From 1200832c6e850a17f36631f6492f953a1b39e6b8 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Fri, 24 Apr 2020 13:15:21 +0200 Subject: mptcp: fix race in msk status update Currently subflow_finish_connect() changes unconditionally any msk socket status other than TCP_ESTABLISHED. If an unblocking connect() races with close(), we can end-up triggering: IPv4: Attempt to release TCP socket in state 1 00000000e32b8b7e when the msk socket is disposed. Be sure to enter the established status only from SYN_SENT. Fixes: c3c123d16c0e ("net: mptcp: don't hang in mptcp_sendmsg() after TCP fallback") Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: David S. Miller --- net/mptcp/subflow.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/mptcp') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index fabd06f2ff45..71256f03707f 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -225,7 +225,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) subflow->icsk_af_ops->sk_rx_dst_set(sk, skb); - if (inet_sk_state_load(parent) != TCP_ESTABLISHED) { + if (inet_sk_state_load(parent) == TCP_SYN_SENT) { inet_sk_state_store(parent, TCP_ESTABLISHED); parent->sk_state_change(parent); } -- cgit v1.2.3-59-g8ed1b From 42c556fef92361bbc58be22f91b1c49db0963c34 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 29 Apr 2020 20:43:20 +0200 Subject: mptcp: replace mptcp_disconnect with a stub Paolo points out that mptcp_disconnect is bogus: "lock_sock(sk); looks suspicious (lock should be already held by the caller) And call to: tcp_disconnect(sk, flags); too, sk is not a tcp socket". ->disconnect() gets called from e.g. inet_stream_connect when one tries to disassociate a connected socket again (to re-connect without closing the socket first). MPTCP however uses mptcp_stream_connect, not inet_stream_connect, for the mptcp-socket connect call. inet_stream_connect only gets called indirectly, for the tcp socket, so any ->disconnect() calls end up calling tcp_disconnect for that tcp subflow sk. This also explains why syzkaller has not yet reported a problem here. So for now replace this with a stub that doesn't do anything. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/14 Acked-by: Paolo Abeni Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- net/mptcp/protocol.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'net/mptcp') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index b22a63ba2348..6e0188f5d3f3 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -1316,11 +1316,12 @@ static void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk) static int mptcp_disconnect(struct sock *sk, int flags) { - lock_sock(sk); - __mptcp_clear_xmit(sk); - release_sock(sk); - mptcp_cancel_work(sk); - return tcp_disconnect(sk, flags); + /* Should never be called. + * inet_stream_connect() calls ->disconnect, but that + * refers to the subflow socket, not the mptcp one. + */ + WARN_ON_ONCE(1); + return 0; } #if IS_ENABLED(CONFIG_MPTCP_IPV6) -- cgit v1.2.3-59-g8ed1b From 263e1201a2c324b60b15ecda5de9ebf1e7293e31 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 30 Apr 2020 15:01:51 +0200 Subject: mptcp: consolidate synack processing. Currently the MPTCP code uses 2 hooks to process syn-ack packets, mptcp_rcv_synsent() and the sk_rx_dst_set() callback. We can drop the first, moving the relevant code into the latter, reducing the hooking into the TCP code. This is also needed by the next patch. v1 -> v2: - use local tcp sock ptr instead of casting the sk variable several times - DaveM Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- include/net/mptcp.h | 1 - net/ipv4/tcp_input.c | 3 --- net/mptcp/options.c | 22 ---------------------- net/mptcp/subflow.c | 27 ++++++++++++++++++++++++--- 4 files changed, 24 insertions(+), 29 deletions(-) (limited to 'net/mptcp') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 0e7c5471010b..4ecfa7d5e0c7 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -72,7 +72,6 @@ void mptcp_parse_option(const struct sk_buff *skb, const unsigned char *ptr, int opsize, struct tcp_options_received *opt_rx); bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb, unsigned int *size, struct mptcp_out_options *opts); -void mptcp_rcv_synsent(struct sock *sk); bool mptcp_synack_options(const struct request_sock *req, unsigned int *size, struct mptcp_out_options *opts); bool mptcp_established_options(struct sock *sk, struct sk_buff *skb, diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index bf4ced9273e8..81425542da44 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5990,9 +5990,6 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); tcp_initialize_rcv_mss(sk); - if (sk_is_mptcp(sk)) - mptcp_rcv_synsent(sk); - /* Remember, tcp_poll() does not lock socket! * Change state from SYN-SENT only after copied_seq * is initialized. */ diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 4a7c467b99db..8fea686a5562 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -344,28 +344,6 @@ bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb, return false; } -void mptcp_rcv_synsent(struct sock *sk) -{ - struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); - struct tcp_sock *tp = tcp_sk(sk); - - if (subflow->request_mptcp && tp->rx_opt.mptcp.mp_capable) { - subflow->mp_capable = 1; - subflow->can_ack = 1; - subflow->remote_key = tp->rx_opt.mptcp.sndr_key; - pr_debug("subflow=%p, remote_key=%llu", subflow, - subflow->remote_key); - } else if (subflow->request_join && tp->rx_opt.mptcp.mp_join) { - subflow->mp_join = 1; - subflow->thmac = tp->rx_opt.mptcp.thmac; - subflow->remote_nonce = tp->rx_opt.mptcp.nonce; - pr_debug("subflow=%p, thmac=%llu, remote_nonce=%u", subflow, - subflow->thmac, subflow->remote_nonce); - } else if (subflow->request_mptcp) { - tcp_sk(sk)->is_mptcp = 0; - } -} - /* MP_JOIN client subflow must wait for 4th ack before sending any data: * TCP can't schedule delack timer before the subflow is fully established. * MPTCP uses the delack timer to do 3rd ack retransmissions diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 71256f03707f..84f6408594c9 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -222,6 +222,7 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); struct sock *parent = subflow->conn; + struct tcp_sock *tp = tcp_sk(sk); subflow->icsk_af_ops->sk_rx_dst_set(sk, skb); @@ -230,14 +231,35 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) parent->sk_state_change(parent); } - if (subflow->conn_finished || !tcp_sk(sk)->is_mptcp) + /* be sure no special action on any packet other than syn-ack */ + if (subflow->conn_finished) + return; + + subflow->conn_finished = 1; + + if (subflow->request_mptcp && tp->rx_opt.mptcp.mp_capable) { + subflow->mp_capable = 1; + subflow->can_ack = 1; + subflow->remote_key = tp->rx_opt.mptcp.sndr_key; + pr_debug("subflow=%p, remote_key=%llu", subflow, + subflow->remote_key); + } else if (subflow->request_join && tp->rx_opt.mptcp.mp_join) { + subflow->mp_join = 1; + subflow->thmac = tp->rx_opt.mptcp.thmac; + subflow->remote_nonce = tp->rx_opt.mptcp.nonce; + pr_debug("subflow=%p, thmac=%llu, remote_nonce=%u", subflow, + subflow->thmac, subflow->remote_nonce); + } else if (subflow->request_mptcp) { + tp->is_mptcp = 0; + } + + if (!tp->is_mptcp) return; if (subflow->mp_capable) { pr_debug("subflow=%p, remote_key=%llu", mptcp_subflow_ctx(sk), subflow->remote_key); mptcp_finish_connect(sk); - subflow->conn_finished = 1; if (skb) { pr_debug("synack seq=%u", TCP_SKB_CB(skb)->seq); @@ -264,7 +286,6 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) if (!mptcp_finish_join(sk)) goto do_reset; - subflow->conn_finished = 1; MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_JOINSYNACKRX); } else { do_reset: -- cgit v1.2.3-59-g8ed1b From cfde141ea3faa30e362bbdb5c28001bbbdb0b8e0 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 30 Apr 2020 15:01:52 +0200 Subject: mptcp: move option parsing into mptcp_incoming_options() The mptcp_options_received structure carries several per packet flags (mp_capable, mp_join, etc.). Such fields must be cleared on each packet, even on dropped ones or packet not carrying any MPTCP options, but