[PATCH] [RFC] tcp: add ability to set a timestamp offset (v2)

[PATCH] [RFC] tcp: add ability to set a timestamp offset (v2)

[Andrey Vagin]

If a TCP socket will get live-migrated from one box to another the
timestamps (which are typically ON) will get screwed up -- the new
kernel will generate TS values that has nothing to do with what they
were on dump. The solution is to yet again fix the kernel and put a
"timestamp offset" on a socket.

v2: We can't adjust tcp_time_stamp only for output packets, because
echoed timestamp values will also no longer have a meaningful
relationship to tcp_time_stamp. That violates assumptions made in
several places in the code, which assumes that we can compare echoed
timestamp values to tcp_time_stamp; //Neal Cardwell

This version is a "bit more" radical. It makes tcp_time_stamp dependent
on tcp_sock. The idea of this patch is in the next two lines:

-#define tcp_time_stamp         ((__u32)(jiffies))
+#define tcp_time_stamp(tp)     ((__u32)(jiffies) + tp->tsoffset)

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: Dave Jones <davej@redhat.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Andrey Vagin <avagin@openvz.org>

Andrey Vagin (2):
  tcp: make tcp_timestamp dependent on tcp_sock
  tcp: add ability to set a timestamp offset (v2)

 include/linux/tcp.h      |  2 ++
 include/net/tcp.h        | 15 +++++++++-----
 include/uapi/linux/tcp.h |  1 +
 net/ipv4/syncookies.c    |  4 ++--
 net/ipv4/tcp.c           |  9 +++++++-
 net/ipv4/tcp_bic.c       | 10 ++++-----
 net/ipv4/tcp_cubic.c     | 14 ++++++-------
 net/ipv4/tcp_htcp.c      |  2 +-
 net/ipv4/tcp_input.c     | 53 +++++++++++++++++++++++++-----------------------
 net/ipv4/tcp_ipv4.c      | 19 +++++++++--------
 net/ipv4/tcp_lp.c        | 10 ++++-----
 net/ipv4/tcp_metrics.c   |  2 +-
 net/ipv4/tcp_output.c    | 35 +++++++++++++++++---------------
 net/ipv4/tcp_timer.c     |  9 ++++----
 net/ipv4/tcp_westwood.c  |  8 +++++---
 net/ipv6/tcp_ipv6.c      | 22 ++++++++++++--------
 16 files changed, 123 insertions(+), 92 deletions(-)

--
1.7.11.7

[PATCH 1/2] tcp: make tcp_timestamp dependent on tcp_sock

[Andrey Vagin]

This will be used for setting timestamp offset.

This patch converts the macros tcp_timestamp to a function with one
argument "struct tcp_sock *tp" and nothing else.

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: Dave Jones <davej@redhat.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Andrey Vagin <avagin@openvz.org>
---
 include/net/tcp.h       | 15 +++++++++-----
 net/ipv4/syncookies.c   |  4 ++--
 net/ipv4/tcp.c          |  2 +-
 net/ipv4/tcp_bic.c      | 10 +++++-----
 net/ipv4/tcp_cubic.c    | 14 ++++++-------
 net/ipv4/tcp_htcp.c     |  2 +-
 net/ipv4/tcp_input.c    | 53 ++++++++++++++++++++++++++-----------------------
 net/ipv4/tcp_ipv4.c     | 19 ++++++++++--------
 net/ipv4/tcp_lp.c       |  8 ++++----
 net/ipv4/tcp_metrics.c  |  2 +-
 net/ipv4/tcp_output.c   | 35 +++++++++++++++++---------------
 net/ipv4/tcp_timer.c    |  9 +++++----
 net/ipv4/tcp_westwood.c |  8 +++++---
 net/ipv6/tcp_ipv6.c     | 22 +++++++++++---------
 14 files changed, 112 insertions(+), 91 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index aed42c7..3e242ba 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -503,7 +503,7 @@ static inline __u32 cookie_v4_init_sequence(struct sock *sk,
 }
 #endif
 
