// SPDX-License-Identifier: GPL-2.0
/*
* INET An implementation of the TCP/IP protocol suite for the LINUX
* operating system. INET is implemented using the BSD Socket
* interface as the means of communication with the user level.
*
* Implementation of the Transmission Control Protocol(TCP).
*
* Authors: Ross Biro
* Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
* Mark Evans, <evansmp@uhura.aston.ac.uk>
* Corey Minyard <wf-rch!minyard@relay.EU.net>
* Florian La Roche, <flla@stud.uni-sb.de>
* Charles Hedrick, <hedrick@klinzhai.rutgers.edu>
* Linus Torvalds, <torvalds@cs.helsinki.fi>
* Alan Cox, <gw4pts@gw4pts.ampr.org>
* Matthew Dillon, <dillon@apollo.west.oic.com>
* Arnt Gulbrandsen, <agulbra@nvg.unit.no>
* Jorge Cwik, <jorge@laser.satlink.net>
*/
/*
* Changes:
* Pedro Roque : Fast Retransmit/Recovery.
* Two receive queues.
* Retransmit queue handled by TCP.
* Better retransmit timer handling.
* New congestion avoidance.
* Header prediction.
* Variable renaming.
*
* Eric : Fast Retransmit.
* Randy Scott : MSS option defines.
* Eric Schenk : Fixes to slow start algorithm.
* Eric Schenk : Yet another double ACK bug.
* Eric Schenk : Delayed ACK bug fixes.
* Eric Schenk : Floyd style fast retrans war avoidance.
* David S. Miller : Don't allow zero congestion window.
* Eric Schenk : Fix retransmitter so that it sends
* next packet on ack of previous packet.
* Andi Kleen : Moved open_request checking here
* and process RSTs for open_requests.
* Andi Kleen : Better prune_queue, and other fixes.
* Andrey Savochkin: Fix RTT measurements in the presence of
* timestamps.
* Andrey Savochkin: Check sequence numbers correctly when
* removing SACKs due to in sequence incoming
* data segments.
* Andi Kleen: Make sure we never ack data there is not
* enough room for. Also make this condition
* a fatal error if it might still happen.
* Andi Kleen: Add tcp_measure_rcv_mss to make
* connections with MSS<min(MTU,ann. MSS)
* work without delayed acks.
* Andi Kleen: Process packets with PSH set in the
* fast path.
* J Hadi Salim: ECN support
* Andrei Gurtov,
* Pasi Sarolahti,
* Panu Kuhlberg: Experimental audit of TCP (re)transmission
* engine. Lots of bugs are found.
* Pasi Sarolahti: F-RTO for dealing with spurious RTOs
*/
#define pr_fmt(fmt) "TCP: " fmt
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/module.h>
#include <linux/sysctl.h>
#include <linux/kernel.h>
#include <linux/prefetch.h>
#include <net/dst.h>
#include <net/tcp.h>
#include <net/inet_common.h>
#include <linux/ipsec.h>
#include <asm/unaligned.h>
#include <linux/errqueue.h>
#include <trace/events/tcp.h>
#include <linux/jump_label_ratelimit.h>
#include <net/busy_poll.h>
#include <net/mptcp.h>
int sysctl_tcp_max_orphans __read_mostly = NR_FILE;
#define FLAG_DATA 0x01 /* Incoming frame contained data. */
#define FLAG_WIN_UPDATE 0x02 /* Incoming ACK was a window update. */
#define FLAG_DATA_ACKED 0x04 /* This ACK acknowledged new data. */
#define FLAG_RETRANS_DATA_ACKED 0x08 /* "" "" some of which was retransmitted. */
#define FLAG_SYN_ACKED 0x10 /* This ACK acknowledged SYN. */
#define FLAG_DATA_SACKED 0x20 /* New SACK. */
#define FLAG_ECE 0x40 /* ECE in this ACK */
#define FLAG_LOST_RETRANS 0x80 /* This ACK marks some retransmission lost */
#define FLAG_SLOWPATH 0x100 /* Do not skip RFC checks for window update.*/
#define FLAG_ORIG_SACK_ACKED 0x200 /* Never retransmitted data are (s)acked */
#define FLAG_SND_UNA_ADVANCED 0x400 /* Snd_una was changed (!= FLAG_DATA_ACKED) */
#define FLAG_DSACKING_ACK 0x800 /* SACK blocks contained D-SACK info */
#define FLAG_SET_XMIT_TIMER 0x1000 /* Set TLP or RTO timer */
#define FLAG_SACK_RENEGING 0x2000 /* snd_una advanced to a sacked seq */
#define FLAG_UPDATE_TS_RECENT 0x4000 /* tcp_replace_ts_recent() */
#define FLAG_NO_CHALLENGE_ACK 0x8000 /* do not call tcp_send_challenge_ack() */
#define FLAG_ACK_MAYBE_DELAYED 0x10000 /* Likely a delayed ACK */
#define FLAG_DSACK_TLP 0x20000 /* DSACK for tail loss probe */
#define FLAG_ACKED (FLAG_DATA_ACKED|FLAG_SYN_ACKED)
#define FLAG_NOT_DUP (FLAG_DATA|FLAG_WIN_UPDATE|FLAG_ACKED)
#define FLAG_CA_ALERT (FLAG_DATA_SACKED|FLAG_ECE|FLAG_DSACKING_ACK)
#define FLAG_FORWARD_PROGRESS (FLAG_ACKED|FLAG_DATA_SACKED)
#define TCP_REMNANT (TCP_FLAG_FIN|TCP_FLAG_URG|TCP_FLAG_SYN|TCP_FLAG_PSH)
