The MPTCP-congestion controls have been removed. This here is a left-over. Signed-off-by: Christoph Paasch <cpaasch(a)apple.com&gt; --- net/ipv4/Kconfig | 19 ------------------- 1 file changed, 19 deletions(-) diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 3620a29db1ef..32cae39cdff6 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -717,21 +717,6 @@ choice config DEFAULT_BBR bool "BBR" if TCP_CONG_BBR=y - config DEFAULT_LIA - bool "Lia" if TCP_CONG_LIA=y - - config DEFAULT_OLIA - bool "Olia" if TCP_CONG_OLIA=y - - config DEFAULT_WVEGAS - bool "Wvegas" if TCP_CONG_WVEGAS=y - - config DEFAULT_BALIA - bool "Balia" if TCP_CONG_BALIA=y - - config DEFAULT_MCTCPDESYNC - bool "Mctcpdesync (EXPERIMENTAL)" if TCP_CONG_MCTCPDESYNC=y - config DEFAULT_RENO bool "Reno" endchoice @@ -752,10 +737,6 @@ config DEFAULT_TCP_CONG default "vegas" if DEFAULT_VEGAS default "westwood" if DEFAULT_WESTWOOD default "veno" if DEFAULT_VENO - default "lia" if DEFAULT_LIA - default "olia" if DEFAULT_OLIA - default "wvegas" if DEFAULT_WVEGAS - default "balia" if DEFAULT_BALIA default "reno" if DEFAULT_RENO default "dctcp" if DEFAULT_DCTCP default "cdg" if DEFAULT_CDG -- 2.16.2

Signed-off-by: Christoph Paasch <cpaasch(a)apple.com&gt; --- Notes: v2: * Make sure that we detect NETDEV_CHANGE events unsetting the IFF_NOMULTIPATH flag include/linux/tcp.h | 1 - include/net/mptcp.h | 42 +----------- include/uapi/linux/if.h | 1 - net/core/dev.c | 2 +- net/mptcp/mptcp_ctrl.c | 6 -- net/mptcp/mptcp_fullmesh.c | 151 +++++++++++-------------------------------- net/mptcp/mptcp_input.c | 36 ----------- net/mptcp/mptcp_ipv4.c | 5 +- net/mptcp/mptcp_ipv6.c | 5 +- net/mptcp/mptcp_ndiffports.c | 4 +- net/mptcp/mptcp_output.c | 27 +------- net/mptcp/mptcp_pm.c | 2 +- net/mptcp/mptcp_sched.c | 20 +----- 13 files changed, 46 insertions(+), 256 deletions(-) diff --git a/include/linux/tcp.h b/include/linux/tcp.h index f38bf1e18e58..cc896f842c31 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -109,7 +109,6 @@ struct tcp_out_options { __u32 sender_nonce; /* random number of the sender */ __u32 token; /* token for mptcp */ - u8 low_prio:1; } mp_join_syns; }; diff --git a/include/net/mptcp.h b/include/net/mptcp.h index a512e9386296..a4f46ac7051b 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -56,7 +56,6 @@ struct mptcp_loc4 { u8 loc4_id; - u8 low_prio:1; int if_idx; struct in_addr addr; }; @@ -69,7 +68,6 @@ struct mptcp_rem4 { struct mptcp_loc6 { u8 loc6_id; - u8 low_prio:1; int if_idx; struct in6_addr addr; }; @@ -103,8 +101,6 @@ struct mptcp_request_sock { u8 loc_id; u8 rem_id; /* Address-id in the MP_JOIN */ u8 is_sub:1, /* Is this a new subflow? */ - low_prio:1, /* Interface set to low-prio? */ - rcv_low_prio:1, mptcp_ver:4; }; @@ -115,11 +111,6 @@ struct mptcp_options_received { is_mp_join:1, join_ack:1, - saw_low_prio:2, /* 0x1 - low-prio set for this subflow - * 0x2 - low-prio set for another subflow - */ - low_prio:1, - saw_add_addr:2, /* Saw at least one add_addr option: * 0x1: IPv4 - 0x2: IPv6 */ @@ -131,7 +122,6 @@ struct mptcp_options_received { mp_fail:1, mp_fclose:1; u8 rem_id; /* Address-id in the MP_JOIN */ - u8 prio_addr_id; /* Address-id in the MP_PRIO */ const unsigned char *add_addr_ptr; /* Pointer to add-address option */ const unsigned char *rem_addr_ptr; /* Pointer to rem-address option */ @@ -169,9 +159,6 @@ struct mptcp_tcp_sock { include_mpc:1, mapping_present:1, map_data_fin:1, - low_prio:1, /* use this socket as backup */ - rcv_low_prio:1, /* Peer sent low-prio option to us */ - send_mp_prio:1, /* Trigger to send mp_prio on this socket */ pre_established:1; /* State between sending 3rd ACK and * receiving the fourth ack of new subflows. */ @@ -220,7 +207,7 @@ struct mptcp_pm_ops { void (*new_remote_address)(struct sock *meta_sk); void (*subflow_error)(struct sock *meta_sk, struct sock *sk); int (*get_local_id)(sa_family_t family, union inet_addr *addr, - struct net *net, bool *low_prio); + struct net *net); void (*addr_signal)(struct sock *sk, unsigned *size, struct tcp_out_options *opts, struct sk_buff *skb); void (*add_raddr)(struct mptcp_cb *mpcb, const union inet_addr *addr, @@ -395,11 +382,6 @@ struct mptcp_cb { #define MPTCP_SUB_REMOVE_ADDR 4 #define MPTCP_SUB_LEN_REMOVE_ADDR 4 -#define MPTCP_SUB_PRIO 5 -#define MPTCP_SUB_LEN_PRIO 3 -#define MPTCP_SUB_LEN_PRIO_ADDR 4 -#define MPTCP_SUB_LEN_PRIO_ALIGN 4 - #define MPTCP_SUB_FAIL 6 #define MPTCP_SUB_LEN_FAIL 12 #define MPTCP_SUB_LEN_FAIL_ALIGN 12 @@ -429,7 +411,6 @@ extern bool mptcp_init_failed; #define OPTION_MP_JOIN (1 << 6) #define OPTION_MP_FCLOSE (1 << 7) #define OPTION_REMOVE_ADDR (1 << 8) -#define OPTION_MP_PRIO (1 << 9) /* MPTCP flags: both TX and RX */ #define MPTCPHDR_SEQ 0x01 /* DSS.M option is present */ @@ -614,23 +595,6 @@ struct mp_fclose { __u64 key; } __attribute__((__packed__)); -struct mp_prio { - __u8 kind; - __u8 len; -#if defined(__LITTLE_ENDIAN_BITFIELD) - __u8 b:1, - rsv:3, - sub:4; -#elif defined(__BIG_ENDIAN_BITFIELD) - __u8 sub:4, - rsv:3, - b:1; -#else -#error "Adjust your <asm/byteorder.h> defines" -#endif - __u8 addr_id; -} __attribute__((__packed__)); - static inline int mptcp_sub_len_dss(const struct mp_dss *m) { return 4 + m->A * (4 + m->a * 4) + m->M * (10 + m->m * 4); @@ -1048,9 +1012,6 @@ static inline void mptcp_init_mp_opt(struct mptcp_options_received *mopt) mopt->is_mp_join = 0; mopt->join_ack = 0; - mopt->saw_low_prio = 0; - mopt->low_prio = 0; - mopt->saw_add_addr = 0; mopt->more_add_addr = 0; @@ -1065,7 +1026,6 @@ static inline void mptcp_reset_mopt(struct tcp_sock *tp) { struct mptcp_options_received *mopt = &tp->mptcp->rx_opt; - mopt->saw_low_prio = 0; mopt->saw_add_addr = 0; mopt->more_add_addr = 0; mopt->saw_rem_addr = 0; diff --git a/include/uapi/linux/if.h b/include/uapi/linux/if.h index 7255e08393db..51540df75bf8 100644 --- a/include/uapi/linux/if.h +++ b/include/uapi/linux/if.h @@ -133,7 +133,6 @@ enum net_device_flags { #endif /* __UAPI_DEF_IF_NET_DEVICE_FLAGS_LOWER_UP_DORMANT_ECHO */ #define IFF_NOMULTIPATH 0x80000 /* Disable for MPTCP */ -#define IFF_MPBACKUP 0x100000 /* Use as backup path for MPTCP */ #define IFF_VOLATILE (IFF_LOOPBACK|IFF_POINTOPOINT|IFF_BROADCAST|IFF_ECHO|\ IFF_MASTER|IFF_SLAVE|IFF_RUNNING|IFF_LOWER_UP|IFF_DORMANT) diff --git a/net/core/dev.c b/net/core/dev.c index 95cff2426ea1..c5cafaac7f58 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -7487,7 +7487,7 @@ int __dev_change_flags(struct net_device *dev, unsigned int flags) dev->flags = (flags & (IFF_DEBUG | IFF_NOTRAILERS | IFF_NOARP | IFF_DYNAMIC | IFF_MULTICAST | IFF_PORTSEL | - IFF_AUTOMEDIA | IFF_NOMULTIPATH | IFF_MPBACKUP)) | + IFF_AUTOMEDIA | IFF_NOMULTIPATH)) | (dev->flags & (IFF_UP | IFF_VOLATILE | IFF_PROMISC | IFF_ALLMULTI)); diff --git a/net/mptcp/mptcp_ctrl.c b/net/mptcp/mptcp_ctrl.c index 8061b934e0d5..927f8c014c0b 100644 --- a/net/mptcp/mptcp_ctrl.c +++ b/net/mptcp/mptcp_ctrl.c @@ -2218,11 +2218,6 @@ struct sock *mptcp_check_req_child(struct sock *meta_sk, goto teardown; } - /* The child is a clone of the meta socket, we must now reset - * some of the fields - */ - child_tp->mptcp->rcv_low_prio = mtreq->rcv_low_prio; - /* We should allow proper increase of the snd/rcv-buffers. Thus, we * use the original values instead of the bloated up ones from the * clone. @@ -2438,7 +2433,6 @@ void mptcp_join_reqsk_init(const struct mptcp_cb *mpcb, mtreq->mptcp_hash_tmac = *(u64 *)mptcp_hash_mac; mtreq->rem_id = mopt.rem_id; - mtreq->rcv_low_prio = mopt.low_prio; inet_rsk(req)->saw_mpc = 1; MPTCP_INC_STATS(sock_net(mpcb->meta_sk), MPTCP_MIB_JOINSYNRX); diff --git a/net/mptcp/mptcp_fullmesh.c b/net/mptcp/mptcp_fullmesh.c index 23f42a06cc70..5043f231bb4c 100644 --- a/net/mptcp/mptcp_fullmesh.c +++ b/net/mptcp/mptcp_fullmesh.c @@ -12,7 +12,6 @@ enum { MPTCP_EVENT_ADD = 1, MPTCP_EVENT_DEL, - MPTCP_EVENT_MOD, }; #define MPTCP_SUBFLOW_RETRY_DELAY 1000 @@ -52,8 +51,7 @@ struct mptcp_loc_addr { struct mptcp_addr_event { struct list_head list; unsigned short family; - u8 code:7, - low_prio:1; + u8 code:7; int if_idx; union inet_addr addr; }; @@ -491,7 +489,6 @@ static void create_subflow_worker(struct work_struct *work) loc.addr.s_addr = inet_sk(meta_sk)->inet_saddr; loc.loc4_id = 0; - loc.low_prio = 0; loc.if_idx = mpcb->master_sk->sk_bound_dev_if; rem.addr.s_addr = inet_sk(meta_sk)->inet_daddr; @@ -541,7 +538,6 @@ static void create_subflow_worker(struct work_struct *work) loc.addr = inet6_sk(meta_sk)->saddr; loc.loc6_id = 0; - loc.low_prio = 0; loc.if_idx = mpcb->master_sk->sk_bound_dev_if; rem.addr = meta_sk->sk_v6_daddr; @@ -763,33 +759,25 @@ static void mptcp_address_worker(struct work_struct *work) &event->addr, event->if_idx); int j = i; - if (j < 0) { - /* Not in the list, so we have to find an empty slot */ - if (event->family == AF_INET) - i = __mptcp_find_free_index(mptcp_local->loc4_bits, - mptcp_local->next_v4_index); - if (event->family == AF_INET6) - i = __mptcp_find_free_index(mptcp_local->loc6_bits, - mptcp_local->next_v6_index); - - if (i < 0) { - mptcp_debug("%s no more space\n", __func__); - goto duno; - } + /* Found the address and it's thus all the same */ + if (j >= 0) + goto duno; - /* It might have been a MOD-event. */ - event->code = MPTCP_EVENT_ADD; - } else { - /* Let's check if anything changes */ - if (event->family == AF_INET && - event->low_prio == mptcp_local->locaddr4[i].low_prio) - goto duno; - - if (event->family == AF_INET6 && - event->low_prio == mptcp_local->locaddr6[i].low_prio) - goto duno; + /* Not in the list, so we have to find an empty slot */ + if (event->family == AF_INET) + i = __mptcp_find_free_index(mptcp_local->loc4_bits, + mptcp_local->next_v4_index); + if (event->family == AF_INET6) + i = __mptcp_find_free_index(mptcp_local->loc6_bits, + mptcp_local->next_v6_index); + + if (i < 0) { + mptcp_debug("%s no more space\n", __func__); + goto duno; } + event->code = MPTCP_EVENT_ADD; + old = mptcp_local; mptcp_local = kmemdup(mptcp_local, sizeof(*mptcp_local), GFP_ATOMIC); @@ -799,21 +787,19 @@ static void mptcp_address_worker(struct work_struct *work) if (event->family == AF_INET) { mptcp_local->locaddr4[i].addr.s_addr = event->addr.in.s_addr; mptcp_local->locaddr4[i].loc4_id = i + 1; - mptcp_local->locaddr4[i].low_prio = event->low_prio; mptcp_local->locaddr4[i].if_idx = event->if_idx; - mptcp_debug("%s updated IP %pI4 on ifidx %u prio %u id %u\n", + mptcp_debug("%s updated IP %pI4 on ifidx %u id %u\n", __func__, &event->addr.in.s_addr, - event->if_idx, event->low_prio, i + 1); + event->if_idx, i + 1); } else { mptcp_local->locaddr6[i].addr = event->addr.in6; mptcp_local->locaddr6[i].loc6_id = i + MPTCP_MAX_ADDR; - mptcp_local->locaddr6[i].low_prio = event->low_prio; mptcp_local->locaddr6[i].if_idx = event->if_idx; - mptcp_debug("%s updated IP %pI6 on ifidx %u prio %u id %u\n", + mptcp_debug("%s updated IP %pI6 on ifidx %u id %u\n", __func__, &event->addr.in6, - event->if_idx, event->low_prio, i + MPTCP_MAX_ADDR); + event->if_idx, i + MPTCP_MAX_ADDR); } if (j < 0) { @@ -965,37 +951,6 @@ static void mptcp_address_worker(struct work_struct *work) announce_remove_addr(loc_id, meta_sk); } } - - if (event->code == MPTCP_EVENT_MOD) { - struct mptcp_tcp_sock *mptcp; - - mptcp_for_each_sub(mpcb, mptcp) { - struct sock *sk = mptcp_to_sock(mptcp); - struct tcp_sock *tp = tcp_sk(sk); - if (event->family == AF_INET && - (sk->sk_family == AF_INET || - mptcp_v6_is_v4_mapped(sk)) && - inet_sk(sk)->inet_saddr == event->addr.in.s_addr) { - if (event->low_prio != tp->mptcp->low_prio) { - tp->mptcp->send_mp_prio = 1; - tp->mptcp->low_prio = event->low_prio; - - tcp_send_ack(sk); - } - } - - if (event->family == AF_INET6 && - sk->sk_family == AF_INET6 && - !ipv6_addr_equal(&inet6_sk(sk)->saddr, &event->addr.in6)) { - if (event->low_prio != tp->mptcp->low_prio) { - tp->mptcp->send_mp_prio = 1; - tp->mptcp->low_prio = event->low_prio; - - tcp_send_ack(sk); - } - } - } - } next: bh_unlock_sock(meta_sk); sock_put(meta_sk); @@ -1042,14 +997,8 @@ static void add_pm_event(struct net *net, const struct mptcp_addr_event *event) break; case MPTCP_EVENT_ADD: mptcp_debug("%s add old_code %u\n", __func__, eventq->code); - eventq->low_prio = event->low_prio; eventq->code = MPTCP_EVENT_ADD; return; - case MPTCP_EVENT_MOD: - mptcp_debug("%s mod old_code %u\n", __func__, eventq->code); - eventq->low_prio = event->low_prio; - eventq->code = MPTCP_EVENT_MOD; - return; } } @@ -1081,7 +1030,6 @@ static void addr4_event_handler(const struct in_ifaddr *ifa, unsigned long event mpevent.family = AF_INET; mpevent.addr.in.s_addr = ifa->ifa_local; - mpevent.low_prio = (netdev->flags & IFF_MPBACKUP) ? 1 : 0; mpevent.if_idx = netdev->ifindex; if (event == NETDEV_DOWN || !netif_running(netdev) || @@ -1089,13 +1037,16 @@ static void addr4_event_handler(const struct in_ifaddr *ifa, unsigned long event mpevent.code = MPTCP_EVENT_DEL; else if (event == NETDEV_UP) mpevent.code = MPTCP_EVENT_ADD; - else if (event == NETDEV_CHANGE) - mpevent.code = MPTCP_EVENT_MOD; + else if (event == NETDEV_CHANGE && !(netdev->flags & IFF_NOMULTIPATH)) + mpevent.code = MPTCP_EVENT_ADD; + else + goto out; - mptcp_debug("%s created event for %pI4, code %u prio %u idx %u\n", __func__, - &ifa->ifa_local, mpevent.code, mpevent.low_prio, mpevent.if_idx); + mptcp_debug("%s created event for %pI4, code %u idx %u\n", __func__, + &ifa->ifa_local, mpevent.code, mpevent.if_idx); add_pm_event(net, &mpevent); +out: spin_unlock_bh(&fm_ns->local_lock); return; } @@ -1143,7 +1094,6 @@ static void addr6_event_handler(const struct inet6_ifaddr *ifa, unsigned long ev mpevent.family = AF_INET6; mpevent.addr.in6 = ifa->addr; - mpevent.low_prio = (netdev->flags & IFF_MPBACKUP) ? 