msk.

Note that the locking is a bit tricky, as we need to acquire the msk socket lock, while still owning the subflow socket one.

Fixes 86e39e04482b ("mptcp: keep track of local endpoint still available for each msk")
---
 net/mptcp/protocol.c |  5 +++++
 net/mptcp/protocol.h |  2 ++
 net/mptcp/subflow.c  | 50 ++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 57 insertions(+)
net/mptcp/protocol.c +5 -0
index 4f4d6c66e190..35e3060233c9 100644
@@ -2318,6 +2318,11 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk,
 		kfree_rcu(subflow, rcu);
 	} else {
 		/* otherwise tcp will dispose of the ssk and subflow ctx */
+		if (ssk->sk_state == TCP_LISTEN) {
+			tcp_set_state(ssk, TCP_CLOSE);
+			mptcp_subflow_queue_clean(ssk);
+			inet_csk_listen_stop(ssk);
+		}
 		__tcp_close(ssk, 0);
 
 		/* close acquired an extra ref */
net/mptcp/protocol.h +2 -0
index ad9b02b1b3e6..95c9ace1437b 100644
@@ -306,6 +306,7 @@ struct mptcp_sock {
 
 	u32 setsockopt_seq;
 	char		ca_name[TCP_CA_NAME_MAX];
+	struct mptcp_sock	*dl_next;
 };
 
 #define mptcp_data_lock(sk) spin_lock_bh(&(sk)->sk_lock.slock)
@@ -610,6 +611,7 @@ void mptcp_close_ssk(struct sock *sk, struct sock *ssk,
 		     struct mptcp_subflow_context *subflow);
 void mptcp_subflow_send_ack(struct sock *ssk);
 void mptcp_subflow_reset(struct sock *ssk);
+void mptcp_subflow_queue_clean(struct sock *ssk);
 void mptcp_sock_graft(struct sock *sk, struct socket *parent);
 struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk);
 
net/mptcp/subflow.c +50 -0

56 lines not on this page · raw message

Signature · 1 line
2.35.3
#10 lore raw

Re: mptcp: fix race on unaccepted mptcp sockets: Tests Results

Hi Paolo,

Thank you for your modifications, that's great!

Our CI did some validations and here is its report:

- KVM Validation: normal:
- Success! ✅:
- Task: https://cirrus-ci.com/task/5240636109488128
- Summary: https://api.cirrus-ci.com/v1/artifact/task/5240636109488128/summary/summary.txt

- KVM Validation: debug:
- Unstable: 1 failed test(s): selftest_diag 🔴:
- Task: https://cirrus-ci.com/task/6366536016330752
- Summary: https://api.cirrus-ci.com/v1/artifact/task/6366536016330752/summary/summary.txt

Initiator: Patchew Applier
Commits: https://github.com/multipath-tcp/mptcp_net-next/commits/bac04f5bed54

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 (Tessares)

#4 lore raw

Re: [PATCH mptcp-net v2 3/6] Squash-to: "mptcp: invoke MP_FAIL response when needed"

In reply to Paolo Abeni
Quoted 91 lines · Paolo Abeni

This tries to address a few issues outstanding in the mentioned patch:
- we explicitly need to reset the timeout timer for mp_fail's sake
- we need to explicitly generate a tcp ack for mp_fail, otherwise there are no guarantees for suck option being sent out
- the timeout timer needs handling need some caring, as it's still shared between mp_fail and msk socket timeout.
- we can re-use msk->first for msk->fail_ssk, as only the first/mpc subflow can fail without reset. That additionally avoid the need to clear fail_ssk on the relevant subflow close.
- fail_tout would need some additional annotation. Just to be on the safe side move its manipulaiton under the ssk socket lock.

Last 2 paragraph of the squash to commit should be replaced with:

"""
It leverages the fact that only the MPC/first subflow can gracefully fail to avoid unneeded subflows traversal: the failing subflow can be only msk->first.

A new 'fail_tout' field is added to the subflow context to record the MP_FAIL response timeout and use such field to reliably share the timeout timer between the MP_FAIL event and the MPTCP socket close timeout.