#define TCP_HP_BITS (~(TCP_RESERVED_BITS|TCP_FLAG_PSH))
#define REXMIT_NONE 0 /* no loss recovery to do */
#define REXMIT_LOST 1 /* retransmit packets marked lost */
#define REXMIT_NEW 2 /* FRTO-style transmit of unsent/new packets */
#if IS_ENABLED(CONFIG_TLS_DEVICE)
static DEFINE_STATIC_KEY_DEFERRED_FALSE(clean_acked_data_enabled, HZ);
void clean_acked_data_enable(struct inet_connection_sock *icsk,
void (*cad)(struct sock *sk, u32 ack_seq))
{
icsk->icsk_clean_acked = cad;
static_branch_deferred_inc(&clean_acked_data_enabled);
}
EXPORT_SYMBOL_GPL(clean_acked_data_enable);
void clean_acked_data_disable(struct inet_connection_sock *icsk)
{
static_branch_slow_dec_deferred(&clean_acked_data_enabled);
icsk->icsk_clean_acked = NULL;
}
EXPORT_SYMBOL_GPL(clean_acked_data_disable);
void clean_acked_data_flush(void)
{
static_key_deferred_flush(&clean_acked_data_enabled);
}
EXPORT_SYMBOL_GPL(clean_acked_data_flush);
#endif
#ifdef CONFIG_CGROUP_BPF
static void bpf_skops_parse_hdr(struct sock *sk, struct sk_buff *skb)
{
bool unknown_opt = tcp_sk(sk)->rx_opt.saw_unknown &&
BPF_SOCK_OPS_TEST_FLAG(tcp_sk(sk),
BPF_SOCK_OPS_PARSE_UNKNOWN_HDR_OPT_CB_FLAG);
bool parse_all_opt = BPF_SOCK_OPS_TEST_FLAG(tcp_sk(sk),
BPF_SOCK_OPS_PARSE_ALL_HDR_OPT_CB_FLAG);
struct bpf_sock_ops_kern sock_ops;
if (likely(!unknown_opt && !parse_all_opt))
return;
/* The skb will be handled in the
* bpf_skops_established() or
* bpf_skops_write_hdr_opt().
*/
switch (sk->sk_state) {
case TCP_SYN_RECV:
case TCP_SYN_SENT:
case TCP_LISTEN:
return;
}
sock_owned_by_me(sk);
memset(&sock_ops, 0, offsetof(struct bpf_sock_ops_kern, temp));
sock_ops.op = BPF_SOCK_OPS_PARSE_HDR_OPT_CB;
sock_ops.is_fullsock = 1;
sock_ops.sk = sk;
bpf_skops_init_skb(&sock_ops, skb, tcp_hdrlen(skb));
BPF_CGROUP_RUN_PROG_SOCK_OPS(&sock_ops);
}
static void bpf_skops_established(struct sock *sk, int bpf_op,
struct sk_buff *skb)
{
struct bpf_sock_ops_kern sock_ops;
sock_owned_by_me(sk);
memset(&sock_ops, 0, offsetof(struct bpf_sock_ops_kern, temp));
sock_ops.op = bpf_op;
sock_ops.is_fullsock = 1;
sock_ops.sk = sk;
/* sk with TCP_REPAIR_ON does not have skb in tcp_finish_connect */
if (skb)
bpf_skops_init_skb(&sock_ops, skb, tcp_hdrlen(skb));
BPF_CGROUP_RUN_PROG_SOCK_OPS(&sock_ops);
}
#else
static void bpf_skops_parse_hdr(struct sock *sk, struct sk_buff *skb)
{
}
static void bpf_skops_established(struct sock *sk, int bpf_op,
struct sk_buff *skb)
{
}
#endif
static void tcp_gro_dev_warn(struct sock *sk, const struct sk_buff *skb,
unsigned int len)
{
static bool __once __read_mostly;
if (!__once) {
struct net_device *dev;
__once = true;
rcu_read_lock();
dev = dev_get_by_index_rcu(sock_net(sk), skb->skb_iif);
if (!dev || len >= dev->mtu)
pr_warn("%s: Driver has suspect GRO implementation, TCP performance may be compromised.\n",
dev ? dev->name : "Unknown driver");
rcu_read_unlock();
}
}
/* Adapt the MSS value used to make delayed ack decision to the
* real world.
*/
static void tcp_measure_rcv_mss(struct sock *sk, const struct sk_buff *skb)
{
struct inet_connection_sock *icsk = inet_csk(sk);
const unsigned int lss = icsk->icsk_ack.last_seg_size;
unsigned int len;
icsk->icsk_ack.last_seg_size = 0;
/* skb->len may jitter because of SACKs, even if peer
* sends good full-sized frames.
*/
len = skb_shinfo(skb)->gso_size ? : skb->len;
if (len >= icsk->icsk_ack.rcv_mss) {
icsk->icsk_ack.rcv_mss = min_t(unsigned int, len,
tcp_sk(sk)->advmss);
/* Account for possibly-removed options */
if (unlikely(len > icsk->icsk_ack.rcv_mss +
MAX_TCP_OPTION_SPACE))
tcp_gro_dev_warn(sk, skb, len);
} else {
/* Otherwise, we make more careful check taking into account,
* that SACKs block is variable.
*
* "len" is invariant segment length, including TCP header.
*/
len += skb->data - skb_transport_header(skb);
if (len >= TCP_MSS_DEFAULT + sizeof(struct tcphdr) ||
/* If PSH is not set, packet should be
* full sized, provided peer TCP is not badly broken.
* This observation (if it is correct 8)) allows
* to handle super-low mtu links fairly.
*/
(len >= TCP_MIN_MSS + sizeof(struct tcphdr) &&
!(tcp_flag_word(tcp_hdr(skb)) & TCP_REMNANT))) {
/* Subtract also invariant (if peer is RFC compliant),
* tcp header plus fixed timestamp option length.
* Resulting "len" is MSS free of SACK jitter.
