^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /* Multipath TCP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) * Copyright (c) 2017 - 2019, Intel Corporation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) #define pr_fmt(fmt) "MPTCP: " fmt
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) #include <linux/kernel.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <linux/module.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/netdevice.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/sched/signal.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/atomic.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <net/sock.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <net/inet_common.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <net/inet_hashtables.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <net/protocol.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <net/tcp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <net/tcp_states.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #if IS_ENABLED(CONFIG_MPTCP_IPV6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <net/transp_v6.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <net/mptcp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include "protocol.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include "mib.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #if IS_ENABLED(CONFIG_MPTCP_IPV6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) struct mptcp6_sock {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) struct mptcp_sock msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) struct ipv6_pinfo np;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) struct mptcp_skb_cb {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) u64 map_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) u64 end_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) u32 offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) #define MPTCP_SKB_CB(__skb) ((struct mptcp_skb_cb *)&((__skb)->cb[0]))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) static struct percpu_counter mptcp_sockets_allocated;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) /* If msk has an initial subflow socket, and the MP_CAPABLE handshake has not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) * completed yet or has failed, return the subflow socket.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) * Otherwise return NULL.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) static struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) if (!msk->subflow || READ_ONCE(msk->can_ack))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) return msk->subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) static bool mptcp_is_tcpsk(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) struct socket *sock = sk->sk_socket;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) if (unlikely(sk->sk_prot == &tcp_prot)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) /* we are being invoked after mptcp_accept() has
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) * accepted a non-mp-capable flow: sk is a tcp_sk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) * not an mptcp one.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) * Hand the socket over to tcp so all further socket ops
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) * bypass mptcp.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) sock->ops = &inet_stream_ops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) #if IS_ENABLED(CONFIG_MPTCP_IPV6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) } else if (unlikely(sk->sk_prot == &tcpv6_prot)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) sock->ops = &inet6_stream_ops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) static struct sock *__mptcp_tcp_fallback(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) sock_owned_by_me((const struct sock *)msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) if (likely(!__mptcp_check_fallback(msk)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) return msk->first;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) static int __mptcp_socket_create(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) struct socket *ssock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) err = mptcp_subflow_create_socket(sk, &ssock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) msk->first = ssock->sk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) msk->subflow = ssock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) subflow = mptcp_subflow_ctx(ssock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) list_add(&subflow->node, &msk->conn_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) subflow->request_mptcp = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) /* accept() will wait on first subflow sk_wq, and we always wakes up
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) * via msk->sk_socket
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) RCU_INIT_POINTER(msk->first->sk_wq, &sk->sk_socket->wq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) static void mptcp_drop(struct sock *sk, struct sk_buff *skb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) sk_drops_add(sk, skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) __kfree_skb(skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) static bool mptcp_try_coalesce(struct sock *sk, struct sk_buff *to,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) struct sk_buff *from)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) bool fragstolen;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) int delta;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) if (MPTCP_SKB_CB(from)->offset ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) !skb_try_coalesce(to, from, &fragstolen, &delta))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) pr_debug("colesced seq %llx into %llx new len %d new end seq %llx",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) MPTCP_SKB_CB(from)->map_seq, MPTCP_SKB_CB(to)->map_seq,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) to->len, MPTCP_SKB_CB(from)->end_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) MPTCP_SKB_CB(to)->end_seq = MPTCP_SKB_CB(from)->end_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) kfree_skb_partial(from, fragstolen);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) atomic_add(delta, &sk->sk_rmem_alloc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) sk_mem_charge(sk, delta);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) static bool mptcp_ooo_try_coalesce(struct mptcp_sock *msk, struct sk_buff *to,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) struct sk_buff *from)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) if (MPTCP_SKB_CB(from)->map_seq != MPTCP_SKB_CB(to)->end_seq)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) return mptcp_try_coalesce((struct sock *)msk, to, from);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) /* "inspired" by tcp_data_queue_ofo(), main differences:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) * - use mptcp seqs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) * - don't cope with sacks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) static void mptcp_data_queue_ofo(struct mptcp_sock *msk, struct sk_buff *skb)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) struct rb_node **p, *parent;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) u64 seq, end_seq, max_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) struct sk_buff *skb1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) int space;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) seq = MPTCP_SKB_CB(skb)->map_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) end_seq = MPTCP_SKB_CB(skb)->end_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) space = tcp_space(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) max_seq = space > 0 ? space + msk->ack_seq : msk->ack_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) pr_debug("msk=%p seq=%llx limit=%llx empty=%d", msk, seq, max_seq,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) RB_EMPTY_ROOT(&msk->out_of_order_queue));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) if (after64(seq, max_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) /* out of window */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) mptcp_drop(sk, skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_NODSSWINDOW);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) p = &msk->out_of_order_queue.rb_node;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_OFOQUEUE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) if (RB_EMPTY_ROOT(&msk->out_of_order_queue)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) rb_link_node(&skb->rbnode, NULL, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) rb_insert_color(&skb->rbnode, &msk->out_of_order_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) msk->ooo_last_skb = skb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) goto end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) /* with 2 subflows, adding at end of ooo queue is quite likely
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) * Use of ooo_last_skb avoids the O(Log(N)) rbtree lookup.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) if (mptcp_ooo_try_coalesce(msk, msk->ooo_last_skb, skb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_OFOMERGE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_OFOQUEUETAIL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) /* Can avoid an rbtree lookup if we are adding skb after ooo_last_skb */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) if (!before64(seq, MPTCP_SKB_CB(msk->ooo_last_skb)->end_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_OFOQUEUETAIL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) parent = &msk->ooo_last_skb->rbnode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) p = &parent->rb_right;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) goto insert;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) /* Find place to insert this segment. Handle overlaps on the way. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) parent = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) while (*p) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) parent = *p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) skb1 = rb_to_skb(parent);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) if (before64(seq, MPTCP_SKB_CB(skb1)->map_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) p = &parent->rb_left;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) if (before64(seq, MPTCP_SKB_CB(skb1)->end_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) if (!after64(end_seq, MPTCP_SKB_CB(skb1)->end_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) /* All the bits are present. Drop. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) mptcp_drop(sk, skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_DUPDATA);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) if (after64(seq, MPTCP_SKB_CB(skb1)->map_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) /* partial overlap:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) * | skb |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) * | skb1 |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) * continue traversing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) /* skb's seq == skb1's seq and skb covers skb1.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) * Replace skb1 with skb.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) rb_replace_node(&skb1->rbnode, &skb->rbnode,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) &msk->out_of_order_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) mptcp_drop(sk, skb1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_DUPDATA);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) goto merge_right;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) } else if (mptcp_ooo_try_coalesce(msk, skb1, skb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_OFOMERGE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) p = &parent->rb_right;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) insert:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) /* Insert segment into RB tree. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) rb_link_node(&skb->rbnode, parent, p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) rb_insert_color(&skb->rbnode, &msk->out_of_order_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) merge_right:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) /* Remove other segments covered by skb. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) while ((skb1 = skb_rb_next(skb)) != NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) if (before64(end_seq, MPTCP_SKB_CB(skb1)->end_seq))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) rb_erase(&skb1->rbnode, &msk->out_of_order_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) mptcp_drop(sk, skb1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_DUPDATA);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) /* If there is no skb after us, we are the last_skb ! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) if (!skb1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) msk->ooo_last_skb = skb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) end:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) skb_condense(skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) skb_set_owner_r(skb, sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) static bool __mptcp_move_skb(struct mptcp_sock *msk, struct sock *ssk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) struct sk_buff *skb, unsigned int offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) size_t copy_len)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) struct sk_buff *tail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) __skb_unlink(skb, &ssk->sk_receive_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) skb_ext_reset(skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) skb_orphan(skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) /* try to fetch required memory from subflow */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) if (!sk_rmem_schedule(sk, skb, skb->truesize)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) int amount = sk_mem_pages(skb->truesize) << SK_MEM_QUANTUM_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) if (ssk->sk_forward_alloc < amount)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) goto drop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) ssk->sk_forward_alloc -= amount;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) sk->sk_forward_alloc += amount;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) /* the skb map_seq accounts for the skb offset:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) * mptcp_subflow_get_mapped_dsn() is based on the current tp->copied_seq
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) * value
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) MPTCP_SKB_CB(skb)->map_seq = mptcp_subflow_get_mapped_dsn(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) MPTCP_SKB_CB(skb)->end_seq = MPTCP_SKB_CB(skb)->map_seq + copy_len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) MPTCP_SKB_CB(skb)->offset = offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) if (MPTCP_SKB_CB(skb)->map_seq == msk->ack_seq) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) /* in sequence */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) WRITE_ONCE(msk->ack_seq, msk->ack_seq + copy_len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) tail = skb_peek_tail(&sk->sk_receive_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) if (tail && mptcp_try_coalesce(sk, tail, skb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) skb_set_owner_r(skb, sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) __skb_queue_tail(&sk->sk_receive_queue, skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) } else if (after64(MPTCP_SKB_CB(skb)->map_seq, msk->ack_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) mptcp_data_queue_ofo(msk, skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) /* old data, keep it simple and drop the whole pkt, sender
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) * will retransmit as needed, if needed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_DUPDATA);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) drop:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) mptcp_drop(sk, skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) static void mptcp_stop_timer(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) struct inet_connection_sock *icsk = inet_csk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) sk_stop_timer(sk, &icsk->icsk_retransmit_timer);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) mptcp_sk(sk)->timer_ival = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) static void mptcp_check_data_fin_ack(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) if (__mptcp_check_fallback(msk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) /* Look for an acknowledged DATA_FIN */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) if (((1 << sk->sk_state) &
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) (TCPF_FIN_WAIT1 | TCPF_CLOSING | TCPF_LAST_ACK)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) msk->write_seq == atomic64_read(&msk->snd_una)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) mptcp_stop_timer(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) WRITE_ONCE(msk->snd_data_fin_enable, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) switch (sk->sk_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) case TCP_FIN_WAIT1:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) inet_sk_state_store(sk, TCP_FIN_WAIT2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) sk->sk_state_change(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) case TCP_CLOSING:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) case TCP_LAST_ACK:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) inet_sk_state_store(sk, TCP_CLOSE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) sk->sk_state_change(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) if (sk->sk_shutdown == SHUTDOWN_MASK ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) sk->sk_state == TCP_CLOSE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) static bool mptcp_pending_data_fin(struct sock *sk, u64 *seq)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) if (READ_ONCE(msk->rcv_data_fin) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) ((1 << sk->sk_state) &
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) (TCPF_ESTABLISHED | TCPF_FIN_WAIT1 | TCPF_FIN_WAIT2))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) u64 rcv_data_fin_seq = READ_ONCE(msk->rcv_data_fin_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) if (msk->ack_seq == rcv_data_fin_seq) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) if (seq)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) *seq = rcv_data_fin_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) static void mptcp_set_timeout(const struct sock *sk, const struct sock *ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) long tout = ssk && inet_csk(ssk)->icsk_pending ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) inet_csk(ssk)->icsk_timeout - jiffies : 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) if (tout <= 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) tout = mptcp_sk(sk)->timer_ival;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) mptcp_sk(sk)->timer_ival = tout > 0 ? tout : TCP_RTO_MIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) static void mptcp_check_data_fin(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) u64 rcv_data_fin_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) if (__mptcp_check_fallback(msk) || !msk->first)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) /* Need to ack a DATA_FIN received from a peer while this side
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) * of the connection is in ESTABLISHED, FIN_WAIT1, or FIN_WAIT2.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) * msk->rcv_data_fin was set when parsing the incoming options
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) * at the subflow level and the msk lock was not held, so this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) * is the first opportunity to act on the DATA_FIN and change
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) * the msk state.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) * If we are caught up to the sequence number of the incoming
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) * DATA_FIN, send the DATA_ACK now and do state transition. If
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) * not caught up, do nothing and let the recv code send DATA_ACK