1 : 0; mpevent.if_idx = netdev->ifindex; if (event == NETDEV_DOWN || !netif_running(netdev) || @@ -1152,12 +1102,13 @@ static void addr6_event_handler(const struct inet6_ifaddr *ifa, unsigned long ev else if (event == NETDEV_UP) mpevent.code = MPTCP_EVENT_ADD; else if (event == NETDEV_CHANGE) - mpevent.code = MPTCP_EVENT_MOD; + goto out; - mptcp_debug("%s created event for %pI6, code %u prio %u idx %u\n", __func__, - &ifa->addr, mpevent.code, mpevent.low_prio, mpevent.if_idx); + mptcp_debug("%s created event for %pI6, code %u idx %u\n", __func__, + &ifa->addr, mpevent.code, mpevent.if_idx); add_pm_event(net, &mpevent); +out: spin_unlock_bh(&fm_ns->local_lock); return; } @@ -1241,7 +1192,6 @@ static void full_mesh_new_session(const struct sock *meta_sk) struct mptcp_cb *mpcb = tcp_sk(meta_sk)->mpcb; struct fullmesh_priv *fmp = fullmesh_get_priv(mpcb); const struct mptcp_fm_ns *fm_ns = fm_get_ns(sock_net(meta_sk)); - struct tcp_sock *master_tp = tcp_sk(mpcb->master_sk); int i, index, if_idx = 0; union inet_addr saddr, daddr; sa_family_t family = AF_INET; @@ -1275,12 +1225,6 @@ static void full_mesh_new_session(const struct sock *meta_sk) if (index < 0) goto fallback; - if (family == AF_INET) - master_tp->mptcp->low_prio = mptcp_local->locaddr4[index].low_prio; - else - master_tp->mptcp->low_prio = mptcp_local->locaddr6[index].low_prio; - master_tp->mptcp->send_mp_prio = master_tp->mptcp->low_prio; - full_mesh_add_raddr(mpcb, &daddr, family, 0, 0); mptcp_set_init_addr_bit(mpcb, &daddr, family, index); @@ -1338,9 +1282,6 @@ static void full_mesh_new_session(const struct sock *meta_sk) for (i = fmp->add_addr; i && fmp->add_addr; i--) tcp_send_ack(mpcb->master_sk); - if (master_tp->mptcp->send_mp_prio) - tcp_send_ack(mpcb->master_sk); - return; fallback: @@ -1414,7 +1355,6 @@ static void full_mesh_release_sock(struct sock *meta_sk) mptcp_for_each_sub(mpcb, mptcp) { struct sock *sk = mptcp_to_sock(mptcp); - struct tcp_sock *tp = tcp_sk(sk); if (sk->sk_family == AF_INET6 && !mptcp_v6_is_v4_mapped(sk)) @@ -1424,13 +1364,6 @@ static void full_mesh_release_sock(struct sock *meta_sk) continue; found = true; - - if (mptcp_local->locaddr4[i].low_prio != tp->mptcp->low_prio) { - tp->mptcp->send_mp_prio = 1; - tp->mptcp->low_prio = mptcp_local->locaddr4[i].low_prio; - - tcp_send_ack(sk); - } } if (!found) { @@ -1458,7 +1391,6 @@ static void full_mesh_release_sock(struct sock *meta_sk) mptcp_for_each_sub(mpcb, mptcp) { struct sock *sk = mptcp_to_sock(mptcp); - struct tcp_sock *tp = tcp_sk(sk); if (sk->sk_family == AF_INET || mptcp_v6_is_v4_mapped(sk)) @@ -1468,13 +1400,6 @@ static void full_mesh_release_sock(struct sock *meta_sk) continue; found = true; - - if (mptcp_local->locaddr6[i].low_prio != tp->mptcp->low_prio) { - tp->mptcp->send_mp_prio = 1; - tp->mptcp->low_prio = mptcp_local->locaddr6[i].low_prio; - - tcp_send_ack(sk); - } } if (!found) { @@ -1535,7 +1460,7 @@ static void full_mesh_release_sock(struct sock *meta_sk) } static int full_mesh_get_local_id(sa_family_t family, union inet_addr *addr, - struct net *net, bool *low_prio) + struct net *net) { struct mptcp_loc_addr *mptcp_local; const struct mptcp_fm_ns *fm_ns = fm_get_ns(net); @@ -1550,10 +1475,8 @@ static int full_mesh_get_local_id(sa_family_t family, union inet_addr *addr, if (index != -1) { if (family == AF_INET) { id = mptcp_local->locaddr4[index].loc4_id; - *low_prio = mptcp_local->locaddr4[index].low_prio; } else { id = mptcp_local->locaddr6[index].loc6_id; - *low_prio = mptcp_local->locaddr6[index].low_prio; } } @@ -1778,8 +1701,8 @@ static int mptcp_fm_seq_show(struct seq_file *seq, void *v) mptcp_for_each_bit_set(mptcp_local->loc4_bits, i) { struct mptcp_loc4 *loc4 = &mptcp_local->locaddr4[i]; - seq_printf(seq, "%u, %u, %u, %pI4 %u\n", i, loc4->loc4_id, - loc4->low_prio, &loc4->addr, loc4->if_idx); + seq_printf(seq, "%u, %u, %pI4 %u\n", i, loc4->loc4_id, + &loc4->addr, loc4->if_idx); } seq_printf(seq, "IPv6, next v6-index: %u\n", mptcp_local->next_v6_index); @@ -1787,8 +1710,8 @@ static int mptcp_fm_seq_show(struct seq_file *seq, void *v) mptcp_for_each_bit_set(mptcp_local->loc6_bits, i) { struct mptcp_loc6 *loc6 = &mptcp_local->locaddr6[i]; - seq_printf(seq, "%u, %u, %u, %pI6 %u\n", i, loc6->loc6_id, - loc6->low_prio, &loc6->addr, loc6->if_idx); + seq_printf(seq, "%u, %u, %pI6 %u\n", i, loc6->loc6_id, + &loc6->addr, loc6->if_idx); } rcu_read_unlock_bh(); diff --git a/net/mptcp/mptcp_input.c b/net/mptcp/mptcp_input.c index 9a4d1609b3b9..652da0439644 100644 --- a/net/mptcp/mptcp_input.c +++ b/net/mptcp/mptcp_input.c @@ -1503,14 +1503,12 @@ void mptcp_parse_options(const uint8_t *ptr, int opsize, case MPTCP_SUB_LEN_JOIN_SYN: mopt->is_mp_join = 1; mopt->saw_mpc = 1; - mopt->low_prio = mpjoin->b; mopt->rem_id = mpjoin->addr_id; mopt->mptcp_rem_token = mpjoin->u.syn.token; mopt->mptcp_recv_nonce = mpjoin->u.syn.nonce; break; case MPTCP_SUB_LEN_JOIN_SYNACK: mopt->saw_mpc = 1; - mopt->low_prio = mpjoin->b; mopt->rem_id = mpjoin->addr_id; mopt->mptcp_recv_tmac = mpjoin->u.synack.mac; mopt->mptcp_recv_nonce = mpjoin->u.synack.nonce; @@ -1613,26 +1611,6 @@ void mptcp_parse_options(const uint8_t *ptr, int opsize, mopt->saw_rem_addr = 1; mopt->rem_addr_ptr = ptr; break; - case MPTCP_SUB_PRIO: - { - const struct mp_prio *mpprio = (struct mp_prio *)ptr; - - if (opsize != MPTCP_SUB_LEN_PRIO && - opsize != MPTCP_SUB_LEN_PRIO_ADDR) { - mptcp_debug("%s: mp_prio: bad option size %d\n", - __func__, opsize); - break; - } - - mopt->saw_low_prio = 1; - mopt->low_prio = mpprio->b; - - if (opsize == MPTCP_SUB_LEN_PRIO_ADDR) { - mopt->saw_low_prio = 2; - mopt->prio_addr_id = mpprio->addr_id; - } - break; - } case MPTCP_SUB_FAIL: if (opsize != MPTCP_SUB_LEN_FAIL) { mptcp_debug("%s: mp_fail: bad option size %d\n", @@ -1963,19 +1941,6 @@ bool mptcp_handle_options(struct sock *sk, const struct tcphdr *th, mopt->more_rem_addr = 0; mopt->saw_rem_addr = 0; } - if (mopt->saw_low_prio) { - if (mopt->saw_low_prio == 1) { - tp->mptcp->rcv_low_prio = mopt->low_prio; - } else { - struct mptcp_tcp_sock *mptcp; - - mptcp_for_each_sub(tp->mpcb, mptcp) { - if (mptcp->rem_id == mopt->prio_addr_id) - mptcp->rcv_low_prio = mopt->low_prio; - } - } - mopt->saw_low_prio = 0; - } mptcp_data_ack(sk, skb); @@ -2077,7 +2042,6 @@ int mptcp_rcv_synsent_state_process(struct sock *sk, struct sock **skptr, * until the 4th ack arrives. */ tp->mptcp->pre_established = 1; - tp->mptcp->rcv_low_prio = tp->mptcp->rx_opt.low_prio; mptcp_hmac_sha1((u8 *)&mpcb->mptcp_loc_key, (u8 *)&mpcb->mptcp_rem_key, diff --git a/net/mptcp/mptcp_ipv4.c b/net/mptcp/mptcp_ipv4.c index c808d10205f5..00c54f421ba7 100644 --- a/net/mptcp/mptcp_ipv4.c +++ b/net/mptcp/mptcp_ipv4.c @@ -104,7 +104,6 @@ static int mptcp_v4_join_init_req(struct request_sock *req, const struct sock *m const struct mptcp_cb *mpcb = tcp_sk(meta_sk)->mpcb; union inet_addr addr; int loc_id; - bool low_prio = false; /* We need to do this as early as possible. Because, if we fail later * (e.g., get_local_id), then reqsk_free tries to remove the @@ -120,11 +119,10 @@ static int mptcp_v4_join_init_req(struct request_sock *req, const struct sock *m tcp_hdr(skb)->source, tcp_hdr(skb)->dest); addr.ip = inet_rsk(req)->ir_loc_addr; - loc_id = mpcb->pm_ops->get_local_id(AF_INET, &addr, sock_net(meta_sk), &low_prio); + loc_id = mpcb->pm_ops->get_local_id(AF_INET, &addr, sock_net(meta_sk)); if (loc_id == -1) return -1; mtreq->loc_id = loc_id; - mtreq->low_prio = low_prio; mptcp_join_reqsk_init(mpcb, req, skb); @@ -292,7 +290,6 @@ int mptcp_init4_subsockets(struct sock *meta_sk, const struct mptcp_loc4 *loc, } tp->mptcp->slave_sk = 1; - tp->mptcp->low_prio = loc->low_prio; /* Initializing the timer for an MPTCP subflow */ timer_setup(&tp->mptcp->mptcp_ack_timer, mptcp_ack_handler, 0); diff --git a/net/mptcp/mptcp_ipv6.c b/net/mptcp/mptcp_ipv6.c index 2d730379d5ed..7249d171c9ef 100644 --- a/net/mptcp/mptcp_ipv6.c +++ b/net/mptcp/mptcp_ipv6.c @@ -133,7 +133,6 @@ static int mptcp_v6_join_init_req(struct request_sock *req, const struct sock *m const struct mptcp_cb *mpcb = tcp_sk(meta_sk)->mpcb; union inet_addr addr; int loc_id; - bool low_prio = false; /* We need to do this as early as possible. Because, if we fail later * (e.g., get_local_id), then reqsk_free tries to remove the @@ -149,11 +148,10 @@ static int mptcp_v6_join_init_req(struct request_sock *req, const struct sock *m tcp_hdr(skb)->source, tcp_hdr(skb)->dest); addr.in6 = inet_rsk(req)->ir_v6_loc_addr; - loc_id = mpcb->pm_ops->get_local_id(AF_INET6, &addr, sock_net(meta_sk), &low_prio); + loc_id = mpcb->pm_ops->get_local_id(AF_INET6, &addr, sock_net(meta_sk)); if (loc_id == -1) return -1; mtreq->loc_id = loc_id; - mtreq->low_prio = low_prio; mptcp_join_reqsk_init(mpcb, req, skb); @@ -319,7 +317,6 @@ int mptcp_init6_subsockets(struct sock *meta_sk, const struct mptcp_loc6 *loc, } tp->mptcp->slave_sk = 1; - tp->mptcp->low_prio = loc->low_prio; /* Initializing the timer for an MPTCP subflow */ timer_setup(&tp->mptcp->mptcp_ack_timer, mptcp_ack_handler, 0); diff --git a/net/mptcp/mptcp_ndiffports.c b/net/mptcp/mptcp_ndiffports.c index 58acf50df752..9faef10d8b64 100644 --- a/net/mptcp/mptcp_ndiffports.c +++ b/net/mptcp/mptcp_ndiffports.c @@ -61,7 +61,6 @@ static void create_subflow_worker(struct work_struct *work) loc.addr.s_addr = inet_sk(meta_sk)->inet_saddr; loc.loc4_id = 0; - loc.low_prio = 0; if (mpcb->master_sk) loc.if_idx = mpcb->master_sk->sk_bound_dev_if; else @@ -79,7 +78,6 @@ static void create_subflow_worker(struct work_struct *work) loc.addr = inet6_sk(meta_sk)->saddr; loc.loc6_id = 0; - loc.low_prio = 0; if (mpcb->master_sk) loc.if_idx = mpcb->master_sk->sk_bound_dev_if; else @@ -128,7 +126,7 @@ static void ndiffports_create_subflows(struct sock *meta_sk) } static int ndiffports_get_local_id(sa_family_t family, union inet_addr *addr, - struct net *net, bool *low_prio) + struct net *net) { return 0; } diff --git a/net/mptcp/mptcp_output.c b/net/mptcp/mptcp_output.c index f9bce20259e9..af695fafaab2 100644 --- a/net/mptcp/mptcp_output.c +++ b/net/mptcp/mptcp_output.c @@ -960,7 +960,6 @@ void mptcp_syn_options(const struct sock *sk, struct tcp_out_options *opts, opts->mptcp_options |= OPTION_MP_JOIN | OPTION_TYPE_SYN; *remaining -= MPTCP_SUB_LEN_JOIN_SYN_ALIGN; opts->mp_join_syns.token = mpcb->mptcp_rem_token; - opts->mp_join_syns.low_prio = tp->mptcp->low_prio; opts->addr_id = tp->mptcp->loc_id; opts->mp_join_syns.sender_nonce = tp->mptcp->mptcp_loc_nonce; } @@ -984,7 +983,6 @@ void mptcp_synack_options(struct request_sock *req, opts->mp_join_syns.sender_truncated_mac = mtreq->mptcp_hash_tmac; opts->mp_join_syns.sender_nonce = mtreq->mptcp_loc_nonce; - opts->mp_join_syns.low_prio = mtreq->low_prio; opts->addr_id = mtreq->loc_id; *remaining -= MPTCP_SUB_LEN_JOIN_SYNACK_ALIGN; } @@ -1085,15 +1083,6 @@ void mptcp_established_options(struct sock *sk, struct sk_buff *skb, mpcb->mptcp_ver < MPTCP_VERSION_1) mpcb->pm_ops->addr_signal(sk, size, opts, skb); - if (unlikely(tp->mptcp->send_mp_prio) && - MAX_TCP_OPTION_SPACE - *size >= MPTCP_SUB_LEN_PRIO_ALIGN) { - opts->options |= OPTION_MPTCP; - opts->mptcp_options |= OPTION_MP_PRIO; - if (skb) - tp->mptcp->send_mp_prio = 0; - *size += MPTCP_SUB_LEN_PRIO_ALIGN; - } - return; } @@ -1149,7 +1138,7 @@ void mptcp_options_write(__be32 *ptr, struct tcp_sock *tp, mpj->len = MPTCP_SUB_LEN_JOIN_SYN; mpj->u.syn.token = opts->mp_join_syns.token; mpj->u.syn.nonce = opts->mp_join_syns.sender_nonce; - mpj->b = opts->mp_join_syns.low_prio; + mpj->b = 0; mpj->addr_id = opts->addr_id; ptr += MPTCP_SUB_LEN_JOIN_SYN_ALIGN >> 2; } else if (OPTION_TYPE_SYNACK & opts->mptcp_options) { @@ -1157,7 +1146,7 @@ void mptcp_options_write(__be32 *ptr, struct tcp_sock *tp, mpj->u.synack.mac = opts->mp_join_syns.sender_truncated_mac; mpj->u.synack.nonce = opts->mp_join_syns.sender_nonce; - mpj->b = opts->mp_join_syns.low_prio; + mpj->b = 0; mpj->addr_id = opts->addr_id; ptr += MPTCP_SUB_LEN_JOIN_SYNACK_ALIGN >> 2; } else if (OPTION_TYPE_ACK & opts->mptcp_options) { @@ -1252,18 +1241,6 @@ void mptcp_options_write(__be32 *ptr, struct tcp_sock *tp, else ptr += mptcp_write_dss_data_seq(tp, skb, ptr); } - if (unlikely(OPTION_MP_PRIO & opts->mptcp_options)) { - struct mp_prio *mpprio = (struct mp_prio *)ptr; - - mpprio->kind = TCPOPT_MPTCP; - mpprio->len = MPTCP_SUB_LEN_PRIO; - mpprio->sub = MPTCP_SUB_PRIO; - mpprio->rsv = 0; - mpprio->b = tp->mptcp->low_prio; - mpprio->addr_id = TCPOPT_NOP; - - ptr += MPTCP_SUB_LEN_PRIO_ALIGN >> 2; - } } /* Sends the datafin */ diff --git a/net/mptcp/mptcp_pm.c b/net/mptcp/mptcp_pm.c index 08e3f05a9e65..fbdf8eb58d17 100644 --- a/net/mptcp/mptcp_pm.c +++ b/net/mptcp/mptcp_pm.c @@ -35,7 +35,7 @@ static DEFINE_SPINLOCK(mptcp_pm_list_lock); static LIST_HEAD(mptcp_pm_list); static int mptcp_default_id(sa_family_t family, union inet_addr *addr, - struct net *net, bool *low_prio) + struct net *net) { return 0; } diff --git a/net/mptcp/mptcp_sched.c b/net/mptcp/mptcp_sched.c index 3662bdf97c0d..82ffe080cb5e 100644 --- a/net/mptcp/mptcp_sched.c +++ b/net/mptcp/mptcp_sched.c @@ -113,22 +113,11 @@ static int mptcp_dont_reinject_skb(const struct tcp_sock *tp, const struct sk_bu mptcp_pi_to_flag(tp->mptcp->path_index) & TCP_SKB_CB(skb)->path_mask; } -static bool subflow_is_backup(const struct tcp_sock *tp) -{ - return tp->mptcp->rcv_low_prio || tp->mptcp->low_prio; -} - -static bool subflow_is_active(const struct tcp_sock *tp) -{ - return !tp->mptcp->rcv_low_prio && !tp->mptcp->low_prio; -} - /* Generic function to iterate over used and unused subflows and to select the * best one */ static struct sock *get_subflow_from_selectors(struct mptcp_cb *mpcb, struct sk_buff *skb, - bool (*selector)(const struct tcp_sock *), bool zero_wnd_test, bool *force) { struct sock *bestsk = NULL; @@ -142,10 +131,6 @@ static struct sock struct tcp_sock *tp = tcp_sk(sk); bool unused = false; - /* First, we choose only the wanted sks */ - if (!(*selector)(tp)) - continue; - if (!mptcp_dont_reinject_skb(tp, skb)) unused = true; else if (found_unused) @@ -233,16 +218,13 @@ static struct sock *get_available_subflow(struct sock *meta_sk, /* Find the best subflow */ restart: - sk = get_subflow_from_selectors(mpcb, skb, &subflow_is_active, - zero_wnd_test, &force); + sk = get_subflow_from_selectors(mpcb, skb, zero_wnd_test, &force); if (force) /* one unused active sk or one NULL sk when there is at least * one temporally unavailable unused active sk */ return sk; - sk = get_subflow_from_selectors(mpcb, skb, &subflow_is_backup, - zero_wnd_test, &force); if (!force && skb) { /* one used backup sk or one NULL sk where there is no one * temporally unavailable unused backup sk -- 2.16.2