Finally, a new ack is generated to send out MP_FAIL notification as soon as we hit the relevant condition, instead of waiting a possibly unbound time for the next data packet.
"""

---

net/mptcp/pm.c       |  4 +++-
net/mptcp/protocol.c | 54 ++++++++++++++++++++++++++++++++++++--------
net/mptcp/protocol.h |  4 ++--
net/mptcp/subflow.c  | 24 ++++++++++++++++++--
4 files changed, 72 insertions(+), 14 deletions(-)
net/mptcp/pm.c +3 -1
index 3c7f07bb124e..45e2a48397b9 100644
@@ -305,13 +305,15 @@ void mptcp_pm_mp_fail_received(struct sock *sk, u64 fail_seq)
 	if (!READ_ONCE(msk->allow_infinite_fallback))
 		return;
 
-	if (!msk->fail_ssk) {
+	if (!subflow->fail_tout) {
 		pr_debug("send MP_FAIL response and infinite map");
 
 		subflow->send_mp_fail = 1;
 		subflow->send_infinite_map = 1;
+		tcp_send_ack(sk);
 	} else {
 		pr_debug("MP_FAIL response received");
+		WRITE_ONCE(subflow->fail_tout, 0);
 	}
 }
 
net/mptcp/protocol.c +17 -1
index a0f9f3831509..50026b8da625 100644
@@ -500,7 +500,7 @@ static void mptcp_set_timeout(struct sock *sk)
 	__mptcp_set_timeout(sk, tout);
 }
 
-static bool tcp_can_send_ack(const struct sock *ssk)
+static inline bool tcp_can_send_ack(const struct sock *ssk)
 {
 	return !((1 << inet_sk_state_load(ssk)) &
 	       (TCPF_SYN_SENT | TCPF_SYN_RECV | TCPF_TIME_WAIT | TCPF_CLOSE | TCPF_LISTEN));
@@ -2490,24 +2490,56 @@ static void __mptcp_retrans(struct sock *sk)
 		mptcp_reset_timer(sk);
 }
 
+/* schedule the timeout timer for the nearest relevant event: either
+ * close timeout or mp_fail timeout. Both of them could be not
+ * scheduled yet
+ */
+void mptcp_reset_timeout(struct mptcp_sock *msk, unsigned long fail_tout)
+{
+	struct sock *sk = (struct sock *)msk;
+	unsigned long timeout, close_timeout;
+
+	if (!fail_tout && !sock_flag(sk, SOCK_DEAD))
+		return;
+
+	close_timeout = inet_csk(sk)->icsk_mtup.probe_timestamp - tcp_jiffies32 + jiffies + TCP_TIMEWAIT_LEN;
+
+	/* the following is basically time_min(close_timeout, fail_tout) */
+	if (!fail_tout)

net/mptcp/protocol.c static void __mptcp_retrans(struct sock *sk) · line 2518

Quoted 10 lines
net/mptcp/protocol.c +10 -0
+		timeout = close_timeout;
+	else if (!sock_flag(sk, SOCK_DEAD))
+		timeout = fail_tout;
+	else if (time_after(close_timeout, fail_tout))
+		timeout = fail_tout;
+	else
+		timeout = close_timeout;
+
+	sk_reset_timer(sk, &sk->sk_timer, timeout);
+}

Hi Paolo -

The above function seems more complex than needed. If mptcp_close() has been called, the fail timeout should be considered canceled and the "close" has exclusive use of sk_timer. subflow->fail_tout can be set to 0 and sk_timer can be set only based on TCP_TIMEWAIT_LEN.

TCP_RTO_MAX for the MP_FAIL timeout was kind of arbitrary - the spec doesn't say what exactly to do. We didn't want the connection to be in the "unacknowledged" MP_FAIL state forever, but also didn't want to give up too fast.

Given that, do you think the complexity is justified to possibly reset the subflow earlier in this "unacknowledged MP_FAIL followed by a close" scenario?