the current mptcp code clears them only on TCP option reset. On several races/corner cases we end-up with stray bits in incoming options, leading to WARN_ON splats. e.g.: [ 171.164906] Bad mapping: ssn=32714 map_seq=1 map_data_len=32713 [ 171.165006] WARNING: CPU: 1 PID: 5026 at net/mptcp/subflow.c:533 warn_bad_map (linux-mptcp/net/mptcp/subflow.c:533 linux-mptcp/net/mptcp/subflow.c:531) [ 171.167632] Modules linked in: ip6_vti ip_vti ip_gre ipip sit tunnel4 ip_tunnel geneve ip6_udp_tunnel udp_tunnel macsec macvtap tap ipvlan macvlan 8021q garp mrp xfrm_interface veth netdevsim nlmon dummy team bonding vcan bridge stp llc ip6_gre gre ip6_tunnel tunnel6 tun binfmt_misc intel_rapl_msr intel_rapl_common rfkill kvm_intel kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel joydev virtio_balloon pcspkr i2c_piix4 sunrpc ip_tables xfs libcrc32c crc32c_intel serio_raw virtio_console ata_generic virtio_blk virtio_net net_failover failover ata_piix libata [ 171.199464] CPU: 1 PID: 5026 Comm: repro Not tainted 5.7.0-rc1.mptcp_f227fdf5d388+ #95 [ 171.200886] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-2.fc30 04/01/2014 [ 171.202546] RIP: 0010:warn_bad_map (linux-mptcp/net/mptcp/subflow.c:533 linux-mptcp/net/mptcp/subflow.c:531) [ 171.206537] Code: c1 ea 03 0f b6 14 02 48 89 f8 83 e0 07 83 c0 03 38 d0 7c 04 84 d2 75 1d 8b 55 3c 44 89 e6 48 c7 c7 20 51 13 95 e8 37 8b 22 fe <0f> 0b 48 83 c4 08 5b 5d 41 5c c3 89 4c 24 04 e8 db d6 94 fe 8b 4c [ 171.220473] RSP: 0018:ffffc90000150560 EFLAGS: 00010282 [ 171.221639] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ 171.223108] RDX: 0000000000000000 RSI: 0000000000000008 RDI: fffff5200002a09e [ 171.224388] RBP: ffff8880aa6e3c00 R08: 0000000000000001 R09: fffffbfff2ec9955 [ 171.225706] R10: ffffffff9764caa7 R11: fffffbfff2ec9954 R12: 0000000000007fca [ 171.227211] R13: ffff8881066f4a7f R14: ffff8880aa6e3c00 R15: 0000000000000020 [ 171.228460] FS: 00007f8623719740(0000) GS:ffff88810be00000(0000) knlGS:0000000000000000 [ 171.230065] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 171.231303] CR2: 00007ffdab190a50 CR3: 00000001038ea006 CR4: 0000000000160ee0 [ 171.232586] Call Trace: [ 171.233109] [ 171.233531] get_mapping_status (linux-mptcp/net/mptcp/subflow.c:691) [ 171.234371] mptcp_subflow_data_available (linux-mptcp/net/mptcp/subflow.c:736 linux-mptcp/net/mptcp/subflow.c:832) [ 171.238181] subflow_state_change (linux-mptcp/net/mptcp/subflow.c:1085 (discriminator 1)) [ 171.239066] tcp_fin (linux-mptcp/net/ipv4/tcp_input.c:4217) [ 171.240123] tcp_data_queue (linux-mptcp/./include/linux/compiler.h:199 linux-mptcp/net/ipv4/tcp_input.c:4822) [ 171.245083] tcp_rcv_established (linux-mptcp/./include/linux/skbuff.h:1785 linux-mptcp/./include/net/tcp.h:1774 linux-mptcp/./include/net/tcp.h:1847 linux-mptcp/net/ipv4/tcp_input.c:5238 linux-mptcp/net/ipv4/tcp_input.c:5730) [ 171.254089] tcp_v4_rcv (linux-mptcp/./include/linux/spinlock.h:393 linux-mptcp/net/ipv4/tcp_ipv4.c:2009) [ 171.258969] ip_protocol_deliver_rcu (linux-mptcp/net/ipv4/ip_input.c:204 (discriminator 1)) [ 171.260214] ip_local_deliver_finish (linux-mptcp/./include/linux/rcupdate.h:651 linux-mptcp/net/ipv4/ip_input.c:232) [ 171.261389] ip_local_deliver (linux-mptcp/./include/linux/netfilter.h:307 linux-mptcp/./include/linux/netfilter.h:301 linux-mptcp/net/ipv4/ip_input.c:252) [ 171.265884] ip_rcv (linux-mptcp/./include/linux/netfilter.h:307 linux-mptcp/./include/linux/netfilter.h:301 linux-mptcp/net/ipv4/ip_input.c:539) [ 171.273666] process_backlog (linux-mptcp/./include/linux/rcupdate.h:651 linux-mptcp/net/core/dev.c:6135) [ 171.275328] net_rx_action (linux-mptcp/net/core/dev.c:6572 linux-mptcp/net/core/dev.c:6640) [ 171.280472] __do_softirq (linux-mptcp/./arch/x86/include/asm/jump_label.h:25 linux-mptcp/./include/linux/jump_label.h:200 linux-mptcp/./include/trace/events/irq.h:142 linux-mptcp/kernel/softirq.c:293) [ 171.281379] do_softirq_own_stack (linux-mptcp/arch/x86/entry/entry_64.S:1083) [ 171.282358] We could address the issue clearing explicitly the relevant fields in several places - tcp_parse_option, tcp_fast_parse_options, possibly others. Instead we move the MPTCP option parsing into the already existing mptcp ingress hook, so that we need to clear the fields in a single place. This allows us dropping an MPTCP hook from the TCP code and removing the quite large mptcp_options_received from the tcp_sock struct. On the flip side, the MPTCP sockets will traverse the option space twice (in tcp_parse_option() and in mptcp_incoming_options(). That looks acceptable: we already do that for syn and 3rd ack packets, plain TCP socket will benefit from it, and even MPTCP sockets will experience better code locality, reducing the jumps between TCP and MPTCP code. v1 -> v2: - rebased on current '-net' tree Fixes: 648ef4b88673 ("mptcp: Implement MPTCP receive path") Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- include/linux/tcp.h | 51 ---------------------------------------- include/net/mptcp.h | 2 -- net/ipv4/tcp_input.c | 4 ---- net/mptcp/options.c | 66 +++++++++++++++++++++++++++++----------------------- net/mptcp/protocol.c | 6 ++--- net/mptcp/protocol.h | 43 ++++++++++++++++++++++++++++++++-- net/mptcp/subflow.c | 65 +++++++++++++++++++++++++++------------------------ 7 files changed, 115 insertions(+), 122 deletions(-) (limited to 'net/mptcp') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 421c99c12291..4f8159e90ce1 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -78,47 +78,6 @@ struct tcp_sack_block { #define TCP_SACK_SEEN (1 << 0) /*1 = peer is SACK capable, */ #define TCP_DSACK_SEEN (1 << 2) /*1 = DSACK was received from peer*/ -#if IS_ENABLED(CONFIG_MPTCP) -struct mptcp_options_received { - u64 sndr_key; - u64 rcvr_key; - u64 data_ack; - u64 data_seq; - u32 subflow_seq; - u16 data_len; - u16 mp_capable : 1, - mp_join : 1, - dss : 1, - add_addr : 1, - rm_addr : 1, - family : 4, - echo : 1, - backup : 1; - u32 token; - u32 nonce; - u64 thmac; - u8 hmac[20]; - u8 join_id; - u8 use_map:1, - dsn64:1, - data_fin:1, - use_ack:1, - ack64:1, - mpc_map:1, - __unused:2; - u8 addr_id; - u8 rm_id; - union { - struct in_addr addr; -#if IS_ENABLED(CONFIG_MPTCP_IPV6) - struct in6_addr addr6; -#endif - }; - u64 ahmac; - u16 port; -}; -#endif - struct tcp_options_received { /* PAWS/RTTM data */ int ts_recent_stamp;/* Time we stored ts_recent (for aging) */ @@ -136,9 +95,6 @@ struct tcp_options_received { u8 num_sacks; /* Number of SACK blocks */ u16 user_mss; /* mss requested by user in ioctl */ u16 mss_clamp; /* Maximal mss, negotiated at connection setup */ -#if IS_ENABLED(CONFIG_MPTCP) - struct mptcp_options_received mptcp; -#endif }; static inline void tcp_clear_options(struct tcp_options_received *rx_opt) @@ -148,13 +104,6 @@ static inline void tcp_clear_options(struct tcp_options_received *rx_opt) #if IS_ENABLED(CONFIG_SMC) rx_opt->smc_ok = 0; #endif -#if IS_ENABLED(CONFIG_MPTCP) - rx_opt->mptcp.mp_capable = 0; - rx_opt->mptcp.mp_join = 0; - rx_opt->mptcp.add_addr = 0; - rx_opt->mptcp.rm_addr = 0; - rx_opt->mptcp.dss = 0; -#endif } /* This is the max number of SACKS that we'll generate and process. It's safe diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 4ecfa7d5e0c7..3bce2019e4da 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -68,8 +68,6 @@ static inline bool rsk_is_mptcp(const struct request_sock *req) return tcp_rsk(req)->is_mptcp; } -void mptcp_parse_option(const struct sk_buff *skb, const unsigned char *ptr, - int opsize, struct tcp_options_received *opt_rx); bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb, unsigned int *size, struct mptcp_out_options *opts); bool mptcp_synack_options(const struct request_sock *req, unsigned int *size, diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 81425542da44..b996dc1069c5 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3926,10 +3926,6 @@ void tcp_parse_options(const struct net *net, */ break; #endif - case TCPOPT_MPTCP: - mptcp_parse_option(skb, ptr, opsize, opt_rx); - break; - case TCPOPT_FASTOPEN: tcp_parse_fastopen_option( opsize - TCPOLEN_FASTOPEN_BASE, diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 