-extern __u32 cookie_init_timestamp(struct request_sock *req);
+extern __u32 cookie_init_timestamp(struct tcp_sock *tp, struct request_sock *req);
 extern bool cookie_check_timestamp(struct tcp_options_received *opt, bool *);
 
 /* From net/ipv6/syncookies.c */
@@ -675,7 +675,10 @@ void tcp_send_window_probe(struct sock *sk);
  * to use only the low 32-bits of jiffies and hide the ugly
  * casts with the following macro.
  */
-#define tcp_time_stamp		((__u32)(jiffies))
+static inline u32 tcp_time_stamp(const struct tcp_sock *tp)
+{
+	return (__u32)jiffies;
+}
 
 #define tcp_flag_byte(th) (((u_int8_t *)th)[13])
 
@@ -1142,9 +1145,11 @@ static inline void tcp_openreq_init(struct request_sock *req,
 static inline void tcp_synack_rtt_meas(struct sock *sk,
 				       struct request_sock *req)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
+
 	if (tcp_rsk(req)->snt_synack)
 		tcp_valid_rtt_meas(sk,
-		    tcp_time_stamp - tcp_rsk(req)->snt_synack);
+		    tcp_time_stamp(tp) - tcp_rsk(req)->snt_synack);
 }
 
 extern void tcp_enter_memory_pressure(struct sock *sk);
@@ -1168,8 +1173,8 @@ static inline u32 keepalive_time_elapsed(const struct tcp_sock *tp)
 {
 	const struct inet_connection_sock *icsk = &tp->inet_conn;
 
-	return min_t(u32, tcp_time_stamp - icsk->icsk_ack.lrcvtime,
-			  tcp_time_stamp - tp->rcv_tstamp);
+	return min_t(u32, tcp_time_stamp(tp) - icsk->icsk_ack.lrcvtime,
+			  tcp_time_stamp(tp) - tp->rcv_tstamp);
 }
 
 static inline int tcp_fin_time(const struct sock *sk)
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index b236ef0..70e14fd 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -64,10 +64,10 @@ static u32 cookie_hash(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport,
  * Since subsequent timestamps use the normal tcp_time_stamp value, we
  * must make sure that the resulting initial timestamp is <= tcp_time_stamp.
  */
-__u32 cookie_init_timestamp(struct request_sock *req)
+__u32 cookie_init_timestamp(struct tcp_sock *tp, struct request_sock *req)
 {
 	struct inet_request_sock *ireq;
-	u32 ts, ts_now = tcp_time_stamp;
+	u32 ts, ts_now = tcp_time_stamp(tp);
 	u32 options = 0;
 
 	ireq = inet_rsk(req);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 1ca2536..086ceda 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2742,7 +2742,7 @@ void tcp_get_info(const struct sock *sk, struct tcp_info *info)
 {
 	const struct tcp_sock *tp = tcp_sk(sk);
 	const struct inet_connection_sock *icsk = inet_csk(sk);
-	u32 now = tcp_time_stamp;
+	u32 now = tcp_time_stamp(tp);
 
 	memset(info, 0, sizeof(*info));
 
diff --git a/net/ipv4/tcp_bic.c b/net/ipv4/tcp_bic.c
index f45e1c2..31faf6b 100644
--- a/net/ipv4/tcp_bic.c
+++ b/net/ipv4/tcp_bic.c
@@ -83,17 +83,17 @@ static void bictcp_init(struct sock *sk)
 /*
  * Compute congestion window to use.
  */
-static inline void bictcp_update(struct bictcp *ca, u32 cwnd)
+static inline void bictcp_update(struct tcp_sock *tp, struct bictcp *ca, u32 cwnd)
 {
 	if (ca->last_cwnd == cwnd &&
-	    (s32)(tcp_time_stamp - ca->last_time) <= HZ / 32)
+	    (s32)(tcp_time_stamp(tp) - ca->last_time) <= HZ / 32)
 		return;
 
 	ca->last_cwnd = cwnd;
-	ca->last_time = tcp_time_stamp;
+	ca->last_time = tcp_time_stamp(tp);
 
 	if (ca->epoch_start == 0) /* record the beginning of an epoch */
-		ca->epoch_start = tcp_time_stamp;
+		ca->epoch_start = tcp_time_stamp(tp);
 
