static void mptcp_ccc_recalc_alpha(struct sock *sk)
{
struct mptcp_cb *mpcb = tcp_sk(sk)->mpcb;
struct sock *sub_sk;
int best_cwnd = 0, best_rtt = 0, can_send = 0;
u64 max_numerator = 0, sum_denominator = 0, alpha = 1;
if (!mpcb)
return;
/* Only one subflow left - fall back to normal reno-behavior
* (set alpha to 1)
*/
if (mpcb->cnt_established <= 1)
goto exit;
/* Do regular alpha-calculation for multiple subflows */
/* Find the max numerator of the alpha-calculation */
mptcp_for_each_sk(mpcb, sub_sk) {
struct tcp_sock *sub_tp = tcp_sk(sub_sk);
u64 tmp;
if (!mptcp_ccc_sk_can_send(sub_sk))
continue;
can_send++;
/* We need to look for the path, that provides the max-value.
* Integer-overflow is not possible here, because
* tmp will be in u64.
*/
tmp = div64_u64(mptcp_ccc_scale(sub_tp->snd_cwnd,
alpha_scale_num), (u64)sub_tp->srtt * sub_tp->srtt);
if (tmp >= max_numerator) {
max_numerator = tmp;
best_cwnd = sub_tp->snd_cwnd;
best_rtt = sub_tp->srtt;
}
}
/* No subflow is able to send - we don't care anymore */
if (unlikely(!can_send))
goto exit;
/* Calculate the denominator */
mptcp_for_each_sk(mpcb, sub_sk) {
struct tcp_sock *sub_tp = tcp_sk(sub_sk);
if (!mptcp_ccc_sk_can_send(sub_sk))
continue;
sum_denominator += div_u64(
mptcp_ccc_scale(sub_tp->snd_cwnd,
alpha_scale_den) * best_rtt,
sub_tp->srtt);
}
sum_denominator *= sum_denominator;
if (unlikely(!sum_denominator)) {
pr_err("%s: sum_denominator == 0, cnt_established:%d\n",
__func__, mpcb->cnt_established);
mptcp_for_each_sk(mpcb, sub_sk) {
struct tcp_sock *sub_tp = tcp_sk(sub_sk);
pr_err("%s: pi:%d, state:%d\n, rtt:%u, cwnd: %u",
__func__, sub_tp->mptcp->path_index,
sub_sk->sk_state, sub_tp->srtt,
sub_tp->snd_cwnd);
}
}
static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
u8 type, u8 code, int offset, __be32 info)
{
const struct ipv6hdr *hdr = (const struct ipv6hdr*)skb->data;
const struct tcphdr *th = (struct tcphdr *)(skb->data+offset);
struct ipv6_pinfo *np;
struct sock *sk;
int err;
struct tcp_sock *tp;
__u32 seq;
struct net *net = dev_net(skb->dev);
sk = inet6_lookup(net, &tcp_hashinfo, &hdr->daddr,
th->dest, &hdr->saddr, th->source, skb->dev->ifindex);
if (sk == NULL) {
ICMP6_INC_STATS_BH(net, __in6_dev_get(skb->dev),
ICMP6_MIB_INERRORS);
return;
}
if (sk->sk_state == TCP_TIME_WAIT) {
inet_twsk_put(inet_twsk(sk));
return;
}
bh_lock_sock(sk);
if (sock_owned_by_user(sk))
NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
if (sk->sk_state == TCP_CLOSE)
goto out;
if (ipv6_hdr(skb)->hop_limit < inet6_sk(sk)->min_hopcount) {
NET_INC_STATS_BH(net, LINUX_MIB_TCPMINTTLDROP);
goto out;
}
tp = tcp_sk(sk);
seq = ntohl(th->seq);
if (sk->sk_state != TCP_LISTEN &&
!between(seq, tp->snd_una, tp->snd_nxt)) {
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
np = inet6_sk(sk);
if (type == ICMPV6_PKT_TOOBIG) {
struct dst_entry *dst;
if (sock_owned_by_user(sk))
goto out;
if ((1 << sk->sk_state) & (TCPF_LISTEN | TCPF_CLOSE))
goto out;
dst = __sk_dst_check(sk, np->dst_cookie);
if (dst == NULL) {
struct inet_sock *inet = inet_sk(sk);
struct flowi6 fl6;
memset(&fl6, 0, sizeof(fl6));
fl6.flowi6_proto = IPPROTO_TCP;
fl6.daddr = np->daddr;
fl6.saddr = np->saddr;
fl6.flowi6_oif = sk->sk_bound_dev_if;
fl6.flowi6_mark = sk->sk_mark;
fl6.fl6_dport = inet->inet_dport;
fl6.fl6_sport = inet->inet_sport;
security_skb_classify_flow(skb, flowi6_to_flowi(&fl6));
dst = ip6_dst_lookup_flow(sk, &fl6, NULL, false);
if (IS_ERR(dst)) {
sk->sk_err_soft = -PTR_ERR(dst);
goto out;
}
} else
dst_hold(dst);
if (inet_csk(sk)->icsk_pmtu_cookie > dst_mtu(dst)) {
tcp_sync_mss(sk, dst_mtu(dst));
tcp_simple_retransmit(sk);
}
dst_release(dst);
goto out;
}
icmpv6_err_convert(type, code, &err);
switch (sk->sk_state) {
struct request_sock *req, **prev;
case TCP_LISTEN:
if (sock_owned_by_user(sk))
goto out;
req = inet6_csk_search_req(sk, &prev, th->dest, &hdr->daddr,
&hdr->saddr, inet6_iif(skb));
if (!req)
goto out;
WARN_ON(req->sk != NULL);
if (seq != tcp_rsk(req)->snt_isn) {
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
inet_csk_reqsk_queue_drop(sk, req, prev);
goto out;
case TCP_SYN_SENT:
case TCP_SYN_RECV:
if (!sock_owned_by_user(sk)) {
sk->sk_err = err;
sk->sk_error_report(sk);
tcp_done(sk);
} else
sk->sk_err_soft = err;
goto out;
}
if (!sock_owned_by_user(sk) && np->recverr) {
sk->sk_err = err;
sk->sk_error_report(sk);
} else
sk->sk_err_soft = err;
out:
bh_unlock_sock(sk);
sock_put(sk);
}
/*
* TCP Westwood
* Here limit is evaluated as Bw estimation*RTTmin (for obtaining it
* in packets we use mss_cache). Rttmin is guaranteed to be >= 2
* so avoids ever returning 0.
*/
static u32 tcp_westwood_bw_rttmin(const struct sock *sk)
{
const struct tcp_sock *tp = tcp_sk(sk);
const struct westwood *w = inet_csk_ca(sk);
return max_t(u32, (w->bw_est * w->rtt_min) / tp->mss_cache, 2);
}
static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack,
u32 seq_rtt, u32 in_flight, int flag)
{
struct tcp_sock *tp = tcp_sk(sk);
struct vegas *vegas = inet_csk_ca(sk);
if (!vegas->doing_vegas_now)
return tcp_reno_cong_avoid(sk, ack, seq_rtt, in_flight, flag);
/* The key players are v_beg_snd_una and v_beg_snd_nxt.
*
* These are so named because they represent the approximate values
* of snd_una and snd_nxt at the beginning of the current RTT. More
* precisely, they represent the amount of data sent during the RTT.
* At the end of the RTT, when we receive an ACK for v_beg_snd_nxt,
* we will calculate that (v_beg_snd_nxt - v_beg_snd_una) outstanding
* bytes of data have been ACKed during the course of the RTT, giving
* an "actual" rate of:
*
* (v_beg_snd_nxt - v_beg_snd_una) / (rtt duration)
*
* Unfortunately, v_beg_snd_una is not exactly equal to snd_una,
* because delayed ACKs can cover more than one segment, so they
* don't line up nicely with the boundaries of RTTs.
*
* Another unfortunate fact of life is that delayed ACKs delay the
* advance of the left edge of our send window, so that the number
* of bytes we send in an RTT is often less than our cwnd will allow.
* So we keep track of our cwnd separately, in v_beg_snd_cwnd.
*/
if (after(ack, vegas->beg_snd_nxt)) {
/* Do the Vegas once-per-RTT cwnd adjustment. */
u32 old_wnd, old_snd_cwnd;
/* Here old_wnd is essentially the window of data that was
* sent during the previous RTT, and has all
* been acknowledged in the course of the RTT that ended
* with the ACK we just received. Likewise, old_snd_cwnd
* is the cwnd during the previous RTT.
*/
old_wnd = (vegas->beg_snd_nxt - vegas->beg_snd_una) /
tp->mss_cache;
old_snd_cwnd = vegas->beg_snd_cwnd;
/* Save the extent of the current window so we can use this
* at the end of the next RTT.
*/
vegas->beg_snd_una = vegas->beg_snd_nxt;
vegas->beg_snd_nxt = tp->snd_nxt;
vegas->beg_snd_cwnd = tp->snd_cwnd;
/* We do the Vegas calculations only if we got enough RTT
* samples that we can be reasonably sure that we got
* at least one RTT sample that wasn't from a delayed ACK.
* If we only had 2 samples total,
* then that means we're getting only 1 ACK per RTT, which
* means they're almost certainly delayed ACKs.
* If we have 3 samples, we should be OK.
*/
if (vegas->cntRTT <= 2) {
/* We don't have enough RTT samples to do the Vegas
* calculation, so we'll behave like Reno.
*/
tcp_reno_cong_avoid(sk, ack, seq_rtt, in_flight, flag);
} else {
u32 rtt, target_cwnd, diff;
/* We have enough RTT samples, so, using the Vegas
* algorithm, we determine if we should increase or
* decrease cwnd, and by how much.
*/
/* Pluck out the RTT we are using for the Vegas
* calculations. This is the min RTT seen during the
* last RTT. Taking the min filters out the effects
* of delayed ACKs, at the cost of noticing congestion
* a bit later.
*/
rtt = vegas->minRTT;
/* Calculate the cwnd we should have, if we weren't
* going too fast.
*
* This is:
* (actual rate in segments) * baseRTT
* We keep it as a fixed point number with
* V_PARAM_SHIFT bits to the right of the binary point.