It's an "optimisation" that we don't need in net-next. Signed-off-by: Christoph Paasch <cpaasch(a)apple.com&gt; --- include/linux/tcp.h | 3 - include/net/mptcp.h | 9 -- net/mptcp/mptcp_ctrl.c | 10 +- net/mptcp/mptcp_fullmesh.c | 12 -- net/mptcp/mptcp_input.c | 20 +-- net/mptcp/mptcp_output.c | 358 +-------------------------------------------- net/mptcp/mptcp_sched.c | 32 ++-- 7 files changed, 21 insertions(+), 423 deletions(-) diff --git a/include/linux/tcp.h b/include/linux/tcp.h index cc896f842c31..6dd9c45c52d6 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -470,9 +470,6 @@ struct tcp_sock { request_mptcp:1, /* Did we send out an MP_CAPABLE? * (this speeds up mptcp_doit() in tcp_recvmsg) */ - pf:1, /* Potentially Failed state: when this flag is set, we - * stop using the subflow - */ mp_killed:1, /* Killed with a tcp_done in mptcp? */ was_meta_sk:1, /* This was a meta sk (in case of reuse) */ is_master_sk:1, diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 914ab2fb3c89..721515ddb252 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -265,12 +265,6 @@ struct mptcp_cb { u8 mptcp_sched[MPTCP_SCHED_DATA_SIZE] __aligned(8); struct mptcp_sched_ops *sched_ops; - struct sk_buff_head reinject_queue; - /* First cache-line boundary is here minus 8 bytes. But from the - * reinject-queue only the next and prev pointers are regularly - * accessed. Thus, the whole data-path is on a single cache-line. - */ - u64 infinite_rcv_seq; /***** Start of fields, used for connection closure */ @@ -422,7 +416,6 @@ extern bool mptcp_init_failed; #define MPTCPHDR_SEQ64_OFO 0x20 /* Is it not in our circular array? */ /* MPTCP flags: TX only */ #define MPTCPHDR_INF 0x08 -#define MPTCP_REINJECT 0x10 /* Did we reinject this segment? */ struct mptcp_option { __u8 kind; @@ -720,7 +713,6 @@ int mptcp_add_sock(struct sock *meta_sk, struct sock *sk, u8 loc_id, u8 rem_id, gfp_t flags); void mptcp_del_sock(struct sock *sk); void mptcp_update_metasocket(const struct sock *meta_sk); -void mptcp_reinject_data(struct sock *orig_sk, int clone_it); void mptcp_update_sndbuf(const struct tcp_sock *tp); void mptcp_send_fin(struct sock *meta_sk); void mptcp_send_active_reset(struct sock *meta_sk, gfp_t priority); @@ -1257,7 +1249,6 @@ static inline int is_master_tp(const struct tcp_sock *tp) } static inline void mptcp_del_sock(const struct sock *sk) {} static inline void mptcp_update_metasocket(const struct sock *meta_sk) {} -static inline void mptcp_reinject_data(struct sock *orig_sk, int clone_it) {} static inline void mptcp_update_sndbuf(const struct tcp_sock *tp) {} static inline void mptcp_clean_rtx_infinite(const struct sk_buff *skb, const struct sock *sk) {} diff --git a/net/mptcp/mptcp_ctrl.c b/net/mptcp/mptcp_ctrl.c index 927f8c014c0b..910268a8ee8e 100644 --- a/net/mptcp/mptcp_ctrl.c +++ b/net/mptcp/mptcp_ctrl.c @@ -562,7 +562,7 @@ struct sock *mptcp_select_ack_sock(const struct sock *meta_sk) struct tcp_sock *tp = tcp_sk(sk); u32 elapsed; - if (!mptcp_sk_can_send_ack(sk) || tp->pf) + if (!mptcp_sk_can_send_ack(sk) || inet_csk(sk)->icsk_ca_state == TCP_CA_Loss) continue; elapsed = keepalive_time_elapsed(tp); @@ -694,8 +694,6 @@ void mptcp_destroy_sock(struct sock *sk) struct mptcp_tcp_sock *mptcp; struct hlist_node *tmp; - __skb_queue_purge(&tcp_sk(sk)->mpcb->reinject_queue); - /* We have to close all remaining subflows. Normally, they * should all be about to get closed. But, if the kernel is * forcing a closure (e.g., tcp_write_err), the subflows might @@ -1231,7 +1229,6 @@ static int mptcp_alloc_mpcb(struct sock *meta_sk, __u64 remote_key, meta_tp->was_meta_sk = 0; /* Initialize the queues */ - skb_queue_head_init(&mpcb->reinject_queue); master_tp->out_of_order_queue = RB_ROOT; master_sk->tcp_rtx_queue = RB_ROOT; INIT_LIST_HEAD(&master_tp->tsq_node); @@ -1462,9 +1459,6 @@ void mptcp_del_sock(struct sock *sk) tp->mptcp->attached = 0; mpcb->path_index_bits &= ~(1 << tp->mptcp->path_index); - if (!tcp_write_queue_empty(sk) || !tcp_rtx_queue_empty(sk)) - mptcp_reinject_data(sk, 0); - if (is_master_tp(tp)) { struct sock *meta_sk = mptcp_meta_sk(sk); struct tcp_sock *meta_tp = tcp_sk(meta_sk); @@ -1922,8 +1916,6 @@ void mptcp_disconnect(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct hlist_node *tmp; - __skb_queue_purge(&tp->mpcb->reinject_queue); - if (tp->inside_tk_table) mptcp_hash_remove_bh(tp); diff --git a/net/mptcp/mptcp_fullmesh.c b/net/mptcp/mptcp_fullmesh.c index 5043f231bb4c..fcfe6e6bd87b 100644 --- a/net/mptcp/mptcp_fullmesh.c +++ b/net/mptcp/mptcp_fullmesh.c @@ -923,12 +923,6 @@ static void mptcp_address_worker(struct work_struct *work) !ipv6_addr_equal(&inet6_sk(sk)->saddr, &event->addr.in6)) continue; - /* Reinject, so that pf = 1 and so we - * won't select this one as the - * ack-sock. - */ - mptcp_reinject_data(sk, 0); - /* We announce the removal of this id */ announce_remove_addr(tcp_sk(sk)->mptcp->loc_id, meta_sk); @@ -1440,12 +1434,6 @@ static void full_mesh_release_sock(struct sock *meta_sk) } if (shall_remove) { - /* Reinject, so that pf = 1 and so we - * won't select this one as the - * ack-sock. - */ - mptcp_reinject_data(sk, 0); - announce_remove_addr(tcp_sk(sk)->mptcp->loc_id, meta_sk); diff --git a/net/mptcp/mptcp_input.c b/net/mptcp/mptcp_input.c index 652da0439644..f247e9ef2853 100644 --- a/net/mptcp/mptcp_input.c +++ b/net/mptcp/mptcp_input.c @@ -92,7 +92,7 @@ static inline int mptcp_tso_acked_reinject(const struct sock *meta_sk, /* Cleans the meta-socket retransmission queue and the reinject-queue. */ static void mptcp_clean_rtx_queue(struct sock *meta_sk, u32 prior_snd_una) { - struct sk_buff *skb, *tmp, *next; + struct sk_buff *skb, *next; struct tcp_sock *meta_tp = tcp_sk(meta_sk); struct mptcp_cb *mpcb = meta_tp->mpcb; bool acked = false; @@ -149,20 +149,6 @@ static void mptcp_clean_rtx_queue(struct sock *meta_sk, u32 prior_snd_una) } sk_wmem_free_skb(meta_sk, skb); } - /* Remove acknowledged data from the reinject queue */ - skb_queue_walk_safe(&mpcb->reinject_queue, skb, tmp) { - if (before(meta_tp->snd_una, TCP_SKB_CB(skb)->end_seq)) { - if (tcp_skb_pcount(skb) == 1 || - !after(meta_tp->snd_una, TCP_SKB_CB(skb)->seq)) - break; - - mptcp_tso_acked_reinject(meta_sk, skb); - break; - } - - __skb_unlink(skb, &mpcb->reinject_queue); - __kfree_skb(skb); - } if (likely(between(meta_tp->snd_up, prior_snd_una, meta_tp->snd_una))) meta_tp->snd_up = meta_tp->snd_una; @@ -1242,9 +1228,6 @@ static void mptcp_data_ack(struct sock *sk, const struct sk_buff *skb) u32 nwin, data_ack, data_seq; u16 data_len = 0; - /* A valid packet came in - subflow is operational again */ - tp->pf = 0; - /* Even if there is no data-ack, we stop retransmitting. * Except if this is a SYN/ACK. Then it is just a retransmission */ @@ -1400,7 +1383,6 @@ static void mptcp_send_reset_rem_id(const struct mptcp_cb *mpcb, u8 rem_id) struct sock *sk_it = mptcp_to_sock(mptcp); if (tcp_sk(sk_it)->mptcp->rem_id == rem_id) { - mptcp_reinject_data(sk_it, 0); mptcp_send_reset(sk_it); } } diff --git a/net/mptcp/mptcp_output.c b/net/mptcp/mptcp_output.c index af695fafaab2..26cf41f75922 100644 --- a/net/mptcp/mptcp_output.c +++ b/net/mptcp/mptcp_output.c @@ -54,309 +54,6 @@ int mptcp_sub_len_remove_addr_align(u16 bitfield) } EXPORT_SYMBOL(mptcp_sub_len_remove_addr_align); -/* get the data-seq and end-data-seq and store them again in the - * tcp_skb_cb - */ -static bool mptcp_reconstruct_mapping(struct sk_buff *skb) -{ - const struct mp_dss *mpdss = (struct mp_dss *)TCP_SKB_CB(skb)->dss; - u32 *p32; - u16 *p16; - - if (!mptcp_is_data_seq(skb)) - return false; - - if (!mpdss->M) - return false; - - /* Move the pointer to the data-seq */ - p32 = (u32 *)mpdss; - p32++; - if (mpdss->A) { - p32++; - if (mpdss->a) - p32++; - } - - TCP_SKB_CB(skb)->seq = ntohl(*p32); - - /* Get the data_len to calculate the end_data_seq */ - p32++; - p32++; - p16 = (u16 *)p32; - TCP_SKB_CB(skb)->end_seq = ntohs(*p16) + TCP_SKB_CB(skb)->seq; - - return true; -} - -static bool mptcp_is_reinjected(const struct sk_buff *skb) -{ - return TCP_SKB_CB(skb)->mptcp_flags & MPTCP_REINJECT; -} - -static void mptcp_find_and_set_pathmask(struct sock *meta_sk, struct sk_buff *skb) -{ - struct rb_node **p = &meta_sk->tcp_rtx_queue.rb_node; - struct rb_node *parent; - struct sk_buff *skb_it; - - while (*p) { - parent = *p; - skb_it = rb_to_skb(parent); - if (before(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb_it)->seq)) { - p = &parent->rb_left; - continue; - } - if (after(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb_it)->seq)) { - p = &parent->rb_right; - continue; - } - - TCP_SKB_CB(skb)->path_mask = TCP_SKB_CB(skb_it)->path_mask; - break; - } -} - -/* Reinject data from one TCP subflow to the meta_sk. If sk == NULL, we are - * coming from the meta-retransmit-timer - */ -static void __mptcp_reinject_data(struct sk_buff *orig_skb, struct sock *meta_sk, - struct sock *sk, int clone_it, - enum tcp_queue tcp_queue) -{ - struct sk_buff *skb, *skb1; - const struct tcp_sock *meta_tp = tcp_sk(meta_sk); - struct mptcp_cb *mpcb = meta_tp->mpcb; - u32 seq, end_seq; - - if (clone_it) { - /* pskb_copy is necessary here, because the TCP/IP-headers - * will be changed when it's going to be reinjected on another - * subflow. - */ - tcp_skb_tsorted_save(orig_skb) { - skb = pskb_copy_for_clone(orig_skb, GFP_ATOMIC); - } tcp_skb_tsorted_restore(orig_skb); - } else { - if (tcp_queue == TCP_FRAG_IN_WRITE_QUEUE) - __skb_unlink(orig_skb, &sk->sk_write_queue); - else - rb_erase(&orig_skb->rbnode, &sk->tcp_rtx_queue); - sock_set_flag(sk, SOCK_QUEUE_SHRUNK); - sk->sk_wmem_queued -= orig_skb->truesize; - sk_mem_uncharge(sk, orig_skb->truesize); - skb = orig_skb; - } - if (unlikely(!skb)) - return; - - if (sk && !mptcp_reconstruct_mapping(skb)) { - __kfree_skb(skb); - return; - } - - skb->sk = meta_sk; - - /* Make sure that this list is clean */ - tcp_skb_tsorted_anchor_cleanup(skb); - - /* Reset subflow-specific TCP control-data */ - TCP_SKB_CB(skb)->sacked = 0; - TCP_SKB_CB(skb)->tcp_flags &= (TCPHDR_ACK | TCPHDR_PSH); - - /* If it reached already the destination, we don't have to reinject it */ - if (!after(TCP_SKB_CB(skb)->end_seq, meta_tp->snd_una)) { - __kfree_skb(skb); - return; - } - - /* Only reinject segments that are fully covered by the mapping */ - if (skb->len + (mptcp_is_data_fin(skb) ? 1 : 0) != - TCP_SKB_CB(skb)->end_seq - TCP_SKB_CB(skb)->seq) { - struct rb_node *parent, **p = &meta_sk->tcp_rtx_queue.rb_node; - u32 end_seq = TCP_SKB_CB(skb)->end_seq; - u32 seq = TCP_SKB_CB(skb)->seq; - - __kfree_skb(skb); - - /* Ok, now we have to look for the full mapping in the meta - * send-queue :S - */ - - /* First, find the first skb that covers us */ - while (*p) { - parent = *p; - skb = rb_to_skb(parent); - - /* Not yet at the mapping? */ - if (!after(end_seq, TCP_SKB_CB(skb)->seq)) { - p = &parent->rb_left; - continue; - } - - if (!before(seq, TCP_SKB_CB(skb)->end_seq)) { - p = &parent->rb_right; - continue; - } - - break; - } - - if (*p) { - /* We found it, now let's reinject everything */ - skb = rb_to_skb(*p); - - skb_rbtree_walk_from(skb) { - if (after(TCP_SKB_CB(skb)->end_seq, end_seq)) - return; - __mptcp_reinject_data(skb, meta_sk, NULL, 1, - TCP_FRAG_IN_RTX_QUEUE); - } - } - return; - } - - /* Segment goes back to the MPTCP-layer. So, we need to zero the - * path_mask/dss. - */ - memset(TCP_SKB_CB(skb)->dss, 0 , mptcp_dss_len); - - /* We need to find out the path-mask from the meta-write-queue - * to properly select a subflow. - */ - mptcp_find_and_set_pathmask(meta_sk, skb); - - /* If it's empty, just add */ - if (skb_queue_empty(&mpcb->reinject_queue)) { - skb_queue_head(&mpcb->reinject_queue, skb); - return; - } - - /* Find place to insert skb - or even we can 'drop' it, as the - * data is already covered by other skb's in the reinject-queue. - * - * This is inspired by code from tcp_data_queue. - */ - - skb1 = skb_peek_tail(&mpcb->reinject_queue); - seq = TCP_SKB_CB(skb)->seq; - while (1) { - if (!after(TCP_SKB_CB(skb1)->seq, seq)) - break; - if (skb_queue_is_first(&mpcb->reinject_queue, skb1)) { - skb1 = NULL; - break; - } - skb1 = skb_queue_prev(&mpcb->reinject_queue, skb1); - } - - /* Do skb overlap to previous one? */ - end_seq = TCP_SKB_CB(skb)->end_seq; - if (skb1 && before(seq, TCP_SKB_CB(skb1)->end_seq)) { - if (!after(end_seq, TCP_SKB_CB(skb1)->end_seq)) { - /* All the bits are present. Don't reinject */ - __kfree_skb(skb); - return; - } - if (seq == TCP_SKB_CB(skb1)->seq) { - if (skb_queue_is_first(&mpcb->reinject_queue, skb1)) - skb1 = NULL; - else - skb1 = skb_queue_prev(&mpcb->reinject_queue, skb1); - } - } - if (!skb1) - __skb_queue_head(&mpcb->reinject_queue, skb); - else - __skb_queue_after(&mpcb->reinject_queue, skb1, skb); - - /* And clean segments covered by new one as whole. */ - while (!skb_queue_is_last(&mpcb->reinject_queue, skb)) { - skb1 = skb_queue_next(&mpcb->reinject_queue, skb); - - if (!after(end_seq, TCP_SKB_CB(skb1)->seq)) - break; - - if (before(end_seq, TCP_SKB_CB(skb1)->end_seq)) - break; - - __skb_unlink(skb1, &mpcb->reinject_queue); - __kfree_skb(skb1); - } - return; -} - -/* Inserts data into the reinject queue */ -void mptcp_reinject_data(struct sock *sk, int clone_it) -{ - struct sock *meta_sk = mptcp_meta_sk(sk); - struct sk_buff *skb_it, *tmp; - enum tcp_queue tcp_queue; - - /* It has already been closed - there is really no point in reinjecting */ - if (meta_sk->sk_state == TCP_CLOSE) - return; - - skb_queue_walk_safe(&sk->sk_write_queue, skb_it, tmp) { - struct tcp_skb_cb *tcb = TCP_SKB_CB(skb_it); - /* Subflow syn's and fin's are not reinjected. - * - * As well as empty subflow-fins with a data-fin. - * They are reinjected below (without the subflow-fin-flag) - */ - if (tcb->tcp_flags & TCPHDR_SYN || - (tcb->tcp_flags & TCPHDR_FIN && !mptcp_is_data_fin(skb_it)) || - (tcb->tcp_flags & TCPHDR_FIN && mptcp_is_data_fin(skb_it) && !skb_it->len)) - continue; - - if (mptcp_is_reinjected(skb_it)) - continue; - - tcb->mptcp_flags |= MPTCP_REINJECT; - __mptcp_reinject_data(skb_it, meta_sk, sk, clone_it, - TCP_FRAG_IN_WRITE_QUEUE); - } - - skb_rbtree_walk_safe(skb_it, &sk->tcp_rtx_queue, tmp) { - struct tcp_skb_cb *tcb = TCP_SKB_CB(skb_it); - /* Subflow syn's and fin's are not reinjected. - * - * As well as empty subflow-fins with a data-fin. - * They are reinjected below (without the subflow-fin-flag) - */ - if (tcb->tcp_flags & TCPHDR_SYN || - (tcb->tcp_flags & TCPHDR_FIN && !mptcp_is_data_fin(skb_it)) || - (tcb->tcp_flags & TCPHDR_FIN && mptcp_is_data_fin(skb_it) && !skb_it->len)) - continue; - - if (mptcp_is_reinjected(skb_it)) - continue; - - tcb->mptcp_flags |= MPTCP_REINJECT; - __mptcp_reinject_data(skb_it, meta_sk, sk, clone_it, - TCP_FRAG_IN_RTX_QUEUE); - } - - skb_it = tcp_write_queue_tail(meta_sk); - tcp_queue = TCP_FRAG_IN_WRITE_QUEUE; - - if (!skb_it) { - skb_it = skb_rb_last(&meta_sk->tcp_rtx_queue); - tcp_queue = TCP_FRAG_IN_RTX_QUEUE; - } - - /* If sk has sent the empty data-fin, we have to reinject it too. */ - if (skb_it && mptcp_is_data_fin(skb_it) && skb_it->len == 0 && - TCP_SKB_CB(skb_it)->path_mask & mptcp_pi_to_flag(tcp_sk(sk)->mptcp->path_index)) { - __mptcp_reinject_data(skb_it, meta_sk, NULL, 1, tcp_queue); - } - - tcp_sk(sk)->pf = 1; - - mptcp_push_pending_frames(meta_sk); -} -EXPORT_SYMBOL(mptcp_reinject_data); - static void mptcp_combine_dfin(const struct sk_buff *skb, const struct sock *meta_sk, struct sock *subsk) @@ -560,8 +257,7 @@ static bool mptcp_skb_entail(struct sock *sk, struct sk_buff *skb, int reinject) * - The reinject-queue (reinject == -1) */ static int mptcp_fragment(struct sock *meta_sk, enum tcp_queue tcp_queue, - struct sk_buff *skb, u32 len, - gfp_t gfp, int reinject) + struct sk_buff *skb, u32 len, gfp_t gfp) { int ret, diff, old_factor; struct sk_buff *buff; @@ -592,34 +288,6 @@ static int mptcp_fragment(struct sock *meta_sk, enum tcp_queue tcp_queue, TCP_SKB_CB(buff)->mptcp_flags = flags; TCP_SKB_CB(buff)->path_mask = TCP_SKB_CB(skb)->path_mask; - /* If reinject == 1, the buff will be added to the reinject - * queue, which is currently not part of memory accounting. So - * undo the changes done by tcp_fragment and update the - * reinject queue. Also, undo changes to the packet counters. - */ - if (reinject == 1) { - int undo = buff->truesize - diff; - meta_sk->sk_wmem_queued -= undo; - sk_mem_uncharge(meta_sk, undo); - - tcp_sk(meta_sk)->mpcb->reinject_queue.qlen++; - if (tcp_queue == TCP_FRAG_IN_WRITE_QUEUE) - meta_sk->sk_write_queue.qlen--; - - if (!before(tcp_sk(meta_sk)->snd_nxt, TCP_SKB_CB(buff)->end_seq)) { - undo = old_factor - tcp_skb_pcount(skb) - - tcp_skb_pcount(buff); - if (undo) - tcp_adjust_pcount(meta_sk, skb, -undo); - } - - /* tcp_fragment's