Quoted 131 lines
net/mptcp/protocol.c +10 -3
+
 static void mptcp_mp_fail_no_response(struct mptcp_sock *msk)
 {
-	struct sock *ssk = msk->fail_ssk;
+	struct sock *ssk = msk->first;
 	bool slow;
 
+	if (!ssk)
+		return;
+
 	pr_debug("MP_FAIL doesn't respond, reset the subflow");
 
 	slow = lock_sock_fast(ssk);
 	mptcp_subflow_reset(ssk);
+	WRITE_ONCE(mptcp_subflow_ctx(ssk)->fail_tout, 0);
 	unlock_sock_fast(ssk, slow);
 
-	msk->fail_ssk = NULL;
+	mptcp_reset_timeout(msk, 0);
 }
 
 static void mptcp_worker(struct work_struct *work)
 {
 	struct mptcp_sock *msk = container_of(work, struct mptcp_sock, work);
 	struct sock *sk = &msk->sk.icsk_inet.sk;
+	unsigned long fail_tout;
 	int state;
 
 	lock_sock(sk);
@@ -2544,7 +2576,8 @@ static void mptcp_worker(struct work_struct *work)
 	if (test_and_clear_bit(MPTCP_WORK_RTX, &msk->flags))
 		__mptcp_retrans(sk);
 
-	if (msk->fail_ssk && time_after(jiffies, msk->fail_tout))
+	fail_tout = msk->first ? READ_ONCE(mptcp_subflow_ctx(msk->first)->fail_tout) : 0;
+	if (fail_tout && time_after(jiffies, fail_tout))
 		mptcp_mp_fail_no_response(msk);
 

unlock:

net/mptcp/protocol.c +2 -2
@@ -2572,8 +2605,6 @@ static int __mptcp_init_sock(struct sock *sk)
 	WRITE_ONCE(msk->csum_enabled, mptcp_is_checksum_enabled(sock_net(sk)));
 	WRITE_ONCE(msk->allow_infinite_fallback, true);
 	msk->recovery = false;
-	msk->fail_ssk = NULL;
-	msk->fail_tout = 0;
 
 	mptcp_pm_data_init(msk);
 
@@ -2804,7 +2835,9 @@ static void __mptcp_destroy_sock(struct sock *sk)
 static void mptcp_close(struct sock *sk, long timeout)
 {
 	struct mptcp_subflow_context *subflow;
+	struct mptcp_sock *msk = mptcp_sk(sk);
 	bool do_cancel_work = false;
+	unsigned long fail_tout = 0;
 
 	lock_sock(sk);
 	sk->sk_shutdown = SHUTDOWN_MASK;
@@ -2822,10 +2855,13 @@ static void mptcp_close(struct sock *sk, long timeout)

cleanup:

	/* orphan all the subflows */
	inet_csk(sk)->icsk_mtup.probe_timestamp = tcp_jiffies32;
net/mptcp/protocol.c +1 -1
-	mptcp_for_each_subflow(mptcp_sk(sk), subflow) {
+	mptcp_for_each_subflow(msk, subflow) {
		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
		bool slow = lock_sock_fast_nested(ssk);
net/mptcp/protocol.c +3 -0
+		if (ssk == msk->first)
+			fail_tout = subflow->fail_tout;
+
		sock_orphan(ssk);
		unlock_sock_fast(ssk, slow);
	}
net/mptcp/protocol.c +2 -2
@@ -2834,13 +2870,13 @@ static void mptcp_close(struct sock *sk, long timeout)
 	sock_hold(sk);
 	pr_debug("msk=%p state=%d", sk, sk->sk_state);
 	if (mptcp_sk(sk)->token)
-		mptcp_event(MPTCP_EVENT_CLOSED, mptcp_sk(sk), NULL, GFP_KERNEL);
+		mptcp_event(MPTCP_EVENT_CLOSED, msk, NULL, GFP_KERNEL);
 
 	if (sk->sk_state == TCP_CLOSE) {
 		__mptcp_destroy_sock(sk);
 		do_cancel_work = true;
 	} else {
-		sk_reset_timer(sk, &sk->sk_timer, jiffies + TCP_TIMEWAIT_LEN);
+		mptcp_reset_timeout(msk, fail_tout);
 	}
 	release_sock(sk);
 	if (do_cancel_work)
net/mptcp/protocol.h +2 -2
index bef7dea9f358..077a717799a0 100644
@@ -306,8 +306,6 @@ struct mptcp_sock {
 
 	u32 setsockopt_seq;
 	char		ca_name[TCP_CA_NAME_MAX];
-	struct sock	*fail_ssk;
-	unsigned long	fail_tout;
 };
 