 	/* start off normal */
 	if (cwnd <= low_window) {
@@ -151,7 +151,7 @@ static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 	if (tp->snd_cwnd <= tp->snd_ssthresh)
 		tcp_slow_start(tp);
 	else {
-		bictcp_update(ca, tp->snd_cwnd);
+		bictcp_update(tp, ca, tp->snd_cwnd);
 		tcp_cong_avoid_ai(tp, ca->cnt);
 	}
 
diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c
index a9077f4..31cfa5d 100644
--- a/net/ipv4/tcp_cubic.c
+++ b/net/ipv4/tcp_cubic.c
@@ -204,7 +204,7 @@ static u32 cubic_root(u64 a)
 /*
  * Compute congestion window to use.
  */
-static inline void bictcp_update(struct bictcp *ca, u32 cwnd)
+static inline void bictcp_update(struct tcp_sock *tp, struct bictcp *ca, u32 cwnd)
 {
 	u64 offs;
 	u32 delta, t, bic_target, max_cnt;
@@ -212,14 +212,14 @@ static inline void bictcp_update(struct bictcp *ca, u32 cwnd)
 	ca->ack_cnt++;	/* count the number of ACKs */
 
 	if (ca->last_cwnd == cwnd &&
-	    (s32)(tcp_time_stamp - ca->last_time) <= HZ / 32)
+	    (s32)(tcp_time_stamp(tp) - ca->last_time) <= HZ / 32)
 		return;
 
 	ca->last_cwnd = cwnd;
-	ca->last_time = tcp_time_stamp;
+	ca->last_time = tcp_time_stamp(tp);
 
 	if (ca->epoch_start == 0) {
-		ca->epoch_start = tcp_time_stamp;	/* record the beginning of an epoch */
+		ca->epoch_start = tcp_time_stamp(tp);	/* record the beginning of an epoch */
 		ca->ack_cnt = 1;			/* start counting */
 		ca->tcp_cwnd = cwnd;			/* syn with cubic */
 
@@ -251,7 +251,7 @@ static inline void bictcp_update(struct bictcp *ca, u32 cwnd)
 	 */
 
 	/* change the unit from HZ to bictcp_HZ */
-	t = ((tcp_time_stamp + msecs_to_jiffies(ca->delay_min>>3)
+	t = ((tcp_time_stamp(tp) + msecs_to_jiffies(ca->delay_min>>3)
 	      - ca->epoch_start) << BICTCP_HZ) / HZ;
 
 	if (t < ca->bic_K)		/* t - K */
@@ -315,7 +315,7 @@ static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 			bictcp_hystart_reset(sk);
 		tcp_slow_start(tp);
 	} else {
-		bictcp_update(ca, tp->snd_cwnd);
+		bictcp_update(tp, ca, tp->snd_cwnd);
 		tcp_cong_avoid_ai(tp, ca->cnt);
 	}
 
@@ -414,7 +414,7 @@ static void bictcp_acked(struct sock *sk, u32 cnt, s32 rtt_us)
 		return;
 
 	/* Discard delay samples right after fast recovery */
-	if ((s32)(tcp_time_stamp - ca->epoch_start) < HZ)
+	if ((s32)(tcp_time_stamp(tp) - ca->epoch_start) < HZ)
 		return;
 
 	delay = (rtt_us << 3) / USEC_PER_MSEC;
diff --git a/net/ipv4/tcp_htcp.c b/net/ipv4/tcp_htcp.c
index c1a8175..5137e1b 100644
--- a/net/ipv4/tcp_htcp.c
+++ b/net/ipv4/tcp_htcp.c
@@ -103,7 +103,7 @@ static void measure_achieved_throughput(struct sock *sk, u32 pkts_acked, s32 rtt
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 	const struct tcp_sock *tp = tcp_sk(sk);
 	struct htcp *ca = inet_csk_ca(sk);
-	u32 now = tcp_time_stamp;
+	u32 now = tcp_time_stamp(tp);
 
 	if (icsk->icsk_ca_state == TCP_CA_Open)
 		ca->pkts_acked = pkts_acked;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a28e4db..1bb2a7d 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -408,7 +408,7 @@ void tcp_init_buffer_space(struct sock *sk)
 		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_time_stamp;
+	tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 }
 