8fea686a5562..eadbd59586e4 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -16,10 +16,10 @@ static bool mptcp_cap_flag_sha256(u8 flags) return (flags & MPTCP_CAP_FLAG_MASK) == MPTCP_CAP_HMAC_SHA256; } -void mptcp_parse_option(const struct sk_buff *skb, const unsigned char *ptr, - int opsize, struct tcp_options_received *opt_rx) +static void mptcp_parse_option(const struct sk_buff *skb, + const unsigned char *ptr, int opsize, + struct mptcp_options_received *mp_opt) { - struct mptcp_options_received *mp_opt = &opt_rx->mptcp; u8 subtype = *ptr >> 4; int expected_opsize; u8 version; @@ -283,12 +283,20 @@ void mptcp_parse_option(const struct sk_buff *skb, const unsigned char *ptr, } void mptcp_get_options(const struct sk_buff *skb, - struct tcp_options_received *opt_rx) + struct mptcp_options_received *mp_opt) { - const unsigned char *ptr; const struct tcphdr *th = tcp_hdr(skb); - int length = (th->doff * 4) - sizeof(struct tcphdr); + const unsigned char *ptr; + int length; + /* initialize option status */ + mp_opt->mp_capable = 0; + mp_opt->mp_join = 0; + mp_opt->add_addr = 0; + mp_opt->rm_addr = 0; + mp_opt->dss = 0; + + length = (th->doff * 4) - sizeof(struct tcphdr); ptr = (const unsigned char *)(th + 1); while (length > 0) { @@ -308,7 +316,7 @@ void mptcp_get_options(const struct sk_buff *skb, if (opsize > length) return; /* don't parse partial options */ if (opcode == TCPOPT_MPTCP) - mptcp_parse_option(skb, ptr, opsize, opt_rx); + mptcp_parse_option(skb, ptr, opsize, mp_opt); ptr += opsize - 2; length -= opsize; } @@ -797,41 +805,41 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb, { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); struct mptcp_sock *msk = mptcp_sk(subflow->conn); - struct mptcp_options_received *mp_opt; + struct mptcp_options_received mp_opt; struct mptcp_ext *mpext; - mp_opt = &opt_rx->mptcp; - if (!check_fully_established(msk, sk, subflow, skb, mp_opt)) + mptcp_get_options(skb, &mp_opt); + if (!check_fully_established(msk, sk, subflow, skb, &mp_opt)) return; - if (mp_opt->add_addr && add_addr_hmac_valid(msk, mp_opt)) { + if (mp_opt.add_addr && add_addr_hmac_valid(msk, &mp_opt)) { struct mptcp_addr_info addr; - addr.port = htons(mp_opt->port); - addr.id = mp_opt->addr_id; - if (mp_opt->family == MPTCP_ADDR_IPVERSION_4) { + addr.port = htons(mp_opt.port); + addr.id = mp_opt.addr_id; + if (mp_opt.family == MPTCP_ADDR_IPVERSION_4) { addr.family = AF_INET; - addr.addr = mp_opt->addr; + addr.addr = mp_opt.addr; } #if IS_ENABLED(CONFIG_MPTCP_IPV6) - else if (mp_opt->family == MPTCP_ADDR_IPVERSION_6) { + else if (mp_opt.family == MPTCP_ADDR_IPVERSION_6) { addr.family = AF_INET6; - addr.addr6 = mp_opt->addr6; + addr.addr6 = mp_opt.addr6; } #endif - if (!mp_opt->echo) + if (!mp_opt.echo) mptcp_pm_add_addr_received(msk, &addr); - mp_opt->add_addr = 0; + mp_opt.add_addr = 0; } - if (!mp_opt->dss) + if (!mp_opt.dss) return; /* we can't wait for recvmsg() to update the ack_seq, otherwise * monodirectional flows will stuck */ - if (mp_opt->use_ack) - update_una(msk, mp_opt); + if (mp_opt.use_ack) + update_una(msk, &mp_opt); mpext = skb_ext_add(skb, SKB_EXT_MPTCP); if (!mpext) @@ -839,8 +847,8 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb, memset(mpext, 0, sizeof(*mpext)); - if (mp_opt->use_map) { - if (mp_opt->mpc_map) { + if (mp_opt.use_map) { + if (mp_opt.mpc_map) { /* this is an MP_CAPABLE carrying MPTCP data * we know this map the first chunk of data */ @@ -851,12 +859,12 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb, mpext->dsn64 = 1; mpext->mpc_map = 1; } else { - mpext->data_seq = mp_opt->data_seq; - mpext->subflow_seq = mp_opt->subflow_seq; - mpext->dsn64 = mp_opt->dsn64; - mpext->data_fin = mp_opt->data_fin; + mpext->data_seq = mp_opt.data_seq; + mpext->subflow_seq = mp_opt.subflow_seq; + mpext->dsn64 = mp_opt.dsn64; + mpext->data_fin = mp_opt.data_fin; } - mpext->data_len = mp_opt->data_len; + mpext->data_len = mp_opt.data_len; mpext->use_map = 1; } } diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 6e0188f5d3f3..e1f23016ed3f 