 #define mptcp_data_lock(sk) spin_lock_bh(&(sk)->sk_lock.slock)
@@ -484,6 +482,7 @@ struct mptcp_subflow_context {
 	u8	stale_count;
 
 	long	delegated_status;
+	unsigned long	fail_tout;
 
 	);
 
@@ -677,6 +676,7 @@ void mptcp_get_options(const struct sk_buff *skb,
 
 void mptcp_finish_connect(struct sock *sk);
 void __mptcp_set_connected(struct sock *sk);
+void mptcp_reset_timeout(struct mptcp_sock *msk, unsigned long fail_tout);
 static inline bool mptcp_is_fully_established(struct sock *sk)
 {
 	return inet_sk_state_load(sk) == TCP_ESTABLISHED &&
net/mptcp/subflow.c +5 -0
index 98b12a9c4eb5..040901c1f40c 100644
@@ -1158,6 +1158,27 @@ static bool subflow_can_fallback(struct mptcp_subflow_context *subflow)
 		return !subflow->fully_established;
 }
 
+static void mptcp_subflow_fail(struct mptcp_sock *msk, struct sock *ssk)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
+	unsigned long fail_tout;
+

net/mptcp/subflow.c static bool subflow_can_fallback(struct mptcp_subflow_context *subflow) · line 1175

Quoted 10 lines
net/mptcp/subflow.c +10 -0
+	/* grecefull failure can happen only on the MPC subflow */
+	if (WARN_ON_ONCE(ssk != READ_ONCE(msk->first)))
+		return;
+
+	/* we don't need extreme accuracy here, use a zero fail_tout as special
+	 * value meaning no fail timeout at all;
+	 */
+	fail_tout = jiffies + TCP_RTO_MAX;
+	if (!fail_tout)
+		fail_tout = 1;

Ah, I was wondering how jiffies wrap-around and the '0' special value were handled! Good idea.

Quoted 24 lines
net/mptcp/subflow.c +7 -2
+	WRITE_ONCE(subflow->fail_tout, fail_tout);
+	tcp_send_ack(ssk);
+
+	mptcp_reset_timeout(msk, subflow->fail_tout);
+}
+
 static bool subflow_check_data_avail(struct sock *ssk)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
@@ -1233,8 +1254,7 @@ static bool subflow_check_data_avail(struct sock *ssk)
 				while ((skb = skb_peek(&ssk->sk_receive_queue)))
 					sk_eat_skb(ssk, skb);
 			} else {
-				msk->fail_ssk = ssk;
-				msk->fail_tout = jiffies + TCP_RTO_MAX;
+				mptcp_subflow_fail(msk, ssk);
 			}
 			WRITE_ONCE(subflow->data_avail, MPTCP_SUBFLOW_NODATA);
 			return true;
Signature · 1 line
2.35.3

--
Mat Martineau
Intel

author maintainer
#5 lore raw

Re: [PATCH mptcp-net v2 3/6] Squash-to: "mptcp: invoke MP_FAIL response when needed"

Quoted 106 lines · Mat Martineau
Quoted 91 lines · Paolo Abeni

This tries to address a few issues outstanding in the mentioned patch:
- we explicitly need to reset the timeout timer for mp_fail's sake
- we need to explicitly generate a tcp ack for mp_fail, otherwise there are no guarantees for suck option being sent out
- the timeout timer needs handling need some caring, as it's still shared between mp_fail and msk socket timeout.
- we can re-use msk->first for msk->fail_ssk, as only the first/mpc subflow can fail without reset. That additionally avoid the need to clear fail_ssk on the relevant subflow close.
- fail_tout would need some additional annotation. Just to be on the safe side move its manipulaiton under the ssk socket lock.