*/
len -= tcp_sk(sk)->tcp_header_len;
icsk->icsk_ack.last_seg_size = len;
if (len == lss) {
icsk->icsk_ack.rcv_mss = len;
return;
}
}
if (icsk->icsk_ack.pending & ICSK_ACK_PUSHED)
icsk->icsk_ack.pending |= ICSK_ACK_PUSHED2;
icsk->icsk_ack.pending |= ICSK_ACK_PUSHED;
}
}
static void tcp_incr_quickack(struct sock *sk, unsigned int max_quickacks)
{
struct inet_connection_sock *icsk = inet_csk(sk);
unsigned int quickacks = tcp_sk(sk)->rcv_wnd / (2 * icsk->icsk_ack.rcv_mss);
if (quickacks == 0)
quickacks = 2;
quickacks = min(quickacks, max_quickacks);
if (quickacks > icsk->icsk_ack.quick)
icsk->icsk_ack.quick = quickacks;
}
void tcp_enter_quickack_mode(struct sock *sk, unsigned int max_quickacks)
{
struct inet_connection_sock *icsk = inet_csk(sk);
tcp_incr_quickack(sk, max_quickacks);
inet_csk_exit_pingpong_mode(sk);
icsk->icsk_ack.ato = TCP_ATO_MIN;
}
EXPORT_SYMBOL(tcp_enter_quickack_mode);
/* Send ACKs quickly, if "quick" count is not exhausted
* and the session is not interactive.
*/
static bool tcp_in_quickack_mode(struct sock *sk)
{
const struct inet_connection_sock *icsk = inet_csk(sk);
const struct dst_entry *dst = __sk_dst_get(sk);
return (dst && dst_metric(dst, RTAX_QUICKACK)) ||
(icsk->icsk_ack.quick && !inet_csk_in_pingpong_mode(sk));
}
static void tcp_ecn_queue_cwr(struct tcp_sock *tp)
{
if (tp->ecn_flags & TCP_ECN_OK)
tp->ecn_flags |= TCP_ECN_QUEUE_CWR;
}
static void tcp_ecn_accept_cwr(struct sock *sk, const struct sk_buff *skb)
{
if (tcp_hdr(skb)->cwr) {
tcp_sk(sk)->ecn_flags &= ~TCP_ECN_DEMAND_CWR;
/* If the sender is telling us it has entered CWR, then its
* cwnd may be very low (even just 1 packet), so we should ACK
* immediately.
*/
if (TCP_SKB_CB(skb)->seq != TCP_SKB_CB(skb)->end_seq)
inet_csk(sk)->icsk_ack.pending |= ICSK_ACK_NOW;
}
}
static void tcp_ecn_withdraw_cwr(struct tcp_sock *tp)
{
tp->ecn_flags &= ~TCP_ECN_QUEUE_CWR;
}
static void __tcp_ecn_check_ce(struct sock *sk, const struct sk_buff *skb)
{
struct tcp_sock *tp = tcp_sk(sk);
switch (TCP_SKB_CB(skb)->ip_dsfield & INET_ECN_MASK) {
case INET_ECN_NOT_ECT:
/* Funny extension: if ECT is not set on a segment,
* and we already seen ECT on a previous segment,
* it is probably a retransmit.
*/
if (tp->ecn_flags & TCP_ECN_SEEN)
tcp_enter_quickack_mode(sk, 2);
break;
case INET_ECN_CE:
if (tcp_ca_needs_ecn(sk))
tcp_ca_event(sk, CA_EVENT_ECN_IS_CE);
if (!(tp->ecn_flags & TCP_ECN_DEMAND_CWR)) {
/* Better not delay acks, sender can have a very low cwnd */
tcp_enter_quickack_mode(sk, 2);
tp->ecn_flags |= TCP_ECN_DEMAND_CWR;
}
tp->ecn_flags |= TCP_ECN_SEEN;
break;
default:
if (tcp_ca_needs_ecn(sk))
tcp_ca_event(sk, CA_EVENT_ECN_NO_CE);
tp->ecn_flags |= TCP_ECN_SEEN;
break;
}
}
static void tcp_ecn_check_ce(struct sock *sk, const struct sk_buff *skb)
{
if (tcp_sk(sk)->ecn_flags & TCP_ECN_OK)
__tcp_ecn_check_ce(sk, skb);
}
static void tcp_ecn_rcv_synack(struct tcp_sock *tp, const struct tcphdr *th)
{
if ((tp->ecn_flags & TCP_ECN_OK) && (!th->ece || th->cwr))
tp->ecn_flags &= ~TCP_ECN_OK;
}
static void tcp_ecn_rcv_syn(struct tcp_sock *tp, const struct tcphdr *th)
{
if ((tp->ecn_flags & TCP_ECN_OK) && (!th->ece || !th->cwr))
tp->ecn_flags &= ~TCP_ECN_OK;
}
static bool tcp_ecn_rcv_ecn_echo(const struct tcp_sock *tp, const struct tcphdr *th)
{
if (th->ece && !th->syn && (tp->ecn_flags & TCP_ECN_OK))
return true;
return false;
}
/* Buffer size and advertised window tuning.
*
* 1. Tuning sk->sk_sndbuf, when connection enters established state.
*/
static void tcp_sndbuf_expand(struct sock *sk)
{
const struct tcp_sock *tp = tcp_sk(sk);
const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops;
int sndmem, per_mss;
u32 nr_segs;
/* Worst case is non GSO/TSO : each frame consumes one skb
* and skb->head is kmalloced using power of two area of memory
*/
per_mss = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache) +
MAX_TCP_HEADER +
SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
per_mss = roundup_pow_of_two(per_mss) +
SKB_DATA_ALIGN(sizeof(struct sk_buff));
nr_segs = max_t(u32, TCP_INIT_CWND, tcp_snd_cwnd(tp));
nr_segs = max_t(u32, nr_segs, tp->reordering + 1);
/* Fast Recovery (RFC 5681 3.2) :
* Cubic needs 1.7 factor, rounded to 2 to include
* extra cushion (application might react slowly to EPOLLOUT)
*/
sndmem = ca_ops->sndbuf_expand ? ca_ops->sndbuf_expand(sk) : 2;
sndmem *= nr_segs * per_mss;
if (sk->sk_sndbuf < sndmem)
WRITE_ONCE(sk->sk_sndbuf,
min(sndmem, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_wmem[2])));
}
/* 2. Tuning advertised window (window_clamp, rcv_ssthresh)
*
* All tcp_full_space() is split to two parts: "network" buffer, allocated
* forward and advertised in receiver window (tp->rcv_wnd) and
* "application buffer", required to isolate scheduling/application
* latencies from network.