*/
target_cwnd = ((old_wnd * vegas->baseRTT)
<< V_PARAM_SHIFT) / rtt;
/* Calculate the difference between the window we had,
* and the window we would like to have. This quantity
* is the "Diff" from the Arizona Vegas papers.
*
* Again, this is a fixed point number with
* V_PARAM_SHIFT bits to the right of the binary
* point.
*/
diff = (old_wnd << V_PARAM_SHIFT) - target_cwnd;
if (tp->snd_cwnd <= tp->snd_ssthresh) {
/* Slow start. */
if (diff > (u32)gamma) {
/* Going too fast. Time to slow down
* and switch to congestion avoidance.
*/
tp->snd_ssthresh = 2;
/* Set cwnd to match the actual rate
* exactly:
* cwnd = (actual rate) * baseRTT
* Then we add 1 because the integer
* truncation robs us of full link
* utilization.
*/
tp->snd_cwnd = min(tp->snd_cwnd,
(target_cwnd >>
V_PARAM_SHIFT)+1);
}
tcp_slow_start(tp);
} else {
struct sock *mptcp_select_ack_sock(const struct mptcp_cb *mpcb, int copied)
{
struct sock *sk, *subsk = NULL;
struct tcp_sock *meta_tp = mpcb_meta_tp(mpcb);
u32 max_data_seq = 0;
/* max_data_seq initialized to correct compiler-warning.
* But the initialization is handled by max_data_seq_set */
short max_data_seq_set = 0;
u32 min_time = 0xffffffff;
/* How do we select the subflow to send the window-update on?
*
* 1. He has to be in a state where he can send an ack.
* 2. He has to be one of those subflow who recently
* contributed to the received stream
* (this guarantees a working subflow)
* a) its latest data_seq received is after the original
* copied_seq.
* We select the one with the lowest rtt, so that the
* window-update reaches our peer the fastest.
* b) if no subflow has this kind of data_seq (e.g., very
* strange meta-level retransmissions going on), we take
* the subflow who last sent the highest data_seq.
*/
mptcp_for_each_sk(mpcb, sk) {
struct tcp_sock *tp = tcp_sk(sk);
if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV |
TCPF_CLOSE | TCPF_LISTEN))
continue;
/* Select among those who contributed to the
* current receive-queue. */
if (copied && after(tp->mptcp->last_data_seq, meta_tp->copied_seq - copied)) {
if (tp->srtt < min_time) {
min_time = tp->srtt;
subsk = sk;
max_data_seq_set = 0;
}
continue;
}
if (!subsk && !max_data_seq_set) {
max_data_seq = tp->mptcp->last_data_seq;
max_data_seq_set = 1;
subsk = sk;
}
/* Otherwise, take the one with the highest data_seq */
if ((!subsk || max_data_seq_set) &&
after(tp->mptcp->last_data_seq, max_data_seq)) {
max_data_seq = tp->mptcp->last_data_seq;
subsk = sk;
}
}
if (!subsk) {
mptcp_debug("%s subsk is null, copied %d, cseq %u\n", __func__,
copied, meta_tp->copied_seq);
mptcp_for_each_sk(mpcb, sk) {
struct tcp_sock *tp = tcp_sk(sk);
mptcp_debug("%s pi %d state %u last_dseq %u\n",
__func__, tp->mptcp->path_index, sk->sk_state,
tp->mptcp->last_data_seq);
}
}
void tcp_v4_err(struct sk_buff *skb, u32 info)
{
struct iphdr *iph = (struct iphdr *)skb->data;
struct tcphdr *th = (struct tcphdr *)(skb->data + (iph->ihl << 2));
struct tcp_sock *tp;
struct inet_sock *inet;
int type = skb->h.icmph->type;
int code = skb->h.icmph->code;
struct sock *sk;
__u32 seq;
int err;
if (skb->len < (iph->ihl << 2) + 8) {
ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
return;
}
sk = inet_lookup(&tcp_hashinfo, iph->daddr, th->dest, iph->saddr,
th->source, inet_iif(skb));
if (!sk) {
ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
return;
}
if (sk->sk_state == TCP_TIME_WAIT) {
inet_twsk_put((struct inet_timewait_sock *)sk);
return;
}
bh_lock_sock(sk);
/* If too many ICMPs get dropped on busy
* servers this needs to be solved differently.
*/
if (sock_owned_by_user(sk))
NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);
if (sk->sk_state == TCP_CLOSE)
goto out;
tp = tcp_sk(sk);
seq = ntohl(th->seq);
if (sk->sk_state != TCP_LISTEN &&
!between(seq, tp->snd_una, tp->snd_nxt)) {
NET_INC_STATS(LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
switch (type) {
case ICMP_SOURCE_QUENCH:
/* Just silently ignore these. */
goto out;
case ICMP_PARAMETERPROB:
err = EPROTO;
break;
case ICMP_DEST_UNREACH:
if (code > NR_ICMP_UNREACH)
goto out;
if (code == ICMP_FRAG_NEEDED) { /* PMTU discovery (RFC1191) */
if (!sock_owned_by_user(sk))
do_pmtu_discovery(sk, iph, info);
goto out;
}
err = icmp_err_convert[code].errno;
break;
case ICMP_TIME_EXCEEDED:
err = EHOSTUNREACH;
break;
default:
goto out;
}
switch (sk->sk_state) {
struct request_sock *req, **prev;
case TCP_LISTEN:
if (sock_owned_by_user(sk))
goto out;
req = inet_csk_search_req(sk, &prev, th->dest,
iph->daddr, iph->saddr);
if (!req)
goto out;
/* ICMPs are not backlogged, hence we cannot get
an established socket here.
*/
BUG_TRAP(!req->sk);
if (seq != tcp_rsk(req)->snt_isn) {
NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
/*
* Still in SYN_RECV, just remove it silently.
* There is no good way to pass the error to the newly
* created socket, and POSIX does not want network
* errors returned from accept().
*/
inet_csk_reqsk_queue_drop(sk, req, prev);
goto out;
case TCP_SYN_SENT:
case TCP_SYN_RECV: /* Cannot happen.
It can f.e. if SYNs crossed.
*/
if (!sock_owned_by_user(sk)) {
TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS);
sk->sk_err = err;
sk->sk_error_report(sk);
tcp_done(sk);
} else {
sk->sk_err_soft = err;
}
goto out;
}
/* If we've already connected we will keep trying
* until we time out, or the user gives up.
*
* rfc1122 4.2.3.9 allows to consider as hard errors
* only PROTO_UNREACH and PORT_UNREACH (well, FRAG_FAILED too,
* but it is obsoleted by pmtu discovery).
*
* Note, that in modern internet, where routing is unreliable
* and in each dark corner broken firewalls sit, sending random
* errors ordered by their masters even this two messages finally lose
* their original sense (even Linux sends invalid PORT_UNREACHs)
*
* Now we are in compliance with RFCs.
* --ANK (980905)
*/
inet = inet_sk(sk);
if (!sock_owned_by_user(sk) && inet->recverr) {
sk->sk_err = err;
sk->sk_error_report(sk);
} else { /* Only an error on timeout */
sk->sk_err_soft = err;
}
out:
bh_unlock_sock(sk);
sock_put(sk);
}
static void ctcp_cwnd_event(struct sock *sk)
{
tcp_sk(sk)->snd_cwnd = instance;
}
static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
int addr_len)
{
struct sockaddr_in6 *usin = (struct sockaddr_in6 *) uaddr;
struct inet_sock *inet = inet_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
struct ipv6_pinfo *np = inet6_sk(sk);
struct tcp_sock *tp = tcp_sk(sk);
struct in6_addr *saddr = NULL, *final_p = NULL, final;
struct flowi fl;
struct dst_entry *dst;
int addr_type;
int err;
if (addr_len < SIN6_LEN_RFC2133)
return -EINVAL;
if (usin->sin6_family != AF_INET6)
return(-EAFNOSUPPORT);
memset(&fl, 0, sizeof(fl));
if (np->sndflow) {
fl.fl6_flowlabel = usin->sin6_flowinfo&IPV6_FLOWINFO_MASK;
IP6_ECN_flow_init(fl.fl6_flowlabel);
if (fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) {
struct ip6_flowlabel *flowlabel;
flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel);
if (flowlabel == NULL)
return -EINVAL;
ipv6_addr_copy(&usin->sin6_addr, &flowlabel->dst);
fl6_sock_release(flowlabel);
}
}
/*
* connect() to INADDR_ANY means loopback (BSD'ism).
*/
if(ipv6_addr_any(&usin->sin6_addr))
usin->sin6_addr.s6_addr[15] = 0x1;
addr_type = ipv6_addr_type(&usin->sin6_addr);
if(addr_type & IPV6_ADDR_MULTICAST)
return -ENETUNREACH;
if (addr_type&IPV6_ADDR_LINKLOCAL) {
if (addr_len >= sizeof(struct sockaddr_in6) &&
usin->sin6_scope_id) {
/* If interface is set while binding, indices
* must coincide.