call to sk_stream_alloc_skb initializes the - * tcp_tsorted_anchor. We need to revert this as it clashes - * with the refdst pointer. - */ - tcp_skb_tsorted_anchor_cleanup(buff); - } - return 0; } @@ -663,7 +331,7 @@ int mptcp_write_wakeup(struct sock *meta_sk, int mib) seg_size = min(seg_size, mss); TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_PSH; if (mptcp_fragment(meta_sk, TCP_FRAG_IN_WRITE_QUEUE, - skb, seg_size, GFP_ATOMIC, 0)) + skb, seg_size, GFP_ATOMIC)) return -1; } else if (!tcp_skb_pcount(skb)) { /* see mptcp_write_xmit on why we use UINT_MAX */ @@ -738,14 +406,7 @@ bool mptcp_write_xmit(struct sock *meta_sk, unsigned int mss_now, int nonagle, subtp = tcp_sk(subsk); mss_now = tcp_current_mss(subsk); - if (reinject == 1) { - if (!after(TCP_SKB_CB(skb)->end_seq, meta_tp->snd_una)) { - /* Segment already reached the peer, take the next one */ - __skb_unlink(skb, &mpcb->reinject_queue); - __kfree_skb(skb); - continue; - } - } else if (reinject == -1) { + if (reinject == -1) { tcp_queue = TCP_FRAG_IN_RTX_QUEUE; } @@ -810,7 +471,7 @@ bool mptcp_write_xmit(struct sock *meta_sk, unsigned int mss_now, int nonagle, if (skb->len > limit && unlikely(mptcp_fragment(meta_sk, tcp_queue, - skb, limit, gfp, reinject))) + skb, limit, gfp))) break; if (!mptcp_skb_entail(subsk, skb, reinject)) @@ -833,11 +494,6 @@ bool mptcp_write_xmit(struct sock *meta_sk, unsigned int mss_now, int nonagle, tcp_minshall_update(meta_tp, mss_now, skb); - if (reinject > 0) { - __skb_unlink(skb, &mpcb->reinject_queue); - kfree_skb(skb); - } - if (push_one) break; } @@ -1488,7 +1144,7 @@ int mptcp_retransmit_skb(struct sock *meta_sk, struct sk_buff *skb) if (skb->len > limit && unlikely(mptcp_fragment(meta_sk, TCP_FRAG_IN_RTX_QUEUE, skb, - limit, GFP_ATOMIC, 0))) + limit, GFP_ATOMIC))) goto failed; if (!mptcp_skb_entail(subsk, skb, -1)) @@ -1640,10 +1296,8 @@ void mptcp_sub_retransmit_timer(struct sock *sk) tcp_retransmit_timer(sk); - if (!tp->fastopen_rsk) { - mptcp_reinject_data(sk, 1); + if (!tp->fastopen_rsk) mptcp_set_rto(sk); - } } /* Modify values to an mptcp-level for the initial window of new subflows */ diff --git a/net/mptcp/mptcp_sched.c b/net/mptcp/mptcp_sched.c index 82ffe080cb5e..74d167e4689b 100644 --- a/net/mptcp/mptcp_sched.c +++ b/net/mptcp/mptcp_sched.c @@ -29,7 +29,7 @@ static bool mptcp_is_def_unavailable(struct sock *sk) if (tp->mptcp->pre_established) return true; - if (tp->pf) + if (inet_csk(sk)->icsk_ca_state == TCP_CA_Loss) return true; return false; @@ -338,25 +338,19 @@ static struct sk_buff *__mptcp_next_segment(struct sock *meta_sk, int *reinject) if (mpcb->infinite_mapping_snd || mpcb->send_infinite_mapping) return tcp_send_head(meta_sk); - skb = skb_peek(&mpcb->reinject_queue); + skb = tcp_send_head(meta_sk); - if (skb) { - *reinject = 1; - } else { - skb = tcp_send_head(meta_sk); - - if (!skb && meta_sk->sk_socket && - test_bit(SOCK_NOSPACE, &meta_sk->sk_socket->flags) && - sk_stream_wspace(meta_sk) < sk_stream_min_wspace(meta_sk)) { - struct sock *subsk = get_available_subflow(meta_sk, NULL, - false); - if (!subsk) - return NULL; - - skb = mptcp_rcv_buf_optimization(subsk, 0); - if (skb) - *reinject = -1; - } + if (!skb && meta_sk->sk_socket && + test_bit(SOCK_NOSPACE, &meta_sk->sk_socket->flags) && + sk_stream_wspace(meta_sk) < sk_stream_min_wspace(meta_sk)) { + struct sock *subsk = get_available_subflow(meta_sk, NULL, + false); + if (!subsk) + return NULL; + + skb = mptcp_rcv_buf_optimization(subsk, 0); + if (skb) + *reinject = -1; } return skb; } -- 2.16.2

Signed-off-by: Christoph Paasch <cpaasch(a)apple.com&gt; --- include/linux/tcp.h | 2 +- include/net/mptcp.h | 9 +----- net/ipv4/tcp.c | 7 ++--- net/ipv4/tcp_fastopen.c | 16 +++-------- net/ipv4/tcp_input.c | 23 ++-------------- net/mptcp/mptcp_ctrl.c | 53 ------------------------------------ net/mptcp/mptcp_input.c | 71 ------------------------------------------------ net/mptcp/mptcp_output.c | 5 +--- 8 files changed, 13 insertions(+), 173 deletions(-) diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 6dd9c45c52d6..01321044047c 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -58,7 +58,7 @@ static inline unsigned int tcp_optlen(const struct sk_buff *skb) /* TCP Fast Open */ #define TCP_FASTOPEN_COOKIE_MIN 4 /* Min Fast Open Cookie size in bytes */ #define TCP_FASTOPEN_COOKIE_MAX 16 /* Max Fast Open Cookie size in bytes */ -#define TCP_FASTOPEN_COOKIE_SIZE 4 /* the size employed by this impl. */ +#define TCP_FASTOPEN_COOKIE_SIZE 8 /* the size employed by this impl. */ /* TCP Fast Open Cookie as stored in memory */ struct tcp_fastopen_cookie { diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 721515ddb252..e1e7619a85ca 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -738,7 +738,6 @@ void mptcp_close(struct sock *meta_sk, long timeout); bool mptcp_doit(struct sock *sk); int mptcp_create_master_sk(struct sock *meta_sk, __u64 remote_key, __u8 mptcp_ver, u32 window); -int mptcp_check_req_fastopen(struct sock *child, struct request_sock *req); int mptcp_check_req_master(struct sock *sk, struct sock *child, struct request_sock *req, const struct sk_buff *skb, int drop); @@ -1054,8 +1053,7 @@ static inline void mptcp_check_rcvseq_wrap(struct tcp_sock *meta_tp, static inline int mptcp_sk_can_send(const struct sock *sk) { - return tcp_passive_fastopen(sk) || - ((1 << sk->sk_state) & (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT) && + return ((1 << sk->sk_state) & (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT) && !tcp_sk(sk)->mptcp->pre_established); } @@ -1278,11 +1276,6 @@ static inline bool mptcp_doit(struct sock *sk) { return false; } -static inline int mptcp_check_req_fastopen(struct sock *child, - struct request_sock *req) -{ - return 1; -} static inline int mptcp_check_req_master(const struct sock *sk, const struct sock *child, const struct request_sock *req, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 87f7616aaf93..b2dac73903be 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1233,7 +1233,7 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size) } if (unlikely(flags & MSG_FASTOPEN || inet_sk(sk)->defer_connect) && - !tp->repair) { + !tp->repair && !sock_flag(sk, SOCK_MPTCP)) { err = tcp_sendmsg_fastopen(sk, msg, &copied_syn, size); if (err == -EINPROGRESS && copied_syn > 0) goto out; @@ -1250,8 +1250,7 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size) * is fully established. */ if (((1 << sk->sk_state) & ~(TCPF_ESTABLISHED | TCPF_CLOSE_WAIT)) && - !tcp_passive_fastopen(mptcp(tp) && tp->mpcb->master_sk ? - tp->mpcb->master_sk : sk)) { + !tcp_passive_fastopen(sk)) { err = sk_stream_wait_connect(sk, &timeo); if (err != 0) goto do_error; @@ -3098,7 +3097,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, case TCP_FASTOPEN: if (val >= 0 && ((1 << sk->sk_state) & (TCPF_CLOSE | - TCPF_LISTEN))) { + TCPF_LISTEN)) && !sock_flag(sk, SOCK_MPTCP)) { tcp_fastopen_init_key_once(net); fastopen_queue_tune(sk, val); diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 370fe7d4b579..018a48477355 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -9,7 +9,6 @@ #include <linux/rculist.h> #include <net/inetpeer.h> #include <net/tcp.h> -#include <net/mptcp.h> void tcp_fastopen_init_key_once(struct net *net) { @@ -219,7 +218,7 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk, { struct tcp_sock *tp; struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue; - struct sock *child, *meta_sk; + struct sock *child; bool own_req; req->num_retrans = 0; @@ -259,22 +258,15 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk, refcount_set(&req->rsk_refcnt, 2); + /* Now finish processing the fastopen child socket. */ + tcp_init_transfer(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); + tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1; tcp_fastopen_add_skb(child, skb); tcp_rsk(req)->rcv_nxt = tp->rcv_nxt; tp->rcv_wup = tp->rcv_nxt; - - meta_sk = child; - if (!mptcp_check_req_fastopen(meta_sk, req)) { - child = tcp_sk(meta_sk)->mpcb->master_sk; - tp = tcp_sk(child); - } - - /* Now finish processing the fastopen child socket. */ - tcp_init_transfer(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); - /* tcp_conn_request() is sending the SYNACK, * and queues the child into listener accept queue. */ diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index ea36a23267cf..49ac1b821ac1 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5784,17 +5784,10 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, struct tcp_fastopen_cookie *cookie) { struct tcp_sock *tp = tcp_sk(sk); - struct sk_buff *data = NULL; + struct sk_buff *data = tp->syn_data ? tcp_rtx_queue_head(sk) : NULL; u16 mss = tp->rx_opt.mss_clamp, try_exp = 0; bool syn_drop = false; - if (tp->syn_data) { - if (mptcp(tp)) - data = tcp_write_queue_head(mptcp_meta_sk(sk)); - else - data = tcp_rtx_queue_head(sk); - } - if (mss == tp->rx_opt.user_mss) { struct tcp_options_received opt; @@ -5825,11 +5818,7 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, tcp_fastopen_cache_set(sk, mss, cookie, syn_drop, try_exp); - /* In mptcp case, we do not rely on "retransmit", but instead on - * "transmit", because if fastopen data is not acked, the retransmission - * becomes the first MPTCP data (see mptcp_rcv_synsent_fastopen). - */ - if (data && !mptcp(tp)) { /* Retransmit unacked data in SYN */ + if (data) { /* Retransmit unacked data in SYN */ skb_rbtree_walk_from(data) { if (__tcp_retransmit_skb(sk, data, 1)) break; @@ -6689,18 +6678,12 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, fastopen_sk = tcp_try_fastopen(sk, skb, req, &foc, dst); } if (fastopen_sk) { - struct sock *meta_sk = fastopen_sk; - - if (mptcp(tcp_sk(fastopen_sk))) - meta_sk = mptcp_meta_sk(fastopen_sk); af_ops->send_synack(fastopen_sk, dst, &fl, req, &foc, TCP_SYNACK_FASTOPEN); /* Add the child socket directly into the accept queue */ - inet_csk_reqsk_queue_add(sk, req, meta_sk); + inet_csk_reqsk_queue_add(sk, req, fastopen_sk); sk->sk_data_ready(sk); bh_unlock_sock(fastopen_sk); - if (meta_sk != fastopen_sk) - bh_unlock_sock(meta_sk); sock_put(fastopen_sk); } else { tcp_rsk(req)->tfo_listener = false; diff --git a/net/mptcp/mptcp_ctrl.c b/net/mptcp/mptcp_ctrl.c index 910268a8ee8e..eee86f84ba31 100644 --- a/net/mptcp/mptcp_ctrl.c +++ b/net/mptcp/mptcp_ctrl.c @@ -2075,59 +2075,6 @@ static int __mptcp_check_req_master(struct sock *child, return 0; } -int mptcp_check_req_fastopen(struct sock *child, struct request_sock *req) -{ - struct sock *meta_sk = child, *master_sk; - struct sk_buff *skb; - u32 new_mapping; - int ret; - - ret = __mptcp_check_req_master(child, req); - if (ret) - return ret; - - master_sk = tcp_sk(meta_sk)->mpcb->master_sk; - - /* We need to rewind copied_seq as it is set to IDSN + 1 and as we have - * pre-MPTCP data in the receive queue. - */ - tcp_sk(meta_sk)->copied_seq -= tcp_sk(master_sk)->rcv_nxt - - tcp_rsk(req)->rcv_isn - 1; - - /* Map subflow sequence number to data sequence numbers. We need to map - * these data to [IDSN - len - 1, IDSN[. - */ - new_mapping = tcp_sk(meta_sk)->copied_seq - tcp_rsk(req)->rcv_isn - 1; - - /* There should be only one skb: the SYN + data. */ - skb_queue_walk(&meta_sk->sk_receive_queue, skb) { - TCP_SKB_CB(skb)->seq += new_mapping; - TCP_SKB_CB(skb)->end_seq += new_mapping; - } - - /* With fastopen we change the semantics of the relative subflow - * sequence numbers to deal with middleboxes that could add/remove - * multiple bytes in the SYN. We chose to start counting at rcv_nxt - 1 - * instead of the regular TCP ISN. - */ - tcp_sk(master_sk)->mptcp->rcv_isn = tcp_sk(master_sk)->rcv_nxt - 1; - - /* We need to update copied_seq of the master_sk to account for the - * already moved data to the meta receive queue. - */ - tcp_sk(master_sk)->copied_seq = tcp_sk(master_sk)->rcv_nxt; - - /* Handled by the master_sk */ - tcp_sk(meta_sk)->fastopen_rsk = NULL; - - /* Subflow establishment is now lockless, drop the lock here it will - * be taken again in tcp_child_process(). - */ - bh_unlock_sock(meta_sk); - - return 0; -} - int mptcp_check_req_master(struct sock *sk, struct sock *child, struct request_sock *req, const struct sk_buff *skb, int drop) diff --git a/net/mptcp/mptcp_input.c b/net/mptcp/mptcp_input.c index f247e9ef2853..88c7a64ef26c 100644 --- a/net/mptcp/mptcp_input.c +++ b/net/mptcp/mptcp_input.c @@ -1934,63 +1934,6 @@ bool mptcp_handle_options(struct sock *sk, const struct tcphdr *th, return false; } -/* In case of fastopen, some data can already be in the write queue. - * We need to update the sequence number of the segments as they - * were initially TCP sequence numbers. - */ -static void mptcp_rcv_synsent_fastopen(struct sock *meta_sk) -{ - struct tcp_sock *meta_tp = tcp_sk(meta_sk); - struct tcp_sock *master_tp = tcp_sk(meta_tp->mpcb->master_sk); - struct sk_buff *skb, *tmp; - u32 new_mapping = meta_tp->write_seq - master_tp->snd_una; - - /* There should only be one skb in write queue: the data not - * acknowledged in the SYN+ACK. In this case, we need to map - * this data to data sequence numbers. - */ - skb_rbtree_walk_safe(skb, &meta_sk->tcp_rtx_queue, tmp) { - /* If the server only acknowledges partially the data sent in - * the SYN, we need to trim the acknowledged part because - * we don't want to retransmit this already received data. - * When we reach this point, tcp_ack() has already cleaned up - * fully acked segments. However, tcp trims partially acked - * segments only when retransmitting. Since MPTCP comes into - * play only now, we will fake an initial transmit, and - * retransmit_skb() will not be called. The following fragment - * comes from __tcp_retransmit_skb(). - */ - if (before(TCP_SKB_CB(skb)->seq, master_tp->snd_una)) { - BUG_ON(before(TCP_SKB_CB(skb)->end_seq, - master_tp->snd_una)); - /* tcp_trim_head can only returns ENOMEM if skb is - * cloned. It is not the case here (see - * tcp_send_syn_data). - */ - BUG_ON(tcp_trim_head(meta_sk, skb, master_tp->snd_una - - TCP_SKB_CB(skb)->seq)); - } - - TCP_SKB_CB(skb)->seq += new_mapping; - TCP_SKB_CB(skb)->end_seq += new_mapping; - - tcp_rtx_queue_unlink(skb, meta_sk); - tcp_add_write_queue_tail(meta_sk, skb); - } - - /* We can advance write_seq by the number of bytes unacknowledged - * and that were mapped in the previous loop. - */ - meta_tp->write_seq += master_tp->write_seq - master_tp->snd_una; - - /* The packets from the master_sk will be entailed to it later - * Until that time, its write queue is empty, and - * write_seq must align with snd_una - */ - master_tp->snd_nxt = master_tp->write_seq = master_tp->snd_una; - master_tp->packets_out = 0; -} - /* The skptr is needed, because if we become MPTCP-capable, we have to switch * from meta-socket to master-socket. * @@ -2049,17 +1992,6 @@ int mptcp_rcv_synsent_state_process(struct sock *sk, struct sock **skptr, *skptr = sk; tp = tcp_sk(sk); - /* If fastopen was used data might be in the send queue. We - * need to update their sequence number to MPTCP-level seqno. - * Note that it can happen in rare cases that fastopen_req is - * NULL and syn_data is 0 but fastopen indeed occurred and - * data has been queued in the write queue (but not sent). - * Example of such rare cases: connect is non-blocking and - * TFO is configured to work without cookies. - */ - if (tcp_rtx_queue_head(meta_sk)) - mptcp_rcv_synsent_fastopen(meta_sk); - /* -1, because the SYN consumed 1 byte. In case of TFO, we * start the subflow-sequence number as if the data of the SYN * is not part of any mapping. @@ -2068,9 +2000,6 @@ int mptcp_rcv_synsent_state_process(struct sock *sk, struct sock **skptr, tp->mptcp->include_mpc = 1; - /* Ensure that fastopen is handled at the meta-level. */ - tp->fastopen_req = NULL; - sk_set_socket(sk, meta_sk->sk_socket); sk->sk_wq = meta_sk->sk_wq; diff --git a/net/mptcp/mptcp_output.c b/net/mptcp/mptcp_output.c index 26cf41f75922..2d84eb2edd97 100644 --- a/net/mptcp/mptcp_output.c +++ b/net/mptcp/mptcp_output.c @@ -1292,12 +1292,9 @@ void mptcp_meta_retransmit_timer(struct sock *meta_sk) void mptcp_sub_retransmit_timer(struct sock *sk) { - struct tcp_sock *tp = tcp_sk(sk); - tcp_retransmit_timer(sk); - if (!tp->fastopen_rsk) - mptcp_set_rto(sk); + mptcp_set_rto(sk); } /* Modify values to an mptcp-level for the initial window of new subflows */ -- 2.16.2

