Merge: mptcp: phase-2 packports for RHEL-9.7

MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-9/-/merge_requests/6795

JIRA: https://issues.redhat.com/browse/RHEL-84571
Upstream Status: all mainline in net-next.git
Conflicts: see individual patches
Tested: boot-tested only

Signed-off-by: Davide Caratti <dcaratti@redhat.com>

Approved-by: Paolo Abeni <pabeni@redhat.com>
Approved-by: Florian Westphal <fwestpha@redhat.com>
Approved-by: CKI KWF Bot <cki-ci-bot+kwf-gitlab-com@redhat.com>

Merged-by: Augusto Caringi <acaringi@redhat.com>

Author: caringi

net/mptcp/mib.c

@@ -15,6 +15,7 @@ static const struct snmp_mib mptcp_snmp_list[] = {
 	SNMP_MIB_ITEM("MPCapableACKRX", MPTCP_MIB_MPCAPABLEPASSIVEACK),
 	SNMP_MIB_ITEM("MPCapableFallbackACK", MPTCP_MIB_MPCAPABLEPASSIVEFALLBACK),
 	SNMP_MIB_ITEM("MPCapableFallbackSYNACK", MPTCP_MIB_MPCAPABLEACTIVEFALLBACK),
+	SNMP_MIB_ITEM("MPCapableEndpAttempt", MPTCP_MIB_MPCAPABLEENDPATTEMPT),
 	SNMP_MIB_ITEM("MPFallbackTokenInit", MPTCP_MIB_TOKENFALLBACKINIT),
 	SNMP_MIB_ITEM("MPTCPRetrans", MPTCP_MIB_RETRANSSEGS),
 	SNMP_MIB_ITEM("MPJoinNoTokenFound", MPTCP_MIB_JOINNOTOKEN),

net/mptcp/mib.h

@@ -10,6 +10,7 @@ enum linux_mptcp_mib_field {
 	MPTCP_MIB_MPCAPABLEPASSIVEACK,	/* Received third ACK with MP_CAPABLE */
 	MPTCP_MIB_MPCAPABLEPASSIVEFALLBACK,/* Server-side fallback during 3-way handshake */
 	MPTCP_MIB_MPCAPABLEACTIVEFALLBACK, /* Client-side fallback during 3-way handshake */
+	MPTCP_MIB_MPCAPABLEENDPATTEMPT,	/* Prohibited MPC to port-based endp */
 	MPTCP_MIB_TOKENFALLBACKINIT,	/* Could not init/allocate token */
 	MPTCP_MIB_RETRANSSEGS,		/* Segments retransmitted at the MPTCP-level */
 	MPTCP_MIB_JOINNOTOKEN,		/* Received MP_JOIN but the token was not found */

net/mptcp/options.c

@@ -156,11 +156,6 @@ static void mptcp_parse_option(const struct sk_buff *skb,
 		pr_debug("DSS\n");
 		ptr++;
 
-		/* we must clear 'mpc_map' be able to detect MP_CAPABLE
-		 * map vs DSS map in mptcp_incoming_options(), and reconstruct
-		 * map info accordingly
-		 */
-		mp_opt->mpc_map = 0;
 		flags = (*ptr++) & MPTCP_DSS_FLAG_MASK;
 		mp_opt->data_fin = (flags & MPTCP_DSS_DATA_FIN) != 0;
 		mp_opt->dsn64 = (flags & MPTCP_DSS_DSN64) != 0;
@@ -368,8 +363,11 @@ void mptcp_get_options(const struct sk_buff *skb,
 	const unsigned char *ptr;
 	int length;
 
-	/* initialize option status */
-	mp_opt->suboptions = 0;
+	/* Ensure that casting the whole status to u32 is efficient and safe */
+	BUILD_BUG_ON(sizeof_field(struct mptcp_options_received, status) != sizeof(u32));
+	BUILD_BUG_ON(!IS_ALIGNED(offsetof(struct mptcp_options_received, status),
+				 sizeof(u32)));
+	*(u32 *)&mp_opt->status = 0;
 
 	length = (th->doff * 4) - sizeof(struct tcphdr);
 	ptr = (const unsigned char *)(th + 1);
@@ -606,7 +604,6 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
 	}
 	opts->ext_copy.use_ack = 1;
 	opts->suboptions = OPTION_MPTCP_DSS;
-	WRITE_ONCE(msk->old_wspace, __mptcp_space((struct sock *)msk));
 