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) * when catching up.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) if (mptcp_pending_data_fin(sk, &rcv_data_fin_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) WRITE_ONCE(msk->ack_seq, msk->ack_seq + 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) WRITE_ONCE(msk->rcv_data_fin, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) sk->sk_shutdown |= RCV_SHUTDOWN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) smp_mb__before_atomic(); /* SHUTDOWN must be visible first */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) set_bit(MPTCP_DATA_READY, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) switch (sk->sk_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) case TCP_ESTABLISHED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) inet_sk_state_store(sk, TCP_CLOSE_WAIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) case TCP_FIN_WAIT1:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) inet_sk_state_store(sk, TCP_CLOSING);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) case TCP_FIN_WAIT2:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) inet_sk_state_store(sk, TCP_CLOSE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) // @@ Close subflows now?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) /* Other states not expected */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) WARN_ON_ONCE(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) mptcp_set_timeout(sk, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) lock_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) tcp_send_ack(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) release_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) sk->sk_state_change(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) if (sk->sk_shutdown == SHUTDOWN_MASK ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) sk->sk_state == TCP_CLOSE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) struct sock *ssk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) unsigned int *bytes)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) unsigned int moved = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) bool more_data_avail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) struct tcp_sock *tp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) u32 old_copied_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) bool done = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) pr_debug("msk=%p ssk=%p", msk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) tp = tcp_sk(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) old_copied_seq = tp->copied_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) u32 map_remaining, offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) u32 seq = tp->copied_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) struct sk_buff *skb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) bool fin;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) /* try to move as much data as available */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) map_remaining = subflow->map_data_len -
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) mptcp_subflow_get_map_offset(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) skb = skb_peek(&ssk->sk_receive_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) if (!skb) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) /* if no data is found, a racing workqueue/recvmsg
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) * already processed the new data, stop here or we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) * can enter an infinite loop
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) if (!moved)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) done = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) if (__mptcp_check_fallback(msk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) /* if we are running under the workqueue, TCP could have
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) * collapsed skbs between dummy map creation and now
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) * be sure to adjust the size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) map_remaining = skb->len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) subflow->map_data_len = skb->len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) offset = seq - TCP_SKB_CB(skb)->seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) fin = TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) if (fin) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) done = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) seq++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) if (offset < skb->len) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) size_t len = skb->len - offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) if (tp->urg_data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) done = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) if (__mptcp_move_skb(msk, ssk, skb, offset, len))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) moved += len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) seq += len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) if (WARN_ON_ONCE(map_remaining < len))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) WARN_ON_ONCE(!fin);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) sk_eat_skb(ssk, skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) done = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) WRITE_ONCE(tp->copied_seq, seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) more_data_avail = mptcp_subflow_data_available(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) if (atomic_read(&sk->sk_rmem_alloc) > READ_ONCE(sk->sk_rcvbuf)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) done = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) } while (more_data_avail);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) *bytes += moved;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) if (tp->copied_seq != old_copied_seq)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) tcp_cleanup_rbuf(ssk, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) return done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) static bool mptcp_ofo_queue(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) struct sk_buff *skb, *tail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) bool moved = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) struct rb_node *p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) u64 end_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) p = rb_first(&msk->out_of_order_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) pr_debug("msk=%p empty=%d", msk, RB_EMPTY_ROOT(&msk->out_of_order_queue));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) while (p) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) skb = rb_to_skb(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) if (after64(MPTCP_SKB_CB(skb)->map_seq, msk->ack_seq))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) p = rb_next(p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) rb_erase(&skb->rbnode, &msk->out_of_order_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) if (unlikely(!after64(MPTCP_SKB_CB(skb)->end_seq,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) msk->ack_seq))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) mptcp_drop(sk, skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_DUPDATA);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) end_seq = MPTCP_SKB_CB(skb)->end_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) tail = skb_peek_tail(&sk->sk_receive_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) if (!tail || !mptcp_ooo_try_coalesce(msk, tail, skb)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) int delta = msk->ack_seq - MPTCP_SKB_CB(skb)->map_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) /* skip overlapping data, if any */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) pr_debug("uncoalesced seq=%llx ack seq=%llx delta=%d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) MPTCP_SKB_CB(skb)->map_seq, msk->ack_seq,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) delta);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) MPTCP_SKB_CB(skb)->offset += delta;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) __skb_queue_tail(&sk->sk_receive_queue, skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) msk->ack_seq = end_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) moved = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) return moved;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) /* In most cases we will be able to lock the mptcp socket. If its already
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) * owned, we need to defer to the work queue to avoid ABBA deadlock.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) static bool move_skbs_to_msk(struct mptcp_sock *msk, struct sock *ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) unsigned int moved = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) if (READ_ONCE(sk->sk_lock.owned))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) if (unlikely(!spin_trylock_bh(&sk->sk_lock.slock)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) /* must re-check after taking the lock */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) if (!READ_ONCE(sk->sk_lock.owned)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) __mptcp_move_skbs_from_subflow(msk, ssk, &moved);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) mptcp_ofo_queue(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) /* If the moves have caught up with the DATA_FIN sequence number
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) * it's time to ack the DATA_FIN and change socket state, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) * this is not a good place to change state. Let the workqueue
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) * do it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) if (mptcp_pending_data_fin(sk, NULL) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) schedule_work(&msk->work))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) sock_hold(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) spin_unlock_bh(&sk->sk_lock.slock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) return moved > 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) void mptcp_data_ready(struct sock *sk, struct sock *ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625) struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) bool wake;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) /* move_skbs_to_msk below can legitly clear the data_avail flag,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) * but we will need later to properly woke the reader, cache its
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) * value
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) wake = subflow->data_avail == MPTCP_SUBFLOW_DATA_AVAIL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) if (wake)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) set_bit(MPTCP_DATA_READY, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) if (atomic_read(&sk->sk_rmem_alloc) < READ_ONCE(sk->sk_rcvbuf) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) move_skbs_to_msk(msk, ssk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) goto wake;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) /* don't schedule if mptcp sk is (still) over limit */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) if (atomic_read(&sk->sk_rmem_alloc) > READ_ONCE(sk->sk_rcvbuf))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) goto wake;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) /* mptcp socket is owned, release_cb should retry */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) if (!test_and_set_bit(TCP_DELACK_TIMER_DEFERRED,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) &sk->sk_tsq_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) sock_hold(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) /* need to try again, its possible release_cb() has already
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) * been called after the test_and_set_bit() above.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) move_skbs_to_msk(msk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) wake:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) if (wake)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) sk->sk_data_ready(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) static void __mptcp_flush_join_list(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) if (likely(list_empty(&msk->join_list)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) spin_lock_bh(&msk->join_list_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666) list_splice_tail_init(&msk->join_list, &msk->conn_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) spin_unlock_bh(&msk->join_list_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) static bool mptcp_timer_pending(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) return timer_pending(&inet_csk(sk)->icsk_retransmit_timer);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) static void mptcp_reset_timer(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) struct inet_connection_sock *icsk = inet_csk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) unsigned long tout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) /* should never be called with mptcp level timer cleared */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) tout = READ_ONCE(mptcp_sk(sk)->timer_ival);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) if (WARN_ON_ONCE(!tout))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) tout = TCP_RTO_MIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) sk_reset_timer(sk, &icsk->icsk_retransmit_timer, jiffies + tout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) void mptcp_data_acked(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) mptcp_reset_timer(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) if ((!test_bit(MPTCP_SEND_SPACE, &mptcp_sk(sk)->flags) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) (inet_sk_state_load(sk) != TCP_ESTABLISHED)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) schedule_work(&mptcp_sk(sk)->work))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) sock_hold(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) void mptcp_subflow_eof(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) if (!test_and_set_bit(MPTCP_WORK_EOF, &msk->flags) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) schedule_work(&msk->work))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) sock_hold(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) static void mptcp_check_for_eof(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) int receivers = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) mptcp_for_each_subflow(msk, subflow)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) receivers += !subflow->rx_eof;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) if (!receivers && !(sk->sk_shutdown & RCV_SHUTDOWN)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) /* hopefully temporary hack: propagate shutdown status
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) * to msk, when all subflows agree on it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) sk->sk_shutdown |= RCV_SHUTDOWN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) smp_mb__before_atomic(); /* SHUTDOWN must be visible first */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) set_bit(MPTCP_DATA_READY, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) sk->sk_data_ready(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) static bool mptcp_ext_cache_refill(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) const struct sock *sk = (const struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) if (!msk->cached_ext)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) msk->cached_ext = __skb_ext_alloc(sk->sk_allocation);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) return !!msk->cached_ext;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) static struct sock *mptcp_subflow_recv_lookup(const struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) sock_owned_by_me(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) if (subflow->data_avail)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) return mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) static bool mptcp_skb_can_collapse_to(u64 write_seq,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) const struct sk_buff *skb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) const struct mptcp_ext *mpext)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) if (!tcp_skb_can_collapse_to(skb))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) /* can collapse only if MPTCP level sequence is in order */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) return mpext && mpext->data_seq + mpext->data_len == write_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) /* we can append data to the given data frag if:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) * - there is space available in the backing page_frag
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) * - the data frag tail matches the current page_frag free offset
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) * - the data frag end sequence number matches the current write seq
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) static bool mptcp_frag_can_collapse_to(const struct mptcp_sock *msk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) const struct page_frag *pfrag,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) const struct mptcp_data_frag *df)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) return df && pfrag->page == df->page &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) pfrag->offset == (df->offset + df->data_len) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) df->data_seq + df->data_len == msk->write_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) static void dfrag_uncharge(struct sock *sk, int len)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) sk_mem_uncharge(sk, len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) sk_wmem_queued_add(sk, -len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) static void dfrag_clear(struct sock *sk, struct mptcp_data_frag *dfrag)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) int len = dfrag->data_len + dfrag->overhead;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) list_del(&dfrag->list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) dfrag_uncharge(sk, len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) put_page(dfrag->page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) static bool mptcp_is_writeable(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) if (!sk_stream_is_writeable((struct sock *)msk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) if (sk_stream_is_writeable(subflow->tcp_sock))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806) static void mptcp_clean_una(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) struct mptcp_data_frag *dtmp, *dfrag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) bool cleaned = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) u64 snd_una;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) /* on fallback we just need to ignore snd_una, as this is really
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) * plain TCP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) if (__mptcp_check_fallback(msk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) atomic64_set(&msk->snd_una, msk->write_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) snd_una = atomic64_read(&msk->snd_una);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) list_for_each_entry_safe(dfrag, dtmp, &msk->rtx_queue, list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) if (after64(dfrag->data_seq + dfrag->data_len, snd_una))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) dfrag_clear(sk, dfrag);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) cleaned = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) dfrag = mptcp_rtx_head(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) if (dfrag && after64(snd_una, dfrag->data_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) u64 delta = snd_una - dfrag->data_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) if (WARN_ON_ONCE(delta > dfrag->data_len))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) dfrag->data_seq += delta;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) dfrag->offset += delta;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) dfrag->data_len -= delta;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) dfrag_uncharge(sk, delta);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) cleaned = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) if (cleaned) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) sk_mem_reclaim_partial(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) /* Only wake up writers if a subflow is ready */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) if (mptcp_is_writeable(msk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) set_bit(MPTCP_SEND_SPACE, &mptcp_sk(sk)->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850) smp_mb__after_atomic();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) /* set SEND_SPACE before sk_stream_write_space clears
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) * NOSPACE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) sk_stream_write_space(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) /* ensure we get enough memory for the frag hdr, beyond some minimal amount of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) * data
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) static bool mptcp_page_frag_refill(struct sock *sk, struct page_frag *pfrag)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) if (likely(skb_page_frag_refill(32U + sizeof(struct mptcp_data_frag),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) pfrag, sk->sk_allocation)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) sk->sk_prot->enter_memory_pressure(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) sk_stream_moderate_sndbuf(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) static struct mptcp_data_frag *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) mptcp_carve_data_frag(const struct mptcp_sock *msk, struct page_frag *pfrag,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) int orig_offset)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) int offset = ALIGN(orig_offset, sizeof(long));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) struct mptcp_data_frag *dfrag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881) dfrag = (struct mptcp_data_frag *)(page_to_virt(pfrag->page) + offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) dfrag->data_len = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) dfrag->data_seq = msk->write_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) dfrag->overhead = offset - orig_offset + sizeof(struct mptcp_data_frag);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) dfrag->offset = offset + sizeof(struct mptcp_data_frag);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) dfrag->page = pfrag->page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) return dfrag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) struct msghdr *msg, struct mptcp_data_frag *dfrag,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) long *timeo, int *pmss_now,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) int *ps_goal)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) int mss_now, avail_size, size_goal, offset, ret, frag_truesize = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) bool dfrag_collapsed, can_collapse = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) struct mptcp_ext *mpext = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) bool retransmission = !!dfrag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) struct sk_buff *skb, *tail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) struct page_frag *pfrag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) u64 *write_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) size_t psize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) /* use the mptcp page cache so that we can easily move the data