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -1334,7 +1334,7 @@ static struct ipv6_pinfo *mptcp_inet6_sk(const struct sock *sk) #endif struct sock *mptcp_sk_clone(const struct sock *sk, - const struct tcp_options_received *opt_rx, + const struct mptcp_options_received *mp_opt, struct request_sock *req) { struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req); @@ -1373,9 +1373,9 @@ struct sock *mptcp_sk_clone(const struct sock *sk, msk->write_seq = subflow_req->idsn + 1; atomic64_set(&msk->snd_una, msk->write_seq); - if (opt_rx->mptcp.mp_capable) { + if (mp_opt->mp_capable) { msk->can_ack = true; - msk->remote_key = opt_rx->mptcp.sndr_key; + msk->remote_key = mp_opt->sndr_key; mptcp_crypto_key_sha(msk->remote_key, NULL, &ack_seq); ack_seq++; msk->ack_seq = ack_seq; diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index a2b3048037d0..e4ca6320ce76 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -91,6 +91,45 @@ #define MPTCP_WORK_RTX 2 #define MPTCP_WORK_EOF 3 +struct mptcp_options_received { + u64 sndr_key; + u64 rcvr_key; + u64 data_ack; + u64 data_seq; + u32 subflow_seq; + u16 data_len; + u16 mp_capable : 1, + mp_join : 1, + dss : 1, + add_addr : 1, + rm_addr : 1, + family : 4, + echo : 1, + backup : 1; + u32 token; + u32 nonce; + u64 thmac; + u8 hmac[20]; + u8 join_id; + u8 use_map:1, + dsn64:1, + data_fin:1, + use_ack:1, + ack64:1, + mpc_map:1, + __unused:2; + u8 addr_id; + u8 rm_id; + union { + struct in_addr addr; +#if IS_ENABLED(CONFIG_MPTCP_IPV6) + struct in6_addr addr6; +#endif + }; + u64 ahmac; + u16 port; +}; + static inline __be32 mptcp_option(u8 subopt, u8 len, u8 nib, u8 field) { return htonl((TCPOPT_MPTCP << 24) | (len << 16) | (subopt << 12) | @@ -331,10 +370,10 @@ int mptcp_proto_v6_init(void); #endif struct sock *mptcp_sk_clone(const struct sock *sk, - const struct tcp_options_received *opt_rx, + const struct mptcp_options_received *mp_opt, struct request_sock *req); void mptcp_get_options(const struct sk_buff *skb, - struct tcp_options_received *opt_rx); + struct mptcp_options_received *mp_opt); void mptcp_finish_connect(struct sock *sk); void mptcp_data_ready(struct sock *sk, struct sock *ssk); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 84f6408594c9..bad998529767 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -124,12 +124,11 @@ static void subflow_init_req(struct request_sock *req, { struct mptcp_subflow_context *listener = mptcp_subflow_ctx(sk_listener); struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req); - struct tcp_options_received rx_opt; + struct mptcp_options_received mp_opt; pr_debug("subflow_req=%p, listener=%p", subflow_req, listener); - memset(&rx_opt.mptcp, 0, sizeof(rx_opt.mptcp)); - mptcp_get_options(skb, &rx_opt); + mptcp_get_options(skb, &mp_opt); subflow_req->mp_capable = 0; subflow_req->mp_join = 0; @@ -142,16 +141,16 @@ static void subflow_init_req(struct request_sock *req, return; #endif - if (rx_opt.mptcp.mp_capable) { + if (mp_opt.mp_capable) { SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_MPCAPABLEPASSIVE); - if (rx_opt.mptcp.mp_join) + if (mp_opt.mp_join) return; - } else if (rx_opt.mptcp.mp_join) { + } else if (mp_opt.mp_join) { SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINSYNRX); } - if (rx_opt.mptcp.mp_capable && listener->request_mptcp) { + if (mp_opt.mp_capable && listener->request_mptcp) { int err; err = mptcp_token_new_request(req); @@ -159,13 +158,13 @@ static void subflow_init_req(struct request_sock *req, subflow_req->mp_capable = 1; subflow_req->ssn_offset = TCP_SKB_CB(skb)->seq; - } else if (rx_opt.mptcp.mp_join && listener->request_mptcp) { + } else if (mp_opt.mp_join && listener->request_mptcp) { subflow_req->ssn_offset = TCP_SKB_CB(skb)->seq; subflow_req->mp_join = 1; - subflow_req->backup = rx_opt.mptcp.backup; - subflow_req->remote_id = rx_opt.mptcp.join_id; - subflow_req->token = rx_opt.mptcp.token; - subflow_req->remote_nonce = rx_opt.mptcp.nonce; + subflow_req->backup = mp_opt.backup; + subflow_req->remote_id = mp_opt.join_id; + subflow_req->token = mp_opt.token; + subflow_req->remote_nonce = mp_opt.nonce; pr_debug("token=%u, remote_nonce=%u", subflow_req->token, subflow_req->remote_nonce); if (!subflow_token_join_request(req, skb)) { @@ -221,6 +220,7 @@ static bool subflow_thmac_valid(struct mptcp_subflow_context *subflow) static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); + struct mptcp_options_received mp_opt; struct sock *parent = subflow->conn; struct tcp_sock *tp = tcp_sk(sk); @@ -237,16 +237,17 @@ static void subflow_finish_connect(struct sock *sk, const struct sk_buff *skb) subflow->conn_finished = 1; - if (subflow->request_mptcp && tp->rx_opt.mptcp.mp_capable) { + mptcp_get_options(skb, &mp_opt); + if (subflow->request_mptcp && mp_opt.mp_capable) { subflow->mp_capable = 1; subflow->can_ack = 1; - subflow->remote_key = tp->rx_opt.mptcp.sndr_key; + subflow->remote_key = mp_opt.sndr_key; pr_debug("subflow=%p, remote_key=%llu", subflow, subflow->remote_key); - } else if (subflow->request_join && tp->rx_opt.mptcp.mp_join) { + } else if (subflow->request_join && mp_opt.mp_join) { subflow->mp_join = 1; - subflow->thmac = tp->rx_opt.mptcp.thmac; - subflow->remote_nonce = tp->rx_opt.mptcp.nonce; + subflow->thmac = mp_opt.thmac; + subflow->remote_nonce = mp_opt.nonce; pr_debug("subflow=%p, thmac=%llu, remote_nonce=%u", subflow, subflow->thmac, subflow->remote_nonce); } else if (subflow->request_mptcp) { @@ -343,7 +344,7 @@ drop: /* validate hmac received in third ACK */ static bool subflow_hmac_valid(const struct request_sock *req, - const struct tcp_options_received *rx_opt) + const struct mptcp_options_received *mp_opt) { const struct mptcp_subflow_request_sock *subflow_req; u8 hmac[MPTCPOPT_HMAC_LEN]; @@ -360,7 +361,7 @@ static bool subflow_hmac_valid(const struct request_sock *req, subflow_req->local_nonce, hmac); ret = true; - if (crypto_memneq(hmac, rx_opt->mptcp.hmac, sizeof(hmac))) + if (crypto_memneq(hmac, mp_opt->hmac, sizeof(hmac))) ret = false; sock_put((struct sock *)msk); @@ -416,7 +417,7 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, { struct mptcp_subflow_context *listener = mptcp_subflow_ctx(sk); struct mptcp_subflow_request_sock *subflow_req; - struct tcp_options_received opt_rx; + struct mptcp_options_received mp_opt; bool fallback_is_fatal = false; struct sock *new_msk = NULL; bool fallback = false; @@ -424,7 +425,10 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, pr_debug("listener=%p, req=%p, conn=%p", listener, req, listener->conn); - opt_rx.mptcp.mp_capable = 0; + /* we need later a valid 'mp_capable' value even when options are not + * parsed + */ + mp_opt.mp_capable = 0; if (tcp_rsk(req)->is_mptcp == 0) goto create_child; @@ -439,22 +443,21 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, goto create_msk; } - mptcp_get_options(skb, &opt_rx); - if (!opt_rx.mptcp.mp_capable) { + mptcp_get_options(skb, &mp_opt); + if (!mp_opt.mp_capable) { fallback = true; goto create_child; } create_msk: - new_msk = mptcp_sk_clone(listener->conn, &opt_rx, req); + new_msk = mptcp_sk_clone(listener->conn, &mp_opt, req); if (!new_msk) fallback = true; } else if (subflow_req->mp_join) { fallback_is_fatal = true; - opt_rx.mptcp.mp_join = 0; - mptcp_get_options(skb, &opt_rx); - if (!opt_rx.mptcp.mp_join || - !subflow_hmac_valid(req, &opt_rx)) { + mptcp_get_options(skb, &mp_opt); + if (!mp_opt.mp_join || + !subflow_hmac_valid(req, &mp_opt)) { SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC); return NULL; } @@ -494,9 +497,9 @@ create_child: /* with OoO packets we can reach here without ingress * mpc option */ - ctx->remote_key = opt_rx.mptcp.sndr_key; - ctx->fully_established = opt_rx.mptcp.mp_capable; - ctx->can_ack = opt_rx.mptcp.mp_capable; + ctx->remote_key = mp_opt.sndr_key; + ctx->fully_established = mp_opt.mp_capable; + ctx->can_ack = mp_opt.mp_capable; } else if (ctx->mp_join) { struct mptcp_sock *owner; -- cgit v1.2.3-59-g8ed1b From d6085fe19b8ec53d8ea2d8a10390b7ab6978ac44 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 30 Apr 2020 15:01:53 +0200 Subject: mptcp: avoid a WARN on bad input. Syzcaller has found a way to trigger the WARN_ON_ONCE condition in check_fully_established(). The root cause is a legit fallback to TCP scenario, so replace the WARN with a plain message on a more strict condition. Fixes: f296234c98a8 ("mptcp: Add handling of incoming MP_JOIN requests") Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- net/mptcp/options.