 	/* Add kind/length/subtype/flag overhead if mapping is not populated */
 	if (dss_size == 0)
@@ -654,6 +651,7 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff *
 	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
 	bool drop_other_suboptions = false;
 	unsigned int opt_size = *size;
+	struct mptcp_addr_info addr;
 	bool echo;
 	int len;
 
@@ -662,7 +660,7 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff *
 	 */
 	if (!mptcp_pm_should_add_signal(msk) ||
 	    (opts->suboptions & (OPTION_MPTCP_MPJ_ACK | OPTION_MPTCP_MPC_ACK)) ||
-	    !mptcp_pm_add_addr_signal(msk, skb, opt_size, remaining, &opts->addr,
+	    !mptcp_pm_add_addr_signal(msk, skb, opt_size, remaining, &addr,
 		    &echo, &drop_other_suboptions))
 		return false;
 
@@ -675,7 +673,7 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff *
 	else if (opts->suboptions & OPTION_MPTCP_DSS)
 		return false;
 
-	len = mptcp_add_addr_len(opts->addr.family, echo, !!opts->addr.port);
+	len = mptcp_add_addr_len(addr.family, echo, !!addr.port);
 	if (remaining < len)
 		return false;
 
@@ -692,6 +690,7 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff *
 		opts->ahmac = 0;
 		*size -= opt_size;
 	}
+	opts->addr = addr;
 	opts->suboptions |= OPTION_MPTCP_ADD_ADDR;
 	if (!echo) {
 		MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ADDADDRTX);
@@ -1286,7 +1285,7 @@ static void mptcp_set_rwin(struct tcp_sock *tp, struct tcphdr *th)
 			}
 			MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_RCVWNDCONFLICT);
 		}
-		return;
+		goto update_wspace;
 	}
 
 	if (rcv_wnd_new != rcv_wnd_old) {
@@ -1311,6 +1310,9 @@ static void mptcp_set_rwin(struct tcp_sock *tp, struct tcphdr *th)
 		th->window = htons(new_win);
 		MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_RCVWNDSHARED);
 	}
+
+update_wspace:
+	WRITE_ONCE(msk->old_wspace, tp->rcv_wnd);
 }
 
 __sum16 __mptcp_make_csum(u64 data_seq, u32 subflow_seq, u16 data_len, __wsum sum)

net/mptcp/pm_netlink.c

@@ -972,7 +972,7 @@ 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,
-					     bool needs_id)
+					     bool needs_id, bool replace)
 {
 	struct mptcp_pm_addr_entry *cur, *del_entry = NULL;
 	unsigned int addr_max;
@@ -1012,6 +1012,17 @@ static int mptcp_pm_nl_append_new_local_addr(struct pm_nl_pernet *pernet,
 			if (entry->addr.id)
 				goto out;
 
+			/* allow callers that only need to look up the local
+			 * addr's id to skip replacement. This allows them to
+			 * avoid calling synchronize_rcu in the packet recv
+			 * path.
+			 */
+			if (!replace) {
+				kfree(entry);
+				ret = cur->addr.id;
+				goto out;
+			}
+
 			pernet->addrs--;
 			entry->addr.id = cur->addr.id;
 			list_del_rcu(&cur->list);
@@ -1126,6 +1137,7 @@ static int mptcp_pm_nl_create_listen_socket(struct sock *sk,
 	 */
 	inet_sk_state_store(newsk, TCP_LISTEN);
 	lock_sock(ssk);
+	WRITE_ONCE(mptcp_subflow_ctx(ssk)->pm_listener, true);
 	err = __inet_listen_sk(ssk, backlog);
 	if (!err)
 		mptcp_event_pm_listener(ssk, MPTCP_EVENT_LISTENER_CREATED);
@@ -1163,7 +1175,7 @@ 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, true);
+	ret = mptcp_pm_nl_append_new_local_addr(pernet, entry, true, false);
 	if (ret < 0)
 		kfree(entry);
 