* window_clamp is maximal advertised window. It can be less than
* tcp_full_space(), in this case tcp_full_space() - window_clamp
* is reserved for "application" buffer. The less window_clamp is
* the smoother our behaviour from viewpoint of network, but the lower
* throughput and the higher sensitivity of the connection to losses. 8)
*
* rcv_ssthresh is more strict window_clamp used at "slow start"
* phase to predict further behaviour of this connection.
* It is used for two goals:
* - to enforce header prediction at sender, even when application
* requires some significant "application buffer". It is check #1.
* - to prevent pruning of receive queue because of misprediction
* of receiver window. Check #2.
*
* The scheme does not work when sender sends good segments opening
* window and then starts to feed us spaghetti. But it should work
* in common situations. Otherwise, we have to rely on queue collapsing.
*/
/* Slow part of check#2. */
static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb,
unsigned int skbtruesize)
{
struct tcp_sock *tp = tcp_sk(sk);
/* Optimize this! */
int truesize = tcp_win_from_space(sk, skbtruesize) >> 1;
int window = tcp_win_from_space(sk, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])) >> 1;
while (tp->rcv_ssthresh <= window) {
if (truesize <= skb->len)
return 2 * inet_csk(sk)->icsk_ack.rcv_mss;
truesize >>= 1;
window >>= 1;
}
return 0;
}
/* Even if skb appears to have a bad len/truesize ratio, TCP coalescing
* can play nice with us, as sk_buff and skb->head might be either
* freed or shared with up to MAX_SKB_FRAGS segments.
* Only give a boost to drivers using page frag(s) to hold the frame(s),
* and if no payload was pulled in skb->head before reaching us.
*/
static u32 truesize_adjust(bool adjust, const struct sk_buff *skb)
{
u32 truesize = skb->truesize;
if (adjust && !skb_headlen(skb)) {
truesize -= SKB_TRUESIZE(skb_end_offset(skb));
/* paranoid check, some drivers might be buggy */
if (unlikely((int)truesize < (int)skb->len))
truesize = skb->truesize;
}
return truesize;
}
static void tcp_grow_window(struct sock *sk, const struct sk_buff *skb,
bool adjust)
{
struct tcp_sock *tp = tcp_sk(sk);
int room;
room = min_t(int, tp->window_clamp, tcp_space(sk)) - tp->rcv_ssthresh;
if (room <= 0)
return;
/* Check #1 */
if (!tcp_under_memory_pressure(sk)) {
unsigned int truesize = truesize_adjust(adjust, skb);
int incr;
/* Check #2. Increase window, if skb with such overhead
* will fit to rcvbuf in future.
*/
if (tcp_win_from_space(sk, truesize) <= skb->len)
incr = 2 * tp->advmss;
else
incr = __tcp_grow_window(sk, skb, truesize);
if (incr) {
incr = max_t(int, incr, 2 * skb->len);
tp->rcv_ssthresh += min(room, incr);
inet_csk(sk)->icsk_ack.quick |= 1;
}
} else {
/* Under pressure:
* Adjust rcv_ssthresh according to reserved mem
*/
tcp_adjust_rcv_ssthresh(sk);
}
}
/* 3. Try to fixup all. It is made immediately after connection enters
* established state.
*/
static void tcp_init_buffer_space(struct sock *sk)
{
int tcp_app_win = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_app_win);
struct tcp_sock *tp = tcp_sk(sk);
int maxwin;
if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK))
tcp_sndbuf_expand(sk);
tcp_mstamp_refresh(tp);
tp->rcvq_space.time = tp->tcp_mstamp;
tp->rcvq_space.seq = tp->copied_seq;
maxwin = tcp_full_space(sk);
if (tp->window_clamp >= maxwin) {
tp->window_clamp = maxwin;
if (tcp_app_win && maxwin > 4 * tp->advmss)
tp->window_clamp = max(maxwin -
(maxwin >> tcp_app_win),
4 * tp->advmss);
}
/* Force reservation of one segment. */
if (tcp_app_win &&
tp->window_clamp > 2 * tp->advmss &&
tp->window_clamp + tp->advmss > maxwin)
tp->window_clamp = max(2 * tp->advmss, maxwin - tp->advmss);
tp->rcv_ssthresh = min(tp->rcv_ssthresh, tp->window_clamp);
tp->snd_cwnd_stamp = tcp_jiffies32;
tp->rcvq_space.space = min3(tp->rcv_ssthresh, tp->rcv_wnd,
(u32)TCP_INIT_CWND * tp->advmss);
}
/* 4. Recalculate window clamp after socket hit its memory bounds. */
static void tcp_clamp_window(struct sock *sk)
{
struct tcp_sock *tp = tcp_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
struct net *net = sock_net(sk);
int rmem2;
icsk->icsk_ack.quick = 0;
rmem2 = READ_ONCE(net->ipv4.sysctl_tcp_rmem[2]);
if (sk->sk_rcvbuf < rmem2 &&
!(sk->sk_userlocks & SOCK_RCVBUF_LOCK) &&
!tcp_under_memory_pressure(sk) &&
sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)) {
WRITE_ONCE(sk->sk_rcvbuf,
min(atomic_read(&sk->sk_rmem_alloc), rmem2));
}
if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf)
tp->rcv_ssthresh = min(tp->window_clamp, 2U * tp->advmss);
}
/* Initialize RCV_MSS value.