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) * from one substream to another, but do per subflow memory accounting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) * Note: pfrag is used only !retransmission, but the compiler if
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) * fooled into a warning if we don't init here
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) pfrag = sk_page_frag(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) if (!retransmission) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) write_seq = &msk->write_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) page = pfrag->page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) write_seq = &dfrag->data_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) page = dfrag->page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) /* compute copy limit */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922) mss_now = tcp_send_mss(ssk, &size_goal, msg->msg_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) *pmss_now = mss_now;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924) *ps_goal = size_goal;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) avail_size = size_goal;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) skb = tcp_write_queue_tail(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) if (skb) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) mpext = skb_ext_find(skb, SKB_EXT_MPTCP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) /* Limit the write to the size available in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) * current skb, if any, so that we create at most a new skb.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) * Explicitly tells TCP internals to avoid collapsing on later
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) * queue management operation, to avoid breaking the ext <->
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) * SSN association set here
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) can_collapse = (size_goal - skb->len > 0) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) mptcp_skb_can_collapse_to(*write_seq, skb, mpext);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) if (!can_collapse)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) TCP_SKB_CB(skb)->eor = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941) avail_size = size_goal - skb->len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) if (!retransmission) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945) /* reuse tail pfrag, if possible, or carve a new one from the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) * page allocator
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) dfrag = mptcp_rtx_tail(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) offset = pfrag->offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) dfrag_collapsed = mptcp_frag_can_collapse_to(msk, pfrag, dfrag);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951) if (!dfrag_collapsed) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) dfrag = mptcp_carve_data_frag(msk, pfrag, offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) offset = dfrag->offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954) frag_truesize = dfrag->overhead;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) psize = min_t(size_t, pfrag->size - offset, avail_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) /* Copy to page */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) pr_debug("left=%zu", msg_data_left(msg));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) psize = copy_page_from_iter(pfrag->page, offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) min_t(size_t, msg_data_left(msg),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) psize),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) &msg->msg_iter);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) pr_debug("left=%zu", msg_data_left(msg));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) if (!psize)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) if (!sk_wmem_schedule(sk, psize + dfrag->overhead)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) iov_iter_revert(&msg->msg_iter, psize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) offset = dfrag->offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) psize = min_t(size_t, dfrag->data_len, avail_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977) /* tell the TCP stack to delay the push so that we can safely
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978) * access the skb after the sendpages call
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) ret = do_tcp_sendpages(ssk, page, offset, psize,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981) msg->msg_flags | MSG_SENDPAGE_NOTLAST | MSG_DONTWAIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) if (ret <= 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) if (!retransmission)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984) iov_iter_revert(&msg->msg_iter, psize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) frag_truesize += ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989) if (!retransmission) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) if (unlikely(ret < psize))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) iov_iter_revert(&msg->msg_iter, psize - ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993) /* send successful, keep track of sent data for mptcp-level
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994) * retransmission
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996) dfrag->data_len += ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) if (!dfrag_collapsed) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) get_page(dfrag->page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) list_add_tail(&dfrag->list, &msk->rtx_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) sk_wmem_queued_add(sk, frag_truesize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) sk_wmem_queued_add(sk, ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) /* charge data on mptcp rtx queue to the master socket
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) * Note: we charge such data both to sk and ssk
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008) sk->sk_forward_alloc -= frag_truesize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) /* if the tail skb extension is still the cached one, collapsing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) * really happened. Note: we can't check for 'same skb' as the sk_buff
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) * hdr on tail can be transmitted, freed and re-allocated by the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) * do_tcp_sendpages() call
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016) tail = tcp_write_queue_tail(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) if (mpext && tail && mpext == skb_ext_find(tail, SKB_EXT_MPTCP)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) WARN_ON_ONCE(!can_collapse);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) mpext->data_len += ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) skb = tcp_write_queue_tail(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) mpext = __skb_ext_set(skb, SKB_EXT_MPTCP, msk->cached_ext);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) msk->cached_ext = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027) memset(mpext, 0, sizeof(*mpext));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) mpext->data_seq = *write_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) mpext->subflow_seq = mptcp_subflow_ctx(ssk)->rel_write_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) mpext->data_len = ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) mpext->use_map = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) mpext->dsn64 = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034) pr_debug("data_seq=%llu subflow_seq=%u data_len=%u dsn64=%d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) mpext->data_seq, mpext->subflow_seq, mpext->data_len,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) mpext->dsn64);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) if (!retransmission)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) pfrag->offset += frag_truesize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) WRITE_ONCE(*write_seq, *write_seq + ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) mptcp_subflow_ctx(ssk)->rel_write_seq += ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) static void mptcp_nospace(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) clear_bit(MPTCP_SEND_SPACE, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) smp_mb__after_atomic(); /* msk->flags is changed by write_space cb */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) struct socket *sock = READ_ONCE(ssk->sk_socket);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) /* enables ssk->write_space() callbacks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) if (sock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) set_bit(SOCK_NOSPACE, &sock->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) static bool mptcp_subflow_active(struct mptcp_subflow_context *subflow)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) /* can't send if JOIN hasn't completed yet (i.e. is usable for mptcp) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069) if (subflow->request_join && !subflow->fully_established)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) /* only send if our side has not closed yet */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) return ((1 << ssk->sk_state) & (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) #define MPTCP_SEND_BURST_SIZE ((1 << 16) - \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) sizeof(struct tcphdr) - \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078) MAX_TCP_OPTION_SPACE - \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) sizeof(struct ipv6hdr) - \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) sizeof(struct frag_hdr))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) struct subflow_send_info {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) struct sock *ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) u64 ratio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) static struct sock *mptcp_subflow_get_send(struct mptcp_sock *msk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) u32 *sndbuf)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) struct subflow_send_info send_info[2];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) int i, nr_active = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) struct sock *ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) u64 ratio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) u32 pace;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) sock_owned_by_me((struct sock *)msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) *sndbuf = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) if (!mptcp_ext_cache_refill(msk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) if (__mptcp_check_fallback(msk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) if (!msk->first)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) *sndbuf = msk->first->sk_sndbuf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107) return sk_stream_memory_free(msk->first) ? msk->first : NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) /* re-use last subflow, if the burst allow that */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) if (msk->last_snd && msk->snd_burst > 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) sk_stream_memory_free(msk->last_snd) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) mptcp_subflow_active(mptcp_subflow_ctx(msk->last_snd))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116) *sndbuf = max(tcp_sk(ssk)->snd_wnd, *sndbuf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) return msk->last_snd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) /* pick the subflow with the lower wmem/wspace ratio */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) for (i = 0; i < 2; ++i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) send_info[i].ssk = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) send_info[i].ratio = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128) if (!mptcp_subflow_active(subflow))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) nr_active += !subflow->backup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132) *sndbuf = max(tcp_sk(ssk)->snd_wnd, *sndbuf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) if (!sk_stream_memory_free(subflow->tcp_sock))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136) pace = READ_ONCE(ssk->sk_pacing_rate);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) if (!pace)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) ratio = div_u64((u64)READ_ONCE(ssk->sk_wmem_queued) << 32,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) pace);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) if (ratio < send_info[subflow->backup].ratio) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) send_info[subflow->backup].ssk = ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144) send_info[subflow->backup].ratio = ratio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148) pr_debug("msk=%p nr_active=%d ssk=%p:%lld backup=%p:%lld",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) msk, nr_active, send_info[0].ssk, send_info[0].ratio,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) send_info[1].ssk, send_info[1].ratio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152) /* pick the best backup if no other subflow is active */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) if (!nr_active)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) send_info[0].ssk = send_info[1].ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) if (send_info[0].ssk) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157) msk->last_snd = send_info[0].ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) msk->snd_burst = min_t(int, MPTCP_SEND_BURST_SIZE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159) sk_stream_wspace(msk->last_snd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) return msk->last_snd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) static void ssk_check_wmem(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167) if (unlikely(!mptcp_is_writeable(msk)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) mptcp_nospace(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) int mss_now = 0, size_goal = 0, ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175) struct page_frag *pfrag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) size_t copied = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) struct sock *ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178) u32 sndbuf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) bool tx_ok;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) long timeo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182) if (msg->msg_flags & ~(MSG_MORE | MSG_DONTWAIT | MSG_NOSIGNAL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) return -EOPNOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185) lock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187) timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) if ((1 << sk->sk_state) & ~(TCPF_ESTABLISHED | TCPF_CLOSE_WAIT)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190) ret = sk_stream_wait_connect(sk, &timeo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) pfrag = sk_page_frag(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196) restart:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197) mptcp_clean_una(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199) if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200) ret = -EPIPE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204) __mptcp_flush_join_list(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205) ssk = mptcp_subflow_get_send(msk, &sndbuf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206) while (!sk_stream_memory_free(sk) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) !ssk ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208) !mptcp_page_frag_refill(ssk, pfrag)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) if (ssk) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210) /* make sure retransmit timer is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211) * running before we wait for memory.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213) * The retransmit timer might be needed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214) * to make the peer send an up-to-date
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) * MPTCP Ack.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217) mptcp_set_timeout(sk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218) if (!mptcp_timer_pending(sk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) mptcp_reset_timer(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) mptcp_nospace(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223) ret = sk_stream_wait_memory(sk, &timeo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) mptcp_clean_una(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229) ssk = mptcp_subflow_get_send(msk, &sndbuf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) if (list_empty(&msk->conn_list)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231) ret = -ENOTCONN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236) /* do auto tuning */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238) sndbuf > READ_ONCE(sk->sk_sndbuf))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239) WRITE_ONCE(sk->sk_sndbuf, sndbuf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) pr_debug("conn_list->subflow=%p", ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243) lock_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244) tx_ok = msg_data_left(msg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245) while (tx_ok) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246) ret = mptcp_sendmsg_frag(sk, ssk, msg, NULL, &timeo, &mss_now,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) &size_goal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) if (ret < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249) if (ret == -EAGAIN && timeo > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) mptcp_set_timeout(sk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251) release_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) goto restart;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) /* burst can be negative, we will try move to the next subflow
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258) * at selection time, if possible.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) msk->snd_burst -= ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) copied += ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) tx_ok = msg_data_left(msg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264) if (!tx_ok)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) if (!sk_stream_memory_free(ssk) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) !mptcp_page_frag_refill(ssk, pfrag) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) !mptcp_ext_cache_refill(msk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) tcp_push(ssk, msg->msg_flags, mss_now,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271) tcp_sk(ssk)->nonagle, size_goal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272) mptcp_set_timeout(sk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273) release_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274) goto restart;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277) /* memory is charged to mptcp level socket as well, i.e.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278) * if msg is very large, mptcp socket may run out of buffer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) * space. mptcp_clean_una() will release data that has
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280) * been acked at mptcp level in the mean time, so there is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281) * a good chance we can continue sending data right away.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283) * Normally, when the tcp subflow can accept more data, then
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284) * so can the MPTCP socket. However, we need to cope with
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285) * peers that might lag behind in their MPTCP-level
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) * acknowledgements, i.e. data might have been acked at
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287) * tcp level only. So, we must also check the MPTCP socket