Last 2 paragraph of the squash to commit should be replaced with:

"""
It leverages the fact that only the MPC/first subflow can gracefully fail to avoid unneeded subflows traversal: the failing subflow can be only msk->first.

A new 'fail_tout' field is added to the subflow context to record the MP_FAIL response timeout and use such field to reliably share the timeout timer between the MP_FAIL event and the MPTCP socket close timeout.

Finally, a new ack is generated to send out MP_FAIL notification as soon as we hit the relevant condition, instead of waiting a possibly unbound time for the next data packet.
"""

---

net/mptcp/pm.c       |  4 +++-
net/mptcp/protocol.c | 54 ++++++++++++++++++++++++++++++++++++--------
net/mptcp/protocol.h |  4 ++--
net/mptcp/subflow.c  | 24 ++++++++++++++++++--
4 files changed, 72 insertions(+), 14 deletions(-)
net/mptcp/pm.c +3 -1
index 3c7f07bb124e..45e2a48397b9 100644
@@ -305,13 +305,15 @@ void mptcp_pm_mp_fail_received(struct sock *sk, u64 fail_seq)
 	if (!READ_ONCE(msk->allow_infinite_fallback))
 		return;
 
-	if (!msk->fail_ssk) {
+	if (!subflow->fail_tout) {
 		pr_debug("send MP_FAIL response and infinite map");
 
 		subflow->send_mp_fail = 1;
 		subflow->send_infinite_map = 1;
+		tcp_send_ack(sk);
 	} else {
 		pr_debug("MP_FAIL response received");
+		WRITE_ONCE(subflow->fail_tout, 0);
 	}
 }
 
net/mptcp/protocol.c +17 -1
index a0f9f3831509..50026b8da625 100644
@@ -500,7 +500,7 @@ static void mptcp_set_timeout(struct sock *sk)
 	__mptcp_set_timeout(sk, tout);
 }
 
-static bool tcp_can_send_ack(const struct sock *ssk)
+static inline bool tcp_can_send_ack(const struct sock *ssk)
 {
 	return !((1 << inet_sk_state_load(ssk)) &
 	       (TCPF_SYN_SENT | TCPF_SYN_RECV | TCPF_TIME_WAIT | TCPF_CLOSE | TCPF_LISTEN));
@@ -2490,24 +2490,56 @@ static void __mptcp_retrans(struct sock *sk)
 		mptcp_reset_timer(sk);
 }
 
+/* schedule the timeout timer for the nearest relevant event: either
+ * close timeout or mp_fail timeout. Both of them could be not
+ * scheduled yet
+ */
+void mptcp_reset_timeout(struct mptcp_sock *msk, unsigned long fail_tout)
+{
+	struct sock *sk = (struct sock *)msk;
+	unsigned long timeout, close_timeout;
+
+	if (!fail_tout && !sock_flag(sk, SOCK_DEAD))
+		return;
+
+	close_timeout = inet_csk(sk)->icsk_mtup.probe_timestamp - tcp_jiffies32 + jiffies + TCP_TIMEWAIT_LEN;
+
+	/* the following is basically time_min(close_timeout, fail_tout) */
+	if (!fail_tout)

net/mptcp/protocol.c static void __mptcp_retrans(struct sock *sk) · line 2518

Quoted 10 lines
net/mptcp/protocol.c +10 -0
+		timeout = close_timeout;
+	else if (!sock_flag(sk, SOCK_DEAD))
+		timeout = fail_tout;
+	else if (time_after(close_timeout, fail_tout))
+		timeout = fail_tout;
+	else
+		timeout = close_timeout;
+
+	sk_reset_timer(sk, &sk->sk_timer, timeout);
+}

Hi Paolo -

Quoted 4 lines

The above function seems more complex than needed. If mptcp_close() has been called, the fail timeout should be considered canceled and the "close" has exclusive use of sk_timer. subflow->fail_tout can be set to 0 and sk_timer can be set only based on TCP_TIMEWAIT_LEN.