* RCV_MSS is an our guess about MSS used by the peer.
* We haven't any direct information about the MSS.
* It's better to underestimate the RCV_MSS rather than overestimate.
* Overestimations make us ACKing less frequently than needed.
* Underestimations are more easy to detect and fix by tcp_measure_rcv_mss().
*/
void tcp_initialize_rcv_mss(struct sock *sk)
{
const struct tcp_sock *tp = tcp_sk(sk);
unsigned int hint = min_t(unsigned int, tp->advmss, tp->mss_cache);
hint = min(hint, tp->rcv_wnd / 2);
hint = min(hint, TCP_MSS_DEFAULT);
hint = max(hint, TCP_MIN_MSS);
inet_csk(sk)->icsk_ack.rcv_mss = hint;
}
EXPORT_SYMBOL(tcp_initialize_rcv_mss);
/* Receiver "autotuning" code.
*
* The algorithm for RTT estimation w/o timestamps is based on
* Dynamic Right-Sizing (DRS) by Wu Feng and Mike Fisk of LANL.
* <https://public.lanl.gov/radiant/pubs.html#DRS>
*
* More detail on this code can be found at
* <http://staff.psc.edu/jheffner/>,
* though this reference is out of date. A new paper
* is pending.
*/
static void tcp_rcv_rtt_update(struct tcp_sock *tp, u32 sample, int win_dep)
{
u32 new_sample = tp->rcv_rtt_est.rtt_us;
long m = sample;
if (new_sample != 0) {
/* If we sample in larger samples in the non-timestamp
* case, we could grossly overestimate the RTT especially
* with chatty applications or bulk transfer apps which
* are stalled on filesystem I/O.
*
* Also, since we are only going for a minimum in the
* non-timestamp case, we do not smooth things out
* else with timestamps disabled convergence takes too
* long.
*/
if (!win_dep) {
m -= (new_sample >> 3);
new_sample += m;
} else {
m <<= 3;
if (m < new_sample)
new_sample = m;
}
} else {
/* No previous measure. */
new_sample = m << 3;
}
tp->rcv_rtt_est.rtt_us = new_sample;
}
static inline void tcp_rcv_rtt_measure(struct tcp_sock *tp)
{
u32 delta_us;
if (tp->rcv_rtt_est.time == 0)
goto new_measure;
if (before(tp->rcv_nxt, tp->rcv_rtt_est.seq))
return;
delta_us = tcp_stamp_us_delta(tp->tcp_mstamp, tp->rcv_rtt_est.time);
if (!delta_us)
delta_us = 1;
tcp_rcv_rtt_update(tp, delta_us, 1);
new_measure:
tp->rcv_rtt_est.seq = tp->rcv_nxt + tp->rcv_wnd;
tp->rcv_rtt_est.time = tp->tcp_mstamp;
}
static inline void tcp_rcv_rtt_measure_ts(struct sock *sk,
const struct sk_buff *skb)
{
struct tcp_sock *tp = tcp_sk(sk);
if (tp->rx_opt.rcv_tsecr == tp->rcv_rtt_last_tsecr)
return;
tp->rcv_rtt_last_tsecr = tp->rx_opt.rcv_tsecr;
if (TCP_SKB_CB(skb)->end_seq -
TCP_SKB_CB(skb)->seq >= inet_csk(sk)->icsk_ack.rcv_mss) {
u32 delta = tcp_time_stamp(tp) - tp->rx_opt.rcv_tsecr;
u32 delta_us;
if (likely(delta < INT_MAX / (USEC_PER_SEC / TCP_TS_HZ))) {
if (!delta)
delta = 1;
delta_us = delta * (USEC_PER_SEC / TCP_TS_HZ);
tcp_rcv_rtt_update(tp, delta_us, 0);
}
}
}
/*
* This function should be called every time data is copied to user space.
* It calculates the appropriate TCP receive buffer space.
*/
void tcp_rcv_space_adjust(struct sock *sk)
{
struct tcp_sock *tp = tcp_sk(sk);
u32 copied;
int time;
trace_tcp_rcv_space_adjust(sk);
tcp_mstamp_refresh(tp);
time = tcp_stamp_us_delta(tp->tcp_mstamp, tp->rcvq_space.time);
if (time < (tp->rcv_rtt_est.rtt_us >> 3) || tp->rcv_rtt_est.rtt_us == 0)
return;
/* Number of bytes copied to user in last RTT */
copied = tp->copied_seq - tp->rcvq_space.seq;
if (copied <= tp->rcvq_space.space)
goto new_measure;
/* A bit of theory :
* copied = bytes received in previous RTT, our base window
* To cope with packet losses, we need a 2x factor
* To cope with slow start, and sender growing its cwin by 100 %
* every RTT, we need a 4x factor, because the ACK we are sending
* now is for the next RTT, not the current one :
* <prev RTT . ><current RTT .. ><next RTT .... >
*/
if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_moderate_rcvbuf) &&
!(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) {
int rcvmem, rcvbuf;
u64 rcvwin, grow;
/* minimal window to cope with packet losses, assuming
* steady state. Add some cushion because of small variations.