Signed-off-by: Christoph Paasch <cpaasch(a)apple.com&gt; --- include/linux/tcp.h | 3 - include/net/mptcp.h | 41 ++------- include/uapi/linux/tcp.h | 1 - net/ipv4/tcp.c | 46 ---------- net/mptcp/Kconfig | 32 ------- net/mptcp/mptcp_ctrl.c | 35 +------- net/mptcp/mptcp_output.c | 7 +- net/mptcp/mptcp_sched.c | 223 ++++------------------------------------------- 8 files changed, 28 insertions(+), 360 deletions(-) diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 01321044047c..02af5c56d2e6 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -476,17 +476,14 @@ struct tcp_sock { close_it:1, /* Must close socket in mptcp_data_ready? */ closing:1, mptcp_ver:4, - mptcp_sched_setsockopt:1, mptcp_pm_setsockopt:1, record_master_info:1; struct mptcp_tcp_sock *mptcp; #ifdef CONFIG_MPTCP -#define MPTCP_SCHED_NAME_MAX 16 #define MPTCP_PM_NAME_MAX 16 struct hlist_nulls_node tk_table; u32 mptcp_loc_token; u64 mptcp_loc_key; - char mptcp_sched_name[MPTCP_SCHED_NAME_MAX]; char mptcp_pm_name[MPTCP_PM_NAME_MAX]; #endif /* CONFIG_MPTCP */ }; diff --git a/include/net/mptcp.h b/include/net/mptcp.h index e1e7619a85ca..90a9e220702a 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -170,8 +170,7 @@ struct mptcp_tcp_sock { u8 loc_id; u8 rem_id; -#define MPTCP_SCHED_SIZE 4 - u8 mptcp_sched[MPTCP_SCHED_SIZE] __aligned(8); + u32 last_rbuf_opti; int init_rcv_wnd; u32 infinite_cutoff_seq; @@ -219,23 +218,6 @@ struct mptcp_pm_ops { struct module *owner; }; -#define MPTCP_SCHED_NAME_MAX 16 -struct mptcp_sched_ops { - struct list_head list; - - struct sock * (*get_subflow)(struct sock *meta_sk, - struct sk_buff *skb, - bool zero_wnd_test); - struct sk_buff * (*next_segment)(struct sock *meta_sk, - int *reinject, - struct sock **subsk, - unsigned int *limit); - void (*init)(struct sock *sk); - - char name[MPTCP_SCHED_NAME_MAX]; - struct module *owner; -}; - struct mptcp_cb { /* list of sockets in this multipath connection */ struct hlist_head conn_list; @@ -261,10 +243,6 @@ struct mptcp_cb { snd_hiseq_index:1, /* Index in snd_high_order of snd_nxt */ rcv_hiseq_index:1; /* Index in rcv_high_order of rcv_nxt */ -#define MPTCP_SCHED_DATA_SIZE 8 - u8 mptcp_sched[MPTCP_SCHED_DATA_SIZE] __aligned(8); - struct mptcp_sched_ops *sched_ops; - u64 infinite_rcv_seq; /***** Start of fields, used for connection closure */ @@ -822,19 +800,18 @@ void mptcp_init_path_manager(struct mptcp_cb *mpcb); void mptcp_cleanup_path_manager(struct mptcp_cb *mpcb); void mptcp_fallback_default(struct mptcp_cb *mpcb); void mptcp_get_default_path_manager(char *name); -int mptcp_set_scheduler(struct sock *sk, const char *name); int mptcp_set_path_manager(struct sock *sk, const char *name); int mptcp_set_default_path_manager(const char *name); extern struct mptcp_pm_ops mptcp_pm_default; -/* MPTCP-scheduler registration/initialization functions */ -int mptcp_register_scheduler(struct mptcp_sched_ops *sched); -void mptcp_unregister_scheduler(struct mptcp_sched_ops *sched); -void mptcp_init_scheduler(struct mptcp_cb *mpcb); -void mptcp_cleanup_scheduler(struct mptcp_cb *mpcb); -void mptcp_get_default_scheduler(char *name); -int mptcp_set_default_scheduler(const char *name); -extern struct mptcp_sched_ops mptcp_sched_default; +void mptcp_sched_init(struct sock *sk); +struct sock *mptcp_get_available_subflow(struct sock *meta_sk, + struct sk_buff *skb, + bool zero_wnd_test); +struct sk_buff *mptcp_next_segment(struct sock *meta_sk, + int *reinject, + struct sock **subsk, + unsigned int *limit); /* Initializes function-pointers and MPTCP-flags */ static inline void mptcp_init_tcp_sock(struct sock *sk) diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h index 072ce786a0fd..2fcfa2257cc9 100644 --- a/include/uapi/linux/tcp.h +++ b/include/uapi/linux/tcp.h @@ -134,7 +134,6 @@ enum { #define TCP_CM_INQ TCP_INQ #define MPTCP_ENABLED 42 -#define MPTCP_SCHEDULER 43 #define MPTCP_PATH_MANAGER 44 #define MPTCP_INFO 45 diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b2dac73903be..3f570b2c7eea 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2802,33 +2802,6 @@ static int do_tcp_setsockopt(struct sock *sk, int level, return tcp_fastopen_reset_cipher(net, sk, key, sizeof(key)); } #ifdef CONFIG_MPTCP - case MPTCP_SCHEDULER: { - char name[MPTCP_SCHED_NAME_MAX]; - - if (optlen < 1) - return -EINVAL; - - /* Cannot be used if MPTCP is not used or we already have - * established an MPTCP-connection. - */ - if (mptcp_init_failed || !sysctl_mptcp_enabled || - sk->sk_state != TCP_CLOSE) - return -EPERM; - - val = strncpy_from_user(name, optval, - min_t(long, MPTCP_SCHED_NAME_MAX - 1, - optlen)); - - if (val < 0) - return -EFAULT; - name[val] = 0; - - lock_sock(sk); - err = mptcp_set_scheduler(sk, name); - release_sock(sk); - return err; - } - case MPTCP_PATH_MANAGER: { char name[MPTCP_PM_NAME_MAX]; @@ -3688,25 +3661,6 @@ static int do_tcp_getsockopt(struct sock *sk, int level, } #endif #ifdef CONFIG_MPTCP - case MPTCP_SCHEDULER: - if (get_user(len, optlen)) - return -EFAULT; - len = min_t(unsigned int, len, MPTCP_SCHED_NAME_MAX); - if (put_user(len, optlen)) - return -EFAULT; - - if (mptcp(tcp_sk(sk))) { - struct mptcp_cb *mpcb = tcp_sk(mptcp_meta_sk(sk))->mpcb; - - if (copy_to_user(optval, mpcb->sched_ops->name, len)) - return -EFAULT; - } else { - if (copy_to_user(optval, tcp_sk(sk)->mptcp_sched_name, - len)) - return -EFAULT; - } - return 0; - case MPTCP_PATH_MANAGER: if (get_user(len, optlen)) return -EFAULT; diff --git a/net/mptcp/Kconfig b/net/mptcp/Kconfig index 9c490ec8fba8..860e9e17f76f 100644 --- a/net/mptcp/Kconfig +++ b/net/mptcp/Kconfig @@ -57,35 +57,3 @@ config DEFAULT_MPTCP_PM default "ndiffports" if DEFAULT_NDIFFPORTS default "default" -menuconfig MPTCP_SCHED_ADVANCED - bool "MPTCP: advanced scheduler control" - depends on MPTCP=y - ---help--- - Support for selection of different schedulers. You should choose 'Y' here, - if you want to choose a different scheduler than the default one. - -if MPTCP_SCHED_ADVANCED - -choice - prompt "Default MPTCP Scheduler" - default DEFAULT - help - Select the Scheduler of your choice - - config DEFAULT_SCHEDULER - bool "Default" - ---help--- - This is the default scheduler, sending first on the subflow - with the lowest RTT. - -endchoice -endif - -config DEFAULT_MPTCP_SCHED - string - depends on (MPTCP=y) - default "default" if DEFAULT_SCHEDULER - default "roundrobin" if DEFAULT_ROUNDROBIN - default "redundant" if DEFAULT_REDUNDANT - default "default" - diff --git a/net/mptcp/mptcp_ctrl.c b/net/mptcp/mptcp_ctrl.c index eee86f84ba31..83639951c5a5 100644 --- a/net/mptcp/mptcp_ctrl.c +++ b/net/mptcp/mptcp_ctrl.c @@ -95,25 +95,6 @@ static int proc_mptcp_path_manager(struct ctl_table *ctl, int write, return ret; } -static int proc_mptcp_scheduler(struct ctl_table *ctl, int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) -{ - char val[MPTCP_SCHED_NAME_MAX]; - struct ctl_table tbl = { - .data = val, - .maxlen = MPTCP_SCHED_NAME_MAX, - }; - int ret; - - mptcp_get_default_scheduler(val); - - ret = proc_dostring(&tbl, write, buffer, lenp, ppos); - if (write && ret == 0) - ret = mptcp_set_default_scheduler(val); - return ret; -} - static struct ctl_table mptcp_table[] = { { .procname = "mptcp_enabled", @@ -151,12 +132,6 @@ static struct ctl_table mptcp_table[] = { .maxlen = MPTCP_PM_NAME_MAX, .proc_handler = proc_mptcp_path_manager, }, - { - .procname = "mptcp_scheduler", - .mode = 0644, - .maxlen = MPTCP_SCHED_NAME_MAX, - .proc_handler = proc_mptcp_scheduler, - }, { } }; @@ -640,7 +615,6 @@ static void mptcp_mpcb_put(struct mptcp_cb *mpcb) { if (refcount_dec_and_test(&mpcb->mpcb_refcnt)) { mptcp_cleanup_path_manager(mpcb); - mptcp_cleanup_scheduler(mpcb); kfree(mpcb->master_info); kmem_cache_free(mptcp_cb_cache, mpcb); } @@ -1312,7 +1286,6 @@ static int mptcp_alloc_mpcb(struct sock *meta_sk, __u64 remote_key, refcount_set(&mpcb->mpcb_refcnt, 1); mptcp_init_path_manager(mpcb); - mptcp_init_scheduler(mpcb); if (!try_module_get(inet_csk(master_sk)->icsk_ca_ops->owner)) tcp_assign_congestion_control(master_sk); @@ -1380,8 +1353,7 @@ int mptcp_add_sock(struct sock *meta_sk, struct sock *sk, u8 loc_id, u8 rem_id, tp->mptcp->loc_id = loc_id; tp->mptcp->rem_id = rem_id; - if (mpcb->sched_ops->init) - mpcb->sched_ops->init(sk); + mptcp_sched_init(sk); /* The corresponding sock_put is in mptcp_sock_destruct(). It cannot be * included in mptcp_del_sock(), because the mpcb must remain alive @@ -2910,17 +2882,12 @@ void __init mptcp_init(void) if (mptcp_register_path_manager(&mptcp_pm_default)) goto register_pm_failed; - if (mptcp_register_scheduler(&mptcp_sched_default)) - goto register_sched_failed; - pr_info("MPTCP: Unstable branch"); mptcp_init_failed = false; return; -register_sched_failed: - mptcp_unregister_path_manager(&mptcp_pm_default); register_pm_failed: unregister_net_sysctl_table(mptcp_sysctl); register_sysctl_failed: diff --git a/net/mptcp/mptcp_output.c b/net/mptcp/mptcp_output.c index 2d84eb2edd97..00a3674ad693 100644 --- a/net/mptcp/mptcp_output.c +++ b/net/mptcp/mptcp_output.c @@ -306,7 +306,7 @@ int mptcp_write_wakeup(struct sock *meta_sk, int mib) before(TCP_SKB_CB(skb)->seq, tcp_wnd_end(meta_tp))) { unsigned int mss; unsigned int seg_size = tcp_wnd_end(meta_tp) - TCP_SKB_CB(skb)->seq; - struct sock *subsk = meta_tp->mpcb->sched_ops->get_subflow(meta_sk, skb, true); + struct sock *subsk = mptcp_get_available_subflow(meta_sk, skb, true); struct tcp_sock *subtp; WARN_ON(TCP_SKB_CB(skb)->sacked); @@ -395,8 +395,7 @@ bool mptcp_write_xmit(struct sock *meta_sk, unsigned int mss_now, int nonagle, tcp_mstamp_refresh(meta_tp); - while ((skb = mpcb->sched_ops->next_segment(meta_sk, &reinject, &subsk, - &sublimit))) { + while ((skb = mptcp_next_segment(meta_sk, &reinject, &subsk, &sublimit))) { enum tcp_queue tcp_queue = TCP_FRAG_IN_WRITE_QUEUE; unsigned int limit; @@ -1111,7 +1110,7 @@ int mptcp_retransmit_skb(struct sock *meta_sk, struct sk_buff *skb) /* We need to make sure that the retransmitted segment can be sent on a * subflow right now. If it is too big, it needs to be fragmented. */ - subsk = meta_tp->mpcb->sched_ops->get_subflow(meta_sk, skb, false); + subsk = mptcp_get_available_subflow(meta_sk, skb, false); if (!subsk) { /* We want to increase icsk_retransmits, thus return 0, so that * mptcp_meta_retransmit_timer enters the desired branch. diff --git a/net/mptcp/mptcp_sched.c b/net/mptcp/mptcp_sched.c index 74d167e4689b..7beff6b762e1 100644 --- a/net/mptcp/mptcp_sched.c +++ b/net/mptcp/mptcp_sched.c @@ -1,20 +1,6 @@ -/* MPTCP Scheduler module selector. Highly inspired by tcp_cong.c */ - #include <linux/module.h> #include <net/mptcp.h> -static DEFINE_SPINLOCK(mptcp_sched_list_lock); -static LIST_HEAD(mptcp_sched_list); - -struct defsched_priv { - u32 last_rbuf_opti; -}; - -static struct defsched_priv *defsched_get_priv(const struct tcp_sock *tp) -{ - return (struct defsched_priv *)&tp->mptcp->mptcp_sched[0]; -} - static bool mptcp_is_def_unavailable(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); @@ -194,9 +180,9 @@ static struct sock * * Additionally, this function is aware of the backup-subflows. */ -static struct sock *get_available_subflow(struct sock *meta_sk, - struct sk_buff *skb, - bool zero_wnd_test) +struct sock *mptcp_get_available_subflow(struct sock *meta_sk, + struct sk_buff *skb, + bool zero_wnd_test) { struct mptcp_cb *mpcb = tcp_sk(meta_sk)->mpcb; struct sock *sk; @@ -244,11 +230,10 @@ static struct sock *get_available_subflow(struct sock *meta_sk, static struct sk_buff *mptcp_rcv_buf_optimization(struct sock *sk, int penal) { - struct sock *meta_sk; const struct tcp_sock *tp = tcp_sk(sk); struct mptcp_tcp_sock *mptcp; struct sk_buff *skb_head; - struct defsched_priv *dsp = defsched_get_priv(tp); + struct sock *meta_sk; meta_sk = mptcp_meta_sk(sk); skb_head = tcp_rtx_queue_head(meta_sk); @@ -265,7 +250,7 @@ static struct sk_buff *mptcp_rcv_buf_optimization(struct sock *sk, int penal) goto retrans; /* Only penalize again after an RTT has elapsed */ - if (tcp_jiffies32 - dsp->last_rbuf_opti < usecs_to_jiffies(tp->srtt_us >> 3)) + if (tcp_jiffies32 - tp->mptcp->last_rbuf_opti < usecs_to_jiffies(tp->srtt_us >> 3)) goto retrans; /* Half the cwnd of the slow flow */ @@ -283,7 +268,7 @@ static struct sk_buff *mptcp_rcv_buf_optimization(struct sock *sk, int penal) if (prior_cwnd >= tp_it->snd_ssthresh) tp_it->snd_ssthresh = max(tp_it->snd_ssthresh >> 1U, 2U); - dsp->last_rbuf_opti = tcp_jiffies32; + tp->mptcp->last_rbuf_opti = tcp_jiffies32; } break; } @@ -343,8 +328,8 @@ static struct sk_buff *__mptcp_next_segment(struct sock *meta_sk, int *reinject) if (!skb && meta_sk->sk_socket && test_bit(SOCK_NOSPACE, &meta_sk->sk_socket->flags) && sk_stream_wspace(meta_sk) < sk_stream_min_wspace(meta_sk)) { - struct sock *subsk = get_available_subflow(meta_sk, NULL, - false); + struct sock *subsk = mptcp_get_available_subflow(meta_sk, NULL, + false); if (!subsk) return NULL; @@ -355,10 +340,10 @@ static struct sk_buff *__mptcp_next_segment(struct sock *meta_sk, int *reinject) return skb; } -static struct sk_buff *mptcp_next_segment(struct sock *meta_sk, - int *reinject, - struct sock **subsk, - unsigned int *limit) +struct sk_buff *mptcp_next_segment(struct sock *meta_sk, + int *reinject, + struct sock **subsk, + unsigned int *limit) { struct sk_buff *skb = __mptcp_next_segment(meta_sk, reinject); unsigned int mss_now; @@ -372,7 +357,7 @@ static struct sk_buff *mptcp_next_segment(struct sock *meta_sk, if (!skb) return NULL; - *subsk = get_available_subflow(meta_sk, skb, false); + *subsk = mptcp_get_available_subflow(meta_sk, skb, false); if (!