I agree the above function is non trivial, on the flip side, it helps keeping all the timeout logic in a single place.

Note that msk->first->fail_tout is under the subflow socket lock protection, and this function is not always invoked under such lock, so we will need to either add more read/write once annotation or split the logic in the caller.

Side note: I agree a bit with David Laight wrt *_once preoliferation, possibly with less divisive language:

https://lore.kernel.org/netdev/[email protected]/

;)

*_ONCE are a bit hard to track and difficult to verify formally, so I tried hard to reduce their usage. Currently we have only a single READ operation outside the relevant lock, which is AFAIK the 'safer' possible scenario with such annotations. All the write operations are under the subflow socket lock.

Quoted 4 lines

TCP_RTO_MAX for the MP_FAIL timeout was kind of arbitrary - the spec doesn't say what exactly to do. We didn't want the connection to be in the "unacknowledged" MP_FAIL state forever, but also didn't want to give up too fast.

This function don't make any assumption on the relative timeout length and should fit the case if we will make them configurable someday

Quoted 3 lines

Given that, do you think the complexity is justified to possibly reset the subflow earlier in this "unacknowledged MP_FAIL followed by a close" scenario?

IMHO yes ;) But if you have strong opinion otherwise I can refactor the code...

Cheers,

Paolo

#6 lore raw

Re: [PATCH mptcp-net v2 3/6] Squash-to: "mptcp: invoke MP_FAIL response when needed"

Quoted 120 lines · Paolo Abeni
Quoted 106 lines · Mat Martineau
Quoted 91 lines · Paolo Abeni

This tries to address a few issues outstanding in the mentioned patch:
- we explicitly need to reset the timeout timer for mp_fail's sake
- we need to explicitly generate a tcp ack for mp_fail, otherwise there are no guarantees for suck option being sent out
- the timeout timer needs handling need some caring, as it's still shared between mp_fail and msk socket timeout.
- we can re-use msk->first for msk->fail_ssk, as only the first/mpc subflow can fail without reset. That additionally avoid the need to clear fail_ssk on the relevant subflow close.
- fail_tout would need some additional annotation. Just to be on the safe side move its manipulaiton under the ssk socket lock.

Last 2 paragraph of the squash to commit should be replaced with:

"""
It leverages the fact that only the MPC/first subflow can gracefully fail to avoid unneeded subflows traversal: the failing subflow can be only msk->first.

A new 'fail_tout' field is added to the subflow context to record the MP_FAIL response timeout and use such field to reliably share the timeout timer between the MP_FAIL event and the MPTCP socket close timeout.

Finally, a new ack is generated to send out MP_FAIL notification as soon as we hit the relevant condition, instead of waiting a possibly unbound time for the next data packet.
"""

---

net/mptcp/pm.c       |  4 +++-
net/mptcp/protocol.c | 54 ++++++++++++++++++++++++++++++++++++--------
net/mptcp/protocol.h |  4 ++--
net/mptcp/subflow.c  | 24 ++++++++++++++++++--
4 files changed, 72 insertions(+), 14 deletions(-)
net/mptcp/pm.c +3 -1
index 3c7f07bb124e..45e2a48397b9 100644
@@ -305,13 +305,15 @@ void mptcp_pm_mp_fail_received(struct sock *sk, u64 fail_seq)
 	if (!READ_ONCE(msk->allow_infinite_fallback))
 		return;
 
-	if (!msk->fail_ssk) {
+	if (!subflow->fail_tout) {
 		pr_debug("send MP_FAIL response and infinite map");
 
 		subflow->send_mp_fail = 1;
 		subflow->send_infinite_map = 1;
+		tcp_send_ack(sk);
 	} else {
 		pr_debug("MP_FAIL response received");
+		WRITE_ONCE(subflow->fail_tout, 0);
 	}
 }
 
net/mptcp/protocol.c +17 -1
index a0f9f3831509..50026b8da625 100644
@@ -500,7 +500,7 @@ static void mptcp_set_timeout(struct sock *sk)
 	__mptcp_set_timeout(sk, tout);
 }
 