@@ -1435,7 +1447,8 @@ int mptcp_pm_nl_add_addr_doit(struct sk_buff *skb, struct genl_info *info)
 		}
 	}
 	ret = mptcp_pm_nl_append_new_local_addr(pernet, entry,
-						!mptcp_pm_has_addr_attr_id(attr, info));
+						!mptcp_pm_has_addr_attr_id(attr, info),
+						true);
 	if (ret < 0) {
 		GENL_SET_ERR_MSG_FMT(info, "too many addresses or duplicate one: %d", ret);
 		goto out_free;
@@ -1538,11 +1551,6 @@ static int mptcp_nl_remove_subflow_and_signal_addr(struct net *net,
 		if (mptcp_pm_is_userspace(msk))
 			goto next;
 
-		if (list_empty(&msk->conn_list)) {
-			mptcp_pm_remove_anno_addr(msk, addr, false);
-			goto next;
-		}
-
 		lock_sock(sk);
 		remove_subflow = lookup_subflow_by_saddr(&msk->conn_list, addr);
 		mptcp_pm_remove_anno_addr(msk, addr, remove_subflow &&
@@ -2045,7 +2053,8 @@ int mptcp_pm_nl_set_flags(struct net *net, struct mptcp_pm_addr_entry *addr, u8
 		return -EINVAL;
 	}
 	if ((addr->flags & MPTCP_PM_ADDR_FLAG_FULLMESH) &&
-	    (entry->flags & MPTCP_PM_ADDR_FLAG_SIGNAL)) {
+	    (entry->flags & (MPTCP_PM_ADDR_FLAG_SIGNAL |
+			     MPTCP_PM_ADDR_FLAG_IMPLICIT))) {
 		spin_unlock_bh(&pernet->lock);
 		return -EINVAL;
 	}

net/mptcp/pm_userspace.c

@@ -316,7 +316,7 @@ int mptcp_pm_nl_remove_doit(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	lock_sock(sk);
-
+	spin_lock_bh(&msk->pm.lock);
 	list_for_each_entry(entry, &msk->pm.userspace_pm_local_addr_list, list) {
 		if (entry->addr.id == id_val) {
 			match = entry;
@@ -326,11 +326,13 @@ int mptcp_pm_nl_remove_doit(struct sk_buff *skb, struct genl_info *info)
 
 	if (!match) {
 		GENL_SET_ERR_MSG(info, "address with specified id not found");
+		spin_unlock_bh(&msk->pm.lock);
 		release_sock(sk);
 		goto remove_err;
 	}
 
 	list_move(&match->list, &free_list);
+	spin_unlock_bh(&msk->pm.lock);
 
 	mptcp_pm_remove_addrs(msk, &free_list);
 

net/mptcp/protocol.c

@@ -1763,8 +1763,10 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg,
 		 * see mptcp_disconnect().
 		 * Attempt it again outside the problematic scope.
 		 */
-		if (!mptcp_disconnect(sk, 0))
+		if (!mptcp_disconnect(sk, 0)) {
+			sk->sk_disconnects++;
 			sk->sk_socket->state = SS_UNCONNECTED;
+		}
 	}
 	inet_sk(sk)->defer_connect = 0;
 