 /* 5. Recalculate window clamp after socket hit its memory bounds. */
@@ -503,11 +503,11 @@ static inline void tcp_rcv_rtt_measure(struct tcp_sock *tp)
 		goto new_measure;
 	if (before(tp->rcv_nxt, tp->rcv_rtt_est.seq))
 		return;
-	tcp_rcv_rtt_update(tp, tcp_time_stamp - tp->rcv_rtt_est.time, 1);
+	tcp_rcv_rtt_update(tp, tcp_time_stamp(tp) - tp->rcv_rtt_est.time, 1);
 
 new_measure:
 	tp->rcv_rtt_est.seq = tp->rcv_nxt + tp->rcv_wnd;
-	tp->rcv_rtt_est.time = tcp_time_stamp;
+	tp->rcv_rtt_est.time = tcp_time_stamp(tp);
 }
 
 static inline void tcp_rcv_rtt_measure_ts(struct sock *sk,
@@ -517,7 +517,7 @@ static inline void tcp_rcv_rtt_measure_ts(struct sock *sk,
 	if (tp->rx_opt.rcv_tsecr &&
 	    (TCP_SKB_CB(skb)->end_seq -
 	     TCP_SKB_CB(skb)->seq >= inet_csk(sk)->icsk_ack.rcv_mss))
-		tcp_rcv_rtt_update(tp, tcp_time_stamp - tp->rx_opt.rcv_tsecr, 0);
+		tcp_rcv_rtt_update(tp, tcp_time_stamp(tp) - tp->rx_opt.rcv_tsecr, 0);
 }
 
 /*
@@ -533,7 +533,7 @@ void tcp_rcv_space_adjust(struct sock *sk)
 	if (tp->rcvq_space.time == 0)
 		goto new_measure;
 
-	time = tcp_time_stamp - tp->rcvq_space.time;
+	time = tcp_time_stamp(tp) - tp->rcvq_space.time;
 	if (time < (tp->rcv_rtt_est.rtt >> 3) || tp->rcv_rtt_est.rtt == 0)
 		return;
 
@@ -573,7 +573,7 @@ void tcp_rcv_space_adjust(struct sock *sk)
 
 new_measure:
 	tp->rcvq_space.seq = tp->copied_seq;
-	tp->rcvq_space.time = tcp_time_stamp;
+	tp->rcvq_space.time = tcp_time_stamp(tp);
 }
 
 /* There is something which you must keep in mind when you analyze the
@@ -598,7 +598,7 @@ static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb)
 
 	tcp_rcv_rtt_measure(tp);
 
-	now = tcp_time_stamp;
+	now = tcp_time_stamp(tp);
 
 	if (!icsk->icsk_ack.ato) {
 		/* The _first_ data packet received, initialize
@@ -2007,7 +2007,7 @@ static void tcp_enter_frto_loss(struct sock *sk, int allowed_segments, int flag)
 
 	tp->snd_cwnd = tcp_packets_in_flight(tp) + allowed_segments;
 	tp->snd_cwnd_cnt = 0;
-	tp->snd_cwnd_stamp = tcp_time_stamp;
+	tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 	tp->frto_counter = 0;
 	tp->bytes_acked = 0;
 