*/
rcvwin = ((u64)copied << 1) + 16 * tp->advmss;
/* Accommodate for sender rate increase (eg. slow start) */
grow = rcvwin * (copied - tp->rcvq_space.space);
do_div(grow, tp->rcvq_space.space);
rcvwin += (grow << 1);
rcvmem = SKB_TRUESIZE(tp->advmss + MAX_TCP_HEADER);
while (tcp_win_from_space(sk, rcvmem) < tp->advmss)
rcvmem += 128;
do_div(rcvwin, tp->advmss);
rcvbuf = min_t(u64, rcvwin * rcvmem,
READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2]));
if (rcvbuf > sk->sk_rcvbuf) {
WRITE_ONCE(sk->sk_rcvbuf, rcvbuf);
/* Make the window clamp follow along. */
tp->window_clamp = tcp_win_from_space(sk, rcvbuf);
}
}
tp->rcvq_space.space = copied;
new_measure:
tp->rcvq_space.seq = tp->copied_seq;
tp->rcvq_space.time = tp->tcp_mstamp;
}
/* There is something which you must keep in mind when you analyze the
* behavior of the tp->ato delayed ack timeout interval. When a
* connection starts up, we want to ack as quickly as possible. The
* problem is that "good" TCP's do slow start at the beginning of data
* transmission. The means that until we send the first few ACK's the
* sender will sit on his end and only queue most of his data, because
* he can only send snd_cwnd unacked packets at any given time. For
* each ACK we send, he increments snd_cwnd and transmits more of his
* queue. -DaveM
*/
static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb)
{
struct tcp_sock *tp = tcp_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
u32 now;
inet_csk_schedule_ack(sk);
tcp_measure_rcv_mss(sk, skb);
tcp_rcv_rtt_measure(tp);
now = tcp_jiffies32;
if (!icsk->icsk_ack.ato) {
/* The _first_ data packet received, initialize
* delayed ACK engine.
*/
tcp_incr_quickack(sk, TCP_MAX_QUICKACKS);
icsk->icsk_ack.ato = TCP_ATO_MIN;
} else {
int m = now - icsk->icsk_ack.lrcvtime;
if (m <= TCP_ATO_MIN / 2) {
/* The fastest case is the first. */
icsk->icsk_ack.ato = (icsk->icsk_ack.ato >> 1) + TCP_ATO_MIN / 2;
} else if (m < icsk->icsk_ack.ato) {
icsk->icsk_ack.ato = (icsk->icsk_ack.ato >> 1) + m;
if (icsk->icsk_ack.ato > icsk->icsk_rto)
icsk->icsk_ack.ato = icsk->icsk_rto;
} else if (m > icsk->icsk_rto) {
/* Too long gap. Apparently sender failed to
* restart window, so that we send ACKs quickly.
*/
tcp_incr_quickack(sk, TCP_MAX_QUICKACKS);
}
}
icsk->icsk_ack.lrcvtime = now;
tcp_ecn_check_ce(sk, skb);
if (skb->len >= 128)
tcp_grow_window(sk, skb, true);
}
/* Called to compute a smoothed rtt estimate. The data fed to this
* routine either comes from timestamps, or from segments that were
* known _not_ to have been retransmitted [see Karn/Partridge
* Proceedings SIGCOMM 87]. The algorithm is from the SIGCOMM 88
* piece by Van Jacobson.
* NOTE: the next three routines used to be one big routine.
* To save cycles in the RFC 1323 implementation it was better to break
* it up into three procedures. -- erics
*/
static void tcp_rtt_estimator(struct sock *sk, long mrtt_us)
{
struct tcp_sock *tp = tcp_sk(sk);
long m = mrtt_us; /* RTT */
u32 srtt = tp->srtt_us;
/* The following amusing code comes from Jacobson's
* article in SIGCOMM '88. Note that rtt and mdev
* are scaled versions of rtt and mean deviation.
* This is designed to be as fast as possible
* m stands for "measurement".
*
* On a 1990 paper the rto value is changed to:
* RTO = rtt + 4 * mdev
*
* Funny. This algorithm seems to be very broken.
* These formulae increase RTO, when it should be decreased, increase
* too slowly, when it should be increased quickly, decrease too quickly
* etc. I guess in BSD RTO takes ONE value, so that it is absolutely
* does not matter how to _calculate_ it. Seems, it was trap
* that VJ failed to avoid. 8)
*/
if (srtt != 0) {
m -= (srtt >> 3); /* m is now error in rtt est */
srtt += m; /* rtt = 7/8 rtt + 1/8 new */
if (m < 0) {
m = -m; /* m is now abs(error) */
m -= (tp->mdev_us >> 2); /* similar update on mdev */
/* This is similar to one of Eifel findings.
* Eifel blocks mdev updates when rtt decreases.
* This solution is a bit different: we use finer gain
* for mdev in this case (alpha*beta).
* Like Eifel it also prevents growth of rto,
* but also it limits too fast rto decreases,
* happening in pure Eifel.
*/
if (m > 0)
m >>= 3;
} else {
m -= (tp->mdev_us >> 2); /* similar update on mdev */
}
tp->mdev_us += m; /* mdev = 3/4 mdev + 1/4 new */
if (tp->mdev_us > tp->mdev_max_us) {
tp->mdev_max_us = tp->mdev_us;
if (tp->mdev_max_us > tp->rttvar_us)
tp->rttvar_us = tp->mdev_max_us;
}
if (after(tp->snd_una, tp->rtt_seq)) {
if (tp->mdev_max_us < tp->rttvar_us)
tp->rttvar_us -= (tp->rttvar_us - tp->mdev_max_us) >> 2;
tp->rtt_seq = tp->snd_nxt;
tp->mdev_max_us = tcp_rto_min_us(sk);
tcp_bpf_rtt(sk);
}
} else {
/* no previous measure. */
srtt = m << 3; /* take the measured time to be rtt */
tp->mdev_us = m << 1; /* make sure rto = 3*rtt */
tp->rttvar_us = max(tp->mdev_us, tcp_rto_min_us(sk));
tp->mdev_max_us = tp->rttvar_us;
tp->rtt_seq = tp->snd_nxt;
tcp_bpf_rtt(sk);
}
tp->srtt_us = max(1U, srtt);
}
static void tcp_update_pacing_rate(struct sock *sk)
{
const struct tcp_sock *tp = tcp_sk(sk);
u64 rate;
/* set sk_pacing_rate to 200 % of current rate (mss * cwnd / srtt) */
rate = (u64)tp->mss_cache * ((USEC_PER_SEC / 100) << 3);
/* current rate is (cwnd * mss) / srtt
* In Slow Start [1], set sk_pacing_rate to 200 % the current rate.