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/mptcp') diff --git a/net/mptcp/options.c b/net/mptcp/options.c index eadbd59586e4..ecf41d52d2fc 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -703,8 +703,6 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *sk, goto fully_established; } - WARN_ON_ONCE(subflow->can_ack); - /* If the first established packet does not contain MP_CAPABLE + data * then fallback to TCP */ @@ -714,6 +712,8 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *sk, return false; } + if (unlikely(!READ_ONCE(msk->pm.server_side))) + pr_warn_once("bogus mpc option on established client sk"); subflow->fully_established = 1; subflow->remote_key = mp_opt->sndr_key; subflow->can_ack = 1; -- cgit v1.2.3-59-g8ed1b From 5a91e32b40af86c790c4348aa3a5926a5c5fbc6d Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 30 Apr 2020 15:01:54 +0200 Subject: mptcp: fix 'use_ack' option access. The mentioned RX option field is initialized only for DSS packet, we must access it only if 'dss' is set too, or the subflow will end-up in a bad status, leading to RFC violations. Fixes: d22f4988ffec ("mptcp: process MP_CAPABLE data option") Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- net/mptcp/options.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/mptcp') diff --git a/net/mptcp/options.c b/net/mptcp/options.c index ecf41d52d2fc..9486720c3256 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -695,7 +695,7 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *sk, if (TCP_SKB_CB(skb)->seq != subflow->ssn_offset + 1) return subflow->mp_capable; - if (mp_opt->use_ack) { + if (mp_opt->dss && mp_opt->use_ack) { /* subflows are fully established as soon as we get any * additional ack. */ -- cgit v1.2.3-59-g8ed1b From a77895dbc0ad0400e377e8f26567820b3658e30e Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 30 Apr 2020 15:01:55 +0200 Subject: mptcp: initialize the data_fin field for mpc packets When parsing MPC+data packets we set the dss field, so we must also initialize the data_fin, or we can find stray value there. Fixes: 9a19371bf029 ("mptcp: fix data_fin handing in RX path") Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- net/mptcp/options.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/mptcp') diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 9486720c3256..45497af23906 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -858,6 +858,7 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb, mpext->subflow_seq = 1; mpext->dsn64 = 1; mpext->mpc_map = 1; + mpext->data_fin = 0; } else { mpext->data_seq = mp_opt.data_seq; mpext->subflow_seq = mp_opt.subflow_seq; -- cgit v1.2.3-59-g8ed1b From ac2b47fb92c50682d89d7350a491a6a392bac5dd Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 30 Apr 2020 15:03:22 +0200 Subject: mptcp: fix uninitialized value access tcp_v{4,6}_syn_recv_sock() set 'own_req' only when returning a not NULL 'child', let's check 'own_req' only if child is available to avoid an - unharmful - UBSAN splat. v1 -> v2: - reference the correct hash Fixes: 4c8941de781c ("mptcp: avoid flipping mp_capable field in syn_recv_sock()") Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- net/mptcp/subflow.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/mptcp') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index bad998529767..67a4e35d4838 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -523,7 +523,7 @@ out: /* check for expected invariant - should never trigger, just help * catching eariler subtle bugs */ - WARN_ON_ONCE(*own_req && child && tcp_sk(child)->is_mptcp && + WARN_ON_ONCE(child && *own_req && tcp_sk(child)->is_mptcp && (!mptcp_subflow_ctx(child) || !mptcp_subflow_ctx(child)->conn)); return child; -- cgit v1.2.3-59-g8ed1b