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288) * limits before we send more data.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290) if (unlikely(!sk_stream_memory_free(sk))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291) tcp_push(ssk, msg->msg_flags, mss_now,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292) tcp_sk(ssk)->nonagle, size_goal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293) mptcp_clean_una(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) if (!sk_stream_memory_free(sk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295) /* can't send more for now, need to wait for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296) * MPTCP-level ACKs from peer.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) * Wakeup will happen via mptcp_clean_una().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300) mptcp_set_timeout(sk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301) release_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302) goto restart;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307) mptcp_set_timeout(sk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) if (copied) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309) tcp_push(ssk, msg->msg_flags, mss_now, tcp_sk(ssk)->nonagle,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310) size_goal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312) /* start the timer, if it's not pending */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) if (!mptcp_timer_pending(sk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314) mptcp_reset_timer(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1315) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1316)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1317) release_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319) ssk_check_wmem(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321) return copied ? : ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324) static void mptcp_wait_data(struct sock *sk, long *timeo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326) DEFINE_WAIT_FUNC(wait, woken_wake_function);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) add_wait_queue(sk_sleep(sk), &wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330) sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332) sk_wait_event(sk, timeo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) test_and_clear_bit(MPTCP_DATA_READY, &msk->flags), &wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335) sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336) remove_wait_queue(sk_sleep(sk), &wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339) static int __mptcp_recvmsg_mskq(struct mptcp_sock *msk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340) struct msghdr *msg,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341) size_t len)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344) struct sk_buff *skb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345) int copied = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347) while ((skb = skb_peek(&sk->sk_receive_queue)) != NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348) u32 offset = MPTCP_SKB_CB(skb)->offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349) u32 data_len = skb->len - offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350) u32 count = min_t(size_t, len - copied, data_len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353) err = skb_copy_datagram_msg(skb, offset, msg, count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) if (unlikely(err < 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355) if (!copied)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) copied += count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362) if (count < data_len) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363) MPTCP_SKB_CB(skb)->offset += count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367) __skb_unlink(skb, &sk->sk_receive_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368) __kfree_skb(skb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370) if (copied >= len)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374) return copied;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377) /* receive buffer autotuning. See tcp_rcv_space_adjust for more information.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379) * Only difference: Use highest rtt estimate of the subflows in use.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385) u32 time, advmss = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386) u64 rtt_us, mstamp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) sock_owned_by_me(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390) if (copied <= 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393) msk->rcvq_space.copied += copied;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) mstamp = div_u64(tcp_clock_ns(), NSEC_PER_USEC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) time = tcp_stamp_us_delta(mstamp, msk->rcvq_space.time);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398) rtt_us = msk->rcvq_space.rtt_us;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399) if (rtt_us && time < (rtt_us >> 3))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402) rtt_us = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404) const struct tcp_sock *tp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405) u64 sf_rtt_us;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406) u32 sf_advmss;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408) tp = tcp_sk(mptcp_subflow_tcp_sock(subflow));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410) sf_rtt_us = READ_ONCE(tp->rcv_rtt_est.rtt_us);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411) sf_advmss = READ_ONCE(tp->advmss);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413) rtt_us = max(sf_rtt_us, rtt_us);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414) advmss = max(sf_advmss, advmss);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417) msk->rcvq_space.rtt_us = rtt_us;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) if (time < (rtt_us >> 3) || rtt_us == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) if (msk->rcvq_space.copied <= msk->rcvq_space.space)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422) goto new_measure;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) if (sock_net(sk)->ipv4.sysctl_tcp_moderate_rcvbuf &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425) !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) int rcvmem, rcvbuf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427) u64 rcvwin, grow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) rcvwin = ((u64)msk->rcvq_space.copied << 1) + 16 * advmss;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431) grow = rcvwin * (msk->rcvq_space.copied - msk->rcvq_space.space);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) do_div(grow, msk->rcvq_space.space);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434) rcvwin += (grow << 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436) rcvmem = SKB_TRUESIZE(advmss + MAX_TCP_HEADER);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437) while (tcp_win_from_space(sk, rcvmem) < advmss)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438) rcvmem += 128;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) do_div(rcvwin, advmss);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) rcvbuf = min_t(u64, rcvwin * rcvmem,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) sock_net(sk)->ipv4.sysctl_tcp_rmem[2]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444) if (rcvbuf > sk->sk_rcvbuf) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) u32 window_clamp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447) window_clamp = tcp_win_from_space(sk, rcvbuf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448) WRITE_ONCE(sk->sk_rcvbuf, rcvbuf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450) /* Make subflows follow along. If we do not do this, we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) * get drops at subflow level if skbs can't be moved to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452) * the mptcp rx queue fast enough (announced rcv_win can
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453) * exceed ssk->sk_rcvbuf).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456) struct sock *ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) bool slow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459) ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460) slow = lock_sock_fast(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461) WRITE_ONCE(ssk->sk_rcvbuf, rcvbuf);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462) tcp_sk(ssk)->window_clamp = window_clamp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463) tcp_cleanup_rbuf(ssk, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464) unlock_sock_fast(ssk, slow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469) msk->rcvq_space.space = msk->rcvq_space.copied;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470) new_measure:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471) msk->rcvq_space.copied = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472) msk->rcvq_space.time = mstamp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1473) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1474)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1475) static bool __mptcp_move_skbs(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) unsigned int moved = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478) bool done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480) /* avoid looping forever below on racing close */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481) if (((struct sock *)msk)->sk_state == TCP_CLOSE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484) __mptcp_flush_join_list(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486) struct sock *ssk = mptcp_subflow_recv_lookup(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) if (!ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491) lock_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492) done = __mptcp_move_skbs_from_subflow(msk, ssk, &moved);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493) release_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) } while (!done);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496) if (mptcp_ofo_queue(msk) || moved > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) mptcp_check_data_fin((struct sock *)msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503) static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504) int nonblock, int flags, int *addr_len)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507) int copied = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508) int target;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509) long timeo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511) if (msg->msg_flags & ~(MSG_WAITALL | MSG_DONTWAIT))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512) return -EOPNOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) lock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515) timeo = sock_rcvtimeo(sk, nonblock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) len = min_t(size_t, len, INT_MAX);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518) target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519) __mptcp_flush_join_list(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521) while (len > (size_t)copied) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522) int bytes_read;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524) bytes_read = __mptcp_recvmsg_mskq(msk, msg, len - copied);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) if (unlikely(bytes_read < 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526) if (!copied)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527) copied = bytes_read;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528) goto out_err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531) copied += bytes_read;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) if (skb_queue_empty(&sk->sk_receive_queue) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) __mptcp_move_skbs(msk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537) /* only the master socket status is relevant here. The exit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538) * conditions mirror closely tcp_recvmsg()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540) if (copied >= target)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543) if (copied) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544) if (sk->sk_err ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545) sk->sk_state == TCP_CLOSE ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) (sk->sk_shutdown & RCV_SHUTDOWN) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) !timeo ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) signal_pending(current))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551) if (sk->sk_err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1552) copied = sock_error(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1553) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1554) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556) if (test_and_clear_bit(MPTCP_WORK_EOF, &msk->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557) mptcp_check_for_eof(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559) if (sk->sk_shutdown & RCV_SHUTDOWN)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562) if (sk->sk_state == TCP_CLOSE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563) copied = -ENOTCONN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567) if (!timeo) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568) copied = -EAGAIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572) if (signal_pending(current)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573) copied = sock_intr_errno(timeo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) pr_debug("block timeout %ld", timeo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579) mptcp_wait_data(sk, &timeo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582) if (skb_queue_empty(&sk->sk_receive_queue)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583) /* entire backlog drained, clear DATA_READY. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) clear_bit(MPTCP_DATA_READY, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586) /* .. race-breaker: ssk might have gotten new data
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587) * after last __mptcp_move_skbs() returned false.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589) if (unlikely(__mptcp_move_skbs(msk)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590) set_bit(MPTCP_DATA_READY, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591) } else if (unlikely(!test_bit(MPTCP_DATA_READY, &msk->flags))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592) /* data to read but mptcp_wait_data() cleared DATA_READY */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593) set_bit(MPTCP_DATA_READY, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) out_err:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596) pr_debug("msk=%p data_ready=%d rx queue empty=%d copied=%d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597) msk, test_bit(MPTCP_DATA_READY, &msk->flags),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) skb_queue_empty(&sk->sk_receive_queue), copied);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599) mptcp_rcv_space_adjust(msk, copied);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602) return copied;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605) static void mptcp_retransmit_handler(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609) if (atomic64_read(&msk->snd_una) == READ_ONCE(msk->write_seq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) mptcp_stop_timer(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612) set_bit(MPTCP_WORK_RTX, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613) if (schedule_work(&msk->work))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614) sock_hold(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618) static void mptcp_retransmit_timer(struct timer_list *t)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620) struct inet_connection_sock *icsk = from_timer(icsk, t,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621) icsk_retransmit_timer);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) struct sock *sk = &icsk->icsk_inet.sk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) bh_lock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625) if (!sock_owned_by_user(sk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1626) mptcp_retransmit_handler(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1627) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1628) /* delegate our work to tcp_release_cb() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1629) if (!test_and_set_bit(TCP_WRITE_TIMER_DEFERRED,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1630) &sk->sk_tsq_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1631) sock_hold(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1632) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1633) bh_unlock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1634) sock_put(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1635) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1636)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1637) /* Find an idle subflow. Return NULL if there is unacked data at tcp
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1638) * level.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1639) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1640) * A backup subflow is returned only if that is the only kind available.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1641) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1642) static struct sock *mptcp_subflow_get_retrans(const struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1643) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1644) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1645) struct sock *backup = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1646)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1647) sock_owned_by_me((const struct sock *)msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1648)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1649) if (__mptcp_check_fallback(msk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1650) return msk->first;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1651)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1652) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1653) struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1654)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1655) if (!mptcp_subflow_active(subflow))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1656) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1657)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1658) /* still data outstanding at TCP level? Don't retransmit. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1659) if (!tcp_write_queue_empty(ssk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1660) if (inet_csk(ssk)->icsk_ca_state >= TCP_CA_Loss)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1661) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1662) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1663) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1664)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1665) if (subflow->backup) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1666) if (!backup)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1667) backup = ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1668) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1669) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1670)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1671) return ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1672) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1673)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1674) return backup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1675) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1676)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1677) /* subflow sockets can be either outgoing (connect) or incoming
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1678) * (accept).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1679) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1680) * Outgoing subflows use in-kernel sockets.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1681) * Incoming subflows do not have their own 'struct socket' allocated,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1682) * so we need to use tcp_close() after detaching them from the mptcp