* In Congestion Avoidance phase, set it to 120 % the current rate.
*
* [1] : Normal Slow Start condition is (tp->snd_cwnd < tp->snd_ssthresh)
* If snd_cwnd >= (tp->snd_ssthresh / 2), we are approaching
* end of slow start and should slow down.
*/
if (tcp_snd_cwnd(tp) < tp->snd_ssthresh / 2)
rate *= READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_pacing_ss_ratio);
else
rate *= READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_pacing_ca_ratio);
rate *= max(tcp_snd_cwnd(tp), tp->packets_out);
if (likely(tp->srtt_us))
do_div(rate, tp->srtt_us);
/* WRITE_ONCE() is needed because sch_fq fetches sk_pacing_rate
* without any lock. We want to make sure compiler wont store
* intermediate values in this location.
*/
WRITE_ONCE(sk->sk_pacing_rate, min_t(u64, rate,
sk->sk_max_pacing_rate));
}
/* Calculate rto without backoff. This is the second half of Van Jacobson's
* routine referred to above.
*/
static void tcp_set_rto(struct sock *sk)
{
const struct tcp_sock *tp = tcp_sk(sk);
/* Old crap is replaced with new one. 8)
*
* More seriously:
* 1. If rtt variance happened to be less 50msec, it is hallucination.
* It cannot be less due to utterly erratic ACK generation made
* at least by solaris and freebsd. "Erratic ACKs" has _nothing_
* to do with delayed acks, because at cwnd>2 true delack timeout
* is invisible. Actually, Linux-2.4 also generates erratic
* ACKs in some circumstances.
*/
inet_csk(sk)->icsk_rto = __tcp_set_rto(tp);
/* 2. Fixups made earlier cannot be right.
* If we do not estimate RTO correctly without them,
* all the algo is pure shit and should be replaced
* with correct one. It is exactly, which we pretend to do.
*/
/* NOTE: clamping at TCP_RTO_MIN is not required, current algo
* guarantees that rto is higher.
*/
tcp_bound_rto(sk);
}
__u32 tcp_init_cwnd(const struct tcp_sock *tp, const struct dst_entry *dst)
{
__u32 cwnd = (dst ? dst_metric(dst, RTAX_INITCWND) : 0);
if (!cwnd)
cwnd = TCP_INIT_CWND;
return min_t(__u32, cwnd, tp->snd_cwnd_clamp);
}
struct tcp_sacktag_state {
/* Timestamps for earliest and latest never-retransmitted segment
* that was SACKed. RTO needs the earliest RTT to stay conservative,
* but congestion control should still get an accurate delay signal.
*/
u64 first_sackt;
u64 last_sackt;
u32 reord;
u32 sack_delivered;
int flag;
unsigned int mss_now;
struct rate_sample *rate;
};
/* Take a notice that peer is sending D-SACKs. Skip update of data delivery
* and spurious retransmission information if this DSACK is unlikely caused by
* sender's action:
* - DSACKed sequence range is larger than maximum receiver's window.
* - Total no. of DSACKed segments exceed the total no. of retransmitted segs.
*/
static u32 tcp_dsack_seen(struct tcp_sock *tp, u32 start_seq,
u32 end_seq, struct tcp_sacktag_state *state)
{
u32 seq_len, dup_segs = 1;
if (!before(start_seq, end_seq))
return 0;
seq_len = end_seq - start_seq;
/* Dubious DSACK: DSACKed range greater than maximum advertised rwnd */
if (seq_len > tp->max_window)
return 0;
if (seq_len > tp->mss_cache)
dup_segs = DIV_ROUND_UP(seq_len, tp->mss_cache);
else if (tp->tlp_high_seq && tp->tlp_high_seq == end_seq)
state->flag |= FLAG_DSACK_TLP;
tp->dsack_dups += dup_segs;
/* Skip the DSACK if dup segs weren't retransmitted by sender */
if (tp->dsack_dups > tp->total_retrans)
return 0;
tp->rx_opt.sack_ok |= TCP_DSACK_SEEN;
/* We increase the RACK ordering window in rounds where we receive
* DSACKs that may have been due to reordering causing RACK to trigger
* a spurious fast recovery. Thus RACK ignores DSACKs that happen
* without having seen reordering, or that match TLP probes (TLP
* is timer-driven, not triggered by RACK).
*/
if (tp->reord_seen && !(state->flag & FLAG_DSACK_TLP))
tp->rack.dsack_seen = 1;
state->flag |= FLAG_DSACKING_ACK;
/* A spurious retransmission is delivered */
state->sack_delivered += dup_segs;
return dup_segs;
}
/* It's reordering when higher sequence was delivered (i.e. sacked) before
* some lower never-retransmitted sequence ("low_seq"). The maximum reordering
* distance is approximated in full-mss packet distance ("reordering").
*/
static void tcp_check_sack_reordering(struct sock *sk, const u32 low_seq,
const int ts)
{
struct tcp_sock *tp = tcp_sk(sk);
const u32 mss = tp->mss_cache;
u32 fack, metric;
fack = tcp_highest_sack_seq(tp);
if (!before(low_seq, fack))
return;
metric = fack - low_seq;
if ((metric > tp->reordering * mss) && mss) {
#if FASTRETRANS_DEBUG > 1
pr_debug("Disorder%d %d %u f%u s%u rr%d\n",
tp->rx_opt.sack_ok, inet_csk(sk)->icsk_ca_state,
tp->reordering,
0,
tp->sacked_out,
tp->undo_marker ? tp->undo_retrans : 0);
#endif
tp->reordering = min_t(u32, (metric + mss - 1) / mss,
READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_max_reordering));
}
/* This exciting event is worth to be remembered. 8) */
tp->reord_seen++;
NET_INC_STATS(sock_net(sk),
ts ? LINUX_MIB_TCPTSREORDER : LINUX_MIB_TCPSACKREORDER);
}
/* This must be called before lost_out or retrans_out are updated
* on a new loss, because we want to know if all skbs previously
* known to be lost have already been retransmitted, indicating
* that this newly lost skb is our next skb to retransmit.