-static bool tcp_can_send_ack(const struct sock *ssk)
+static inline bool tcp_can_send_ack(const struct sock *ssk)
 {
 	return !((1 << inet_sk_state_load(ssk)) &
 	       (TCPF_SYN_SENT | TCPF_SYN_RECV | TCPF_TIME_WAIT | TCPF_CLOSE | TCPF_LISTEN));
@@ -2490,24 +2490,56 @@ static void __mptcp_retrans(struct sock *sk)
 		mptcp_reset_timer(sk);
 }
 
+/* schedule the timeout timer for the nearest relevant event: either
+ * close timeout or mp_fail timeout. Both of them could be not
+ * scheduled yet
+ */
+void mptcp_reset_timeout(struct mptcp_sock *msk, unsigned long fail_tout)
+{
+	struct sock *sk = (struct sock *)msk;
+	unsigned long timeout, close_timeout;
+
+	if (!fail_tout && !sock_flag(sk, SOCK_DEAD))
+		return;
+
+	close_timeout = inet_csk(sk)->icsk_mtup.probe_timestamp - tcp_jiffies32 + jiffies + TCP_TIMEWAIT_LEN;
+
+	/* the following is basically time_min(close_timeout, fail_tout) */
+	if (!fail_tout)

net/mptcp/protocol.c static void __mptcp_retrans(struct sock *sk) · line 2518

Quoted 10 lines
net/mptcp/protocol.c +10 -0
+		timeout = close_timeout;
+	else if (!sock_flag(sk, SOCK_DEAD))
+		timeout = fail_tout;
+	else if (time_after(close_timeout, fail_tout))
+		timeout = fail_tout;
+	else
+		timeout = close_timeout;
+
+	sk_reset_timer(sk, &sk->sk_timer, timeout);
+}

Hi Paolo -

Quoted 4 lines

The above function seems more complex than needed. If mptcp_close() has been called, the fail timeout should be considered canceled and the "close" has exclusive use of sk_timer. subflow->fail_tout can be set to 0 and sk_timer can be set only based on TCP_TIMEWAIT_LEN.

I agree the above function is non trivial, on the flip side, it helps keeping all the timeout logic in a single place.

Note that msk->first->fail_tout is under the subflow socket lock protection, and this function is not always invoked under such lock, so we will need to either add more read/write once annotation or split the logic in the caller.

Quoted 5 lines

Side note: I agree a bit with David Laight wrt *_once preoliferation, possibly with less divisive language:
https://lore.kernel.org/netdev/[email protected]/

;)

Heh - yeah, I saw that exchange too :)

Quoted 15 lines

*_ONCE are a bit hard to track and difficult to verify formally, so I tried hard to reduce their usage. Currently we have only a single READ operation outside the relevant lock, which is AFAIK the 'safer' possible scenario with such annotations. All the write operations are under the subflow socket lock.

Quoted 4 lines

TCP_RTO_MAX for the MP_FAIL timeout was kind of arbitrary - the spec doesn't say what exactly to do. We didn't want the connection to be in the "unacknowledged" MP_FAIL state forever, but also didn't want to give up too fast.

This function don't make any assumption on the relative timeout length and should fit the case if we will make them configurable someday

Quoted 7 lines
Quoted 3 lines

Given that, do you think the complexity is justified to possibly reset the subflow earlier in this "unacknowledged MP_FAIL followed by a close" scenario?

IMHO yes ;) But if you have strong opinion otherwise I can refactor the code...

I do think it's better to cancel the MP_FAIL timeout once mptcp_close() is called. It's not just about the code in mptcp_reset_timeout(), it doesn't seem meaningful to continue MP_FAIL handling after close and to have various scenarios to consider/test/etc. for timeout or MP_FAIL echo ordering.

--
Mat Martineau
Intel