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1683) * parent socket.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1684) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1685) void __mptcp_close_ssk(struct sock *sk, struct sock *ssk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1686) struct mptcp_subflow_context *subflow,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1687) long timeout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1688) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1689) struct socket *sock = READ_ONCE(ssk->sk_socket);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1690)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1691) list_del(&subflow->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1692)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1693) if (sock && sock != sk->sk_socket) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1694) /* outgoing subflow */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1695) sock_release(sock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1696) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1697) /* incoming subflow */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1698) tcp_close(ssk, timeout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1699) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1700) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1701)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1702) static unsigned int mptcp_sync_mss(struct sock *sk, u32 pmtu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1703) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1704) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1705) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1706)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1707) static void pm_work(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1708) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1709) struct mptcp_pm_data *pm = &msk->pm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1710)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1711) spin_lock_bh(&msk->pm.lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1712)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1713) pr_debug("msk=%p status=%x", msk, pm->status);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1714) if (pm->status & BIT(MPTCP_PM_ADD_ADDR_RECEIVED)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1715) pm->status &= ~BIT(MPTCP_PM_ADD_ADDR_RECEIVED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1716) mptcp_pm_nl_add_addr_received(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1717) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1718) if (pm->status & BIT(MPTCP_PM_RM_ADDR_RECEIVED)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1719) pm->status &= ~BIT(MPTCP_PM_RM_ADDR_RECEIVED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1720) mptcp_pm_nl_rm_addr_received(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1721) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1722) if (pm->status & BIT(MPTCP_PM_ESTABLISHED)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1723) pm->status &= ~BIT(MPTCP_PM_ESTABLISHED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1724) mptcp_pm_nl_fully_established(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1725) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1726) if (pm->status & BIT(MPTCP_PM_SUBFLOW_ESTABLISHED)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1727) pm->status &= ~BIT(MPTCP_PM_SUBFLOW_ESTABLISHED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1728) mptcp_pm_nl_subflow_established(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1729) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1730)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1731) spin_unlock_bh(&msk->pm.lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1732) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1733)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1734) static void __mptcp_close_subflow(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1735) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1736) struct mptcp_subflow_context *subflow, *tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1737)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1738) list_for_each_entry_safe(subflow, tmp, &msk->conn_list, node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1739) struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1740)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1741) if (inet_sk_state_load(ssk) != TCP_CLOSE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1742) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1744) __mptcp_close_ssk((struct sock *)msk, ssk, subflow, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1745) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1746) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1747)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1748) static void mptcp_worker(struct work_struct *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1749) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1750) struct mptcp_sock *msk = container_of(work, struct mptcp_sock, work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1751) struct sock *ssk, *sk = &msk->sk.icsk_inet.sk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1752) int orig_len, orig_offset, mss_now = 0, size_goal = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1753) struct mptcp_data_frag *dfrag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1754) u64 orig_write_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1755) size_t copied = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1756) struct msghdr msg = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1757) .msg_flags = MSG_DONTWAIT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1758) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1759) long timeo = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1760)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1761) lock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1762) mptcp_clean_una(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1763) mptcp_check_data_fin_ack(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1764) __mptcp_flush_join_list(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1765) if (test_and_clear_bit(MPTCP_WORK_CLOSE_SUBFLOW, &msk->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1766) __mptcp_close_subflow(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1767)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1768) __mptcp_move_skbs(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1769)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1770) if (msk->pm.status)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1771) pm_work(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1772)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1773) if (test_and_clear_bit(MPTCP_WORK_EOF, &msk->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1774) mptcp_check_for_eof(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1775)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1776) mptcp_check_data_fin(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1777)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1778) if (!test_and_clear_bit(MPTCP_WORK_RTX, &msk->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1779) goto unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1780)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1781) dfrag = mptcp_rtx_head(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1782) if (!dfrag)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1783) goto unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1784)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1785) if (!mptcp_ext_cache_refill(msk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1786) goto reset_unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1787)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1788) ssk = mptcp_subflow_get_retrans(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1789) if (!ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1790) goto reset_unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1791)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1792) lock_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1793)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1794) orig_len = dfrag->data_len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1795) orig_offset = dfrag->offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1796) orig_write_seq = dfrag->data_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1797) while (dfrag->data_len > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1798) int ret = mptcp_sendmsg_frag(sk, ssk, &msg, dfrag, &timeo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1799) &mss_now, &size_goal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1800) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1801) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1802)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1803) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_RETRANSSEGS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1804) copied += ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1805) dfrag->data_len -= ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1806) dfrag->offset += ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1807)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1808) if (!mptcp_ext_cache_refill(msk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1809) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1810) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1811) if (copied)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1812) tcp_push(ssk, msg.msg_flags, mss_now, tcp_sk(ssk)->nonagle,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1813) size_goal);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1814)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1815) dfrag->data_seq = orig_write_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1816) dfrag->offset = orig_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1817) dfrag->data_len = orig_len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1818)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1819) mptcp_set_timeout(sk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1820) release_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1821)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1822) reset_unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1823) if (!mptcp_timer_pending(sk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1824) mptcp_reset_timer(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1825)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1826) unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1827) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1828) sock_put(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1829) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1830)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1831) static int __mptcp_init_sock(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1832) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1833) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1834)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1835) spin_lock_init(&msk->join_list_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1836)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1837) INIT_LIST_HEAD(&msk->conn_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1838) INIT_LIST_HEAD(&msk->join_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1839) INIT_LIST_HEAD(&msk->rtx_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1840) __set_bit(MPTCP_SEND_SPACE, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1841) INIT_WORK(&msk->work, mptcp_worker);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1842) msk->out_of_order_queue = RB_ROOT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1843)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1844) msk->first = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1845) inet_csk(sk)->icsk_sync_mss = mptcp_sync_mss;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1846)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1847) mptcp_pm_data_init(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1848)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1849) /* re-use the csk retrans timer for MPTCP-level retrans */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1850) timer_setup(&msk->sk.icsk_retransmit_timer, mptcp_retransmit_timer, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1851)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1852) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1853) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1854)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1855) static int mptcp_init_sock(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1856) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1857) struct net *net = sock_net(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1858) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1859)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1860) ret = __mptcp_init_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1861) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1862) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1863)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1864) if (!mptcp_is_enabled(net))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1865) return -ENOPROTOOPT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1866)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1867) if (unlikely(!net->mib.mptcp_statistics) && !mptcp_mib_alloc(net))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1868) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1869)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1870) ret = __mptcp_socket_create(mptcp_sk(sk));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1871) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1872) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1873)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1874) sk_sockets_allocated_inc(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1875) sk->sk_rcvbuf = sock_net(sk)->ipv4.sysctl_tcp_rmem[1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1876) sk->sk_sndbuf = sock_net(sk)->ipv4.sysctl_tcp_wmem[1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1877)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1878) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1879) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1880)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1881) static void __mptcp_clear_xmit(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1882) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1883) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1884) struct mptcp_data_frag *dtmp, *dfrag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1885)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1886) sk_stop_timer(sk, &msk->sk.icsk_retransmit_timer);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1887)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1888) list_for_each_entry_safe(dfrag, dtmp, &msk->rtx_queue, list)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1889) dfrag_clear(sk, dfrag);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1890) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1891)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1892) static void mptcp_cancel_work(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1893) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1894) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1895)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1896) if (cancel_work_sync(&msk->work))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1897) sock_put(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1898) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1899)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1900) void mptcp_subflow_shutdown(struct sock *sk, struct sock *ssk, int how)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1901) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1902) lock_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1903)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1904) switch (ssk->sk_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1905) case TCP_LISTEN:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1906) if (!(how & RCV_SHUTDOWN))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1907) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1908) fallthrough;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1909) case TCP_SYN_SENT:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1910) tcp_disconnect(ssk, O_NONBLOCK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1911) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1912) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1913) if (__mptcp_check_fallback(mptcp_sk(sk))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1914) pr_debug("Fallback");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1915) ssk->sk_shutdown |= how;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1916) tcp_shutdown(ssk, how);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1917) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1918) pr_debug("Sending DATA_FIN on subflow %p", ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1919) mptcp_set_timeout(sk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1920) tcp_send_ack(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1921) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1922) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1923) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1924)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1925) release_sock(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1926) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1927)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1928) static const unsigned char new_state[16] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1929) /* current state: new state: action: */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1930) [0 /* (Invalid) */] = TCP_CLOSE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1931) [TCP_ESTABLISHED] = TCP_FIN_WAIT1 | TCP_ACTION_FIN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1932) [TCP_SYN_SENT] = TCP_CLOSE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1933) [TCP_SYN_RECV] = TCP_FIN_WAIT1 | TCP_ACTION_FIN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1934) [TCP_FIN_WAIT1] = TCP_FIN_WAIT1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1935) [TCP_FIN_WAIT2] = TCP_FIN_WAIT2,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1936) [TCP_TIME_WAIT] = TCP_CLOSE, /* should not happen ! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1937) [TCP_CLOSE] = TCP_CLOSE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1938) [TCP_CLOSE_WAIT] = TCP_LAST_ACK | TCP_ACTION_FIN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1939) [TCP_LAST_ACK] = TCP_LAST_ACK,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1940) [TCP_LISTEN] = TCP_CLOSE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1941) [TCP_CLOSING] = TCP_CLOSING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1942) [TCP_NEW_SYN_RECV] = TCP_CLOSE, /* should not happen ! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1943) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1944)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1945) static int mptcp_close_state(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1946) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1947) int next = (int)new_state[sk->sk_state];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1948) int ns = next & TCP_STATE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1949)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1950) inet_sk_state_store(sk, ns);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1951)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1952) return next & TCP_ACTION_FIN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1953) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1954)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1955) static void mptcp_close(struct sock *sk, long timeout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1956) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1957) struct mptcp_subflow_context *subflow, *tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1958) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1959) LIST_HEAD(conn_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1960)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1961) lock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1962) sk->sk_shutdown = SHUTDOWN_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1963)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1964) if (sk->sk_state == TCP_LISTEN) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1965) inet_sk_state_store(sk, TCP_CLOSE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1966) goto cleanup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1967) } else if (sk->sk_state == TCP_CLOSE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1968) goto cleanup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1969) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1970)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1971) if (__mptcp_check_fallback(msk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1972) goto update_state;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1973) } else if (mptcp_close_state(sk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1974) pr_debug("Sending DATA_FIN sk=%p", sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1975) WRITE_ONCE(msk->write_seq, msk->write_seq + 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1976) WRITE_ONCE(msk->snd_data_fin_enable, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1977)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1978) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1979) struct sock *tcp_sk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1980)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1981) mptcp_subflow_shutdown(sk, tcp_sk, SHUTDOWN_MASK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1982) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1983) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1984)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1985) sk_stream_wait_close(sk, timeout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1986)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1987) update_state:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1988) inet_sk_state_store(sk, TCP_CLOSE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1989)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1990) cleanup:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1991) /* be sure to always acquire the join list lock, to sync vs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1992) * mptcp_finish_join().