*/
if (sk->sk_bound_dev_if &&
sk->sk_bound_dev_if != usin->sin6_scope_id)
return -EINVAL;
sk->sk_bound_dev_if = usin->sin6_scope_id;
}
/* Connect to link-local address requires an interface */
if (!sk->sk_bound_dev_if)
return -EINVAL;
}
if (tp->rx_opt.ts_recent_stamp &&
!ipv6_addr_equal(&np->daddr, &usin->sin6_addr)) {
tp->rx_opt.ts_recent = 0;
tp->rx_opt.ts_recent_stamp = 0;
tp->write_seq = 0;
}
ipv6_addr_copy(&np->daddr, &usin->sin6_addr);
np->flow_label = fl.fl6_flowlabel;
/*
* TCP over IPv4
*/
if (addr_type == IPV6_ADDR_MAPPED) {
u32 exthdrlen = icsk->icsk_ext_hdr_len;
struct sockaddr_in sin;
SOCK_DEBUG(sk, "connect: ipv4 mapped\n");
if (__ipv6_only_sock(sk))
return -ENETUNREACH;
sin.sin_family = AF_INET;
sin.sin_port = usin->sin6_port;
sin.sin_addr.s_addr = usin->sin6_addr.s6_addr32[3];
icsk->icsk_af_ops = &ipv6_mapped;
sk->sk_backlog_rcv = tcp_v4_do_rcv;
err = tcp_v4_connect(sk, (struct sockaddr *)&sin, sizeof(sin));
if (err) {
icsk->icsk_ext_hdr_len = exthdrlen;
icsk->icsk_af_ops = &ipv6_specific;
sk->sk_backlog_rcv = tcp_v6_do_rcv;
goto failure;
} else {
ipv6_addr_set(&np->saddr, 0, 0, htonl(0x0000FFFF),
inet->saddr);
ipv6_addr_set(&np->rcv_saddr, 0, 0, htonl(0x0000FFFF),
inet->rcv_saddr);
}
return err;
}
if (!ipv6_addr_any(&np->rcv_saddr))
saddr = &np->rcv_saddr;
fl.proto = IPPROTO_TCP;
ipv6_addr_copy(&fl.fl6_dst, &np->daddr);
ipv6_addr_copy(&fl.fl6_src,
(saddr ? saddr : &np->saddr));
fl.oif = sk->sk_bound_dev_if;
fl.fl_ip_dport = usin->sin6_port;
fl.fl_ip_sport = inet->sport;
if (np->opt && np->opt->srcrt) {
struct rt0_hdr *rt0 = (struct rt0_hdr *)np->opt->srcrt;
ipv6_addr_copy(&final, &fl.fl6_dst);
ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
final_p = &final;
}
err = ip6_dst_lookup(sk, &dst, &fl);
if (err)
goto failure;
if (final_p) {
ipv6_addr_copy(&fl.fl6_dst, final_p);
fl.flags |= FLOWI_FLAG_NOTROUTE;
}
if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0)
goto failure;
if (saddr == NULL) {
saddr = &fl.fl6_src;
ipv6_addr_copy(&np->rcv_saddr, saddr);
}
/* set the source address */
ipv6_addr_copy(&np->saddr, saddr);
inet->rcv_saddr = LOOPBACK4_IPV6;
ip6_dst_store(sk, dst, NULL);
sk->sk_route_caps = dst->dev->features &
~(NETIF_F_IP_CSUM | NETIF_F_TSO);
icsk->icsk_ext_hdr_len = 0;
if (np->opt)
icsk->icsk_ext_hdr_len = (np->opt->opt_flen +
np->opt->opt_nflen);
tp->rx_opt.mss_clamp = IPV6_MIN_MTU - sizeof(struct tcphdr) - sizeof(struct ipv6hdr);
inet->dport = usin->sin6_port;
tcp_set_state(sk, TCP_SYN_SENT);
err = inet6_hash_connect(&tcp_death_row, sk);
if (err)
goto late_failure;
if (!tp->write_seq)
tp->write_seq = secure_tcpv6_sequence_number(np->saddr.s6_addr32,
np->daddr.s6_addr32,
inet->sport,
inet->dport);
err = tcp_connect(sk);
if (err)
goto late_failure;
return 0;
late_failure:
tcp_set_state(sk, TCP_CLOSE);
__sk_dst_reset(sk);
failure:
inet->dport = 0;
sk->sk_route_caps = 0;
return err;
}
/*
* BST_TMR_DELAY_SOCK_SYNC timeout.
* If sock is ready, get sock sync properties and post them to daemon
*/
static void delay_sock_bastet_timeout(struct sock *sk)
{
int err;
struct bst_set_sock_sync_prop sock_p;
struct bastet_sock *bsk = sk->bastet;
/* Accurating time */
if (time_after(bsk->bastet_timeout, jiffies)) {
sk_reset_timer(sk, &bsk->bastet_timer, bsk->bastet_timeout);
return;
}
/* We must reset timer event, bastet_delay_sock_sync_notify depends on it
* this code must be put after accurating time
*/
bsk->bastet_timer_event = BST_TMR_EVT_INVALID;
/* In repair mode or userspace needs repair, do not sync sock */
if (unlikely(tcp_sk(sk)->repair || bsk->need_repair)) {
BASTET_LOGE("sk: %p in repair mode", sk);
return;
}
if (TCP_ESTABLISHED != sk->sk_state) {
BASTET_LOGE("sk: %p sk_state is not TCP_ESTABLISHED", sk);
return;
}
if (BST_SOCK_VALID != bsk->bastet_sock_state) {
BASTET_LOGE("sk: %p state: %d not expected", sk, bsk->bastet_sock_state);
return;
}
/* Sock owner has used since last setup */
if (time_after(bsk->last_sock_active_time_point + bsk->delay_sync_time_section, jiffies)) {
setup_sock_sync_delay_timer(sk);
return;
}
/* Sock owner has some data unacked,
* Coming ack would trigger delay timer again */
if (!tcp_write_queue_empty(sk)) {
BASTET_LOGI("sk: %p has sent data not acked", sk);
post_indicate_packet(BST_IND_TRIGGER_THAW, &bsk->pid, sizeof(pid_t));
return;
}
/* Sock owner has some data to recv, do not sync.
* If sock owner has none recv action,
* delay timer should be stopped. */
if (!skb_queue_empty(&sk->sk_receive_queue)) {
BASTET_LOGI("sk: %p has received data in queue", sk);
bsk->last_sock_active_time_point = jiffies;
setup_sock_sync_delay_timer(sk);
post_indicate_packet(BST_IND_TRIGGER_THAW, &bsk->pid, sizeof(pid_t));
return;
}
memset(&sock_p, 0, sizeof(struct bst_set_sock_sync_prop));
bastet_get_comm_prop(sk, &sock_p.guide);
bastet_get_sock_prop(sk, &sock_p.sync_prop);
err = post_indicate_packet(BST_IND_SOCK_SYNC_PROP, &sock_p, sizeof(sock_p));
if (!err) {
/* if post success */
bsk->bastet_sock_state = BST_SOCK_INVALID;
}
}
static bool tcp_fastopen_create_child(struct sock *sk,
struct sk_buff *skb,
struct dst_entry *dst,
struct request_sock *req)
{
struct tcp_sock *tp = tcp_sk(sk);
struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
struct sock *child;
req->num_retrans = 0;
req->num_timeout = 0;
req->sk = NULL;
child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb, req, NULL);
if (child == NULL)
return false;
spin_lock(&queue->fastopenq->lock);
queue->fastopenq->qlen++;
spin_unlock(&queue->fastopenq->lock);
/* Initialize the child socket. Have to fix some values to take
* into account the child is a Fast Open socket and is created
* only out of the bits carried in the SYN packet.
*/
tp = tcp_sk(child);
tp->fastopen_rsk = req;
/* Do a hold on the listner sk so that if the listener is being
* closed, the child that has been accepted can live on and still
* access listen_lock.
*/
sock_hold(sk);
tcp_rsk(req)->listener = sk;
/* RFC1323: The window in SYN & SYN/ACK segments is never
* scaled. So correct it appropriately.
*/
tp->snd_wnd = ntohs(tcp_hdr(skb)->window);
/* Activate the retrans timer so that SYNACK can be retransmitted.
* The request socket is not added to the SYN table of the parent
* because it's been added to the accept queue directly.
*/
inet_csk_reset_xmit_timer(child, ICSK_TIME_RETRANS,
TCP_TIMEOUT_INIT, TCP_RTO_MAX);
/* Add the child socket directly into the accept queue */
inet_csk_reqsk_queue_add(sk, req, child);
/* Now finish processing the fastopen child socket. */
inet_csk(child)->icsk_af_ops->rebuild_header(child);
tcp_init_congestion_control(child);
tcp_mtup_init(child);
tcp_init_metrics(child);
tcp_init_buffer_space(child);
/* Queue the data carried in the SYN packet. We need to first
* bump skb's refcnt because the caller will attempt to free it.
*
* XXX (TFO) - we honor a zero-payload TFO request for now,
* (any reason not to?) but no need to queue the skb since
* there is no data. How about SYN+FIN?
*/
if (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq + 1) {
skb = skb_get(skb);
skb_dst_drop(skb);
__skb_pull(skb, tcp_hdr(skb)->doff * 4);
skb_set_owner_r(skb, child);
__skb_queue_tail(&child->sk_receive_queue, skb);
tp->syn_data_acked = 1;
}
tcp_rsk(req)->rcv_nxt = tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
sk->sk_data_ready(sk, 0);
bh_unlock_sock(child);
sock_put(child);
WARN_ON(req->sk == NULL);
return true;
}
static inline int mptcp_olia_sk_can_send(const struct sock *sk)
{
return mptcp_sk_can_send(sk) && tcp_sk(sk)->srtt;
}
static struct sock *tcp_fastopen_create_child(struct sock *sk,
struct sk_buff *skb,
struct dst_entry *dst,
struct request_sock *req)
{
struct tcp_sock *tp;
struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
struct sock *child;
bool own_req;
req->num_retrans = 0;
req->num_timeout = 0;
req->sk = NULL;
child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb, req, NULL,
NULL, &own_req);
if (!child)
return NULL;
spin_lock(&queue->fastopenq.lock);
queue->fastopenq.qlen++;
spin_unlock(&queue->fastopenq.lock);
/* Initialize the child socket. Have to fix some values to take
* into account the child is a Fast Open socket and is created
* only out of the bits carried in the SYN packet.
*/
tp = tcp_sk(child);
tp->fastopen_rsk = req;
tcp_rsk(req)->tfo_listener = true;
/* RFC1323: The window in SYN & SYN/ACK segments is never
* scaled. So correct it appropriately.
*/
tp->snd_wnd = ntohs(tcp_hdr(skb)->window);
/* Activate the retrans timer so that SYNACK can be retransmitted.
* The request socket is not added to the ehash
* because it's been added to the accept queue directly.
*/
inet_csk_reset_xmit_timer(child, ICSK_TIME_RETRANS,
TCP_TIMEOUT_INIT, TCP_RTO_MAX);
atomic_set(&req->rsk_refcnt, 2);
/* Now finish processing the fastopen child socket. */
inet_csk(child)->icsk_af_ops->rebuild_header(child);
tcp_init_congestion_control(child);
tcp_mtup_init(child);
tcp_init_metrics(child);
tcp_init_buffer_space(child);
tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1;
tcp_fastopen_add_skb(child, skb);
tcp_rsk(req)->rcv_nxt = tp->rcv_nxt;
tp->rcv_wup = tp->rcv_nxt;
/* tcp_conn_request() is sending the SYNACK,
* and queues the child into listener accept queue.