net/mptcp/protocol.h

@@ -149,22 +149,24 @@ struct mptcp_options_received {
 	u32	subflow_seq;
 	u16	data_len;
 	__sum16	csum;
-	u16	suboptions;
+	struct_group(status,
+		u16 suboptions;
+		u16 use_map:1,
+		    dsn64:1,
+		    data_fin:1,
+		    use_ack:1,
+		    ack64:1,
+		    mpc_map:1,
+		    reset_reason:4,
+		    reset_transient:1,
+		    echo:1,
+		    backup:1,
+		    deny_join_id0:1,
+		    __unused:2;
+	);
+	u8	join_id;
 	u32	token;
 	u32	nonce;
-	u16	use_map:1,
-		dsn64:1,
-		data_fin:1,
-		use_ack:1,
-		ack64:1,
-		mpc_map:1,
-		reset_reason:4,
-		reset_transient:1,
-		echo:1,
-		backup:1,
-		deny_join_id0:1,
-		__unused:2;
-	u8	join_id;
 	u64	thmac;
 	u8	hmac[MPTCPOPT_HMAC_LEN];
 	struct mptcp_addr_info addr;
@@ -518,6 +520,7 @@ struct mptcp_subflow_context {
 		__unused : 9;
 	bool	data_avail;
 	bool	scheduled;
+	bool	pm_listener;	    /* a listener managed by the kernel PM? */
 	u32	remote_nonce;
 	u64	thmac;
 	u32	local_nonce;
@@ -736,10 +739,15 @@ static inline u64 mptcp_data_avail(const struct mptcp_sock *msk)
 
 static inline bool mptcp_epollin_ready(const struct sock *sk)
 {
+	u64 data_avail = mptcp_data_avail(mptcp_sk(sk));
+
+	if (!data_avail)
+		return false;
+
 	/* mptcp doesn't have to deal with small skbs in the receive queue,
-	 * at it can always coalesce them
+	 * as it can always coalesce them
 	 */
-	return (mptcp_data_avail(mptcp_sk(sk)) >= sk->sk_rcvlowat) ||
+	return (data_avail >= sk->sk_rcvlowat) ||
 	       (mem_cgroup_sockets_enabled && sk->sk_memcg &&
 		mem_cgroup_under_socket_pressure(sk->sk_memcg)) ||
 	       READ_ONCE(tcp_memory_pressure);

net/mptcp/subflow.c

@@ -134,6 +134,13 @@ static void subflow_add_reset_reason(struct sk_buff *skb, u8 reason)
 	}
 }
 
+static int subflow_reset_req_endp(struct request_sock *req, struct sk_buff *skb)
+{
+	SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_MPCAPABLEENDPATTEMPT);
+	subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT);
+	return -EPERM;
+}
+
 /* Init mptcp request socket.
  *
  * Returns an error code if a JOIN has failed and a TCP reset
@@ -165,13 +172,17 @@ static int subflow_check_req(struct request_sock *req,
 	if (opt_mp_capable) {
 		SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_MPCAPABLEPASSIVE);
 
+		if (unlikely(listener->pm_listener))
+			return subflow_reset_req_endp(req, skb);
 		if (opt_mp_join)
 			return 0;
 	} else if (opt_mp_join) {
 		SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINSYNRX);
 
 		if (mp_opt.backup)
 			SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINSYNBACKUPRX);
+	} else if (unlikely(listener->pm_listener)) {
+		return subflow_reset_req_endp(req, skb);
 	}
 
 	if (opt_mp_capable && listener->request_mptcp) {
@@ -736,8 +747,6 @@ static bool subflow_hmac_valid(const struct request_sock *req,
 
 	subflow_req = mptcp_subflow_rsk(req);
 	msk = subflow_req->msk;
-	if (!msk)
-		return false;
 
 	subflow_generate_hmac(READ_ONCE(msk->remote_key),
 			      READ_ONCE(msk->local_key),
@@ -835,12 +844,8 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 
 	} else if (subflow_req->mp_join) {
 		mptcp_get_options(skb, &mp_opt);
-		if (!(mp_opt.suboptions & OPTION_MPTCP_MPJ_ACK) ||
-		    !subflow_hmac_valid(req, &mp_opt) ||
-		    !mptcp_can_accept_new_subflow(subflow_req->msk)) {
-			SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC);
+		if (!(mp_opt.suboptions & OPTION_MPTCP_MPJ_ACK))
 			fallback = true;
-		}
 	}
 
 create_child:
@@ -890,6 +895,13 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,
 				goto dispose_child;
 			}
 
+			if (!subflow_hmac_valid(req, &mp_opt) ||
+			    !mptcp_can_accept_new_subflow(subflow_req->msk)) {
+				SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC);
+				subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT);
+				goto dispose_child;
+			}
+
 			/* move the msk reference ownership to the subflow */
 			subflow_req->msk = NULL;
 			ctx->conn = (struct sock *)owner;