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1993) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1994) spin_lock_bh(&msk->join_list_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1995) list_splice_tail_init(&msk->join_list, &msk->conn_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1996) spin_unlock_bh(&msk->join_list_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1997) list_splice_init(&msk->conn_list, &conn_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1998)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1999) __mptcp_clear_xmit(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2000)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2001) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2002)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2003) list_for_each_entry_safe(subflow, tmp, &conn_list, node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2004) struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2005) __mptcp_close_ssk(sk, ssk, subflow, timeout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2006) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2007)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2008) mptcp_cancel_work(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2009)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2010) __skb_queue_purge(&sk->sk_receive_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2011)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2012) sk_common_release(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2013) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2014)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2015) static void mptcp_copy_inaddrs(struct sock *msk, const struct sock *ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2016) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2017) #if IS_ENABLED(CONFIG_MPTCP_IPV6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2018) const struct ipv6_pinfo *ssk6 = inet6_sk(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2019) struct ipv6_pinfo *msk6 = inet6_sk(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2020)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2021) msk->sk_v6_daddr = ssk->sk_v6_daddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2022) msk->sk_v6_rcv_saddr = ssk->sk_v6_rcv_saddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2023)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2024) if (msk6 && ssk6) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2025) msk6->saddr = ssk6->saddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2026) msk6->flow_label = ssk6->flow_label;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2027) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2028) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2029)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2030) inet_sk(msk)->inet_num = inet_sk(ssk)->inet_num;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2031) inet_sk(msk)->inet_dport = inet_sk(ssk)->inet_dport;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2032) inet_sk(msk)->inet_sport = inet_sk(ssk)->inet_sport;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2033) inet_sk(msk)->inet_daddr = inet_sk(ssk)->inet_daddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2034) inet_sk(msk)->inet_saddr = inet_sk(ssk)->inet_saddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2035) inet_sk(msk)->inet_rcv_saddr = inet_sk(ssk)->inet_rcv_saddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2036) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2037)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2038) static int mptcp_disconnect(struct sock *sk, int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2039) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2040) /* Should never be called.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2041) * inet_stream_connect() calls ->disconnect, but that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2042) * refers to the subflow socket, not the mptcp one.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2043) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2044) WARN_ON_ONCE(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2045) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2046) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2047)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2048) #if IS_ENABLED(CONFIG_MPTCP_IPV6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2049) static struct ipv6_pinfo *mptcp_inet6_sk(const struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2050) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2051) unsigned int offset = sizeof(struct mptcp6_sock) - sizeof(struct ipv6_pinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2052)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2053) return (struct ipv6_pinfo *)(((u8 *)sk) + offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2054) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2055) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2056)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2057) struct sock *mptcp_sk_clone(const struct sock *sk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2058) const struct mptcp_options_received *mp_opt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2059) struct request_sock *req)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2060) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2061) struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2062) struct sock *nsk = sk_clone_lock(sk, GFP_ATOMIC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2063) struct mptcp_sock *msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2064) u64 ack_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2065)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2066) if (!nsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2067) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2068)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2069) #if IS_ENABLED(CONFIG_MPTCP_IPV6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2070) if (nsk->sk_family == AF_INET6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2071) inet_sk(nsk)->pinet6 = mptcp_inet6_sk(nsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2072) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2073)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2074) __mptcp_init_sock(nsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2075)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2076) msk = mptcp_sk(nsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2077) msk->local_key = subflow_req->local_key;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2078) msk->token = subflow_req->token;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2079) msk->subflow = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2080) WRITE_ONCE(msk->fully_established, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2081)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2082) msk->write_seq = subflow_req->idsn + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2083) atomic64_set(&msk->snd_una, msk->write_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2084) if (mp_opt->mp_capable) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2085) msk->can_ack = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2086) msk->remote_key = mp_opt->sndr_key;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2087) mptcp_crypto_key_sha(msk->remote_key, NULL, &ack_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2088) ack_seq++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2089) WRITE_ONCE(msk->ack_seq, ack_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2090) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2091)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2092) sock_reset_flag(nsk, SOCK_RCU_FREE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2093) /* will be fully established after successful MPC subflow creation */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2094) inet_sk_state_store(nsk, TCP_SYN_RECV);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2095)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2096) security_inet_csk_clone(nsk, req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2097) bh_unlock_sock(nsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2098)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2099) /* keep a single reference */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2100) __sock_put(nsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2101) return nsk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2102) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2103)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2104) void mptcp_rcv_space_init(struct mptcp_sock *msk, const struct sock *ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2105) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2106) const struct tcp_sock *tp = tcp_sk(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2107)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2108) msk->rcvq_space.copied = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2109) msk->rcvq_space.rtt_us = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2110)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2111) msk->rcvq_space.time = tp->tcp_mstamp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2112)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2113) /* initial rcv_space offering made to peer */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2114) msk->rcvq_space.space = min_t(u32, tp->rcv_wnd,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2115) TCP_INIT_CWND * tp->advmss);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2116) if (msk->rcvq_space.space == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2117) msk->rcvq_space.space = TCP_INIT_CWND * TCP_MSS_DEFAULT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2118) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2119)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2120) static struct sock *mptcp_accept(struct sock *sk, int flags, int *err,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2121) bool kern)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2122) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2123) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2124) struct socket *listener;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2125) struct sock *newsk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2126)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2127) listener = __mptcp_nmpc_socket(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2128) if (WARN_ON_ONCE(!listener)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2129) *err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2130) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2131) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2132)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2133) pr_debug("msk=%p, listener=%p", msk, mptcp_subflow_ctx(listener->sk));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2134) newsk = inet_csk_accept(listener->sk, flags, err, kern);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2135) if (!newsk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2136) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2137)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2138) pr_debug("msk=%p, subflow is mptcp=%d", msk, sk_is_mptcp(newsk));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2139) if (sk_is_mptcp(newsk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2140) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2141) struct sock *new_mptcp_sock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2142) struct sock *ssk = newsk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2143)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2144) subflow = mptcp_subflow_ctx(newsk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2145) new_mptcp_sock = subflow->conn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2147) /* is_mptcp should be false if subflow->conn is missing, see
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2148) * subflow_syn_recv_sock()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2149) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2150) if (WARN_ON_ONCE(!new_mptcp_sock)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2151) tcp_sk(newsk)->is_mptcp = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2152) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2153) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2154)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2155) /* acquire the 2nd reference for the owning socket */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2156) sock_hold(new_mptcp_sock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2157)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2158) local_bh_disable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2159) bh_lock_sock(new_mptcp_sock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2160) msk = mptcp_sk(new_mptcp_sock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2161) msk->first = newsk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2162)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2163) newsk = new_mptcp_sock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2164) mptcp_copy_inaddrs(newsk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2165) list_add(&subflow->node, &msk->conn_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2167) mptcp_rcv_space_init(msk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2168) bh_unlock_sock(new_mptcp_sock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2169)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2170) __MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPCAPABLEPASSIVEACK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2171) local_bh_enable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2172) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2173) MPTCP_INC_STATS(sock_net(sk),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2174) MPTCP_MIB_MPCAPABLEPASSIVEFALLBACK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2175) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2176)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2177) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2178) newsk->sk_kern_sock = kern;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2179) return newsk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2180) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2182) void mptcp_destroy_common(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2183) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2184) skb_rbtree_purge(&msk->out_of_order_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2185) mptcp_token_destroy(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2186) mptcp_pm_free_anno_list(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2187) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2188)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2189) static void mptcp_destroy(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2190) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2191) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2192)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2193) if (msk->cached_ext)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2194) __skb_ext_put(msk->cached_ext);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2195)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2196) mptcp_destroy_common(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2197) sk_sockets_allocated_dec(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2198) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2199)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2200) static int mptcp_setsockopt_sol_socket(struct mptcp_sock *msk, int optname,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2201) sockptr_t optval, unsigned int optlen)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2202) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2203) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2204) struct socket *ssock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2205) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2206)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2207) switch (optname) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2208) case SO_REUSEPORT:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2209) case SO_REUSEADDR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2210) lock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2211) ssock = __mptcp_nmpc_socket(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2212) if (!ssock) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2213) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2214) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2215) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2217) ret = sock_setsockopt(ssock, SOL_SOCKET, optname, optval, optlen);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2218) if (ret == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2219) if (optname == SO_REUSEPORT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2220) sk->sk_reuseport = ssock->sk->sk_reuseport;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2221) else if (optname == SO_REUSEADDR)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2222) sk->sk_reuse = ssock->sk->sk_reuse;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2223) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2224) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2225) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2226) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2227)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2228) return sock_setsockopt(sk->sk_socket, SOL_SOCKET, optname, optval, optlen);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2229) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2230)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2231) static int mptcp_setsockopt_v6(struct mptcp_sock *msk, int optname,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2232) sockptr_t optval, unsigned int optlen)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2233) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2234) struct sock *sk = (struct sock *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2235) int ret = -EOPNOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2236) struct socket *ssock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2237)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2238) switch (optname) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2239) case IPV6_V6ONLY:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2240) lock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2241) ssock = __mptcp_nmpc_socket(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2242) if (!ssock) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2243) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2244) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2245) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2246)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2247) ret = tcp_setsockopt(ssock->sk, SOL_IPV6, optname, optval, optlen);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2248) if (ret == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2249) sk->sk_ipv6only = ssock->sk->sk_ipv6only;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2250)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2251) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2252) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2253) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2254)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2255) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2256) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2257)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2258) static bool mptcp_unsupported(int level, int optname)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2259) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2260) if (level == SOL_IP) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2261) switch (optname) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2262) case IP_ADD_MEMBERSHIP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2263) case IP_ADD_SOURCE_MEMBERSHIP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2264) case IP_DROP_MEMBERSHIP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2265) case IP_DROP_SOURCE_MEMBERSHIP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2266) case IP_BLOCK_SOURCE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2267) case IP_UNBLOCK_SOURCE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2268) case MCAST_JOIN_GROUP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2269) case MCAST_LEAVE_GROUP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2270) case MCAST_JOIN_SOURCE_GROUP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2271) case MCAST_LEAVE_SOURCE_GROUP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2272) case MCAST_BLOCK_SOURCE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2273) case MCAST_UNBLOCK_SOURCE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2274) case MCAST_MSFILTER:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2275) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2276) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2277) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2278) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2279) if (level == SOL_IPV6) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2280) switch (optname) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2281) case IPV6_ADDRFORM:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2282) case IPV6_ADD_MEMBERSHIP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2283) case IPV6_DROP_MEMBERSHIP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2284) case IPV6_JOIN_ANYCAST:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2285) case IPV6_LEAVE_ANYCAST:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2286) case MCAST_JOIN_GROUP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2287) case MCAST_LEAVE_GROUP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2288) case MCAST_JOIN_SOURCE_GROUP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2289) case MCAST_LEAVE_SOURCE_GROUP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2290) case MCAST_BLOCK_SOURCE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2291) case MCAST_UNBLOCK_SOURCE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2292) case MCAST_MSFILTER:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2293) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2294) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2295) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2296) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2297) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2298) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2299)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2300) static int mptcp_setsockopt(struct sock *sk, int level, int optname,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2301) sockptr_t optval, unsigned int optlen)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2302) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2303) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2304) struct sock *ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2305)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2306) pr_debug("msk=%p", msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2307)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2308) if (mptcp_unsupported(level, optname))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2309) return -ENOPROTOOPT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2311) if (level == SOL_SOCKET)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2312) return mptcp_setsockopt_sol_socket(msk, optname, optval, optlen);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2313)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2314) /* @@ the meaning of setsockopt() when the socket is connected and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2315) * there are multiple subflows is not yet defined. It is up to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2316) * MPTCP-level socket to configure the subflows until the subflow