*/
return child;
}
int ipv6_setsockopt(struct sock *sk, int level, int optname,
char __user *optval, int optlen)
{
struct ipv6_pinfo *np = inet6_sk(sk);
int val, valbool;
int retv = -ENOPROTOOPT;
if (level == SOL_IP && sk->sk_type != SOCK_RAW)
return udp_prot.setsockopt(sk, level, optname, optval, optlen);
if(level!=SOL_IPV6)
goto out;
if (optval == NULL)
val=0;
else if (get_user(val, (int __user *) optval))
return -EFAULT;
valbool = (val!=0);
lock_sock(sk);
switch (optname) {
case IPV6_ADDRFORM:
if (val == PF_INET) {
struct ipv6_txoptions *opt;
struct sk_buff *pktopt;
if (sk->sk_protocol != IPPROTO_UDP &&
sk->sk_protocol != IPPROTO_TCP)
break;
if (sk->sk_state != TCP_ESTABLISHED) {
retv = -ENOTCONN;
break;
}
if (ipv6_only_sock(sk) ||
!(ipv6_addr_type(&np->daddr) & IPV6_ADDR_MAPPED)) {
retv = -EADDRNOTAVAIL;
break;
}
fl6_free_socklist(sk);
ipv6_sock_mc_close(sk);
if (sk->sk_protocol == IPPROTO_TCP) {
struct tcp_sock *tp = tcp_sk(sk);
local_bh_disable();
sock_prot_dec_use(sk->sk_prot);
sock_prot_inc_use(&tcp_prot);
local_bh_enable();
sk->sk_prot = &tcp_prot;
tp->af_specific = &ipv4_specific;
sk->sk_socket->ops = &inet_stream_ops;
sk->sk_family = PF_INET;
tcp_sync_mss(sk, tp->pmtu_cookie);
} else {
local_bh_disable();
sock_prot_dec_use(sk->sk_prot);
sock_prot_inc_use(&udp_prot);
local_bh_enable();
sk->sk_prot = &udp_prot;
sk->sk_socket->ops = &inet_dgram_ops;
sk->sk_family = PF_INET;
}
opt = xchg(&np->opt, NULL);
if (opt)
sock_kfree_s(sk, opt, opt->tot_len);
pktopt = xchg(&np->pktoptions, NULL);
if (pktopt)
kfree_skb(pktopt);
sk->sk_destruct = inet_sock_destruct;
#ifdef INET_REFCNT_DEBUG
atomic_dec(&inet6_sock_nr);
#endif
module_put(THIS_MODULE);
retv = 0;
break;
}
goto e_inval;
case IPV6_V6ONLY:
if (inet_sk(sk)->num)
goto e_inval;
np->ipv6only = valbool;
retv = 0;
break;
case IPV6_PKTINFO:
np->rxopt.bits.rxinfo = valbool;
retv = 0;
break;
case IPV6_HOPLIMIT:
np->rxopt.bits.rxhlim = valbool;
retv = 0;
break;
case IPV6_RTHDR:
if (val < 0 || val > 2)
goto e_inval;
np->rxopt.bits.srcrt = val;
retv = 0;
break;
case IPV6_HOPOPTS:
np->rxopt.bits.hopopts = valbool;
retv = 0;
break;
case IPV6_DSTOPTS:
np->rxopt.bits.dstopts = valbool;
retv = 0;
break;
case IPV6_FLOWINFO:
np->rxopt.bits.rxflow = valbool;
retv = 0;
break;
case IPV6_PKTOPTIONS:
{
struct ipv6_txoptions *opt = NULL;
struct msghdr msg;
struct flowi fl;
int junk;
fl.fl6_flowlabel = 0;
fl.oif = sk->sk_bound_dev_if;
if (optlen == 0)
goto update;
/* 1K is probably excessive
* 1K is surely not enough, 2K per standard header is 16K.
*/
retv = -EINVAL;
if (optlen > 64*1024)
break;
opt = sock_kmalloc(sk, sizeof(*opt) + optlen, GFP_KERNEL);
retv = -ENOBUFS;
if (opt == NULL)
break;
memset(opt, 0, sizeof(*opt));
opt->tot_len = sizeof(*opt) + optlen;
retv = -EFAULT;
if (copy_from_user(opt+1, optval, optlen))
goto done;
msg.msg_controllen = optlen;
msg.msg_control = (void*)(opt+1);
retv = datagram_send_ctl(&msg, &fl, opt, &junk);
if (retv)
goto done;
update:
retv = 0;
if (sk->sk_type == SOCK_STREAM) {
if (opt) {
struct tcp_sock *tp = tcp_sk(sk);
if (!((1 << sk->sk_state) &
(TCPF_LISTEN | TCPF_CLOSE))
&& inet_sk(sk)->daddr != LOOPBACK4_IPV6) {
tp->ext_header_len = opt->opt_flen + opt->opt_nflen;
tcp_sync_mss(sk, tp->pmtu_cookie);
}
}
opt = xchg(&np->opt, opt);
sk_dst_reset(sk);
} else {
write_lock(&sk->sk_dst_lock);
opt = xchg(&np->opt, opt);
write_unlock(&sk->sk_dst_lock);
sk_dst_reset(sk);
}
done:
if (opt)
sock_kfree_s(sk, opt, opt->tot_len);
break;
}
case IPV6_UNICAST_HOPS:
if (val > 255 || val < -1)
goto e_inval;
np->hop_limit = val;
retv = 0;
break;
case IPV6_MULTICAST_HOPS:
if (sk->sk_type == SOCK_STREAM)
goto e_inval;
if (val > 255 || val < -1)
goto e_inval;
np->mcast_hops = val;
retv = 0;
break;
case IPV6_MULTICAST_LOOP:
np->mc_loop = valbool;
retv = 0;
break;
case IPV6_MULTICAST_IF:
if (sk->sk_type == SOCK_STREAM)
goto e_inval;
if (sk->sk_bound_dev_if && sk->sk_bound_dev_if != val)
goto e_inval;
if (__dev_get_by_index(val) == NULL) {
retv = -ENODEV;
break;
}
np->mcast_oif = val;
retv = 0;
break;
case IPV6_ADD_MEMBERSHIP:
case IPV6_DROP_MEMBERSHIP:
{
struct ipv6_mreq mreq;
retv = -EFAULT;
if (copy_from_user(&mreq, optval, sizeof(struct ipv6_mreq)))
break;
if (optname == IPV6_ADD_MEMBERSHIP)
retv = ipv6_sock_mc_join(sk, mreq.ipv6mr_ifindex, &mreq.ipv6mr_multiaddr);
else
retv = ipv6_sock_mc_drop(sk, mreq.ipv6mr_ifindex, &mreq.ipv6mr_multiaddr);
break;
}
case IPV6_JOIN_ANYCAST:
case IPV6_LEAVE_ANYCAST:
{
struct ipv6_mreq mreq;
if (optlen != sizeof(struct ipv6_mreq))
goto e_inval;
retv = -EFAULT;
if (copy_from_user(&mreq, optval, sizeof(struct ipv6_mreq)))
break;
if (optname == IPV6_JOIN_ANYCAST)
retv = ipv6_sock_ac_join(sk, mreq.ipv6mr_ifindex, &mreq.ipv6mr_acaddr);
else
retv = ipv6_sock_ac_drop(sk, mreq.ipv6mr_ifindex, &mreq.ipv6mr_acaddr);
break;
}
case MCAST_JOIN_GROUP:
case MCAST_LEAVE_GROUP:
{
struct group_req greq;
struct sockaddr_in6 *psin6;
retv = -EFAULT;
if (copy_from_user(&greq, optval, sizeof(struct group_req)))
break;
if (greq.gr_group.ss_family != AF_INET6) {
retv = -EADDRNOTAVAIL;
break;
}
psin6 = (struct sockaddr_in6 *)&greq.gr_group;
if (optname == MCAST_JOIN_GROUP)
retv = ipv6_sock_mc_join(sk, greq.gr_interface,
&psin6->sin6_addr);
else
retv = ipv6_sock_mc_drop(sk, greq.gr_interface,
&psin6->sin6_addr);
break;
}
case MCAST_JOIN_SOURCE_GROUP:
case MCAST_LEAVE_SOURCE_GROUP:
case MCAST_BLOCK_SOURCE:
case MCAST_UNBLOCK_SOURCE:
{
struct group_source_req greqs;
int omode, add;
if (optlen != sizeof(struct group_source_req))
goto e_inval;
if (copy_from_user(&greqs, optval, sizeof(greqs))) {
retv = -EFAULT;
break;
}
if (greqs.gsr_group.ss_family != AF_INET6 ||
greqs.gsr_source.ss_family != AF_INET6) {
retv = -EADDRNOTAVAIL;
break;
}
if (optname == MCAST_BLOCK_SOURCE) {
omode = MCAST_EXCLUDE;
add = 1;
} else if (optname == MCAST_UNBLOCK_SOURCE) {
omode = MCAST_EXCLUDE;
add = 0;
} else if (optname == MCAST_JOIN_SOURCE_GROUP) {
struct sockaddr_in6 *psin6;
psin6 = (struct sockaddr_in6 *)&greqs.gsr_group;
retv = ipv6_sock_mc_join(sk, greqs.gsr_interface,
&psin6->sin6_addr);
if (retv)
break;
omode = MCAST_INCLUDE;
add = 1;
} else /*IP_DROP_SOURCE_MEMBERSHIP */ {
omode = MCAST_INCLUDE;
add = 0;
}
retv = ip6_mc_source(add, omode, sk, &greqs);
break;
}
case MCAST_MSFILTER:
{
extern int sysctl_optmem_max;
extern int sysctl_mld_max_msf;
struct group_filter *gsf;
if (optlen < GROUP_FILTER_SIZE(0))
goto e_inval;
if (optlen > sysctl_optmem_max) {
retv = -ENOBUFS;
break;
}
gsf = (struct group_filter *)kmalloc(optlen,GFP_KERNEL);
if (gsf == 0) {
retv = -ENOBUFS;
break;
}
retv = -EFAULT;
if (copy_from_user(gsf, optval, optlen)) {
kfree(gsf);
break;
}
/* numsrc >= (4G-140)/128 overflow in 32 bits */
if (gsf->gf_numsrc >= 0x1ffffffU ||
gsf->gf_numsrc > sysctl_mld_max_msf) {
kfree(gsf);
retv = -ENOBUFS;
break;
}
if (GROUP_FILTER_SIZE(gsf->gf_numsrc) > optlen) {
kfree(gsf);
retv = -EINVAL;
break;
}
retv = ip6_mc_msfilter(sk, gsf);
kfree(gsf);
break;
}
case IPV6_ROUTER_ALERT:
retv = ip6_ra_control(sk, val, NULL);
break;
case IPV6_MTU_DISCOVER:
if (val<0 || val>2)
goto e_inval;
np->pmtudisc = val;
retv = 0;
break;
case IPV6_MTU:
if (val && val < IPV6_MIN_MTU)
goto e_inval;
np->frag_size = val;
retv = 0;
break;
case IPV6_RECVERR:
np->recverr = valbool;
if (!val)
skb_queue_purge(&sk->sk_error_queue);
retv = 0;
break;
case IPV6_FLOWINFO_SEND:
np->sndflow = valbool;
retv = 0;
break;
case IPV6_FLOWLABEL_MGR:
retv = ipv6_flowlabel_opt(sk, optval, optlen);
break;
case IPV6_IPSEC_POLICY:
case IPV6_XFRM_POLICY:
retv = -EPERM;
if (!capable(CAP_NET_ADMIN))
break;
retv = xfrm_user_policy(sk, optname, optval, optlen);
break;
#ifdef CONFIG_NETFILTER
default:
retv = nf_setsockopt(sk, PF_INET6, optname, optval,
optlen);
break;
#endif
}
release_sock(sk);
out:
return retv;
e_inval:
release_sock(sk);
return -EINVAL;
}
/**
* xs_tcp_connect_worker - connect a TCP socket to a remote endpoint
* @args: RPC transport to connect
*
* Invoked by a work queue tasklet.