*subsk) return NULL; @@ -420,185 +405,7 @@ static struct sk_buff *mptcp_next_segment(struct sock *meta_sk, return skb; } -static void defsched_init(struct sock *sk) -{ - struct defsched_priv *dsp = defsched_get_priv(tcp_sk(sk)); - - dsp->last_rbuf_opti = tcp_jiffies32; -} - -struct mptcp_sched_ops mptcp_sched_default = { - .get_subflow = get_available_subflow, - .next_segment = mptcp_next_segment, - .init = defsched_init, - .name = "default", - .owner = THIS_MODULE, -}; - -static struct mptcp_sched_ops *mptcp_sched_find(const char *name) -{ - struct mptcp_sched_ops *e; - - list_for_each_entry_rcu(e, &mptcp_sched_list, list) { - if (strcmp(e->name, name) == 0) - return e; - } - - return NULL; -} - -int mptcp_register_scheduler(struct mptcp_sched_ops *sched) -{ - int ret = 0; - - if (!sched->get_subflow || !sched->next_segment) - return -EINVAL; - - spin_lock(&mptcp_sched_list_lock); - if (mptcp_sched_find(sched->name)) { - pr_notice("%s already registered\n", sched->name); - ret = -EEXIST; - } else { - list_add_tail_rcu(&sched->list, &mptcp_sched_list); - pr_info("%s registered\n", sched->name); - } - spin_unlock(&mptcp_sched_list_lock); - - return ret; -} -EXPORT_SYMBOL_GPL(mptcp_register_scheduler); - -void mptcp_unregister_scheduler(struct mptcp_sched_ops *sched) -{ - spin_lock(&mptcp_sched_list_lock); - list_del_rcu(&sched->list); - spin_unlock(&mptcp_sched_list_lock); - - /* Wait for outstanding readers to complete before the - * module gets removed entirely. - * - * A try_module_get() should fail by now as our module is - * in "going" state since no refs are held anymore and - * module_exit() handler being called. - */ - synchronize_rcu(); -} -EXPORT_SYMBOL_GPL(mptcp_unregister_scheduler); - -void mptcp_get_default_scheduler(char *name) -{ - struct mptcp_sched_ops *sched; - - BUG_ON(list_empty(&mptcp_sched_list)); - - rcu_read_lock(); - sched = list_entry(mptcp_sched_list.next, struct mptcp_sched_ops, list); - strncpy(name, sched->name, MPTCP_SCHED_NAME_MAX); - rcu_read_unlock(); -} - -int mptcp_set_default_scheduler(const char *name) -{ - struct mptcp_sched_ops *sched; - int ret = -ENOENT; - - spin_lock(&mptcp_sched_list_lock); - sched = mptcp_sched_find(name); -#ifdef CONFIG_MODULES - if (!sched && capable(CAP_NET_ADMIN)) { - spin_unlock(&mptcp_sched_list_lock); - - request_module("mptcp_%s", name); - spin_lock(&mptcp_sched_list_lock); - sched = mptcp_sched_find(name); - } -#endif - - if (sched) { - list_move(&sched->list, &mptcp_sched_list); - ret = 0; - } else { - pr_info("%s is not available\n", name); - } - spin_unlock(&mptcp_sched_list_lock); - - return ret; -} - -/* Must be called with rcu lock held */ -static struct mptcp_sched_ops *__mptcp_sched_find_autoload(const char *name) -{ - struct mptcp_sched_ops *sched = mptcp_sched_find(name); -#ifdef CONFIG_MODULES - if (!sched && capable(CAP_NET_ADMIN)) { - rcu_read_unlock(); - request_module("mptcp_%s", name); - rcu_read_lock(); - sched = mptcp_sched_find(name); - } -#endif - return sched; -} - -void mptcp_init_scheduler(struct mptcp_cb *mpcb) -{ - struct mptcp_sched_ops *sched; - struct sock *meta_sk = mpcb->meta_sk; - struct tcp_sock *meta_tp = tcp_sk(meta_sk); - - rcu_read_lock(); - /* if scheduler was set using socket option */ - if (meta_tp->mptcp_sched_setsockopt) { - sched = __mptcp_sched_find_autoload(meta_tp->mptcp_sched_name); - if (sched && try_module_get(sched->owner)) { - mpcb->sched_ops = sched; - goto out; - } - } - - list_for_each_entry_rcu(sched, &mptcp_sched_list, list) { - if (try_module_get(sched->owner)) { - mpcb->sched_ops = sched; - break; - } - } -out: - rcu_read_unlock(); -} - -/* Change scheduler for socket */ -int mptcp_set_scheduler(struct sock *sk, const char *name) -{ - struct mptcp_sched_ops *sched; - int err = 0; - - rcu_read_lock(); - sched = __mptcp_sched_find_autoload(name); - - if (!sched) { - err = -ENOENT; - } else if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) { - err = -EPERM; - } else { - strcpy(tcp_sk(sk)->mptcp_sched_name, name); - tcp_sk(sk)->mptcp_sched_setsockopt = 1; - } - rcu_read_unlock(); - - return err; -} - -/* Manage refcounts on socket close. */ -void mptcp_cleanup_scheduler(struct mptcp_cb *mpcb) +void mptcp_sched_init(struct sock *sk) { - module_put(mpcb->sched_ops->owner); -} - -/* Set default value from kernel configuration at bootup */ -static int __init mptcp_scheduler_default(void) -{ - BUILD_BUG_ON(sizeof(struct defsched_priv) > MPTCP_SCHED_SIZE); - - return mptcp_set_default_scheduler(CONFIG_DEFAULT_MPTCP_SCHED); + tcp_sk(sk)->mptcp->last_rbuf_opti = tcp_jiffies32; } -late_initcall(mptcp_scheduler_default); -- 2.16.2

Signed-off-by: Christoph Paasch <cpaasch(a)apple.com&gt; --- Notes: v2: * Added fullmesh_initialized flag to indicate that we went through _new_session() and it's all setup. Necessary because the removal got rid of the function-pointers which were guarding that implicitly. * Fixed some oversights in the removal of mptcp_become_fully_estab(). include/linux/tcp.h | 3 - include/net/mptcp.h | 93 ++++++++----- include/net/netns/mptcp.h | 8 +- net/ipv4/tcp.c | 47 ------- net/ipv4/tcp_input.c | 4 +- net/ipv4/tcp_output.c | 3 +- net/mptcp/Kconfig | 37 ----- net/mptcp/Makefile | 4 +- net/mptcp/mptcp_ctrl.c | 49 +------ net/mptcp/mptcp_fullmesh.c | 326 +++++++-------------------------------------- net/mptcp/mptcp_input.c | 32 ++--- net/mptcp/mptcp_ipv4.c | 2 +- net/mptcp/mptcp_ipv6.c | 2 +- net/mptcp/mptcp_output.c | 8 +- net/mptcp/mptcp_pm.c | 226 ------------------------------- 15 files changed, 136 insertions(+), 708 deletions(-) delete mode 100644 net/mptcp/mptcp_pm.c diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 02af5c56d2e6..84ec344fe184 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -476,15 +476,12 @@ struct tcp_sock { close_it:1, /* Must close socket in mptcp_data_ready? */ closing:1, mptcp_ver:4, - mptcp_pm_setsockopt:1, record_master_info:1; struct mptcp_tcp_sock *mptcp; #ifdef CONFIG_MPTCP -#define MPTCP_PM_NAME_MAX 16 struct hlist_nulls_node tk_table; u32 mptcp_loc_token; u64 mptcp_loc_key; - char mptcp_pm_name[MPTCP_PM_NAME_MAX]; #endif /* CONFIG_MPTCP */ }; diff --git a/include/net/mptcp.h b/include/net/mptcp.h index 90a9e220702a..ac655e828fa5 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -195,27 +195,49 @@ struct mptcp_tw { in_list:1; }; -#define MPTCP_PM_NAME_MAX 16 -struct mptcp_pm_ops { - struct list_head list; +struct fullmesh_rem4 { + u8 rem4_id; + u8 bitfield; + u8 retry_bitfield; + __be16 port; + struct in_addr addr; +}; - /* Signal the creation of a new MPTCP-session. */ - void (*new_session)(const struct sock *meta_sk); - void (*release_sock)(struct sock *meta_sk); - void (*fully_established)(struct sock *meta_sk); - void (*new_remote_address)(struct sock *meta_sk); - void (*subflow_error)(struct sock *meta_sk, struct sock *sk); - int (*get_local_id)(sa_family_t family, union inet_addr *addr, - struct net *net); - void (*addr_signal)(struct sock *sk, unsigned *size, - struct tcp_out_options *opts, struct sk_buff *skb); - void (*add_raddr)(struct mptcp_cb *mpcb, const union inet_addr *addr, - sa_family_t family, __be16 port, u8 id); - void (*rem_raddr)(struct mptcp_cb *mpcb, u8 rem_id); - void (*delete_subflow)(struct sock *sk); - - char name[MPTCP_PM_NAME_MAX]; - struct module *owner; +struct fullmesh_rem6 { + u8 rem6_id; + u8 bitfield; + u8 retry_bitfield; + __be16 port; + struct in6_addr addr; +}; + +/* Max number of local or remote addresses we can store. + * When changing, see the bitfield below in fullmesh_rem4/6. + */ +#define MPTCP_MAX_ADDR 8 + +struct fullmesh_priv { + /* Worker struct for subflow establishment */ + struct work_struct subflow_work; + /* Delayed worker, when the routing-tables are not yet ready. */ + struct delayed_work subflow_retry_work; + + /* Remote addresses */ + struct fullmesh_rem4 remaddr4[MPTCP_MAX_ADDR]; + struct fullmesh_rem6 remaddr6[MPTCP_MAX_ADDR]; + + struct mptcp_cb *mpcb; + + u16 remove_addrs; /* Addresses to remove */ + u8 announced_addrs_v4; /* IPv4 Addresses we did announce */ + u8 announced_addrs_v6; /* IPv6 Addresses we did announce */ + + u8 add_addr; /* Are we sending an add_addr? */ + + u8 rem4_bits; + u8 rem6_bits; + + u8 fullmesh_initialized:1; }; struct mptcp_cb { @@ -273,9 +295,7 @@ struct mptcp_cb { __u32 mptcp_loc_token; __u32 mptcp_rem_token; -#define MPTCP_PM_SIZE 608 - u8 mptcp_pm[MPTCP_PM_SIZE] __aligned(8); - struct mptcp_pm_ops *pm_ops; + struct fullmesh_priv fmp; unsigned long path_index_bits; @@ -690,7 +710,6 @@ void mptcp_cleanup_rbuf(struct sock *meta_sk, int copied); int mptcp_add_sock(struct sock *meta_sk, struct sock *sk, u8 loc_id, u8 rem_id, gfp_t flags); void mptcp_del_sock(struct sock *sk); -void mptcp_update_metasocket(const struct sock *meta_sk); void mptcp_update_sndbuf(const struct tcp_sock *tp); void mptcp_send_fin(struct sock *meta_sk); void mptcp_send_active_reset(struct sock *meta_sk, gfp_t priority); @@ -793,16 +812,19 @@ int mptcp_finish_handshake(struct sock *child, struct sk_buff *skb); int mptcp_get_info(const struct sock *meta_sk, char __user *optval, int optlen); void mptcp_clear_sk(struct sock *sk, int size); -/* MPTCP-path-manager registration/initialization functions */ -int mptcp_register_path_manager(struct mptcp_pm_ops *pm); -void mptcp_unregister_path_manager(struct mptcp_pm_ops *pm); -void mptcp_init_path_manager(struct mptcp_cb *mpcb); -void mptcp_cleanup_path_manager(struct mptcp_cb *mpcb); -void mptcp_fallback_default(struct mptcp_cb *mpcb); -void mptcp_get_default_path_manager(char *name); -int mptcp_set_path_manager(struct sock *sk, const char *name); -int mptcp_set_default_path_manager(const char *name); -extern struct mptcp_pm_ops mptcp_pm_default; +int __init full_mesh_register(void); +void full_mesh_new_session(const struct sock *meta_sk); +void full_mesh_release_sock(struct sock *meta_sk); +void full_mesh_create_subflows(struct sock *meta_sk); +int full_mesh_get_local_id(sa_family_t family, union inet_addr *addr, + struct net *net); +void full_mesh_addr_signal(struct sock *sk, unsigned *size, + struct tcp_out_options *opts, + struct sk_buff *skb); +void full_mesh_add_raddr(struct mptcp_cb *mpcb, + const union inet_addr *addr, + sa_family_t family, __be16 port, u8 id); +void full_mesh_rem_raddr(struct mptcp_cb *mpcb, u8 rem_id); void mptcp_sched_init(struct sock *sk); struct sock *mptcp_get_available_subflow(struct sock *meta_sk, @@ -1223,7 +1245,6 @@ static inline int is_master_tp(const struct tcp_sock *tp) return 0; } static inline void mptcp_del_sock(const struct sock *sk) {} -static inline void mptcp_update_metasocket(const struct sock *meta_sk) {} static inline void mptcp_update_sndbuf(const struct tcp_sock *tp) {} static inline void mptcp_clean_rtx_infinite(const struct sk_buff *skb, const struct sock *sk) {} @@ -1336,6 +1357,8 @@ static inline bool mptcp_can_new_subflow(const struct sock *meta_sk) { return false; } +static inline void full_mesh_new_session(const struct sock *meta_sk) {} +static inline void full_mesh_release_sock(struct sock *meta_sk) {} #endif /* CONFIG_MPTCP */ diff --git a/include/net/netns/mptcp.h b/include/net/netns/mptcp.h index 6680f3bbcfc8..c9dd96de4db5 100644 --- a/include/net/netns/mptcp.h +++ b/include/net/netns/mptcp.h @@ -32,12 +32,8 @@ #include <linux/compiler.h> -enum { - MPTCP_PM_FULLMESH = 0, - MPTCP_PM_MAX -}; - struct mptcp_mib; +struct mptcp_fm_ns; struct netns_mptcp { DEFINE_SNMP_STAT(struct mptcp_mib, mptcp_statistics); @@ -46,7 +42,7 @@ struct netns_mptcp { struct proc_dir_entry *proc_net_mptcp; #endif - void *path_managers[MPTCP_PM_MAX]; + struct mptcp_fm_ns *fm_ns; }; #endif /* __NETNS_MPTCP_H__ */ diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 3f570b2c7eea..76555f8b26fe 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2801,34 +2801,6 @@ static int do_tcp_setsockopt(struct sock *sk, int level, return tcp_fastopen_reset_cipher(net, sk, key, sizeof(key)); } -#ifdef CONFIG_MPTCP - case MPTCP_PATH_MANAGER: { - char name[MPTCP_PM_NAME_MAX]; - - if (optlen < 1) - return -EINVAL; - - /* Cannot be used if MPTCP is not used or we already have - * established an MPTCP-connection. - */ - if (mptcp_init_failed || !sysctl_mptcp_enabled || - sk->sk_state != TCP_CLOSE) - return -EPERM; - - val = strncpy_from_user(name, optval, - min_t(long, MPTCP_PM_NAME_MAX - 1, - optlen)); - - if (val < 0) - return -EFAULT; - name[val] = 0; - - lock_sock(sk); - err = mptcp_set_path_manager(sk, name); - release_sock(sk); - return err; - } -#endif default: /* fallthru */ break; @@ -3661,25 +3633,6 @@ static int do_tcp_getsockopt(struct sock *sk, int level, } #endif #ifdef CONFIG_MPTCP - case MPTCP_PATH_MANAGER: - if (get_user(len, optlen)) - return -EFAULT; - len = min_t(unsigned int, len, MPTCP_PM_NAME_MAX); - if (put_user(len, optlen)) - return -EFAULT; - - if (mptcp(tcp_sk(sk))) { - struct mptcp_cb *mpcb = tcp_sk(mptcp_meta_sk(sk))->mpcb; - - if (copy_to_user(optval, mpcb->pm_ops->name, len)) - return -EFAULT; - } else { - if (copy_to_user(optval, tcp_sk(sk)->mptcp_pm_name, - len)) - return -EFAULT; - } - return 0; - case MPTCP_ENABLED: if (sk->sk_state != TCP_SYN_SENT) val = mptcp(tp) ? 1 : 0; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 49ac1b821ac1..5cf9161db844 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -6177,7 +6177,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) * addresses, which can only be done after the third ack * of the 3-way handshake. */ - mptcp_update_metasocket(tp->meta_sk); + full_mesh_new_session(tp->meta_sk); } if (queued >= 0) return queued; @@ -6280,7 +6280,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) */ if (mptcp(tp)) { if (is_master_tp(tp)) - mptcp_update_metasocket(mptcp_meta_sk(sk)); + full_mesh_new_session(mptcp_meta_sk(sk)); else tcp_send_ack(sk); } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index a1cb6a3c0dda..e41db7407a90 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -921,8 +921,7 @@ void tcp_release_cb(struct sock *sk) __sock_put(sk); } if (flags & TCPF_PATH_MANAGER_DEFERRED) { - if (tcp_sk(sk)->mpcb->pm_ops->release_sock) - tcp_sk(sk)->mpcb->pm_ops->release_sock(sk); + full_mesh_release_sock(sk); __sock_put(sk); } if (flags & TCPF_SUB_DEFERRED) diff --git a/net/mptcp/Kconfig b/net/mptcp/Kconfig index 60c07fde5dec..c3b835e40d8b 100644 --- a/net/mptcp/Kconfig +++ b/net/mptcp/Kconfig @@ -8,40 +8,3 @@ config MPTCP This replaces the normal TCP stack with a Multipath TCP stack, able to use several paths at once. -menuconfig MPTCP_PM_ADVANCED - bool "MPTCP: advanced path-manager control" - depends on MPTCP=y - ---help--- - Support for selection of different path-managers. You should choose 'Y' here, - because otherwise you will not actively create new MPTCP-subflows. - -if MPTCP_PM_ADVANCED - -config MPTCP_FULLMESH - tristate "MPTCP Full-Mesh Path-Manager" - depends on MPTCP=y - ---help--- - This path-management module will create a full-mesh among all IP-addresses. - -choice - prompt "Default MPTCP Path-Manager" - default DEFAULT - help - Select the Path-Manager of your choice - - config DEFAULT_FULLMESH - bool "Full mesh" if MPTCP_FULLMESH=y - - config DEFAULT_DUMMY - bool "Default" - -endchoice - -endif - -config DEFAULT_MPTCP_PM - string - default "default" if DEFAULT_DUMMY - default "fullmesh" if DEFAULT_FULLMESH - default "default" - diff --git a/net/mptcp/Makefile b/net/mptcp/Makefile index 433fd17b2987..87901e6b3843 100644 --- a/net/mptcp/Makefile +++ b/net/mptcp/Makefile @@ -5,10 +5,8 @@ obj-$(CONFIG_MPTCP) += mptcp.o -mptcp-y := mptcp_ctrl.o mptcp_ipv4.o mptcp_pm.o \ +mptcp-y := mptcp_ctrl.o mptcp_ipv4.o mptcp_fullmesh.o \ mptcp_output.o mptcp_input.o mptcp_sched.o -obj-$(CONFIG_MPTCP_FULLMESH) += mptcp_fullmesh.o - mptcp-$(subst m,y,$(CONFIG_IPV6)) += mptcp_ipv6.o diff --git a/net/mptcp/mptcp_ctrl.c b/net/mptcp/mptcp_ctrl.c index 83639951c5a5..f170de1c72d4 100644 --- a/net/mptcp/mptcp_ctrl.c +++ b/net/mptcp/mptcp_ctrl.c @@ -76,25 +76,6 @@ bool mptcp_init_failed __read_mostly; struct static_key mptcp_static_key = STATIC_KEY_INIT_FALSE; EXPORT_SYMBOL(mptcp_static_key); -static int proc_mptcp_path_manager(struct ctl_table *ctl, int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) -{ - char val[MPTCP_PM_NAME_MAX]; - struct ctl_table tbl = { - .data = val, - .maxlen = MPTCP_PM_NAME_MAX, - }; - int ret; - - mptcp_get_default_path_manager(val); - - ret = proc_dostring(&tbl, write, buffer, lenp, ppos); - if (write && ret == 0) - ret = mptcp_set_default_path_manager(val); - return ret; -} - static struct ctl_table mptcp_table[] = { { .procname = "mptcp_enabled", @@ -126,12 +107,6 @@ static struct ctl_table mptcp_table[] = { .mode = 0644, .proc_handler = &proc_dointvec }, - { - .procname = "mptcp_path_manager", - .mode = 0644, - .maxlen = MPTCP_PM_NAME_MAX, - .proc_handler = proc_mptcp_path_manager, - }, { } }; @@ -604,9 +579,6 @@ static void mptcp_sock_def_error_report(struct sock *sk) tcp_done(meta_sk); } - if (mpcb->pm_ops->subflow_error) - mpcb->pm_ops->subflow_error(meta_sk, sk); - sk->sk_err = 0; return; } @@ -614,7 +586,6 @@ static void mptcp_sock_def_error_report(struct sock *sk) static void mptcp_mpcb_put(struct mptcp_cb *mpcb) { if (refcount_dec_and_test(&mpcb->mpcb_refcnt)) { - mptcp_cleanup_path_manager(mpcb); kfree(mpcb->master_info); kmem_cache_free(mptcp_cb_cache, mpcb); } @@ -1285,8 +1256,6 @@ static int mptcp_alloc_mpcb(struct sock *meta_sk, __u64 remote_key, /* The meta is directly linked - set refcnt to 1 */ refcount_set(&mpcb->mpcb_refcnt, 1); - mptcp_init_path_manager(mpcb); - if (!try_module_get(inet_csk(master_sk)->icsk_ca_ops->owner)) tcp_assign_congestion_control(master_sk); @@ -1417,9 +1386,6 @@ void mptcp_del_sock(struct sock *sk) mpcb = tp->mpcb; - if (mpcb->pm_ops->delete_subflow) - mpcb->pm_ops->delete_subflow(sk); - mptcp_debug("%s: Removing subsock tok %#x pi:%d state %d is_meta? %d\n", __func__, mpcb->mptcp_loc_token, tp->mptcp->path_index, sk->sk_state, is_meta_sk(sk)); @@ -1452,15 +1418,6 @@ void mptcp_del_sock(struct sock *sk) rcu_assign_pointer(inet_sk(sk)->inet_opt, NULL); } -/* Updates the MPTCP-session based on path-manager information (e.g., addresses, - * low-prio flows,...). - */ -void mptcp_update_metasocket(const struct sock *meta_sk) -{ - if (tcp_sk(meta_sk)->mpcb->pm_ops->new_session) - tcp_sk(meta_sk)->mpcb->pm_ops->new_session(meta_sk); -} - /* Clean up the receive buffer for full frames taken by the user, * then send an ACK if necessary. COPIED is the number of bytes * tcp_recvmsg has given to the user so far, it speeds up the @@ -2879,8 +2836,8 @@ void __init mptcp_init(void) if (!mptcp_sysctl) goto register_sysctl_failed; - if (mptcp_register_path_manager(&mptcp_pm_default)) - goto register_pm_failed; + if (full_mesh_register()) + goto register_full_mesh_failed; pr_info("MPTCP: Unstable branch"); @@ -2888,7 +2845,7 @@ void __init mptcp_init(void) return; -register_pm_failed: +register_full_mesh_failed: unregister_net_sysctl_table(mptcp_sysctl); register_sysctl_failed: mptcp_pm_v4_undo(); diff --git a/net/mptcp/mptcp_fullmesh.c b/net/mptcp/mptcp_fullmesh.c index fcfe6e6bd87b..20e0e9bbc532 100644 --- a/net/mptcp/mptcp_fullmesh.c +++ b/net/mptcp/mptcp_fullmesh.c @@ -16,27 +16,6 @@ enum { #define MPTCP_SUBFLOW_RETRY_DELAY 1000 -/* Max number of local or remote addresses we can store. - * When changing, see the bitfield below in fullmesh_rem4/6. - */ -#define MPTCP_MAX_ADDR 8 - -struct fullmesh_rem4 { - u8 rem4_id; - u8 bitfield; - u8 retry_bitfield; - __be16 port; - struct in_addr addr; -}; - -struct fullmesh_rem6 { - u8 rem6_id; - u8 bitfield; - u8 retry_bitfield; - __be16 port; - struct in6_addr addr; -}; - struct mptcp_loc_addr { struct mptcp_loc4 locaddr4[MPTCP_MAX_ADDR]; u8 loc4_bits; @@ -56,31 +35,6 @@ struct mptcp_addr_event { union inet_addr addr; }; -struct fullmesh_priv { - /* Worker struct for subflow establishment */ - struct work_struct subflow_work; - /* Delayed worker, when the routing-tables are not yet ready. */ - struct delayed_work subflow_retry_work; - - /* Remote addresses */ - struct fullmesh_rem4 remaddr4[MPTCP_MAX_ADDR]; - struct fullmesh_rem6 remaddr6[MPTCP_MAX_ADDR]; - - struct mptcp_cb *mpcb; - - u16 remove_addrs; /* Addresses to remove */ - u8 announced_addrs_v4; /* IPv4 Addresses we did announce */ - u8 announced_addrs_v6; /* IPv6 Addresses we did announce */ - - u8 add_addr; /* Are we sending an add_addr? */ - - u8 rem4_bits; - u8 rem6_bits; - - /* Have we established the additional subflows for primary pair? */ - u8 first_pair:1; -}; - struct mptcp_fm_ns { struct mptcp_loc_addr __rcu *local; spinlock_t local_lock; /* Protecting the above pointer */ @@ -90,26 +44,14 @@ struct mptcp_fm_ns { struct net *net; }; -static int num_subflows __read_mostly = 1; -module_param(num_subflows, int, 0644); -MODULE_PARM_DESC(num_subflows, "choose the number of subflows per pair of IP addresses of MPTCP connection"); - -static int create_on_err __read_mostly; -module_param(create_on_err, int, 0644); -MODULE_PARM_DESC(create_on_err, "recreate the subflow upon a timeout"); - -static struct mptcp_pm_ops full_mesh __read_mostly; - -static void full_mesh_create_subflows(struct sock *meta_sk); - -static struct mptcp_fm_ns *fm_get_ns(const struct net *net) +static struct mptcp_fm_ns *fm_get_ns(struct net *net) { - return (struct mptcp_fm_ns *)net->mptcp.path_managers[MPTCP_PM_FULLMESH]; + return net->mptcp.fm_ns; } -static struct fullmesh_priv *fullmesh_get_priv(const struct mptcp_cb *mpcb) +static struct fullmesh_priv *fullmesh_get_priv(struct mptcp_cb *mpcb) { - return (struct fullmesh_priv *)&mpcb->mptcp_pm[0]; + return &mpcb->fmp; } /* Find the first free index in the bitfield */ @@ -265,7 +207,7 @@ static void mptcp_v4_rem_raddress(struct mptcp_cb *mpcb, u8 id) } } -static void mptcp_v6_rem_raddress(const struct mptcp_cb *mpcb, u8 id) +static void mptcp_v6_rem_raddress(struct mptcp_cb *mpcb, u8 id) { int i; struct fullmesh_priv *fmp = fullmesh_get_priv(mpcb); @@ -281,7 +223,7 @@ static void mptcp_v6_rem_raddress(const struct mptcp_cb *mpcb, u8 id) } /* Sets the bitfield of the remote-address field */ -static void mptcp_v4_set_init_addr_bit(const struct mptcp_cb *mpcb, +static void mptcp_v4_set_init_addr_bit(struct mptcp_cb *mpcb, const struct in_addr *addr, u8 index) { int i; @@ -320,28 +262,6 @@ static void mptcp_set_init_addr_bit(struct mptcp_cb *mpcb, mptcp_v6_set_init_addr_bit(mpcb, &addr->in6, id); } -static void mptcp_v4_subflows(struct sock *meta_sk, - const struct mptcp_loc4 *loc, - struct mptcp_rem4 *rem) -{ - int i; - - for (i = 1; i < num_subflows; i++) - mptcp_init4_subsockets(meta_sk, loc, rem); -} - -#if IS_ENABLED(CONFIG_IPV6) -static void mptcp_v6_subflows(struct sock *meta_sk, - const struct mptcp_loc6 *loc, - struct mptcp_rem6 *rem) -{ - int i; - - for (i = 1; i < num_subflows; i++) - mptcp_init6_subsockets(meta_sk, loc, rem); -} -#endif - static void retry_subflow_worker(struct work_struct *work) { struct delayed_work *delayed_work = container_of(work, @@ -397,9 +317,6 @@ static void retry_subflow_worker(struct work_struct *work) rem4.rem4_id = rem->rem4_id; mptcp_init4_subsockets(meta_sk, &mptcp_local->locaddr4[i], &rem4); - mptcp_v4_subflows(meta_sk, - &mptcp_local->locaddr4[i], - &rem4); goto next_subflow; } } @@ -421,9 +338,6 @@ static void retry_subflow_worker(struct work_struct *work) rem6.rem6_id = rem->rem6_id; mptcp_init6_subsockets(meta_sk, &mptcp_local->locaddr6[i], &rem6); - mptcp_v6_subflows(meta_sk, - &mptcp_local->locaddr6[i], - &rem6); goto next_subflow; } } @@ -482,23 +396,6 @@ static void create_subflow_worker(struct work_struct *work) !tcp_sk(mpcb->master_sk)->mptcp->fully_established) goto exit; - /* Create the additional subflows for the first pair */ - if (fmp->first_pair == 0 && mpcb->master_sk) { - struct mptcp_loc4 loc; - struct mptcp_rem4 rem; - - loc.addr.s_addr = inet_sk(meta_sk)->inet_saddr; - loc.loc4_id = 0; - loc.if_idx = mpcb->master_sk->sk_bound_dev_if; - - rem.addr.s_addr = inet_sk(meta_sk)->inet_daddr; - rem.port = inet_sk(meta_sk)->inet_dport; - rem.rem4_id = 0; /* Default 0 */ - - mptcp_v4_subflows(meta_sk, &loc, &rem); - - fmp->first_pair = 1; - } iter++; mptcp_for_each_bit_set(fmp->rem4_bits, i) { @@ -523,31 +420,11 @@ static void create_subflow_worker(struct work_struct *work) if (mptcp_init4_subsockets(meta_sk, &mptcp_local->locaddr4[i], &rem4) == -ENETUNREACH) retry = rem->retry_bitfield |= (1 << i); - else - mptcp_v4_subflows(meta_sk, - &mptcp_local->locaddr4[i], - &rem4); goto next_subflow; } } #if IS_ENABLED(CONFIG_IPV6) - if (fmp->first_pair == 0 && mpcb->master_sk) { - struct mptcp_loc6 loc; - struct mptcp_rem6 rem; - - loc.addr = inet6_sk(meta_sk)->saddr; - loc.loc6_id = 0; - loc.if_idx = mpcb->master_sk->sk_bound_dev_if; - - rem.addr = meta_sk->sk_v6_daddr; - rem.port = inet_sk(meta_sk)->inet_dport; - rem.rem6_id = 0; /* Default 0 */ - - mptcp_v6_subflows(meta_sk, &loc, &rem); - - fmp->first_pair = 1; - } mptcp_for_each_bit_set(fmp->rem6_bits, i) { struct fullmesh_rem6 *rem; u8 remaining_bits; @@ -570,10 +447,6 @@ static void create_subflow_worker(struct work_struct *work) if (mptcp_init6_subsockets(meta_sk, &mptcp_local->locaddr6[i], &rem6) == -ENETUNREACH) retry = rem->retry_bitfield |= (1 << i); - else - mptcp_v6_subflows(meta_sk, - &mptcp_local->locaddr6[i], - &rem6); goto next_subflow; } } @@ -834,6 +707,7 @@ static void mptcp_address_worker(struct work_struct *work) tk_table) { struct sock *meta_sk = (struct sock *)meta_tp, *sk; bool meta_v4 = meta_sk->sk_family == AF_INET; + struct fullmesh_priv *fmp; struct mptcp_cb *mpcb; if (sock_net(meta_sk) != net) @@ -863,8 +737,8 @@ static void mptcp_address_worker(struct work_struct *work) mpcb->send_infinite_mapping) goto next; - /* May be that the pm has changed in-between */ - if (mpcb->pm_ops != &full_mesh) + fmp = fullmesh_get_priv(mpcb); + if (!fmp->fullmesh_initialized) goto next; if (sock_owned_by_user(meta_sk)) { @@ -954,7 +828,7 @@ static void mptcp_address_worker(struct work_struct *work) goto next_event; } -static struct mptcp_addr_event *lookup_similar_event(const struct net *net, +static struct mptcp_addr_event *lookup_similar_event(struct net *net, const struct mptcp_addr_event *event) { struct mptcp_addr_event *eventq; @@ -1170,17 +1044,22 @@ static struct notifier_block mptcp_pm_netdev_notifier = { .notifier_call = netdev_event, }; -static void full_mesh_add_raddr(struct mptcp_cb *mpcb, - const union inet_addr *addr, - sa_family_t family, __be16 port, u8 id) +void full_mesh_add_raddr(struct mptcp_cb *mpcb, + const union inet_addr *addr, + sa_family_t family, __be16 port, u8 id) { + struct fullmesh_priv *fmp = fullmesh_get_priv(mpcb); + + if (!fmp->fullmesh_initialized) + return; + if (family == AF_INET) mptcp_addv4_raddr(mpcb, &addr->in, port, id); else mptcp_addv6_raddr(mpcb, &addr->in6, port, id); } -static void full_mesh_new_session(const struct sock *meta_sk) +void full_mesh_new_session(const struct sock *meta_sk) { struct mptcp_loc_addr *mptcp_local; struct mptcp_cb *mpcb = tcp_sk(meta_sk)->mpcb; @@ -1217,7 +1096,10 @@ static void full_mesh_new_session(const struct sock *meta_sk) else index = mptcp_find_address(mptcp_local, family, &saddr, if_idx); if (index < 0) - goto fallback; + goto exit; + + /* Point of no-return */ + fmp->fullmesh_initialized = 1; full_mesh_add_raddr(mpcb, &daddr, family, 0, 0); mptcp_set_init_addr_bit(mpcb, &daddr, family, index); @@ -1278,17 +1160,19 @@ static void full_mesh_new_session(const struct sock *meta_sk) return; -fallback: +exit: rcu_read_unlock_bh(); - mptcp_fallback_default(mpcb); return; } -static void full_mesh_create_subflows(struct sock *meta_sk) +void full_mesh_create_subflows(struct sock *meta_sk) { - const struct mptcp_cb *mpcb = tcp_sk(meta_sk)->mpcb; + struct mptcp_cb *mpcb = tcp_sk(meta_sk)->mpcb; struct fullmesh_priv *fmp = fullmesh_get_priv(mpcb); + if (!fmp->fullmesh_initialized) + return; + if (mpcb->infinite_mapping_snd || mpcb->infinite_mapping_rcv || mpcb->send_infinite_mapping || mpcb->server_side || sock_flag(meta_sk, SOCK_DEAD)) @@ -1304,28 +1188,10 @@ static void full_mesh_create_subflows(struct sock *meta_sk) } } -static void full_mesh_subflow_error(struct sock *meta_sk, struct sock *sk) -{ - const struct mptcp_cb *mpcb = tcp_sk(meta_sk)->mpcb; - - if (!create_on_err) - return; - - if (mpcb->infinite_mapping_snd || mpcb->infinite_mapping_rcv || - mpcb->send_infinite_mapping || - mpcb->server_side || sock_flag(meta_sk, SOCK_DEAD)) - return; - - if (sk->sk_err != ETIMEDOUT) - return; - - full_mesh_create_subflows(meta_sk); -} - /* Called upon release_sock, if the socket was owned by the user during * a path-management event. */ -static void full_mesh_release_sock(struct sock *meta_sk) +void full_mesh_release_sock(struct sock *meta_sk) { struct mptcp_loc_addr *mptcp_local; struct mptcp_cb *mpcb = tcp_sk(meta_sk)->mpcb; @@ -1336,6 +1202,9 @@ static void full_mesh_release_sock(struct sock *meta_sk) struct hlist_node *tmp; int i; + if (!fmp->fullmesh_initialized) + return; + rcu_read_lock_bh(); mptcp_local = rcu_dereference(fm_ns->local); @@ -1447,8 +1316,8 @@ static void full_mesh_release_sock(struct sock *meta_sk) rcu_read_unlock_bh(); } -static int full_mesh_get_local_id(sa_family_t family, union inet_addr *addr, - struct net *net) +int full_mesh_get_local_id(sa_family_t family, union inet_addr *addr, + struct net *net) { struct mptcp_loc_addr *mptcp_local; const struct mptcp_fm_ns *fm_ns = fm_get_ns(net); @@ -1474,9 +1343,9 @@ static int full_mesh_get_local_id(sa_family_t family, union inet_addr *addr, return id; } -static void full_mesh_addr_signal(struct sock *sk, unsigned *size, - struct tcp_out_options *opts, - struct sk_buff *skb) +void full_mesh_addr_signal(struct sock *sk, unsigned *size, + struct tcp_out_options *opts, + struct sk_buff *skb) { const struct tcp_sock *tp = tcp_sk(sk); struct mptcp_cb *mpcb = tp->mpcb; @@ -1488,6 +1357,9 @@ static void full_mesh_addr_signal(struct sock *sk, unsigned *size, u8 unannouncedv4 = 0, unannouncedv6 = 0; bool meta_v4 = meta_sk->sk_family == AF_INET; + if (!fmp->fullmesh_initialized) + return; + mpcb->addr_signal = 0; if (likely(!fmp->add_addr)) @@ -1604,77 +1476,21 @@ static void full_mesh_addr_signal(struct sock *sk, unsigned *size, mpcb->addr_signal = !!