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2317) * is in TCP fallback, when TCP socket options are passed through
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2318) * to the one remaining subflow.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2319) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2320) lock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2321) ssk = __mptcp_tcp_fallback(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2322) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2323) if (ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2324) return tcp_setsockopt(ssk, level, optname, optval, optlen);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2326) if (level == SOL_IPV6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2327) return mptcp_setsockopt_v6(msk, optname, optval, optlen);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2328)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2329) return -EOPNOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2330) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2331)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2332) static int mptcp_getsockopt(struct sock *sk, int level, int optname,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2333) char __user *optval, int __user *option)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2334) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2335) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2336) struct sock *ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2337)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2338) pr_debug("msk=%p", msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2339)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2340) /* @@ the meaning of setsockopt() when the socket is connected and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2341) * there are multiple subflows is not yet defined. It is up to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2342) * MPTCP-level socket to configure the subflows until the subflow
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2343) * is in TCP fallback, when socket options are passed through
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2344) * to the one remaining subflow.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2345) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2346) lock_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2347) ssk = __mptcp_tcp_fallback(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2348) release_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2349) if (ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2350) return tcp_getsockopt(ssk, level, optname, optval, option);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2351)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2352) return -EOPNOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2353) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2354)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2355) #define MPTCP_DEFERRED_ALL (TCPF_DELACK_TIMER_DEFERRED | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2356) TCPF_WRITE_TIMER_DEFERRED)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2357)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2358) /* this is very alike tcp_release_cb() but we must handle differently a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2359) * different set of events
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2360) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2361) static void mptcp_release_cb(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2362) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2363) unsigned long flags, nflags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2364)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2365) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2366) flags = sk->sk_tsq_flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2367) if (!(flags & MPTCP_DEFERRED_ALL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2368) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2369) nflags = flags & ~MPTCP_DEFERRED_ALL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2370) } while (cmpxchg(&sk->sk_tsq_flags, flags, nflags) != flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2371)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2372) sock_release_ownership(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2374) if (flags & TCPF_DELACK_TIMER_DEFERRED) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2375) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2376) struct sock *ssk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2377)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2378) ssk = mptcp_subflow_recv_lookup(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2379) if (!ssk || !schedule_work(&msk->work))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2380) __sock_put(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2381) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2382)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2383) if (flags & TCPF_WRITE_TIMER_DEFERRED) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2384) mptcp_retransmit_handler(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2385) __sock_put(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2386) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2387) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2388)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2389) static int mptcp_hash(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2390) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2391) /* should never be called,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2392) * we hash the TCP subflows not the master socket
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2393) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2394) WARN_ON_ONCE(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2395) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2396) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2397)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2398) static void mptcp_unhash(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2399) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2400) /* called from sk_common_release(), but nothing to do here */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2401) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2402)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2403) static int mptcp_get_port(struct sock *sk, unsigned short snum)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2404) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2405) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2406) struct socket *ssock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2407)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2408) ssock = __mptcp_nmpc_socket(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2409) pr_debug("msk=%p, subflow=%p", msk, ssock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2410) if (WARN_ON_ONCE(!ssock))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2411) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2412)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2413) return inet_csk_get_port(ssock->sk, snum);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2414) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2415)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2416) void mptcp_finish_connect(struct sock *ssk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2417) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2418) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2419) struct mptcp_sock *msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2420) struct sock *sk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2421) u64 ack_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2422)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2423) subflow = mptcp_subflow_ctx(ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2424) sk = subflow->conn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2425) msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2426)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2427) pr_debug("msk=%p, token=%u", sk, subflow->token);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2428)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2429) mptcp_crypto_key_sha(subflow->remote_key, NULL, &ack_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2430) ack_seq++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2431) subflow->map_seq = ack_seq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2432) subflow->map_subflow_seq = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2433)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2434) /* the socket is not connected yet, no msk/subflow ops can access/race
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2435) * accessing the field below
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2436) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2437) WRITE_ONCE(msk->remote_key, subflow->remote_key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2438) WRITE_ONCE(msk->local_key, subflow->local_key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2439) WRITE_ONCE(msk->write_seq, subflow->idsn + 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2440) WRITE_ONCE(msk->ack_seq, ack_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2441) WRITE_ONCE(msk->can_ack, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2442) atomic64_set(&msk->snd_una, msk->write_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2444) mptcp_pm_new_connection(msk, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2445)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2446) mptcp_rcv_space_init(msk, ssk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2447) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2448)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2449) static void mptcp_sock_graft(struct sock *sk, struct socket *parent)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2450) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2451) write_lock_bh(&sk->sk_callback_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2452) rcu_assign_pointer(sk->sk_wq, &parent->wq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2453) sk_set_socket(sk, parent);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2454) sk->sk_uid = SOCK_INODE(parent)->i_uid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2455) write_unlock_bh(&sk->sk_callback_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2456) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2457)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2458) bool mptcp_finish_join(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2459) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2460) struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2461) struct mptcp_sock *msk = mptcp_sk(subflow->conn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2462) struct sock *parent = (void *)msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2463) struct socket *parent_sock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2464) bool ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2465)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2466) pr_debug("msk=%p, subflow=%p", msk, subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2467)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2468) /* mptcp socket already closing? */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2469) if (!mptcp_is_fully_established(parent))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2470) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2471)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2472) if (!msk->pm.server_side)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2473) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2474)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2475) if (!mptcp_pm_allow_new_subflow(msk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2476) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2477)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2478) /* active connections are already on conn_list, and we can't acquire
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2479) * msk lock here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2480) * use the join list lock as synchronization point and double-check
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2481) * msk status to avoid racing with mptcp_close()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2482) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2483) spin_lock_bh(&msk->join_list_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2484) ret = inet_sk_state_load(parent) == TCP_ESTABLISHED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2485) if (ret && !WARN_ON_ONCE(!list_empty(&subflow->node)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2486) list_add_tail(&subflow->node, &msk->join_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2487) spin_unlock_bh(&msk->join_list_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2488) if (!ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2489) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2491) /* attach to msk socket only after we are sure he will deal with us
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2492) * at close time
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2493) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2494) parent_sock = READ_ONCE(parent->sk_socket);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2495) if (parent_sock && !sk->sk_socket)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2496) mptcp_sock_graft(sk, parent_sock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2497) subflow->map_seq = READ_ONCE(msk->ack_seq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2498) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2499) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2500)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2501) static bool mptcp_memory_free(const struct sock *sk, int wake)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2502) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2503) struct mptcp_sock *msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2505) return wake ? test_bit(MPTCP_SEND_SPACE, &msk->flags) : true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2506) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2507)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2508) static struct proto mptcp_prot = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2509) .name = "MPTCP",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2510) .owner = THIS_MODULE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2511) .init = mptcp_init_sock,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2512) .disconnect = mptcp_disconnect,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2513) .close = mptcp_close,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2514) .accept = mptcp_accept,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2515) .setsockopt = mptcp_setsockopt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2516) .getsockopt = mptcp_getsockopt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2517) .shutdown = tcp_shutdown,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2518) .destroy = mptcp_destroy,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2519) .sendmsg = mptcp_sendmsg,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2520) .recvmsg = mptcp_recvmsg,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2521) .release_cb = mptcp_release_cb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2522) .hash = mptcp_hash,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2523) .unhash = mptcp_unhash,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2524) .get_port = mptcp_get_port,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2525) .sockets_allocated = &mptcp_sockets_allocated,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2526) .memory_allocated = &tcp_memory_allocated,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2527) .memory_pressure = &tcp_memory_pressure,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2528) .stream_memory_free = mptcp_memory_free,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2529) .sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_tcp_wmem),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2530) .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_tcp_rmem),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2531) .sysctl_mem = sysctl_tcp_mem,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2532) .obj_size = sizeof(struct mptcp_sock),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2533) .slab_flags = SLAB_TYPESAFE_BY_RCU,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2534) .no_autobind = true,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2535) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2536)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2537) static int mptcp_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2538) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2539) struct mptcp_sock *msk = mptcp_sk(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2540) struct socket *ssock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2541) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2543) lock_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2544) ssock = __mptcp_nmpc_socket(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2545) if (!ssock) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2546) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2547) goto unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2548) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2549)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2550) err = ssock->ops->bind(ssock, uaddr, addr_len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2551) if (!err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2552) mptcp_copy_inaddrs(sock->sk, ssock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2553)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2554) unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2555) release_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2556) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2557) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2558)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2559) static void mptcp_subflow_early_fallback(struct mptcp_sock *msk,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2560) struct mptcp_subflow_context *subflow)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2561) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2562) subflow->request_mptcp = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2563) __mptcp_do_fallback(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2564) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2565)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2566) static int mptcp_stream_connect(struct socket *sock, struct sockaddr *uaddr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2567) int addr_len, int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2568) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2569) struct mptcp_sock *msk = mptcp_sk(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2570) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2571) struct socket *ssock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2572) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2573)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2574) lock_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2575) if (sock->state != SS_UNCONNECTED && msk->subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2576) /* pending connection or invalid state, let existing subflow