@@ -2056,7 +2056,7 @@ void tcp_enter_loss(struct sock *sk, int how)
 	}
 	tp->snd_cwnd	   = 1;
 	tp->snd_cwnd_cnt   = 0;
-	tp->snd_cwnd_stamp = tcp_time_stamp;
+	tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 
 	tp->bytes_acked = 0;
 	tcp_clear_retrans_partial(tp);
@@ -2170,7 +2170,9 @@ static bool tcp_pause_early_retransmit(struct sock *sk, int flag)
 static inline int tcp_skb_timedout(const struct sock *sk,
 				   const struct sk_buff *skb)
 {
-	return tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto;
+	const struct tcp_sock *tp = tcp_sk(sk);
+
+	return tcp_time_stamp(tp) - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto;
 }
 
 static inline int tcp_head_timedout(const struct sock *sk)
@@ -2467,7 +2469,7 @@ static inline void tcp_moderate_cwnd(struct tcp_sock *tp)
 {
 	tp->snd_cwnd = min(tp->snd_cwnd,
 			   tcp_packets_in_flight(tp) + tcp_max_burst(tp));
-	tp->snd_cwnd_stamp = tcp_time_stamp;
+	tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 }
 
 /* Nothing was retransmitted or returned timestamp is less
@@ -2531,7 +2533,7 @@ static void tcp_undo_cwr(struct sock *sk, const bool undo_ssthresh)
 	} else {
 		tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh);
 	}
-	tp->snd_cwnd_stamp = tcp_time_stamp;
+	tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 }
 
 static inline bool tcp_may_undo(const struct tcp_sock *tp)
@@ -2726,7 +2728,7 @@ static inline void tcp_end_cwnd_reduction(struct sock *sk)
 	if (inet_csk(sk)->icsk_ca_state == TCP_CA_CWR ||
 	    (tp->undo_marker && tp->snd_ssthresh < TCP_INFINITE_SSTHRESH)) {
 		tp->snd_cwnd = tp->snd_ssthresh;
-		tp->snd_cwnd_stamp = tcp_time_stamp;
+		tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 	}
 	tcp_ca_event(sk, CA_EVENT_COMPLETE_CWR);
 }
@@ -2799,7 +2801,7 @@ static void tcp_mtup_probe_success(struct sock *sk)
 		       tcp_mss_to_mtu(sk, tp->mss_cache) /
 		       icsk->icsk_mtup.probe_size;
 	tp->snd_cwnd_cnt = 0;
-	tp->snd_cwnd_stamp = tcp_time_stamp;
+	tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 	tp->snd_ssthresh = tcp_current_ssthresh(sk);
 
 	icsk->icsk_mtup.search_low = icsk->icsk_mtup.probe_size;
@@ -3045,7 +3047,7 @@ static void tcp_ack_saw_tstamp(struct sock *sk, int flag)
 	 */
 	struct tcp_sock *tp = tcp_sk(sk);
 
-	tcp_valid_rtt_meas(sk, tcp_time_stamp - tp->rx_opt.rcv_tsecr);
+	tcp_valid_rtt_meas(sk, tcp_time_stamp(tp) - tp->rx_opt.rcv_tsecr);
 }
 
 static void tcp_ack_no_tstamp(struct sock *sk, u32 seq_rtt, int flag)
@@ -3078,9 +3080,10 @@ static inline void tcp_ack_update_rtt(struct sock *sk, const int flag,
 
 static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 	icsk->icsk_ca_ops->cong_avoid(sk, ack, in_flight);
-	tcp_sk(sk)->snd_cwnd_stamp = tcp_time_stamp;
+	tcp_sk(sk)->snd_cwnd_stamp = tcp_time_stamp(tp);
 }
 