*/
static void xs_tcp_connect_worker(void *args)
{
struct rpc_xprt *xprt = (struct rpc_xprt *)args;
struct socket *sock = xprt->sock;
int err, status = -EIO;
if (xprt->shutdown || xprt->addr.sin_port == 0)
goto out;
dprintk("RPC: xs_tcp_connect_worker for xprt %p\n", xprt);
if (!xprt->sock) {
/* start from scratch */
if ((err = sock_create_kern(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock)) < 0) {
dprintk("RPC: can't create TCP transport socket (%d).\n", -err);
goto out;
}
if (xprt->resvport && xs_bindresvport(xprt, sock) < 0) {
sock_release(sock);
goto out;
}
} else
/* "close" the socket, preserving the local port */
xs_tcp_reuse_connection(xprt);
if (!xprt->inet) {
struct sock *sk = sock->sk;
write_lock_bh(&sk->sk_callback_lock);
sk->sk_user_data = xprt;
xprt->old_data_ready = sk->sk_data_ready;
xprt->old_state_change = sk->sk_state_change;
xprt->old_write_space = sk->sk_write_space;
sk->sk_data_ready = xs_tcp_data_ready;
sk->sk_state_change = xs_tcp_state_change;
sk->sk_write_space = xs_tcp_write_space;
sk->sk_allocation = GFP_ATOMIC;
/* socket options */
sk->sk_userlocks |= SOCK_BINDPORT_LOCK;
sock_reset_flag(sk, SOCK_LINGER);
tcp_sk(sk)->linger2 = 0;
tcp_sk(sk)->nonagle |= TCP_NAGLE_OFF;
xprt_clear_connected(xprt);
/* Reset to new socket */
xprt->sock = sock;
xprt->inet = sk;
write_unlock_bh(&sk->sk_callback_lock);
}
/* Tell the socket layer to start connecting... */
status = sock->ops->connect(sock, (struct sockaddr *) &xprt->addr,
sizeof(xprt->addr), O_NONBLOCK);
dprintk("RPC: %p connect status %d connected %d sock state %d\n",
xprt, -status, xprt_connected(xprt), sock->sk->sk_state);
if (status < 0) {
switch (status) {
case -EINPROGRESS:
case -EALREADY:
goto out_clear;
case -ECONNREFUSED:
case -ECONNRESET:
/* retry with existing socket, after a delay */
break;
default:
/* get rid of existing socket, and retry */
xs_close(xprt);
break;
}
}
out:
xprt_wake_pending_tasks(xprt, status);
out_clear:
xprt_clear_connecting(xprt);
}
static void tcp_veno_cong_avoid(struct sock *sk, u32 ack,
u32 seq_rtt, u32 in_flight, int flag)
{
struct tcp_sock *tp = tcp_sk(sk);
struct veno *veno = inet_csk_ca(sk);
if (!veno->doing_veno_now)
return tcp_reno_cong_avoid(sk, ack, seq_rtt, in_flight, flag);
/* limited by applications */
if (!tcp_is_cwnd_limited(sk, in_flight))
return;
/* We do the Veno calculations only if we got enough rtt samples */
if (veno->cntrtt <= 2) {
/* We don't have enough rtt samples to do the Veno
* calculation, so we'll behave like Reno.
*/
tcp_reno_cong_avoid(sk, ack, seq_rtt, in_flight, flag);
} else {
u32 rtt, target_cwnd;
/* We have enough rtt samples, so, using the Veno
* algorithm, we determine the state of the network.
*/
rtt = veno->minrtt;
target_cwnd = ((tp->snd_cwnd * veno->basertt)
<< V_PARAM_SHIFT) / rtt;
veno->diff = (tp->snd_cwnd << V_PARAM_SHIFT) - target_cwnd;
if (tp->snd_cwnd <= tp->snd_ssthresh) {
/* Slow start. */
tcp_slow_start(tp);
} else {
/* Congestion avoidance. */
if (veno->diff < (u32)beta) {
/* In the "non-congestive state", increase cwnd
* every rtt.
*/
if (tp->snd_cwnd_cnt >= tp->snd_cwnd) {
if (tp->snd_cwnd < tp->snd_cwnd_clamp)
tp->snd_cwnd++;
tp->snd_cwnd_cnt = 0;
} else
tp->snd_cwnd_cnt++;
} else {
/* In the "congestive state", increase cwnd
* every other rtt.
*/
if (tp->snd_cwnd_cnt >= tp->snd_cwnd) {
if (veno->inc
&& tp->snd_cwnd <
tp->snd_cwnd_clamp) {
tp->snd_cwnd++;
veno->inc = 0;
} else
veno->inc = 1;
tp->snd_cwnd_cnt = 0;
} else
tp->snd_cwnd_cnt++;
}
}
if (tp->snd_cwnd < 2)
tp->snd_cwnd = 2;
else if (tp->snd_cwnd > tp->snd_cwnd_clamp)
tp->snd_cwnd = tp->snd_cwnd_clamp;
}
/* Wipe the slate clean for the next rtt. */
/* veno->cntrtt = 0; */
veno->minrtt = 0x7fffffff;
}
/* Save metrics learned by this TCP session. This function is called
* only, when TCP finishes successfully i.e. when it enters TIME-WAIT
* or goes from LAST-ACK to CLOSE.
*/
void tcp_update_metrics(struct sock *sk)
{
const struct inet_connection_sock *icsk = inet_csk(sk);
struct dst_entry *dst = __sk_dst_get(sk);
struct tcp_sock *tp = tcp_sk(sk);
struct tcp_metrics_block *tm;
unsigned long rtt;
u32 val;
int m;
if (sysctl_tcp_nometrics_save || !dst)
return;
if (dst->flags & DST_HOST)
dst_confirm(dst);
rcu_read_lock();
if (icsk->icsk_backoff || !tp->srtt_us) {
/* This session failed to estimate rtt. Why?
* Probably, no packets returned in time. Reset our
* results.
*/
tm = tcp_get_metrics(sk, dst, false);
if (tm && !tcp_metric_locked(tm, TCP_METRIC_RTT))
tcp_metric_set(tm, TCP_METRIC_RTT, 0);
goto out_unlock;
} else
tm = tcp_get_metrics(sk, dst, true);
if (!tm)
goto out_unlock;
rtt = tcp_metric_get(tm, TCP_METRIC_RTT);
m = rtt - tp->srtt_us;
/* If newly calculated rtt larger than stored one, store new
* one. Otherwise, use EWMA. Remember, rtt overestimation is
* always better than underestimation.
*/
if (!tcp_metric_locked(tm, TCP_METRIC_RTT)) {
if (m <= 0)
rtt = tp->srtt_us;
else
rtt -= (m >> 3);
tcp_metric_set(tm, TCP_METRIC_RTT, rtt);
}
if (!tcp_metric_locked(tm, TCP_METRIC_RTTVAR)) {
unsigned long var;
if (m < 0)
m = -m;
/* Scale deviation to rttvar fixed point */
m >>= 1;
if (m < tp->mdev_us)
m = tp->mdev_us;
var = tcp_metric_get(tm, TCP_METRIC_RTTVAR);
if (m >= var)
var = m;
else
var -= (var - m) >> 2;
tcp_metric_set(tm, TCP_METRIC_RTTVAR, var);
}
//start_listening
void moon_listen(void)
{
//For test purposes
moonraker_socket_t *listen;
listen = kmalloc(sizeof(moonraker_socket_t),GFP_KERNEL);
listen->proto = kmalloc(sizeof(moonraker_proto_t),GFP_KERNEL);
listen->proto->name = kmalloc(5,GFP_KERNEL);
listen->ip = 2130706433;
listen->port = 80;
listen->proto->defer_accept=0;
listen->keepalive_timeout=0;
listen->ack_pingpong=1;
listen->max_backlog=2048;
listen->defer_accept=1;
strcpy(listen->proto->name,"http");
//end for test purpose
struct sockaddr_in sin;
struct socket *sock = NULL;
struct sock *sk;
struct tcp_sock *tp;
struct inet_connection_sock *icsk;
moonraker_proto_t *proto = listen->proto;
u16 port = listen->port;
u32 addr = listen->ip; //127.0.0.1
int err = 0;
err = sock_create_kern(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock);
if (err) {
printk(KERN_ERR "Moonraker: error %d creating socket.\n", err);
goto error;
}
sin.sin_family = AF_INET;
sin.sin_addr.s_addr = htonl(addr);
sin.sin_port = htons(port);
sk = sock->sk;
icsk = inet_csk(sk);
sk->sk_reuse = 1;
sock_set_flag(sk, SOCK_URGINLINE);
err = sock->ops->bind(sock, (struct sockaddr*)&sin, sizeof(sin));
if (err){
printk(KERN_ERR "Moonraker: error %d binding socket. This means that probably some other process is (or was a short time ago) using addr %d\n",err,sin.sin_addr.s_addr);
goto error;
}
tp = tcp_sk(sk);
printk("listen sk accept_queue: %d.\n",!reqsk_queue_empty(&icsk->icsk_accept_queue));
icsk->icsk_ack.pingpong = listen->ack_pingpong;
sock_reset_flag(sk, SOCK_LINGER);
sk->sk_lingertime = 0;
tp->linger2 = listen->keepalive_timeout * HZ;
if (proto->defer_accept && !listen->keepalive_timeout && listen->defer_accept)
icsk->icsk_accept_queue.rskq_defer_accept = 1;
err = sock->ops->listen(sock, listen->max_backlog);
if (err) {
printk(KERN_ERR "Moonraker: error %d listening on socket.\n", err);
goto error;
}
printk(KERN_NOTICE "Moonraker: thread %d listens on %s://%d.%d.%d.%d:%d.\n",
1, proto->name, HIPQUAD(addr), port);
// return sock;
return;
error:
if (sock)
sock_release(sock);
return;
return NULL;
}
static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
int type, int code, int offset, __u32 info)
{
struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data;
const struct tcphdr *th = (struct tcphdr *)(skb->data+offset);
struct ipv6_pinfo *np;
struct sock *sk;
int err;
struct tcp_sock *tp;
__u32 seq;
sk = inet6_lookup(&tcp_hashinfo, &hdr->daddr, th->dest, &hdr->saddr,
th->source, skb->dev->ifindex);
if (sk == NULL) {
ICMP6_INC_STATS_BH(__in6_dev_get(skb->dev), ICMP6_MIB_INERRORS);
return;
}
if (sk->sk_state == TCP_TIME_WAIT) {
inet_twsk_put((struct inet_timewait_sock *)sk);
return;
}
bh_lock_sock(sk);
if (sock_owned_by_user(sk))
NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);
if (sk->sk_state == TCP_CLOSE)
goto out;
tp = tcp_sk(sk);
seq = ntohl(th->seq);
if (sk->sk_state != TCP_LISTEN &&
!between(seq, tp->snd_una, tp->snd_nxt)) {
NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
np = inet6_sk(sk);
if (type == ICMPV6_PKT_TOOBIG) {
struct dst_entry *dst = NULL;
if (sock_owned_by_user(sk))
goto out;
if ((1 << sk->sk_state) & (TCPF_LISTEN | TCPF_CLOSE))
goto out;
/* icmp should have updated the destination cache entry */
dst = __sk_dst_check(sk, np->dst_cookie);
if (dst == NULL) {
struct inet_sock *inet = inet_sk(sk);
struct flowi fl;
/* BUGGG_FUTURE: Again, it is not clear how
to handle rthdr case. Ignore this complexity
for now.
*/
memset(&fl, 0, sizeof(fl));
fl.proto = IPPROTO_TCP;
ipv6_addr_copy(&fl.fl6_dst, &np->daddr);
ipv6_addr_copy(&fl.fl6_src, &np->saddr);
fl.oif = sk->sk_bound_dev_if;
fl.fl_ip_dport = inet->dport;
fl.fl_ip_sport = inet->sport;
if ((err = ip6_dst_lookup(sk, &dst, &fl))) {
sk->sk_err_soft = -err;
goto out;
}
if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0) {
sk->sk_err_soft = -err;
goto out;
}
} else
dst_hold(dst);
if (inet_csk(sk)->icsk_pmtu_cookie > dst_mtu(dst)) {
tcp_sync_mss(sk, dst_mtu(dst));
tcp_simple_retransmit(sk);
} /* else let the usual retransmit timer handle it */
dst_release(dst);
goto out;
}
icmpv6_err_convert(type, code, &err);
/* Might be for an request_sock */
switch (sk->sk_state) {
struct request_sock *req, **prev;
case TCP_LISTEN:
if (sock_owned_by_user(sk))
goto out;
req = inet6_csk_search_req(sk, &prev, th->dest, &hdr->daddr,
&hdr->saddr, inet6_iif(skb));
if (!req)
goto out;
/* ICMPs are not backlogged, hence we cannot get
* an established socket here.
*/
BUG_TRAP(req->sk == NULL);
if (seq != tcp_rsk(req)->snt_isn) {
NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
inet_csk_reqsk_queue_drop(sk, req, prev);
goto out;
case TCP_SYN_SENT:
case TCP_SYN_RECV: /* Cannot happen.
It can, it SYNs are crossed. --ANK */
if (!sock_owned_by_user(sk)) {
TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS);
sk->sk_err = err;
sk->sk_error_report(sk); /* Wake people up to see the error (see connect in sock.c) */
tcp_done(sk);
} else
sk->sk_err_soft = err;
goto out;
}
if (!sock_owned_by_user(sk) && np->recverr) {
sk->sk_err = err;
sk->sk_error_report(sk);
} else
sk->sk_err_soft = err;
out:
bh_unlock_sock(sk);
sock_put(sk);
}
static void bictcp_init(struct sock *sk)
{
bictcp_reset(inet_csk_ca(sk));
if (initial_ssthresh)
tcp_sk(sk)->snd_ssthresh = initial_ssthresh;
}
/* This will initiate an outgoing connection. */
int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{
struct inet_sock *inet = inet_sk(sk);
struct tcp_sock *tp = tcp_sk(sk);
struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
struct rtable *rt;
u32 daddr, nexthop;
int tmp;
int err;
if (addr_len < sizeof(struct sockaddr_in))
return -EINVAL;
if (usin->sin_family != AF_INET)
return -EAFNOSUPPORT;
nexthop = daddr = usin->sin_addr.s_addr;
if (inet->opt && inet->opt->srr) {
if (!daddr)
return -EINVAL;
nexthop = inet->opt->faddr;
}
tmp = ip_route_connect(&rt, nexthop, inet->saddr,
RT_CONN_FLAGS(sk), sk->sk_bound_dev_if,
IPPROTO_TCP,
inet->sport, usin->sin_port, sk);
if (tmp < 0)
return tmp;
if (rt->rt_flags & (RTCF_MULTICAST | RTCF_BROADCAST)) {
ip_rt_put(rt);
return -ENETUNREACH;
}
if (!inet->opt || !inet->opt->srr)
daddr = rt->rt_dst;
if (!inet->saddr)
inet->saddr = rt->rt_src;
inet->rcv_saddr = inet->saddr;
if (tp->rx_opt.ts_recent_stamp && inet->daddr != daddr) {
/* Reset inherited state */
tp->rx_opt.ts_recent = 0;
tp->rx_opt.ts_recent_stamp = 0;
tp->write_seq = 0;
}
if (tcp_death_row.sysctl_tw_recycle &&
!tp->rx_opt.ts_recent_stamp && rt->rt_dst == daddr) {
struct inet_peer *peer = rt_get_peer(rt);
/* VJ's idea. We save last timestamp seen from
* the destination in peer table, when entering state TIME-WAIT
* and initialize rx_opt.ts_recent from it, when trying new connection.
*/
if (peer && peer->tcp_ts_stamp + TCP_PAWS_MSL >= xtime.tv_sec) {
tp->rx_opt.ts_recent_stamp = peer->tcp_ts_stamp;
tp->rx_opt.ts_recent = peer->tcp_ts;
}
}
inet->dport = usin->sin_port;
inet->daddr = daddr;
inet_csk(sk)->icsk_ext_hdr_len = 0;
if (inet->opt)
inet_csk(sk)->icsk_ext_hdr_len = inet->opt->optlen;
tp->rx_opt.mss_clamp = 536;
/* Socket identity is still unknown (sport may be zero).
* However we set state to SYN-SENT and not releasing socket
* lock select source port, enter ourselves into the hash tables and
* complete initialization after this.
*/
tcp_set_state(sk, TCP_SYN_SENT);
err = inet_hash_connect(&tcp_death_row, sk);
if (err)
goto failure;
err = ip_route_newports(&rt, IPPROTO_TCP, inet->sport, inet->dport, sk);
if (err)
goto failure;
/* OK, now commit destination to socket. */
sk_setup_caps(sk, &rt->u.dst);
if (!tp->write_seq)
tp->write_seq = secure_tcp_sequence_number(inet->saddr,
inet->daddr,
inet->sport,
usin->sin_port);
inet->id = tp->write_seq ^ jiffies;
err = tcp_connect(sk);
rt = NULL;
if (err)
goto failure;
return 0;
failure:
/* This unhashes the socket and releases the local port, if necessary. */
tcp_set_state(sk, TCP_CLOSE);
ip_rt_put(rt);
sk->sk_route_caps = 0;
inet->dport = 0;
return err;
}
static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked)
{
struct tcp_sock *tp = tcp_sk(sk);
struct vegas *vegas = inet_csk_ca(sk);
if (!vegas->doing_vegas_now) {
tcp_reno_cong_avoid(sk, ack, acked);
return;
}
if (after(ack, vegas->beg_snd_nxt)) {
/* Do the Vegas once-per-RTT cwnd adjustment. */
/* Save the extent of the current window so we can use this
* at the end of the next RTT.
*/
vegas->beg_snd_nxt = tp->snd_nxt;
/* We do the Vegas calculations only if we got enough RTT
* samples that we can be reasonably sure that we got
* at least one RTT sample that wasn't from a delayed ACK.
* If we only had 2 samples total,
* then that means we're getting only 1 ACK per RTT, which
* means they're almost certainly delayed ACKs.
* If we have 3 samples, we should be OK.
*/
if (vegas->cntRTT <= 2) {
/* We don't have enough RTT samples to do the Vegas
* calculation, so we'll behave like Reno.
*/
tcp_reno_cong_avoid(sk, ack, acked);
} else {
u32 rtt, diff;
u64 target_cwnd;
/* We have enough RTT samples, so, using the Vegas
* algorithm, we determine if we should increase or
* decrease cwnd, and by how much.
*/
/* Pluck out the RTT we are using for the Vegas
* calculations. This is the min RTT seen during the
* last RTT. Taking the min filters out the effects
* of delayed ACKs, at the cost of noticing congestion
* a bit later.
*/
rtt = vegas->minRTT;
/* Calculate the cwnd we should have, if we weren't
* going too fast.
*
* This is:
* (actual rate in segments) * baseRTT
*/
target_cwnd = (u64)tp->snd_cwnd * vegas->baseRTT;
do_div(target_cwnd, rtt);
/* Calculate the difference between the window we had,
* and the window we would like to have. This quantity
* is the "Diff" from the Arizona Vegas papers.
*/
diff = tp->snd_cwnd * (rtt-vegas->baseRTT) / vegas->baseRTT;
if (diff > gamma && tp->snd_cwnd <= tp->snd_ssthresh) {
/* Going too fast. Time to slow down
* and switch to congestion avoidance.
*/
/* Set cwnd to match the actual rate
* exactly:
* cwnd = (actual rate) * baseRTT
* Then we add 1 because the integer
* truncation robs us of full link
* utilization.
*/
tp->snd_cwnd = min(tp->snd_cwnd, (u32)target_cwnd+1);
tp->snd_ssthresh = tcp_vegas_ssthresh(tp);
} else if (tp->snd_cwnd <= tp->snd_ssthresh) {
/* Slow start. */
tcp_slow_start(tp, acked);
} else {
/* Congestion avoidance. */
/* Figure out where we would like cwnd
* to be.
*/
if (diff > beta) {
/* The old window was too fast, so
* we slow down.
*/
tp->snd_cwnd--;
tp->snd_ssthresh
= tcp_vegas_ssthresh(tp);
} else if (diff < alpha) {
/* We don't have enough extra packets
* in the network, so speed up.
*/
tp->snd_cwnd++;
} else {
/* Sending just as fast as we
* should be.
*/
}
}
if (tp->snd_cwnd < 2)
tp->snd_cwnd = 2;
else if (tp->snd_cwnd > tp->snd_cwnd_clamp)
tp->snd_cwnd = tp->snd_cwnd_clamp;
tp->snd_ssthresh = tcp_current_ssthresh(sk);
}
/* Wipe the slate clean for the next RTT. */
vegas->cntRTT = 0;
vegas->minRTT = 0x7fffffff;
}
/* Use normal slow start */
else if (tp->snd_cwnd <= tp->snd_ssthresh)
tcp_slow_start(tp, acked);
}
static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked)
{
struct tcp_sock *tp = tcp_sk(sk);
struct yeah *yeah = inet_csk_ca(sk);
if (!tcp_is_cwnd_limited(sk))
return;
if (tcp_in_slow_start(tp))
tcp_slow_start(tp, acked);
else if (!yeah->doing_reno_now) {
/* Scalable */
tp->snd_cwnd_cnt += yeah->pkts_acked;
if (tp->snd_cwnd_cnt > min(tp->snd_cwnd, TCP_SCALABLE_AI_CNT)) {
if (tp->snd_cwnd < tp->snd_cwnd_clamp)
tp->snd_cwnd++;
tp->snd_cwnd_cnt = 0;
}
yeah->pkts_acked = 1;
} else {
/* Reno */
tcp_cong_avoid_ai(tp, tp->snd_cwnd, 1);
}
/* The key players are v_vegas.beg_snd_una and v_beg_snd_nxt.
*
* These are so named because they represent the approximate values
* of snd_una and snd_nxt at the beginning of the current RTT. More
* precisely, they represent the amount of data sent during the RTT.
* At the end of the RTT, when we receive an ACK for v_beg_snd_nxt,
* we will calculate that (v_beg_snd_nxt - v_vegas.beg_snd_una) outstanding
* bytes of data have been ACKed during the course of the RTT, giving
* an "actual" rate of:
*
* (v_beg_snd_nxt - v_vegas.beg_snd_una) / (rtt duration)
*
* Unfortunately, v_vegas.beg_snd_una is not exactly equal to snd_una,
* because delayed ACKs can cover more than one segment, so they
* don't line up yeahly with the boundaries of RTTs.
*
* Another unfortunate fact of life is that delayed ACKs delay the
* advance of the left edge of our send window, so that the number
* of bytes we send in an RTT is often less than our cwnd will allow.
* So we keep track of our cwnd separately, in v_beg_snd_cwnd.
*/
if (after(ack, yeah->vegas.beg_snd_nxt)) {
/* We do the Vegas calculations only if we got enough RTT
* samples that we can be reasonably sure that we got
* at least one RTT sample that wasn't from a delayed ACK.
* If we only had 2 samples total,
* then that means we're getting only 1 ACK per RTT, which
* means they're almost certainly delayed ACKs.
* If we have 3 samples, we should be OK.
*/
if (yeah->vegas.cntRTT > 2) {
u32 rtt, queue;
u64 bw;
/* We have enough RTT samples, so, using the Vegas
* algorithm, we determine if we should increase or
* decrease cwnd, and by how much.
*/
/* Pluck out the RTT we are using for the Vegas
* calculations. This is the min RTT seen during the
* last RTT. Taking the min filters out the effects
* of delayed ACKs, at the cost of noticing congestion
* a bit later.
*/
rtt = yeah->vegas.minRTT;
/* Compute excess number of packets above bandwidth
* Avoid doing full 64 bit divide.
*/
bw = tp->snd_cwnd;
bw *= rtt - yeah->vegas.baseRTT;
do_div(bw, rtt);
queue = bw;
if (queue > TCP_YEAH_ALPHA ||
rtt - yeah->vegas.baseRTT > (yeah->vegas.baseRTT / TCP_YEAH_PHY)) {
if (queue > TCP_YEAH_ALPHA &&
tp->snd_cwnd > yeah->reno_count) {
u32 reduction = min(queue / TCP_YEAH_GAMMA ,
tp->snd_cwnd >> TCP_YEAH_EPSILON);
tp->snd_cwnd -= reduction;
tp->snd_cwnd = max(tp->snd_cwnd,
yeah->reno_count);
tp->snd_ssthresh = tp->snd_cwnd;
}
if (yeah->reno_count <= 2)
yeah->reno_count = max(tp->snd_cwnd>>1, 2U);
else
yeah->reno_count++;
yeah->doing_reno_now = min(yeah->doing_reno_now + 1,
0xffffffU);
} else {
static u32 tcp_illinois_cwnd_undo(struct sock *sk)
{
const struct illinois *ca = inet_csk_ca(sk);
return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd);
}
static u32 tcp_scalable_ssthresh(struct sock *sk)
{
const struct tcp_sock *tp = tcp_sk(sk);
return max(tp->snd_cwnd - (tp->snd_cwnd>>TCP_SCALABLE_MD_SCALE), 2U);
}
static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
int addr_len)
{
struct sockaddr_in6 *usin = (struct sockaddr_in6 *) uaddr;
struct inet_sock *inet = inet_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
struct ipv6_pinfo *np = inet6_sk(sk);
struct tcp_sock *tp = tcp_sk(sk);
struct in6_addr *saddr = NULL, *final_p, final;
struct rt6_info *rt;
struct flowi6 fl6;
struct dst_entry *dst;
int addr_type;
int err;
if (addr_len < SIN6_LEN_RFC2133)
return -EINVAL;
if (usin->sin6_family != AF_INET6)
return -EAFNOSUPPORT;
memset(&fl6, 0, sizeof(fl6));
if (np->sndflow) {
fl6.flowlabel = usin->sin6_flowinfo&IPV6_FLOWINFO_MASK;
IP6_ECN_flow_init(fl6.flowlabel);
if (fl6.flowlabel&IPV6_FLOWLABEL_MASK) {
struct ip6_flowlabel *flowlabel;
flowlabel = fl6_sock_lookup(sk, fl6.flowlabel);
if (flowlabel == NULL)
return -EINVAL;
usin->sin6_addr = flowlabel->dst;
fl6_sock_release(flowlabel);
}
}
if(ipv6_addr_any(&usin->sin6_addr))
usin->sin6_addr.s6_addr[15] = 0x1;
addr_type = ipv6_addr_type(&usin->sin6_addr);
if(addr_type & IPV6_ADDR_MULTICAST)
return -ENETUNREACH;
if (addr_type&IPV6_ADDR_LINKLOCAL) {
if (addr_len >= sizeof(struct sockaddr_in6) &&
usin->sin6_scope_id) {
if (sk->sk_bound_dev_if &&
sk->sk_bound_dev_if != usin->sin6_scope_id)
return -EINVAL;
sk->sk_bound_dev_if = usin->sin6_scope_id;
}
if (!sk->sk_bound_dev_if)
return -EINVAL;
}
if (tp->rx_opt.ts_recent_stamp &&
!ipv6_addr_equal(&np->daddr, &usin->sin6_addr)) {
tp->rx_opt.ts_recent = 0;
tp->rx_opt.ts_recent_stamp = 0;
tp->write_seq = 0;
}
np->daddr = usin->sin6_addr;
np->flow_label = fl6.flowlabel;
if (addr_type == IPV6_ADDR_MAPPED) {
u32 exthdrlen = icsk->icsk_ext_hdr_len;
struct sockaddr_in sin;
SOCK_DEBUG(sk, "connect: ipv4 mapped\n");
if (__ipv6_only_sock(sk))
return -ENETUNREACH;
sin.sin_family = AF_INET;
sin.sin_port = usin->sin6_port;
sin.sin_addr.s_addr = usin->sin6_addr.s6_addr32[3];
icsk->icsk_af_ops = &ipv6_mapped;
sk->sk_backlog_rcv = tcp_v4_do_rcv;
#ifdef CONFIG_TCP_MD5SIG
tp->af_specific = &tcp_sock_ipv6_mapped_specific;
#endif
err = tcp_v4_connect(sk, (struct sockaddr *)&sin, sizeof(sin));
if (err) {
icsk->icsk_ext_hdr_len = exthdrlen;
icsk->icsk_af_ops = &ipv6_specific;
sk->sk_backlog_rcv = tcp_v6_do_rcv;
#ifdef CONFIG_TCP_MD5SIG
tp->af_specific = &tcp_sock_ipv6_specific;
#endif
goto failure;
} else {
ipv6_addr_set_v4mapped(inet->inet_saddr, &np->saddr);
ipv6_addr_set_v4mapped(inet->inet_rcv_saddr,
&np->rcv_saddr);
}
return err;
}
if (!ipv6_addr_any(&np->rcv_saddr))
saddr = &np->rcv_saddr;
fl6.flowi6_proto = IPPROTO_TCP;
fl6.daddr = np->daddr;
fl6.saddr = saddr ? *saddr : np->saddr;
fl6.flowi6_oif = sk->sk_bound_dev_if;
fl6.flowi6_mark = sk->sk_mark;
fl6.fl6_dport = usin->sin6_port;
fl6.fl6_sport = inet->inet_sport;
final_p = fl6_update_dst(&fl6, np->opt, &final);
security_sk_classify_flow(sk, flowi6_to_flowi(&fl6));
dst = ip6_dst_lookup_flow(sk, &fl6, final_p, true);
if (IS_ERR(dst)) {
err = PTR_ERR(dst);
goto failure;
}
if (saddr == NULL) {
saddr = &fl6.saddr;
np->rcv_saddr = *saddr;
}
np->saddr = *saddr;
inet->inet_rcv_saddr = LOOPBACK4_IPV6;
sk->sk_gso_type = SKB_GSO_TCPV6;
__ip6_dst_store(sk, dst, NULL, NULL);
rt = (struct rt6_info *) dst;
if (tcp_death_row.sysctl_tw_recycle &&
!tp->rx_opt.ts_recent_stamp &&
ipv6_addr_equal(&rt->rt6i_dst.addr, &np->daddr)) {
struct inet_peer *peer = rt6_get_peer(rt);
if (peer) {
inet_peer_refcheck(peer);
if ((u32)get_seconds() - peer->tcp_ts_stamp <= TCP_PAWS_MSL) {
tp->rx_opt.ts_recent_stamp = peer->tcp_ts_stamp;
tp->rx_opt.ts_recent = peer->tcp_ts;
}
}
}
icsk->icsk_ext_hdr_len = 0;
if (np->opt)
icsk->icsk_ext_hdr_len = (np->opt->opt_flen +
np->opt->opt_nflen);
tp->rx_opt.mss_clamp = IPV6_MIN_MTU - sizeof(struct tcphdr) - sizeof(struct ipv6hdr);
inet->inet_dport = usin->sin6_port;
tcp_set_state(sk, TCP_SYN_SENT);
err = inet6_hash_connect(&tcp_death_row, sk);
if (err)
goto late_failure;
if (!tp->write_seq)
tp->write_seq = secure_tcpv6_sequence_number(np->saddr.s6_addr32,
np->daddr.s6_addr32,
inet->inet_sport,
inet->inet_dport);
err = tcp_connect(sk);
if (err)
goto late_failure;
return 0;
late_failure:
tcp_set_state(sk, TCP_CLOSE);
__sk_dst_reset(sk);
failure:
inet->inet_dport = 0;
sk->sk_route_caps = 0;
return err;
}
/* Create a new IPv4 subflow.
*
* We are in user-context and meta-sock-lock is hold.
*/
int mptcp_init4_subsockets(struct sock *meta_sk, const struct mptcp_loc4 *loc,
struct mptcp_rem4 *rem)
{
struct tcp_sock *tp;
struct sock *sk;
struct sockaddr_in loc_in, rem_in;
struct socket_alloc sock_full;
struct socket *sock = (struct socket *)&sock_full;
int ret;
/** First, create and prepare the new socket */
memcpy(&sock_full, meta_sk->sk_socket, sizeof(sock_full));
sock->state = SS_UNCONNECTED;
sock->ops = NULL;
ret = inet_create(sock_net(meta_sk), sock, IPPROTO_TCP, 1);
if (unlikely(ret < 0)) {
net_err_ratelimited("%s inet_create failed ret: %d\n",
__func__, ret);
return ret;
}
sk = sock->sk;
tp = tcp_sk(sk);
/* All subsockets need the MPTCP-lock-class */
lockdep_set_class_and_name(&(sk)->sk_lock.slock, &meta_slock_key, meta_slock_key_name);
lockdep_init_map(&(sk)->sk_lock.dep_map, meta_key_name, &meta_key, 0);
ret = mptcp_add_sock(meta_sk, sk, loc->loc4_id, rem->rem4_id, GFP_KERNEL);
if (ret) {
net_err_ratelimited("%s mptcp_add_sock failed ret: %d\n",
__func__, ret);
goto error;
}
tp->mptcp->slave_sk = 1;
/* Initializing the timer for an MPTCP subflow */
timer_setup(&tp->mptcp->mptcp_ack_timer, mptcp_ack_handler, 0);
/** Then, connect the socket to the peer */
loc_in.sin_family = AF_INET;
rem_in.sin_family = AF_INET;
loc_in.sin_port = 0;
if (rem->port)
rem_in.sin_port = rem->port;
else
rem_in.sin_port = inet_sk(meta_sk)->inet_dport;
loc_in.sin_addr = loc->addr;
rem_in.sin_addr = rem->addr;
if (loc->if_idx)
sk->sk_bound_dev_if = loc->if_idx;
ret = kernel_bind(sock, (struct sockaddr *)&loc_in,
sizeof(struct sockaddr_in));
if (ret < 0) {
net_err_ratelimited("%s: token %#x bind() to %pI4 index %d failed, error %d\n",
__func__, tcp_sk(meta_sk)->mpcb->mptcp_loc_token,
&loc_in.sin_addr, loc->if_idx, ret);
goto error;
}
mptcp_debug("%s: token %#x pi %d src_addr:%pI4:%d dst_addr:%pI4:%d ifidx: %d\n",
__func__, tcp_sk(meta_sk)->mpcb->mptcp_loc_token,
tp->mptcp->path_index, &loc_in.sin_addr,
ntohs(loc_in.sin_port), &rem_in.sin_addr,
ntohs(rem_in.sin_port), loc->if_idx);
ret = kernel_connect(sock, (struct sockaddr *)&rem_in,
sizeof(struct sockaddr_in), O_NONBLOCK);
if (ret < 0 && ret != -EINPROGRESS) {
net_err_ratelimited("%s: MPTCP subsocket connect() failed, error %d\n",
__func__, ret);
goto error;
}
MPTCP_INC_STATS(sock_net(meta_sk), MPTCP_MIB_JOINSYNTX);
sk_set_socket(sk, meta_sk->sk_socket);
sk->sk_wq = meta_sk->sk_wq;
return 0;
error:
/* May happen if mptcp_add_sock fails first */
if (!mptcp(tp)) {
tcp_close(sk, 0);
} else {
local_bh_disable();
mptcp_sub_force_close(sk);
local_bh_enable();
}
return ret;
}
int generic_send_file (tux_req_t *req, struct socket *sock, int cachemiss)
{
sock_send_desc_t sock_desc;
int len, want, nonblock = !cachemiss;
struct tcp_opt *tp = tcp_sk(sock->sk);
tp->nonagle = 2;
sock_desc.sock = sock;
sock_desc.req = req;
repeat:
Dprintk("generic_send_file(%p,%d,%p) called, f_pos: %Ld, output_len: %Ld.\n", req, nonblock, sock, req->in_file->f_pos, req->output_len);
if (req->proto->check_req_err(req, cachemiss))
return -1;
if (connection_too_fast(req) == 2) {
len = -5;
goto out;
}
if (req->total_file_len < req->in_file->f_pos)
TUX_BUG();
req->desc.written = 0;
/*
* Careful, output_len can be 64-bit, while 'want' can be 32-bit.
*/
if (req->output_len > SEND_BLOCKSIZE)
want = SEND_BLOCKSIZE;
else
want = req->output_len;
req->desc.count = want;
req->desc.arg.buf = (char *) &sock_desc;
req->desc.error = 0;
Dprintk("sendfile(), desc.count: %d.\n", req->desc.count);
do_generic_file_read(req->in_file, &req->in_file->f_pos, &req->desc, sock_send_actor, nonblock);
if (req->desc.written > 0) {
req->bytes_sent += req->desc.written;
req->output_len -= req->desc.written;
}
if (!nonblock && (req->desc.error == -EWOULDBLOCKIO))
TUX_BUG();
Dprintk("sendfile() wrote: %d bytes.\n", req->desc.written);
if (req->output_len && !req->desc.written && !req->desc.error) {
#if CONFIG_TUX_DEBUG
req->bytes_expected = 0;
#endif
req->in_file->f_pos = 0;
req->error = TUX_ERROR_CONN_CLOSE;
zap_request(req, cachemiss);
return -1;
}
switch (req->desc.error) {
case -EWOULDBLOCKIO:
len = -3;
break;
case -EAGAIN:
no_write_space:
Dprintk("sk->wmem_queued: %d, sk->sndbuf: %d.\n",
sock->sk->sk_wmem_queued, sock->sk->sk_sndbuf);
len = -4;
break;
default:
len = req->desc.written;
#if CONFIG_TUX_DEBUG
if (req->desc.error)
TDprintk("TUX: sendfile() returned error %d (signals pending: %08lx)!\n", req->desc.error, current->pending.signal.sig[0]);
#endif
if (!req->desc.error) {
if (req->output_len < 0)
BUG();
if (req->output_len) {
if (test_bit(SOCK_NOSPACE, &sock->flags))
goto no_write_space;
goto repeat;
}
}
#if CONFIG_TUX_DEBUG
if (req->desc.written != want)
TDprintk("TUX: sendfile() wrote %d bytes, wanted %d! (pos %Ld) (signals pending: %08lx).\n", req->desc.written, want, req->in_file->f_pos, current->pending.signal.sig[0]);
else
Dprintk("TUX: sendfile() FINISHED for req %p, wrote %d bytes.\n", req, req->desc.written);
req->bytes_expected = 0;
#endif
break;
}
out:
Dprintk("sendfile() wrote %d bytes.\n", len);
return len;
}