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2577) * cope with that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2578) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2579) ssock = msk->subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2580) goto do_connect;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2581) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2582)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2583) ssock = __mptcp_nmpc_socket(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2584) if (!ssock) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2585) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2586) goto unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2587) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2588)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2589) mptcp_token_destroy(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2590) inet_sk_state_store(sock->sk, TCP_SYN_SENT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2591) subflow = mptcp_subflow_ctx(ssock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2592) #ifdef CONFIG_TCP_MD5SIG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2593) /* no MPTCP if MD5SIG is enabled on this socket or we may run out of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2594) * TCP option space.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2595) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2596) if (rcu_access_pointer(tcp_sk(ssock->sk)->md5sig_info))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2597) mptcp_subflow_early_fallback(msk, subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2598) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2599) if (subflow->request_mptcp && mptcp_token_new_connect(ssock->sk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2600) mptcp_subflow_early_fallback(msk, subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2601)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2602) do_connect:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2603) err = ssock->ops->connect(ssock, uaddr, addr_len, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2604) sock->state = ssock->state;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2605)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2606) /* on successful connect, the msk state will be moved to established by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2607) * subflow_finish_connect()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2608) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2609) if (!err || err == -EINPROGRESS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2610) mptcp_copy_inaddrs(sock->sk, ssock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2611) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2612) inet_sk_state_store(sock->sk, inet_sk_state_load(ssock->sk));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2613)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2614) unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2615) release_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2616) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2617) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2618)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2619) static int mptcp_listen(struct socket *sock, int backlog)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2620) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2621) struct mptcp_sock *msk = mptcp_sk(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2622) struct socket *ssock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2623) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2624)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2625) pr_debug("msk=%p", msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2626)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2627) lock_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2628) ssock = __mptcp_nmpc_socket(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2629) if (!ssock) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2630) err = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2631) goto unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2632) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2633)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2634) mptcp_token_destroy(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2635) inet_sk_state_store(sock->sk, TCP_LISTEN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2636) sock_set_flag(sock->sk, SOCK_RCU_FREE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2637)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2638) err = ssock->ops->listen(ssock, backlog);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2639) inet_sk_state_store(sock->sk, inet_sk_state_load(ssock->sk));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2640) if (!err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2641) mptcp_copy_inaddrs(sock->sk, ssock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2642)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2643) unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2644) release_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2645) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2646) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2647)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2648) static int mptcp_stream_accept(struct socket *sock, struct socket *newsock,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2649) int flags, bool kern)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2650) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2651) struct mptcp_sock *msk = mptcp_sk(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2652) struct socket *ssock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2653) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2654)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2655) pr_debug("msk=%p", msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2656)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2657) lock_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2658) if (sock->sk->sk_state != TCP_LISTEN)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2659) goto unlock_fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2660)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2661) ssock = __mptcp_nmpc_socket(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2662) if (!ssock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2663) goto unlock_fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2664)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2665) clear_bit(MPTCP_DATA_READY, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2666) sock_hold(ssock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2667) release_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2668)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2669) err = ssock->ops->accept(sock, newsock, flags, kern);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2670) if (err == 0 && !mptcp_is_tcpsk(newsock->sk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2671) struct mptcp_sock *msk = mptcp_sk(newsock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2672) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2673)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2674) /* set ssk->sk_socket of accept()ed flows to mptcp socket.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2675) * This is needed so NOSPACE flag can be set from tcp stack.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2676) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2677) __mptcp_flush_join_list(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2678) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2679) struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2680)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2681) if (!ssk->sk_socket)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2682) mptcp_sock_graft(ssk, newsock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2683) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2684) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2686) if (inet_csk_listen_poll(ssock->sk))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2687) set_bit(MPTCP_DATA_READY, &msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2688) sock_put(ssock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2689) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2690)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2691) unlock_fail:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2692) release_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2693) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2694) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2695)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2696) static __poll_t mptcp_check_readable(struct mptcp_sock *msk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2697) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2698) return test_bit(MPTCP_DATA_READY, &msk->flags) ? EPOLLIN | EPOLLRDNORM :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2699) 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2700) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2701)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2702) static __poll_t mptcp_poll(struct file *file, struct socket *sock,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2703) struct poll_table_struct *wait)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2704) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2705) struct sock *sk = sock->sk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2706) struct mptcp_sock *msk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2707) __poll_t mask = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2708) int state;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2709)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2710) msk = mptcp_sk(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2711) sock_poll_wait(file, sock, wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2712)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2713) state = inet_sk_state_load(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2714) pr_debug("msk=%p state=%d flags=%lx", msk, state, msk->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2715) if (state == TCP_LISTEN)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2716) return mptcp_check_readable(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2717)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2718) if (state != TCP_SYN_SENT && state != TCP_SYN_RECV) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2719) mask |= mptcp_check_readable(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2720) if (test_bit(MPTCP_SEND_SPACE, &msk->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2721) mask |= EPOLLOUT | EPOLLWRNORM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2722) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2723) if (sk->sk_shutdown & RCV_SHUTDOWN)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2724) mask |= EPOLLIN | EPOLLRDNORM | EPOLLRDHUP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2725)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2726) return mask;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2727) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2728)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2729) static int mptcp_shutdown(struct socket *sock, int how)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2730) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2731) struct mptcp_sock *msk = mptcp_sk(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2732) struct mptcp_subflow_context *subflow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2733) int ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2734)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2735) pr_debug("sk=%p, how=%d", msk, how);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2736)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2737) lock_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2738)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2739) how++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2740) if ((how & ~SHUTDOWN_MASK) || !how) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2741) ret = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2742) goto out_unlock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2743) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2744)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2745) if (sock->state == SS_CONNECTING) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2746) if ((1 << sock->sk->sk_state) &
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2747) (TCPF_SYN_SENT | TCPF_SYN_RECV | TCPF_CLOSE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2748) sock->state = SS_DISCONNECTING;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2749) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2750) sock->state = SS_CONNECTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2751) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2752)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2753) /* If we've already sent a FIN, or it's a closed state, skip this. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2754) if (__mptcp_check_fallback(msk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2755) if (how == SHUT_WR || how == SHUT_RDWR)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2756) inet_sk_state_store(sock->sk, TCP_FIN_WAIT1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2757)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2758) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2759) struct sock *tcp_sk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2760)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2761) mptcp_subflow_shutdown(sock->sk, tcp_sk, how);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2762) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2763) } else if ((how & SEND_SHUTDOWN) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2764) ((1 << sock->sk->sk_state) &
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2765) (TCPF_ESTABLISHED | TCPF_SYN_SENT |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2766) TCPF_SYN_RECV | TCPF_CLOSE_WAIT)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2767) mptcp_close_state(sock->sk)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2768) __mptcp_flush_join_list(msk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2769)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2770) WRITE_ONCE(msk->write_seq, msk->write_seq + 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2771) WRITE_ONCE(msk->snd_data_fin_enable, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2772)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2773) mptcp_for_each_subflow(msk, subflow) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2774) struct sock *tcp_sk = mptcp_subflow_tcp_sock(subflow);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2775)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2776) mptcp_subflow_shutdown(sock->sk, tcp_sk, how);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2777) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2778) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2779)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2780) /* Wake up anyone sleeping in poll. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2781) sock->sk->sk_state_change(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2782)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2783) out_unlock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2784) release_sock(sock->sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2785)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2786) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2787) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2788)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2789) static const struct proto_ops mptcp_stream_ops = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2790) .family = PF_INET,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2791) .owner = THIS_MODULE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2792) .release = inet_release,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2793) .bind = mptcp_bind,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2794) .connect = mptcp_stream_connect,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2795) .socketpair = sock_no_socketpair,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2796) .accept = mptcp_stream_accept,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2797) .getname = inet_getname,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2798) .poll = mptcp_poll,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2799) .ioctl = inet_ioctl,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2800) .gettstamp = sock_gettstamp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2801) .listen = mptcp_listen,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2802) .shutdown = mptcp_shutdown,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2803) .setsockopt = sock_common_setsockopt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2804) .getsockopt = sock_common_getsockopt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2805) .sendmsg = inet_sendmsg,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2806) .recvmsg = inet_recvmsg,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2807) .mmap = sock_no_mmap,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2808) .sendpage = inet_sendpage,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2809) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2810)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2811) static struct inet_protosw mptcp_protosw = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2812) .type = SOCK_STREAM,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2813) .protocol = IPPROTO_MPTCP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2814) .prot = &mptcp_prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2815) .ops = &mptcp_stream_ops,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2816) .flags = INET_PROTOSW_ICSK,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2817) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2818)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2819) void __init mptcp_proto_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2820) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2821) mptcp_prot.h.hashinfo = tcp_prot.h.hashinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2822)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2823) if (percpu_counter_init(&mptcp_sockets_allocated, 0, GFP_KERNEL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2824) panic("Failed to allocate MPTCP pcpu counter\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2825)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2826) mptcp_subflow_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2827) mptcp_pm_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2828) mptcp_token_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2829)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2830) if (proto_register(&mptcp_prot, 1) != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2831) panic("Failed to register MPTCP proto.\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2832)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2833) inet_register_protosw(&mptcp_protosw);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2834)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2835) BUILD_BUG_ON(sizeof(struct mptcp_skb_cb) > sizeof_field(struct sk_buff, cb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2836) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2837)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2838) #if IS_ENABLED(CONFIG_MPTCP_IPV6)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2839) static const struct proto_ops mptcp_v6_stream_ops = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2840) .family = PF_INET6,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2841) .owner = THIS_MODULE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2842) .release = inet6_release,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2843) .bind = mptcp_bind,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2844) .connect = mptcp_stream_connect,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2845) .socketpair = sock_no_socketpair,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2846) .accept = mptcp_stream_accept,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2847) .getname = inet6_getname,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2848) .poll = mptcp_poll,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2849) .ioctl = inet6_ioctl,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2850) .gettstamp = sock_gettstamp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2851) .listen = mptcp_listen,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2852) .shutdown = mptcp_shutdown,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2853) .setsockopt = sock_common_setsockopt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2854) .getsockopt = sock_common_getsockopt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2855) .sendmsg = inet6_sendmsg,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2856) .recvmsg = inet6_recvmsg,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2857) .mmap = sock_no_mmap,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2858) .sendpage = inet_sendpage,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2859) #ifdef CONFIG_COMPAT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2860) .compat_ioctl = inet6_compat_ioctl,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2861) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2862) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2863)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2864) static struct proto mptcp_v6_prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2865)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2866) static void mptcp_v6_destroy(struct sock *sk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2867) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2868) mptcp_destroy(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2869) inet6_destroy_sock(sk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2870) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2871)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2872) static struct inet_protosw mptcp_v6_protosw = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2873) .type = SOCK_STREAM,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2874) .protocol = IPPROTO_MPTCP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2875) .prot = &mptcp_v6_prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2876) .ops = &mptcp_v6_stream_ops,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2877) .flags = INET_PROTOSW_ICSK,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2878) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2879)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2880) int __init mptcp_proto_v6_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2881) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2882) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2883)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2884) mptcp_v6_prot = mptcp_prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2885) strcpy(mptcp_v6_prot.name, "MPTCPv6");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2886) mptcp_v6_prot.slab = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2887) mptcp_v6_prot.destroy = mptcp_v6_destroy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2888) mptcp_v6_prot.obj_size = sizeof(struct mptcp6_sock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2889)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2890) err = proto_register(&mptcp_v6_prot, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2891) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2892) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2893)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2894) err = inet6_register_protosw(&mptcp_v6_protosw);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2895) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2896) proto_unregister(&mptcp_v6_prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2897)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2898) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2899) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2900) #endif
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags