diff options
Diffstat (limited to 'net/mptcp/options.c')
| -rw-r--r-- | net/mptcp/options.c | 426 |
1 files changed, 301 insertions, 125 deletions
diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 645dd984fef0..f24ae7d40e88 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -26,6 +26,7 @@ static void mptcp_parse_option(const struct sk_buff *skb, { u8 subtype = *ptr >> 4; int expected_opsize; + u16 subopt; u8 version; u8 flags; u8 i; @@ -38,11 +39,15 @@ static void mptcp_parse_option(const struct sk_buff *skb, expected_opsize = TCPOLEN_MPTCP_MPC_ACK_DATA; else expected_opsize = TCPOLEN_MPTCP_MPC_ACK; + subopt = OPTION_MPTCP_MPC_ACK; } else { - if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_ACK) + if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_ACK) { expected_opsize = TCPOLEN_MPTCP_MPC_SYNACK; - else + subopt = OPTION_MPTCP_MPC_SYNACK; + } else { expected_opsize = TCPOLEN_MPTCP_MPC_SYN; + subopt = OPTION_MPTCP_MPC_SYN; + } } /* Cfr RFC 8684 Section 3.3.0: @@ -85,7 +90,7 @@ static void mptcp_parse_option(const struct sk_buff *skb, mp_opt->deny_join_id0 = !!(flags & MPTCP_CAP_DENY_JOIN_ID0); - mp_opt->suboptions |= OPTIONS_MPTCP_MPC; + mp_opt->suboptions |= subopt; if (opsize >= TCPOLEN_MPTCP_MPC_SYNACK) { mp_opt->sndr_key = get_unaligned_be64(ptr); ptr += 8; @@ -107,55 +112,50 @@ static void mptcp_parse_option(const struct sk_buff *skb, ptr += 2; } if (opsize == TCPOLEN_MPTCP_MPC_ACK_DATA_CSUM) { - mp_opt->csum = (__force __sum16)get_unaligned_be16(ptr); + mp_opt->csum = get_unaligned((__force __sum16 *)ptr); mp_opt->suboptions |= OPTION_MPTCP_CSUMREQD; ptr += 2; } - pr_debug("MP_CAPABLE version=%x, flags=%x, optlen=%d sndr=%llu, rcvr=%llu len=%d csum=%u", + pr_debug("MP_CAPABLE version=%x, flags=%x, optlen=%d sndr=%llu, rcvr=%llu len=%d csum=%u\n", version, flags, opsize, mp_opt->sndr_key, mp_opt->rcvr_key, mp_opt->data_len, mp_opt->csum); break; case MPTCPOPT_MP_JOIN: - mp_opt->suboptions |= OPTIONS_MPTCP_MPJ; if (opsize == TCPOLEN_MPTCP_MPJ_SYN) { + mp_opt->suboptions |= OPTION_MPTCP_MPJ_SYN; mp_opt->backup = *ptr++ & MPTCPOPT_BACKUP; mp_opt->join_id = *ptr++; mp_opt->token = get_unaligned_be32(ptr); ptr += 4; mp_opt->nonce = get_unaligned_be32(ptr); ptr += 4; - pr_debug("MP_JOIN bkup=%u, id=%u, token=%u, nonce=%u", + pr_debug("MP_JOIN bkup=%u, id=%u, token=%u, nonce=%u\n", mp_opt->backup, mp_opt->join_id, mp_opt->token, mp_opt->nonce); } else if (opsize == TCPOLEN_MPTCP_MPJ_SYNACK) { + mp_opt->suboptions |= OPTION_MPTCP_MPJ_SYNACK; mp_opt->backup = *ptr++ & MPTCPOPT_BACKUP; mp_opt->join_id = *ptr++; mp_opt->thmac = get_unaligned_be64(ptr); ptr += 8; mp_opt->nonce = get_unaligned_be32(ptr); ptr += 4; - pr_debug("MP_JOIN bkup=%u, id=%u, thmac=%llu, nonce=%u", + pr_debug("MP_JOIN bkup=%u, id=%u, thmac=%llu, nonce=%u\n", mp_opt->backup, mp_opt->join_id, mp_opt->thmac, mp_opt->nonce); } else if (opsize == TCPOLEN_MPTCP_MPJ_ACK) { + mp_opt->suboptions |= OPTION_MPTCP_MPJ_ACK; ptr += 2; memcpy(mp_opt->hmac, ptr, MPTCPOPT_HMAC_LEN); - pr_debug("MP_JOIN hmac"); - } else { - mp_opt->suboptions &= ~OPTIONS_MPTCP_MPJ; + pr_debug("MP_JOIN hmac\n"); } break; case MPTCPOPT_DSS: - pr_debug("DSS"); + pr_debug("DSS\n"); ptr++; - /* we must clear 'mpc_map' be able to detect MP_CAPABLE - * map vs DSS map in mptcp_incoming_options(), and reconstruct - * map info accordingly - */ - mp_opt->mpc_map = 0; flags = (*ptr++) & MPTCP_DSS_FLAG_MASK; mp_opt->data_fin = (flags & MPTCP_DSS_DATA_FIN) != 0; mp_opt->dsn64 = (flags & MPTCP_DSS_DSN64) != 0; @@ -163,7 +163,7 @@ static void mptcp_parse_option(const struct sk_buff *skb, mp_opt->ack64 = (flags & MPTCP_DSS_ACK64) != 0; mp_opt->use_ack = (flags & MPTCP_DSS_HAS_ACK); - pr_debug("data_fin=%d dsn64=%d use_map=%d ack64=%d use_ack=%d", + pr_debug("data_fin=%d dsn64=%d use_map=%d ack64=%d use_ack=%d\n", mp_opt->data_fin, mp_opt->dsn64, mp_opt->use_map, mp_opt->ack64, mp_opt->use_ack); @@ -201,7 +201,7 @@ static void mptcp_parse_option(const struct sk_buff *skb, ptr += 