*/
static void tcp_verify_retransmit_hint(struct tcp_sock *tp, struct sk_buff *skb)
{
if ((!tp->retransmit_skb_hint && tp->retrans_out >= tp->lost_out) ||
(tp->retransmit_skb_hint &&
before(TCP_SKB_CB(skb)->seq,
TCP_SKB_CB(tp->retransmit_skb_hint)->seq)))
tp->retransmit_skb_hint = skb;
}
/* Sum the number of packets on the wire we have marked as lost, and
* notify the congestion control module that the given skb was marked lost.
*/
static void tcp_notify_skb_loss_event(struct tcp_sock *tp, const struct sk_buff *skb)
{
tp->lost += tcp_skb_pcount(skb);
}
void tcp_mark_skb_lost(struct sock *sk, struct sk_buff *skb)
{
__u8 sacked = TCP_SKB_CB(skb)->sacked;
struct tcp_sock *tp = tcp_sk(sk);
if (sacked & TCPCB_SACKED_ACKED)
return;
tcp_verify_retransmit_hint(tp, skb);
if (sacked & TCPCB_LOST) {
if (sacked & TCPCB_SACKED_RETRANS) {
/* Account for retransmits that are lost again */
TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
tp->retrans_out -= tcp_skb_pcount(skb);
NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPLOSTRETRANSMIT,
tcp_skb_pcount(skb));
tcp_notify_skb_loss_event(tp, skb);
}
} else {
tp->lost_out += tcp_skb_pcount(skb);
TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
tcp_notify_skb_loss_event(tp, skb);
}
}
/* Updates the delivered and delivered_ce counts */
static void tcp_count_delivered(struct tcp_sock *tp, u32 delivered,
bool ece_ack)
{
tp->delivered += delivered;
if (ece_ack)
tp->delivered_ce += delivered;
}
/* This procedure tags the retransmission queue when SACKs arrive.
*
* We have three tag bits: SACKED(S), RETRANS(R) and LOST(L).
* Packets in queue with these bits set are counted in variables
* sacked_out, retrans_out and lost_out, correspondingly.
*
* Valid combinations are:
* Tag InFlight Description
* 0 1 - orig segment is in flight.
* S 0 - nothing flies, orig reached receiver.
* L 0 - nothing flies, orig lost by net.
* R 2 - both orig and retransmit are in flight.
* L|R 1 - orig is lost, retransmit is in flight.
* S|R 1 - orig reached receiver, retrans is still in flight.
* (L|S|R is logically valid, it could occur when L|R is sacked,
* but it is equivalent to plain S and code short-curcuits it to S.
* L|S is logically invalid, it would mean -1 packet in flight 8))
*
* These 6 states form finite state machine, controlled by the following events:
* 1. New ACK (+SACK) arrives. (tcp_sacktag_write_queue())
* 2. Retransmission. (tcp_retransmit_skb(), tcp_xmit_retransmit_queue())
* 3. Loss detection event of two flavors:
* A. Scoreboard estimator decided the packet is lost.
* A'. Reno "three dupacks" marks head of queue lost.
* B. SACK arrives sacking SND.NXT at the moment, when the
* segment was retransmitted.
* 4. D-SACK added new rule: D-SACK changes any tag to S.
*
* It is pleasant to note, that state diagram turns out to be commutative,
* so that we are allowed not to be bothered by order of our actions,
* when multiple events arrive simultaneously. (see the function below).
*
* Reordering detection.
* --------------------
* Reordering metric is maximal distance, which a packet can be displaced
* in packet stream. With SACKs we can estimate it:
*
* 1. SACK fills old hole and the corresponding segment was not
* ever retransmitted -> reordering. Alas, we cannot use it
* when segment was retransmitted.
* 2. The last flaw is solved with D-SACK. D-SACK arrives
* for retransmitted and already SACKed segment -> reordering..
* Both of these heuristics are not used in Loss state, when we cannot
* account for retransmits accurately.
*
* SACK block validation.
* ----------------------
*
* SACK block range validation checks that the received SACK block fits to
* the expected sequence limits, i.e., it is between SND.UNA and SND.NXT.
* Note that SND.UNA is not included to the range though being valid because
* it means that the receiver is rather inconsistent with itself reporting
* SACK reneging when it should advance SND.UNA. Such SACK block this is
* perfectly valid, however, in light of RFC2018 which explicitly states
* that "SACK block MUST reflect the newest segment. Even if the newest
* segment is going to be discarded ...", not that it looks very clever
* in case of head skb. Due to potentional receiver driven attacks, we
* choose to avoid immediate execution of a walk in write queue due to
* reneging and defer head skb's loss recovery to standard loss recovery
* procedure that will eventually trigger (nothing forbids us doing this).
*
* Implements also blockage to start_seq wrap-around. Problem lies in the
* fact that though start_seq (s) is before end_seq (i.e., not reversed),
* there's no guarantee that it will be before snd_nxt (n). The problem
* happens when start_seq resides between end_seq wrap (e_w) and snd_nxt
* wrap (s_w):
*
* <- outs wnd -> <- wrapzone ->
* u e n u_w e_w s n_w
* | | | | | | |
* |<------------+------+----- TCP seqno space --------------+---------->|
* ...-- <2^31 ->| |<--------...
* ...---- >2^31 ------>| |<--------...
*
* Current code wouldn't be vulnerable but it's better still to discard such
* crazy SACK blocks. Doing this check for start_seq alone closes somewhat
* similar case (end_seq after snd_nxt wrap) as earlier reversed check in
* snd_nxt wrap -> snd_una region will then become "well d