(fmp->add_addr || fmp->remove_addrs); } -static void full_mesh_rem_raddr(struct mptcp_cb *mpcb, u8 rem_id) +void full_mesh_rem_raddr(struct mptcp_cb *mpcb, u8 rem_id) { - mptcp_v4_rem_raddress(mpcb, rem_id); - mptcp_v6_rem_raddress(mpcb, rem_id); -} - -static void full_mesh_delete_subflow(struct sock *sk) -{ - struct fullmesh_priv *fmp = fullmesh_get_priv(tcp_sk(sk)->mpcb); - struct mptcp_fm_ns *fm_ns = fm_get_ns(sock_net(sk)); - struct mptcp_loc_addr *mptcp_local; - int index, i; + struct fullmesh_priv *fmp = fullmesh_get_priv(mpcb); - if (!create_on_err) + if (!fmp->fullmesh_initialized) return; - rcu_read_lock_bh(); - mptcp_local = rcu_dereference_bh(fm_ns->local); - - if (sk->sk_family == AF_INET || mptcp_v6_is_v4_mapped(sk)) { - union inet_addr saddr; - - saddr.ip = inet_sk(sk)->inet_saddr; - index = mptcp_find_address(mptcp_local, AF_INET, &saddr, - sk->sk_bound_dev_if); - if (index < 0) - goto out; - - mptcp_for_each_bit_set(fmp->rem4_bits, i) { - struct fullmesh_rem4 *rem4 = &fmp->remaddr4[i]; - - if (rem4->addr.s_addr != sk->sk_daddr) - continue; - - if (rem4->port && rem4->port != inet_sk(sk)->inet_dport) - continue; - - rem4->bitfield &= ~(1 << index); - } -#if IS_ENABLED(CONFIG_IPV6) - } else { - union inet_addr saddr; - - saddr.in6 = inet6_sk(sk)->saddr; - index = mptcp_find_address(mptcp_local, AF_INET6, &saddr, - sk->sk_bound_dev_if); - if (index < 0) - goto out; - - mptcp_for_each_bit_set(fmp->rem6_bits, i) { - struct fullmesh_rem6 *rem6 = &fmp->remaddr6[i]; - - if (!ipv6_addr_equal(&rem6->addr, &sk->sk_v6_daddr)) - continue; - - if (rem6->port && rem6->port != inet_sk(sk)->inet_dport) - continue; - - rem6->bitfield &= ~(1 << index); - } -#endif - } - -out: - rcu_read_unlock_bh(); + mptcp_v4_rem_raddress(mpcb, rem_id); + mptcp_v6_rem_raddress(mpcb, rem_id); } /* Output /proc/net/mptcp_fullmesh */ static int mptcp_fm_seq_show(struct seq_file *seq, void *v) { - const struct net *net = seq->private; + struct net *net = seq->private; struct mptcp_loc_addr *mptcp_local; const struct mptcp_fm_ns *fm_ns = fm_get_ns(net); int i; @@ -1735,7 +1551,7 @@ static int mptcp_fm_init_net(struct net *net) INIT_LIST_HEAD(&fm_ns->events); spin_lock_init(&fm_ns->local_lock); fm_ns->net = net; - net->mptcp.path_managers[MPTCP_PM_FULLMESH] = fm_ns; + net->mptcp.fm_ns = fm_ns; return 0; err_seq_fops: @@ -1778,28 +1594,11 @@ static struct pernet_operations full_mesh_net_ops = { .exit = mptcp_fm_exit_net, }; -static struct mptcp_pm_ops full_mesh __read_mostly = { - .new_session = full_mesh_new_session, - .release_sock = full_mesh_release_sock, - .fully_established = full_mesh_create_subflows, - .new_remote_address = full_mesh_create_subflows, - .subflow_error = full_mesh_subflow_error, - .get_local_id = full_mesh_get_local_id, - .addr_signal = full_mesh_addr_signal, - .add_raddr = full_mesh_add_raddr, - .rem_raddr = full_mesh_rem_raddr, - .delete_subflow = full_mesh_delete_subflow, - .name = "fullmesh", - .owner = THIS_MODULE, -}; - /* General initialization of MPTCP_PM */ -static int __init full_mesh_register(void) +int __init full_mesh_register(void) { int ret; - BUILD_BUG_ON(sizeof(struct fullmesh_priv) > MPTCP_PM_SIZE); - ret = register_pernet_subsys(&full_mesh_net_ops); if (ret) goto out; @@ -1816,21 +1615,14 @@ static int __init full_mesh_register(void) if (ret) goto err_reg_inet6addr; #endif - - ret = mptcp_register_path_manager(&full_mesh); - if (ret) - goto err_reg_pm; - out: return ret; -err_reg_pm: #if IS_ENABLED(CONFIG_IPV6) - unregister_inet6addr_notifier(&inet6_addr_notifier); err_reg_inet6addr: -#endif unregister_netdevice_notifier(&mptcp_pm_netdev_notifier); +#endif err_reg_netdev: unregister_inetaddr_notifier(&mptcp_pm_inetaddr_notifier); err_reg_inetaddr: @@ -1838,21 +1630,3 @@ static int __init full_mesh_register(void) goto out; } -static void full_mesh_unregister(void) -{ -#if IS_ENABLED(CONFIG_IPV6) - unregister_inet6addr_notifier(&inet6_addr_notifier); -#endif - unregister_netdevice_notifier(&mptcp_pm_netdev_notifier); - unregister_inetaddr_notifier(&mptcp_pm_inetaddr_notifier); - unregister_pernet_subsys(&full_mesh_net_ops); - mptcp_unregister_path_manager(&full_mesh); -} - -module_init(full_mesh_register); -module_exit(full_mesh_unregister); - -MODULE_AUTHOR("Christoph Paasch"); -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("Full-Mesh MPTCP"); -MODULE_VERSION("0.88"); diff --git a/net/mptcp/mptcp_input.c b/net/mptcp/mptcp_input.c index 88c7a64ef26c..68c10ac0072b 100644 --- a/net/mptcp/mptcp_input.c +++ b/net/mptcp/mptcp_input.c @@ -44,15 +44,6 @@ static inline bool before64(const u64 seq1, const u64 seq2) /* is seq1 > seq2 ? */ #define after64(seq1, seq2) before64(seq2, seq1) -static inline void mptcp_become_fully_estab(struct sock *sk) -{ - tcp_sk(sk)->mptcp->fully_established = 1; - - if (is_master_tp(tcp_sk(sk)) && - tcp_sk(sk)->mpcb->pm_ops->fully_established) - tcp_sk(sk)->mpcb->pm_ops->fully_established(mptcp_meta_sk(sk)); -} - /* Similar to tcp_tso_acked without any memory accounting */ static inline int mptcp_tso_acked_reinject(const struct sock *meta_sk, struct sk_buff *skb) @@ -436,8 +427,11 @@ static int mptcp_prevalidate_skb(struct sock *sk, struct sk_buff *skb) */ if (!tp->mptcp->fully_established) { tp->mptcp->init_rcv_wnd -= skb->len; - if (tp->mptcp->init_rcv_wnd < 0) - mptcp_become_fully_estab(sk); + if (tp->mptcp->init_rcv_wnd < 0) { + tp->mptcp->fully_established = 1; + if (is_master_tp(tp)) + full_mesh_create_subflows(tp->meta_sk); + } } return 0; @@ -1243,11 +1237,14 @@ static void mptcp_data_ack(struct sock *sk, const struct sk_buff *skb) return; if (unlikely(!tp->mptcp->fully_established) && - tp->mptcp->snt_isn + 1 != TCP_SKB_CB(skb)->ack_seq) + tp->mptcp->snt_isn + 1 != TCP_SKB_CB(skb)->ack_seq) { /* As soon as a subflow-data-ack (not acking syn, thus snt_isn + 1) * includes a data-ack, we are fully established */ - mptcp_become_fully_estab(sk); + tp->mptcp->fully_established = 1; + if (is_master_tp(tp)) + full_mesh_create_subflows(meta_sk); + } /* After we did the subflow-only processing (stopping timer and marking * subflow as established), check if we can proceed with MPTCP-level @@ -1757,8 +1754,7 @@ static void mptcp_handle_add_addr(const unsigned char *ptr, struct sock *sk) return; } - if (mpcb->pm_ops->add_raddr) - mpcb->pm_ops->add_raddr(mpcb, &addr, family, port, mpadd->addr_id); + full_mesh_add_raddr(mpcb, &addr, family, port, mpadd->addr_id); MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ADDADDRRX); } @@ -1773,8 +1769,7 @@ static void mptcp_handle_rem_addr(const unsigned char *ptr, struct sock *sk) for (i = 0; i <= mprem->len - MPTCP_SUB_LEN_REMOVE_ADDR; i++) { rem_id = (&mprem->addrs_id)[i]; - if (mpcb->pm_ops->rem_raddr) - mpcb->pm_ops->rem_raddr(mpcb, rem_id); + full_mesh_rem_raddr(mpcb, rem_id); mptcp_send_reset_rem_id(mpcb, rem_id); MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_REMADDRSUB); @@ -1877,8 +1872,7 @@ static inline void mptcp_path_array_check(struct sock *meta_sk) if (unlikely(mpcb->list_rcvd)) { mpcb->list_rcvd = 0; - if (mpcb->pm_ops->new_remote_address) - mpcb->pm_ops->new_remote_address(meta_sk); + full_mesh_create_subflows(meta_sk); } } diff --git a/net/mptcp/mptcp_ipv4.c b/net/mptcp/mptcp_ipv4.c index 94bc1416af77..8153b1d7e4bf 100644 --- a/net/mptcp/mptcp_ipv4.c +++ b/net/mptcp/mptcp_ipv4.c @@ -105,7 +105,7 @@ static int mptcp_v4_join_init_req(struct request_sock *req, const struct sock *m tcp_hdr(skb)->source, tcp_hdr(skb)->dest); addr.ip = inet_rsk(req)->ir_loc_addr; - loc_id = mpcb->pm_ops->get_local_id(AF_INET, &addr, sock_net(meta_sk)); + loc_id = full_mesh_get_local_id(AF_INET, &addr, sock_net(meta_sk)); if (loc_id == -1) return -1; mtreq->loc_id = loc_id; diff --git a/net/mptcp/mptcp_ipv6.c b/net/mptcp/mptcp_ipv6.c index 297dd872fbfb..19ce256a7679 100644 --- a/net/mptcp/mptcp_ipv6.c +++ b/net/mptcp/mptcp_ipv6.c @@ -134,7 +134,7 @@ static int mptcp_v6_join_init_req(struct request_sock *req, const struct sock *m tcp_hdr(skb)->source, tcp_hdr(skb)->dest); addr.in6 = inet_rsk(req)->ir_v6_loc_addr; - loc_id = mpcb->pm_ops->get_local_id(AF_INET6, &addr, sock_net(meta_sk)); + loc_id = full_mesh_get_local_id(AF_INET6, &addr, sock_net(meta_sk)); if (loc_id == -1) return -1; mtreq->loc_id = loc_id; diff --git a/net/mptcp/mptcp_output.c b/net/mptcp/mptcp_output.c index 00a3674ad693..b3978a7e9254 100644 --- a/net/mptcp/mptcp_output.c +++ b/net/mptcp/mptcp_output.c @@ -709,9 +709,9 @@ void mptcp_established_options(struct sock *sk, struct sk_buff *skb, *size += MPTCP_SUB_LEN_JOIN_ACK_ALIGN; } - if (unlikely(mpcb->addr_signal) && mpcb->pm_ops->addr_signal && + if (unlikely(mpcb->addr_signal) && mpcb->mptcp_ver >= MPTCP_VERSION_1 && skb && !mptcp_is_data_seq(skb)) { - mpcb->pm_ops->addr_signal(sk, size, opts, skb); + full_mesh_addr_signal(sk, size, opts, skb); if (opts->add_addr_v6) /* Skip subsequent options */ @@ -734,9 +734,9 @@ void mptcp_established_options(struct sock *sk, struct sk_buff *skb, *size += MPTCP_SUB_LEN_DSS_ALIGN; } - if (unlikely(mpcb->addr_signal) && mpcb->pm_ops->addr_signal && + if (unlikely(mpcb->addr_signal) && mpcb->mptcp_ver < MPTCP_VERSION_1) - mpcb->pm_ops->addr_signal(sk, size, opts, skb); + full_mesh_addr_signal(sk, size, opts, skb); return; } diff --git a/net/mptcp/mptcp_pm.c b/net/mptcp/mptcp_pm.c deleted file mode 100644 index fbdf8eb58d17..000000000000 --- a/net/mptcp/mptcp_pm.c +++ /dev/null @@ -1,226 +0,0 @@ -/* - * MPTCP implementation - MPTCP-subflow-management - * - * Initial Design & Implementation: - * Sébastien Barré <sebastien.barre(a)uclouvain.be&gt; - * - * Current Maintainer & Author: - * Christoph Paasch <christoph.paasch(a)uclouvain.be&gt; - * - * Additional authors: - * Jaakko Korkeaniemi <jaakko.korkeaniemi(a)aalto.fi&gt; - * Gregory Detal <gregory.detal(a)uclouvain.be&gt; - * Fabien Duchêne <fabien.duchene(a)uclouvain.be&gt; - * Andreas Seelinger <Andreas.Seelinger(a)rwth-aachen.de&gt; - * Lavkesh Lahngir <lavkesh51(a)gmail.com&gt; - * Andreas Ripke <ripke(a)neclab.eu&gt; - * Vlad Dogaru <vlad.dogaru(a)intel.com&gt; - * Octavian Purdila <octavian.purdila(a)intel.com&gt; - * John Ronan <jronan(a)tssg.org&gt; - * Catalin Nicutar <catalin.nicutar(a)gmail.com&gt; - * Brandon Heller <brandonh(a)stanford.edu&gt; - * - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - - -#include <linux/module.h> -#include <net/mptcp.h> - -static DEFINE_SPINLOCK(mptcp_pm_list_lock); -static LIST_HEAD(mptcp_pm_list); - -static int mptcp_default_id(sa_family_t family, union inet_addr *addr, - struct net *net) -{ - return 0; -} - -struct mptcp_pm_ops mptcp_pm_default = { - .get_local_id = mptcp_default_id, /* We do not care */ - .name = "default", - .owner = THIS_MODULE, -}; - -static struct mptcp_pm_ops *mptcp_pm_find(const char *name) -{ - struct mptcp_pm_ops *e; - - list_for_each_entry_rcu(e, &mptcp_pm_list, list) { - if (strcmp(e->name, name) == 0) - return e; - } - - return NULL; -} - -int mptcp_register_path_manager(struct mptcp_pm_ops *pm) -{ - int ret = 0; - - if (!pm->get_local_id) - return -EINVAL; - - spin_lock(&mptcp_pm_list_lock); - if (mptcp_pm_find(pm->name)) { - pr_notice("%s already registered\n", pm->name); - ret = -EEXIST; - } else { - list_add_tail_rcu(&pm->list, &mptcp_pm_list); - pr_info("%s registered\n", pm->name); - } - spin_unlock(&mptcp_pm_list_lock); - - return ret; -} -EXPORT_SYMBOL_GPL(mptcp_register_path_manager); - -void mptcp_unregister_path_manager(struct mptcp_pm_ops *pm) -{ - spin_lock(&mptcp_pm_list_lock); - list_del_rcu(&pm->list); - spin_unlock(&mptcp_pm_list_lock); - - /* Wait for outstanding readers to complete before the - * module gets removed entirely. - * - * A try_module_get() should fail by now as our module is - * in "going" state since no refs are held anymore and - * module_exit() handler being called. - */ - synchronize_rcu(); -} -EXPORT_SYMBOL_GPL(mptcp_unregister_path_manager); - -void mptcp_get_default_path_manager(char *name) -{ - struct mptcp_pm_ops *pm; - - BUG_ON(list_empty(&mptcp_pm_list)); - - rcu_read_lock(); - pm = list_entry(mptcp_pm_list.next, struct mptcp_pm_ops, list); - strncpy(name, pm->name, MPTCP_PM_NAME_MAX); - rcu_read_unlock(); -} - -int mptcp_set_default_path_manager(const char *name) -{ - struct mptcp_pm_ops *pm; - int ret = -ENOENT; - - spin_lock(&mptcp_pm_list_lock); - pm = mptcp_pm_find(name); -#ifdef CONFIG_MODULES - if (!pm && capable(CAP_NET_ADMIN)) { - spin_unlock(&mptcp_pm_list_lock); - - request_module("mptcp_%s", name); - spin_lock(&mptcp_pm_list_lock); - pm = mptcp_pm_find(name); - } -#endif - - if (pm) { - list_move(&pm->list, &mptcp_pm_list); - ret = 0; - } else { - pr_info("%s is not available\n", name); - } - spin_unlock(&mptcp_pm_list_lock); - - return ret; -} - -static struct mptcp_pm_ops *__mptcp_pm_find_autoload(const char *name) -{ - struct mptcp_pm_ops *pm = mptcp_pm_find(name); -#ifdef CONFIG_MODULES - if (!pm && capable(CAP_NET_ADMIN)) { - rcu_read_unlock(); - request_module("mptcp_%s", name); - rcu_read_lock(); - pm = mptcp_pm_find(name); - } -#endif - return pm; -} - -void mptcp_init_path_manager(struct mptcp_cb *mpcb) -{ - struct mptcp_pm_ops *pm; - struct sock *meta_sk = mpcb->meta_sk; - struct tcp_sock *meta_tp = tcp_sk(meta_sk); - - rcu_read_lock(); - /* if path manager was set using socket option */ - if (meta_tp->mptcp_pm_setsockopt) { - pm = __mptcp_pm_find_autoload(meta_tp->mptcp_pm_name); - if (pm && try_module_get(pm->owner)) { - mpcb->pm_ops = pm; - goto out; - } - } - - list_for_each_entry_rcu(pm, &mptcp_pm_list, list) { - if (try_module_get(pm->owner)) { - mpcb->pm_ops = pm; - break; - } - } -out: - rcu_read_unlock(); -} - -/* Change path manager for socket */ -int mptcp_set_path_manager(struct sock *sk, const char *name) -{ - struct mptcp_pm_ops *pm; - int err = 0; - - rcu_read_lock(); - pm = __mptcp_pm_find_autoload(name); - - if (!pm) { - err = -ENOENT; - } else if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) { - err = -EPERM; - } else { - strcpy(tcp_sk(sk)->mptcp_pm_name, name); - tcp_sk(sk)->mptcp_pm_setsockopt = 1; - } - rcu_read_unlock(); - - return err; -} - -/* Manage refcounts on socket close. */ -void mptcp_cleanup_path_manager(struct mptcp_cb *mpcb) -{ - module_put(mpcb->pm_ops->owner); -} - -/* Fallback to the default path-manager. */ -void mptcp_fallback_default(struct mptcp_cb *mpcb) -{ - struct mptcp_pm_ops *pm; - - mptcp_cleanup_path_manager(mpcb); - pm = mptcp_pm_find("default"); - - /* Cannot fail - it's the default module */ - try_module_get(pm->owner); - mpcb->pm_ops = pm; -} -EXPORT_SYMBOL_GPL(mptcp_fallback_default); - -/* Set default value from kernel configuration at bootup */ -static int __init mptcp_path_manager_default(void) -{ - return mptcp_set_default_path_manager(CONFIG_DEFAULT_MPTCP_PM); -} -late_initcall(mptcp_path_manager_default); -- 2.16.2