4; } - pr_debug("data_ack=%llu", mp_opt->data_ack); + pr_debug("data_ack=%llu\n", mp_opt->data_ack); } if (mp_opt->use_map) { @@ -221,11 +221,11 @@ static void mptcp_parse_option(const struct sk_buff *skb, if (opsize == expected_opsize + TCPOLEN_MPTCP_DSS_CHECKSUM) { mp_opt->suboptions |= OPTION_MPTCP_CSUMREQD; - mp_opt->csum = (__force __sum16)get_unaligned_be16(ptr); + mp_opt->csum = get_unaligned((__force __sum16 *)ptr); ptr += 2; } - pr_debug("data_seq=%llu subflow_seq=%u data_len=%u csum=%d:%u", + pr_debug("data_seq=%llu subflow_seq=%u data_len=%u csum=%d:%u\n", mp_opt->data_seq, mp_opt->subflow_seq, mp_opt->data_len, !!(mp_opt->suboptions & OPTION_MPTCP_CSUMREQD), mp_opt->csum); @@ -287,7 +287,7 @@ static void mptcp_parse_option(const struct sk_buff *skb, mp_opt->ahmac = get_unaligned_be64(ptr); ptr += 8; } - pr_debug("ADD_ADDR%s: id=%d, ahmac=%llu, echo=%d, port=%d", + pr_debug("ADD_ADDR%s: id=%d, ahmac=%llu, echo=%d, port=%d\n", (mp_opt->addr.family == AF_INET6) ? "6" : "", mp_opt->addr.id, mp_opt->ahmac, mp_opt->echo, ntohs(mp_opt->addr.port)); break; @@ -303,7 +303,7 @@ static void mptcp_parse_option(const struct sk_buff *skb, mp_opt->rm_list.nr = opsize - TCPOLEN_MPTCP_RM_ADDR_BASE; for (i = 0; i < mp_opt->rm_list.nr; i++) mp_opt->rm_list.ids[i] = *ptr++; - pr_debug("RM_ADDR: rm_list_nr=%d", mp_opt->rm_list.nr); + pr_debug("RM_ADDR: rm_list_nr=%d\n", mp_opt->rm_list.nr); break; case MPTCPOPT_MP_PRIO: @@ -312,7 +312,7 @@ static void mptcp_parse_option(const struct sk_buff *skb, mp_opt->suboptions |= OPTION_MPTCP_PRIO; mp_opt->backup = *ptr++ & MPTCP_PRIO_BKUP; - pr_debug("MP_PRIO: prio=%d", mp_opt->backup); + pr_debug("MP_PRIO: prio=%d\n", mp_opt->backup); break; case MPTCPOPT_MP_FASTCLOSE: @@ -323,6 +323,7 @@ static void mptcp_parse_option(const struct sk_buff *skb, mp_opt->rcvr_key = get_unaligned_be64(ptr); ptr += 8; mp_opt->suboptions |= OPTION_MPTCP_FASTCLOSE; + pr_debug("MP_FASTCLOSE: recv_key=%llu\n", mp_opt->rcvr_key); break; case MPTCPOPT_RST: @@ -336,6 +337,8 @@ static void mptcp_parse_option(const struct sk_buff *skb, flags = *ptr++; mp_opt->reset_transient = flags & MPTCP_RST_TRANSIENT; mp_opt->reset_reason = *ptr; + pr_debug("MP_RST: transient=%u reason=%u\n", + mp_opt->reset_transient, mp_opt->reset_reason); break; case MPTCPOPT_MP_FAIL: @@ -345,7 +348,7 @@ static void mptcp_parse_option(const struct sk_buff *skb, ptr += 2; mp_opt->suboptions |= OPTION_MPTCP_FAIL; mp_opt->fail_seq = get_unaligned_be64(ptr); - pr_debug("MP_FAIL: data_seq=%llu", mp_opt->fail_seq); + pr_debug("MP_FAIL: data_seq=%llu\n", mp_opt->fail_seq); break; default: @@ -353,16 +356,18 @@ static void mptcp_parse_option(const struct sk_buff *skb, } } -void mptcp_get_options(const struct sock *sk, - const struct sk_buff *skb, +void mptcp_get_options(const struct sk_buff *skb, struct mptcp_options_received *mp_opt) { const struct tcphdr *th = tcp_hdr(skb); const unsigned char *ptr; int length; - /* initialize option status */ - mp_opt->suboptions = 0; + /* Ensure that casting the whole status to u32 is efficient and safe */ + BUILD_BUG_ON(sizeof_field(struct mptcp_options_received, status) != sizeof(u32)); + BUILD_BUG_ON(!IS_ALIGNED(offsetof(struct mptcp_options_received, status), + sizeof(u32))); + *(u32 *)&mp_opt->status = 0; length = (th->doff * 4) - sizeof(struct tcphdr); ptr = (const unsigned char *)(th + 1); @@ -409,7 +414,7 @@ bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb, *size = TCPOLEN_MPTCP_MPC_SYN; return true; } else if (subflow->request_join) { - pr_debug("remote_token=%u, nonce=%u", subflow->remote_token, + pr_debug("remote_token=%u, nonce=%u\n", subflow->remote_token, subflow->local_nonce); opts->suboptions = OPTION_MPTCP_MPJ_SYN; opts->join_id = subflow->local_id; @@ -427,7 +432,6 @@ static void clear_3rdack_retransmission(struct sock *sk) struct inet_connection_sock *icsk = inet_csk(sk); sk_stop_timer(sk, &icsk->icsk_delack_timer); - icsk->icsk_ack.timeout = 0; icsk->icsk_ack.ato = 0; icsk->icsk_ack.pending &= ~(ICSK_ACK_SCHED | ICSK_ACK_TIMER); } @@ -435,7 +439,6 @@ static void clear_3rdack_retransmission(struct sock *sk) static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb, bool snd_data_fin_enable, unsigned int *size, - unsigned int remaining, struct mptcp_out_options *opts) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); @@ -454,7 +457,7 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb, return false; /* MPC/MPJ needed only on 3rd ack packet, DATA_FIN and TCP shutdown take precedence */ - if (subflow->fully_established || snd_data_fin_enable || + if (READ_ONCE(subflow->fully_established) || snd_data_fin_enable || subflow->snd_isn != TCP_SKB_CB(skb)->seq || sk->sk_state != TCP_ESTABLISHED) return false; @@ -493,7 +496,7 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb, *size = TCPOLEN_MPTCP_MPC_ACK; } - pr_debug("subflow=%p, local_key=%llu, remote_key=%llu map_len=%d", + pr_debug("subflow=%p, local_key=%llu, remote_key=%llu map_len=%d\n", subflow, subflow->local_key, subflow->remote_key, data_len); @@ -502,7 +505,7 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb, opts->suboptions = OPTION_MPTCP_MPJ_ACK; memcpy(opts->hmac, subflow->hmac, MPTCPOPT_HMAC_LEN); *size = TCPOLEN_MPTCP_MPJ_ACK; - pr_debug("subflow=%p", subflow); + pr_debug("subflow=%p\n", subflow); /* we can use the full delegate action helper only from BH context * If we are in process context - sk is flushing the backlog at @@ -549,7 +552,6 @@ static void mptcp_write_data_fin(struct mptcp_subflow_context *subflow, static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, bool snd_data_fin_enable, unsigned int *size, - unsigned int remaining, struct mptcp_out_options *opts) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); @@ -573,7 +575,6 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, opts->ext_copy = *mpext; } - remaining -= map_size; dss_size = map_size; if (skb && snd_data_fin_enable) mptcp_write_data_fin(subflow, skb, &opts->ext_copy); @@ -602,7 +603,6 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, } opts->ext_copy.use_ack = 1; opts->suboptions = OPTION_MPTCP_DSS; - WRITE_ONCE(msk->old_wspace, __mptcp_space((struct sock *)msk)); /* Add kind/length/subtype/flag overhead if mapping is not populated */ if (dss_size == 0) @@ -650,8 +650,8 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff * struct mptcp_sock *msk = mptcp_sk(subflow->conn); bool drop_other_suboptions = false; unsigned int opt_size = *size; + struct mptcp_addr_info addr; bool echo; - bool port; int len; /* add addr will strip the existing options, be sure to avoid breaking @@ -659,19 +659,26 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff * */ if (!mptcp_pm_should_add_signal(msk) || (opts->suboptions & (OPTION_MPTCP_MPJ_ACK | OPTION_MPTCP_MPC_ACK)) || - !mptcp_pm_add_addr_signal(msk, skb, opt_size, remaining, &opts->addr, - &echo, &port, &drop_other_suboptions)) + !mptcp_pm_add_addr_signal(msk, skb, opt_size, remaining, &addr, + &echo, &drop_other_suboptions)) return false; + /* + * Later on, mptcp_write_options() will enforce mutually exclusion with + * DSS, bail out if such option is set and we can't drop it. + */ if (drop_other_suboptions) remaining += opt_size; - len = mptcp_add_addr_len(opts->addr.family, echo, port); + else if (opts->suboptions & OPTION_MPTCP_DSS) + return false; + + len = mptcp_add_addr_len(addr.family, echo, !!addr.port); if (remaining < len) return false; *size = len; if (drop_other_suboptions) { - pr_debug("drop other suboptions"); + pr_debug("drop other suboptions\n"); opts->suboptions = 0; /* note that e.g. DSS could have written into the memory @@ -682,13 +689,17 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff * opts->ahmac = 0; *size -= opt_size; } + opts->addr = addr; opts->suboptions |= OPTION_MPTCP_ADD_ADDR; if (!echo) { - opts->ahmac = add_addr_generate_hmac(msk->local_key, - msk->remote_key, + MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ADDADDRTX); + opts->ahmac = add_addr_generate_hmac(READ_ONCE(msk->local_key), + READ_ONCE(msk->remote_key), &opts->addr); + } else { + MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ECHOADDTX); } - pr_debug("addr_id=%d, ahmac=%llu, echo=%d, port=%d", + pr_debug("addr_id=%d, ahmac=%llu, echo=%d, port=%d\n", opts->addr.id, opts->ahmac, echo, ntohs(opts->addr.port)); return true; @@ -719,8 +730,8 @@ static bool mptcp_established_options_rm_addr(struct sock *sk, opts->rm_list = rm_list; for (i = 0; i < opts->rm_list.nr; i++) - pr_debug("rm_list_ids[%d]=%d", i, opts->rm_list.ids[i]); - + pr_debug("rm_list_ids[%d]=%d\n", i, opts->rm_list.ids[i]); + MPTCP_ADD_STATS(sock_net(sk), MPTCP_MIB_RMADDRTX, opts->rm_list.nr); return true; } @@ -745,7 +756,7 @@ static bool mptcp_established_options_mp_prio(struct sock *sk, opts->suboptions |= OPTION_MPTCP_PRIO; opts->backup = subflow->request_bkup; - pr_debug("prio=%d", opts->backup); + pr_debug("prio=%d\n", opts->backup); return true; } @@ -764,6 +775,7 @@ static noinline bool mptcp_established_options_rst(struct sock *sk, struct sk_bu opts->suboptions |= OPTION_MPTCP_RST; opts->reset_transient = subflow->reset_transient; opts->reset_reason = subflow->reset_reason; + MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPRSTTX); return true; } @@ -784,9 +796,10 @@ static bool mptcp_established_options_fastclose(struct sock *sk, *size = TCPOLEN_MPTCP_FASTCLOSE; opts->suboptions |= OPTION_MPTCP_FASTCLOSE; - opts->rcvr_key = msk->remote_key; + opts->rcvr_key = READ_ONCE(msk->remote_key); - pr_debug("FASTCLOSE key=%llu", opts->rcvr_key); + pr_debug("FASTCLOSE key=%llu\n", opts->rcvr_key); + MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPFASTCLOSETX); return true; } @@ -807,7 +820,8 @@ static bool mptcp_established_options_mp_fail(struct sock *sk, opts->suboptions |= OPTION_MPTCP_FAIL; opts->fail_seq = subflow->map_seq; - pr_debug("MP_FAIL fail_seq=%llu", opts->fail_seq); + pr_debug("MP_FAIL fail_seq=%llu\n", opts->fail_seq); + MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPFAILTX); return true; } @@ -824,8 +838,11 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb, opts->suboptions = 0; - if (unlikely(__mptcp_check_fallback(msk))) - return false; + /* Force later mptcp_write_options(), but do not use any actual + * option space. + */ + if (unlikely(__mptcp_check_fallback(msk) && !mptcp_check_infinite_map(skb))) + return true; if (unlikely(skb && TCP_SKB_CB(skb)->tcp_flags & TCPHDR_RST)) { if (mptcp_established_options_fastclose(sk, &opt_size, remaining, opts) || @@ -842,9 +859,9 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb, } snd_data_fin = mptcp_data_fin_enabled(msk); - if (mptcp_established_options_mp(sk, skb, snd_data_fin, &opt_size, remaining, opts)) + if (mptcp_established_options_mp(sk, skb, snd_data_fin, &opt_size, opts)) ret = true; - else if (mptcp_established_options_dss(sk, skb, snd_data_fin, &opt_size, remaining, opts)) { + else if (mptcp_established_options_dss(sk, skb, snd_data_fin, &opt_size, opts)) { unsigned int mp_fail_size; ret = true; @@ -894,16 +911,16 @@ bool mptcp_synack_options(const struct request_sock *req, unsigned int *size, opts->csum_reqd = subflow_req->csum_reqd; opts->allow_join_id0 = subflow_req->allow_join_id0; *size = TCPOLEN_MPTCP_MPC_SYNACK; - pr_debug("subflow_req=%p, local_key=%llu", + pr_debug("subflow_req=%p, local_key=%llu\n", subflow_req, subflow_req->local_key); return true; } else if (subflow_req->mp_join) { opts->suboptions = OPTION_MPTCP_MPJ_SYNACK; - opts->backup = subflow_req->backup; + opts->backup = subflow_req->request_bkup; opts->join_id = subflow_req->local_id; opts->thmac = subflow_req->thmac; opts->nonce = subflow_req->local_nonce; - pr_debug("req=%p, bkup=%u, id=%u, thmac=%llu, nonce=%u", + pr_debug("req=%p, bkup=%u, id=%u, thmac=%llu, nonce=%u\n", subflow_req, opts->backup, opts->join_id, opts->thmac, opts->nonce); *size = TCPOLEN_MPTCP_MPJ_SYNACK; @@ -920,7 +937,7 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *ssk, /* here we can process OoO, in-window pkts, only in-sequence 4th ack * will make the subflow fully established */ - if (likely(subflow->fully_established)) { + if (likely(READ_ONCE(subflow->fully_established))) { /* on passive sockets, check for 3rd ack retransmission * note that msk is always set by subflow_syn_recv_sock() * for mp_join subflows @@ -928,9 +945,9 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *ssk, if (TCP_SKB_CB(skb)->seq == subflow->ssn_offset + 1 && TCP_SKB_CB(skb)->end_seq == TCP_SKB_CB(skb)->seq && subflow->mp_join && (mp_opt->suboptions & OPTIONS_MPTCP_MPJ) && - READ_ONCE(msk->pm.server_side)) + !subflow->request_join) tcp_send_ack(ssk); - goto fully_established; + goto check_notify; } /* we must process OoO packets before the first subflow is fully @@ -941,17 +958,19 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *ssk, if (TCP_SKB_CB(skb)->seq != subflow->ssn_offset + 1) { if (subflow->mp_join) goto reset; + if (subflow->is_mptfo && mp_opt->suboptions & OPTION_MPTCP_MPC_ACK) + goto set_fully_established; return subflow->mp_capable; } - if (((mp_opt->suboptions & OPTION_MPTCP_DSS) && mp_opt->use_ack) || - ((mp_opt->suboptions & OPTION_MPTCP_ADD_ADDR) && !mp_opt->echo)) { + if (subflow->remote_key_valid && + (((mp_opt->suboptions & OPTION_MPTCP_DSS) && mp_opt->use_ack) || + ((mp_opt->suboptions & OPTION_MPTCP_ADD_ADDR) && + (!mp_opt->echo || subflow->mp_join)))) { /* subflows are fully established as soon as we get any * additional ack, including ADD_ADDR. */ - subflow->fully_established = 1; - WRITE_ONCE(msk->fully_established, true); - goto fully_established; + goto set_fully_established; } /* If the first established packet does not contain MP_CAPABLE + data @@ -962,19 +981,25 @@ static bool check_fully_established(struct mptcp_sock *msk, struct sock *ssk, if (subflow->mp_join) goto reset; subflow->mp_capable = 0; - pr_fallback(msk); - __mptcp_do_fallback(msk); + if (!mptcp_try_fallback(ssk, MPTCP_MIB_MPCAPABLEDATAFALLBACK)) { + MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_FALLBACKFAILED); + goto reset; + } return false; } + if (unlikely(!READ_ONCE(msk->pm.server_side))) + pr_warn_once("bogus mpc option on established client sk"); + +set_fully_established: if (mp_opt->deny_join_id0) WRITE_ONCE(msk->pm.remote_deny_join_id0, true); - if (unlikely(!READ_ONCE(msk->pm.server_side))) - pr_warn_once("bogus mpc option on established client sk"); - mptcp_subflow_fully_established(subflow, mp_opt); + mptcp_data_lock((struct sock *)msk); + __mptcp_subflow_fully_established(msk, subflow, mp_opt); + mptcp_data_unlock((struct sock *)msk); -fully_established: +check_notify: /* if the subflow is not already linked into the conn_list, we can't * notify the PM: this subflow is still on the listener queue * and the PM possibly acquiring the subflow lock could race with @@ -988,7 +1013,7 @@ fully_established: clear_3rdack_retransmission(ssk); mptcp_pm_subflow_established(msk); } else { - mptcp_pm_fully_established(msk, ssk, GFP_ATOMIC); + mptcp_pm_fully_established(msk, ssk); } return true; @@ -1013,6 +1038,37 @@ u64 __mptcp_expand_seq(u64 old_seq, u64 cur_seq) return cur_seq; } +static void __mptcp_snd_una_update(struct mptcp_sock *msk, u64 new_snd_una) +{ + msk->bytes_acked += new_snd_una - msk->snd_una; + WRITE_ONCE(msk->snd_una, new_snd_una); +} + +static void rwin_update(struct mptcp_sock *msk, struct sock *ssk, + struct sk_buff *skb) +{ + struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); + struct tcp_sock *tp = tcp_sk(ssk); + u64 mptcp_rcv_wnd; + + /* Avoid touching extra cachelines if TCP is going to accept this + * skb without filling the TCP-level window even with a possibly + * outdated mptcp-level rwin. + */ + if (!skb->len || skb->len < tcp_receive_window(tp)) + return; + + mptcp_rcv_wnd = atomic64_read(&msk->rcv_wnd_sent); + if (!after64(mptcp_rcv_wnd, subflow->rcv_wnd_sent)) + return; + + /* Some other subflow grew the mptcp-level rwin since rcv_wup, + * resync. + */ + tp->rcv_wnd += mptcp_rcv_wnd - subflow->rcv_wnd_sent; + subflow->rcv_wnd_sent = mptcp_rcv_wnd; +} + static void ack_update_msk(struct mptcp_sock *msk, struct sock *ssk, struct mptcp_options_received *mp_opt) @@ -1037,21 +1093,22 @@ static void ack_update_msk(struct mptcp_sock *msk, new_wnd_end = new_snd_una + tcp_sk(ssk)->snd_wnd; if (after64(new_wnd_end, msk->wnd_end)) - msk->wnd_end = new_wnd_end; + WRITE_ONCE(msk->wnd_end, new_wnd_end); /* this assumes mptcp_incoming_options() is invoked after tcp_ack() */ - if (after64(msk->wnd_end, READ_ONCE(msk->snd_nxt))) + if (after64(msk->wnd_end, snd_nxt)) __mptcp_check_push(sk, ssk); if (after64(new_snd_una, old_snd_una)) { - msk->snd_una = new_snd_una; + __mptcp_snd_una_update(msk, new_snd_una); __mptcp_data_acked(sk); } + msk->last_ack_recv = tcp_jiffies32; mptcp_data_unlock(sk); trace_ack_update_msk(mp_opt->data_ack, old_snd_una, new_snd_una, - new_wnd_end, msk->wnd_end); + new_wnd_end, READ_ONCE(msk->wnd_end)); } bool mptcp_update_rcv_data_fin(struct mptcp_sock *msk, u64 data_fin_seq, bool use_64bit) @@ -1079,18 +1136,19 @@ static bool add_addr_hmac_valid(struct mptcp_sock *msk, if (mp_opt->echo) return true; - hmac = add_addr_generate_hmac(msk->remote_key, - msk->local_key, + hmac = add_addr_generate_hmac(READ_ONCE(msk->remote_key), + READ_ONCE(msk->local_key), &mp_opt->addr); pr_debug("msk=%p, ahmac=%llu, mp_opt->ahmac=%llu\n", - msk, (unsigned long long)hmac, - (unsigned long long)mp_opt->ahmac); + msk, hmac, mp_opt->ahmac); return hmac == mp_opt->ahmac; } -/* Return false if a subflow has been reset, else return true */ +/* Return false in case of error (or subflow has been reset), + * else return true. + */ bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); @@ -1107,12 +1165,18 @@ bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) mptcp_data_lock(subflow->conn); if (sk_stream_memory_free(sk)) __mptcp_check_push(subflow->conn, sk); + + /* on fallback we just need to ignore the msk-level snd_una, as + * this is really plain TCP + */ + __mptcp_snd_una_update(msk, READ_ONCE(msk->snd_nxt)); + __mptcp_data_acked(subflow->conn); mptcp_data_unlock(subflow->conn); return true; } - mptcp_get_options(sk, skb, &mp_opt); + mptcp_get_options(skb, &mp_opt); /* The subflow can be in close state only if check_fully_established() * just sent a reset. If so, tell the caller to ignore the current packet. @@ -1122,15 +1186,16 @@ bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) if (unlikely(mp_opt.suboptions != OPTION_MPTCP_DSS)) { if ((mp_opt.suboptions & OPTION_MPTCP_FASTCLOSE) && - msk->local_key == mp_opt.rcvr_key) { + READ_ONCE(msk->local_key) == mp_opt.rcvr_key) { WRITE_ONCE(msk->rcv_fastclose, true); mptcp_schedule_work((struct sock *)msk); + MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPFASTCLOSERX); } if ((mp_opt.suboptions & OPTION_MPTCP_ADD_ADDR) && add_addr_hmac_valid(msk, &mp_opt)) { if (!mp_opt.echo) { - mptcp_pm_add_addr_received(msk, &mp_opt.addr); + mptcp_pm_add_addr_received(sk, &mp_opt.addr); MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ADDADDR); } else { mptcp_pm_add_addr_echoed(msk, &mp_opt.addr); @@ -1159,6 +1224,7 @@ bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) subflow->reset_seen = 1; subflow->reset_reason = mp_opt.reset_reason; subflow->reset_transient = mp_opt.reset_transient; + MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPRSTRX); } if (!(mp_opt.suboptions & OPTION_MPTCP_DSS)) @@ -1170,6 +1236,7 @@ bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) */ if (mp_opt.use_ack) ack_update_msk(msk, sk, &mp_opt); + rwin_update(msk, sk, skb); /* Zero-data-length packets are dropped by the caller and not * propagated to the MPTCP layer, so the skb extension does not @@ -1178,16 +1245,15 @@ bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) */ if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) { if (mp_opt.data_fin && mp_opt.data_len == 1 && - mptcp_update_rcv_data_fin(msk, mp_opt.data_seq, mp_opt.dsn64) && - schedule_work(&msk->work)) - sock_hold(subflow->conn); + mptcp_update_rcv_data_fin(msk, mp_opt.data_seq, mp_opt.dsn64)) + mptcp_schedule_work((struct sock *)msk); return true; } mpext = skb_ext_add(skb, SKB_EXT_MPTCP); if (!mpext) - return true; + return false; memset(mpext, 0, sizeof(*mpext)); @@ -1220,23 +1286,88 @@ bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) return true; } -static void mptcp_set_rwin(const struct tcp_sock *tp) +static void mptcp_set_rwin(struct tcp_sock *tp, struct tcphdr *th) { const struct sock *ssk = (const struct sock *)tp; - const struct mptcp_subflow_context *subflow; + struct mptcp_subflow_context *subflow; + u64 ack_seq, rcv_wnd_old, rcv_wnd_new; struct mptcp_sock *msk; - u64 ack_seq; + u32 new_win; + u64 win; subflow = mptcp_subflow_ctx(ssk); msk = mptcp_sk(subflow->conn); - ack_seq = READ_ONCE(msk->ack_seq) + tp->rcv_wnd; + ack_seq = READ_ONCE(msk->ack_seq); + rcv_wnd_new = ack_seq + tp->rcv_wnd; + + rcv_wnd_old = atomic64_read(&msk->rcv_wnd_sent); + if (after64(rcv_wnd_new, rcv_wnd_old)) { + u64 rcv_wnd; + + for (;;) { + rcv_wnd = atomic64_cmpxchg(&msk->rcv_wnd_sent, rcv_wnd_old, rcv_wnd_new); + + if (rcv_wnd == rcv_wnd_old) + break; + + rcv_wnd_old = rcv_wnd; + if (before64(rcv_wnd_new, rcv_wnd_old)) { + MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_RCVWNDCONFLICTUPDATE); + goto raise_win; + } + MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_RCVWNDCONFLICT); + } + goto update_wspace; + } + + if (rcv_wnd_new != rcv_wnd_old) { +raise_win: + /* The msk-level rcv wnd is after the tcp level one, + * sync the latter. + */ + rcv_wnd_new = rcv_wnd_old; + win = rcv_wnd_old - ack_seq; + tp->rcv_wnd = min_t(u64, win, U32_MAX); + new_win = tp->rcv_wnd; - if (after64(ack_seq, READ_ONCE(msk->rcv_wnd_sent))) - WRITE_ONCE(msk->rcv_wnd_sent, ack_seq); + /* Make sure we do not exceed the maximum possible + * scaled window. + */ + if (unlikely(th->syn)) + new_win = min(new_win, 65535U) << tp->rx_opt.rcv_wscale; + if (!tp->rx_opt.rcv_wscale && + READ_ONCE(sock_net(ssk)->ipv4.sysctl_tcp_workaround_signed_windows)) + new_win = min(new_win, MAX_TCP_WINDOW); + else + new_win = min(new_win, (65535U << tp->rx_opt.rcv_wscale)); + + /* RFC1323 scaling applied */ + new_win >>= tp->rx_opt.rcv_wscale; + th->window = htons(new_win); + MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_RCVWNDSHARED); + } + +update_wspace: + WRITE_ONCE(msk->old_wspace, tp->rcv_wnd); + subflow->rcv_wnd_sent = rcv_wnd_new; } -u16 __mptcp_make_csum(u64 data_seq, u32 subflow_seq, u16 data_len, __wsum sum) +static void mptcp_track_rwin(struct tcp_sock *tp) +{ + const struct sock *ssk = (const struct sock *)tp; + struct mptcp_subflow_context *subflow; + struct mptcp_sock *msk; + + if (!ssk) + return; + + subflow = mptcp_subflow_ctx(ssk); + msk = mptcp_sk(subflow->conn); + WRITE_ONCE(msk->old_wspace, tp->rcv_wnd); +} + +__sum16 __mptcp_make_csum(u64 data_seq, u32 subflow_seq, u16 data_len, __wsum sum) { struct csum_pseudo_header header; __wsum csum; @@ -1252,34 +1383,52 @@ u16 __mptcp_make_csum(u64 data_seq, u32 subflow_seq, u16 data_len, __wsum sum) header.csum = 0; csum = csum_partial(&header, sizeof(header), sum); - return (__force u16)csum_fold(csum); + return csum_fold(csum); } -static u16 mptcp_make_csum(const struct mptcp_ext *mpext) +static __sum16 mptcp_make_csum(const struct mptcp_ext *mpext) { return __mptcp_make_csum(mpext->data_seq, mpext->subflow_seq, mpext->data_len, ~csum_unfold(mpext->csum)); } -void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp, - struct mptcp_out_options *opts) +static void put_len_csum(u16 len, __sum16 csum, void *data) { - if (unlikely(OPTION_MPTCP_FAIL & opts->suboptions)) { - const struct sock *ssk = (const struct sock *)tp; - struct mptcp_subflow_context *subflow; + __sum16 *sumptr = data + 2; + __be16 *ptr = data; - subflow = mptcp_subflow_ctx(ssk); - subflow->send_mp_fail = 0; + put_unaligned_be16(len, ptr); - *ptr++ = mptcp_option(MPTCPOPT_MP_FAIL, - TCPOLEN_MPTCP_FAIL, - 0, 0); - put_unaligned_be64(opts->fail_seq, ptr); - ptr += 2; - } + put_unaligned(csum, sumptr); +} - /* DSS, MPC, MPJ, ADD_ADDR, FASTCLOSE and RST are mutually exclusive, - * see mptcp_established_options*() +void mptcp_write_options(struct tcphdr *th, __be32 *ptr, struct tcp_sock *tp, + struct mptcp_out_options *opts) +{ + const struct sock *ssk = (const struct sock *)tp; + struct mptcp_subflow_context *subflow; + + /* Which options can be used together? + * + * X: mutually exclusive + * O: often used together + * C: can be used together in some cases + * P: could be used together but we prefer not to (optimisations) + * + * Opt: | MPC | MPJ | DSS | ADD | RM | PRIO | FAIL | FC | + * ------|------|------|------|------|------|------|------|------| + * MPC |------|------|------|------|------|------|------|------| + * MPJ | X |------|------|------|------|------|------|------| + * DSS | X | X |------|------|------|------|------|------| + * ADD | X | X | P |------|------|------|------|------| + * RM | C | C | C | P |------|------|------|------| + * PRIO | X | C | C | C | C |------|------|------| + * FAIL | X | X | C | X | X | X |------|------| + * FC | X | X | X | X | X | X | X |------| + * RST | X | X | X | X | X | X | O | O | + * ------|------|------|------|------|------|------|------|------| + * + * The same applies in mptcp_established_options() function. */ if (likely(OPTION_MPTCP_DSS & opts->suboptions)) { struct mptcp_ext *mpext = &opts->ext_copy; @@ -1328,14 +1477,22 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp, put_unaligned_be32(mpext->subflow_seq, ptr); ptr += 1; if (opts->csum_reqd) { - put_unaligned_be32(mpext->data_len << 16 | - mptcp_make_csum(mpext), ptr); + /* data_len == 0 is reserved for the infinite mapping, + * the checksum will also be set to 0. + */ + put_len_csum(mpext->data_len, + (mpext->data_len ? mptcp_make_csum(mpext) : 0), + ptr); } else { put_unaligned_be32(mpext->data_len << 16 | TCPOPT_NOP << 8 | TCPOPT_NOP, ptr); } ptr += 1; } + + /* We might need to add MP_FAIL options in rare cases */ + if (unlikely(OPTION_MPTCP_FAIL & opts->suboptions)) + goto mp_fail; } else if (OPTIONS_MPTCP_MPC & opts->suboptions) { u8 len, flag = MPTCP_CAP_HMAC_SHA256; @@ -1376,11 +1533,12 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp, goto mp_capable_done; if (opts->csum_reqd) { - put_unaligned_be32(opts->data_len << 16 | - __mptcp_make_csum(opts->data_seq, - opts->subflow_seq, - opts->data_len, - ~csum_unfold(opts->csum)), ptr); + put_len_csum(opts->data_len, + __mptcp_make_csum(opts->data_seq, + opts->subflow_seq, + opts->data_len, + ~csum_unfold(opts->csum)), + ptr); } else { put_unaligned_be32(opts->data_len << 16 | TCPOPT_NOP << 8 | TCPOPT_NOP, ptr); @@ -1479,6 +1637,21 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp, if (OPTION_MPTCP_RST & opts->suboptions) goto mp_rst; return; + } else if (unlikely(OPTION_MPTCP_FAIL & opts->suboptions)) { +mp_fail: + /* MP_FAIL is mutually exclusive with others except RST */ + subflow = mptcp_subflow_ctx(ssk); + subflow->send_mp_fail = 0; + + *ptr++ = mptcp_option(MPTCPOPT_MP_FAIL, + TCPOLEN_MPTCP_FAIL, + 0, 0); + put_unaligned_be64(opts->fail_seq, ptr); + ptr += 2; + + if (OPTION_MPTCP_RST & opts->suboptions) + goto mp_rst; + return; } else if (unlikely(OPTION_MPTCP_RST & opts->suboptions)) { mp_rst: *ptr++ = mptcp_option(MPTCPOPT_RST, @@ -1486,18 +1659,21 @@ mp_rst: opts->reset_transient, opts->reset_reason); return; + } else if (unlikely(!opts->suboptions)) { + /* Fallback to TCP */ + mptcp_track_rwin(tp); + return; } if (OPTION_MPTCP_PRIO & opts->suboptions) { - const struct sock *ssk = (const struct sock *)tp; - struct mptcp_subflow_context *subflow; - subflow = mptcp_subflow_ctx(ssk); subflow->send_mp_prio = 0; *ptr++ = mptcp_option(MPTCPOPT_MP_PRIO, TCPOLEN_MPTCP_PRIO, opts->backup, TCPOPT_NOP); + + MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_MPPRIOTX); } mp_capable_done: @@ -1522,7 +1698,7 @@ mp_capable_done: } if (tp) - mptcp_set_rwin(tp); + mptcp_set_rwin(tp, th); } __be32 mptcp_get_reset_option(const struct sk_buff *skb) |