 /* Restart timer after forward progress on connection.
@@ -3104,7 +3107,7 @@ void tcp_rearm_rto(struct sock *sk)
 		if (tp->early_retrans_delayed) {
 			struct sk_buff *skb = tcp_write_queue_head(sk);
 			const u32 rto_time_stamp = TCP_SKB_CB(skb)->when + rto;
-			s32 delta = (s32)(rto_time_stamp - tcp_time_stamp);
+			s32 delta = (s32)(rto_time_stamp - tcp_time_stamp(tp));
 			/* delta may not be positive if the socket is locked
 			 * when the delayed ER timer fires and is rescheduled.
 			 */
@@ -3166,7 +3169,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,
 	struct tcp_sock *tp = tcp_sk(sk);
 	const struct inet_connection_sock *icsk = inet_csk(sk);
 	struct sk_buff *skb;
-	u32 now = tcp_time_stamp;
+	u32 now = tcp_time_stamp(tp);
 	int fully_acked = true;
 	int flag = 0;
 	u32 pkts_acked = 0;
@@ -3656,7 +3659,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
 	 */
 	sk->sk_err_soft = 0;
 	icsk->icsk_probes_out = 0;
-	tp->rcv_tstamp = tcp_time_stamp;
+	tp->rcv_tstamp = tcp_time_stamp(tp);
 	prior_packets = tp->packets_out;
 	if (!prior_packets)
 		goto no_queue;
@@ -4971,7 +4974,7 @@ void tcp_cwnd_application_limited(struct sock *sk)
 		}
 		tp->snd_cwnd_used = 0;
 	}
-	tp->snd_cwnd_stamp = tcp_time_stamp;
+	tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 }
 
 static bool tcp_should_expand_sndbuf(const struct sock *sk)
@@ -5019,7 +5022,7 @@ static void tcp_new_space(struct sock *sk)
 		sndmem *= 2 * demanded;
 		if (sndmem > sk->sk_sndbuf)
 			sk->sk_sndbuf = min(sndmem, sysctl_tcp_wmem[2]);
-		tp->snd_cwnd_stamp = tcp_time_stamp;
+		tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 	}
 
 	sk->sk_write_space(sk);
@@ -5605,7 +5608,7 @@ void tcp_finish_connect(struct sock *sk, struct sk_buff *skb)
 	/* Prevent spurious tcp_cwnd_restart() on first data
 	 * packet.
 	 */
-	tp->lsndtime = tcp_time_stamp;
+	tp->lsndtime = tcp_time_stamp(tp);
 
 	tcp_init_buffer_space(sk);
 
@@ -5694,7 +5697,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
 
 		if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr &&
 		    !between(tp->rx_opt.rcv_tsecr, tp->retrans_stamp,
-			     tcp_time_stamp)) {
+			     tcp_time_stamp(tp))) {
 			NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSACTIVEREJECTED);
 			goto reset_and_undo;
 		}
@@ -5815,7 +5818,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
 			 * to stand against the temptation 8)     --ANK
 			 */
 			inet_csk_schedule_ack(sk);
-			icsk->icsk_ack.lrcvtime = tcp_time_stamp;
+			icsk->icsk_ack.lrcvtime = tcp_time_stamp(tp);
 			tcp_enter_quickack_mode(sk);
 			inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK,
 						  TCP_DELACK_MAX, TCP_RTO_MAX);
@@ -6059,7 +6062,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 				/* Prevent spurious tcp_cwnd_restart() on
 				 * first data packet.
 				 */
-				tp->lsndtime = tcp_time_stamp;
+				tp->lsndtime = tcp_time_stamp(tp);
 
 				tcp_initialize_rcv_mss(sk);
 				tcp_fast_path_on(tp);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 54139fa..7103540 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -442,7 +442,7 @@ void tcp_v4_err(struct sk_buff *icmp_skb, u32 info)
 		BUG_ON(!skb);
 
 		remaining = icsk->icsk_rto - min(icsk->icsk_rto,
-				tcp_time_stamp - TCP_SKB_CB(skb)->when);
+				tcp_time_stamp(tp) - TCP_SKB_CB(skb)->when);
 
 		if (remaining) {
 			inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
@@ -724,11 +724,13 @@ release_sk1:
    outside socket context is ugly, certainly. What can I do?
  */
 
-static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack,
+static void tcp_v4_send_ack(struct sock *sk, struct sk_buff *skb,
+			    u32 seq, u32 ack,
 			    u32 win, u32 ts, int oif,
 			    struct tcp_md5sig_key *key,
 			    int reply_flags, u8 tos)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	const struct tcphdr *th = tcp_hdr(skb);
 	struct {
 		struct tcphdr th;
@@ -750,7 +752,7 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack,
 		rep.opt[0] = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
 				   (TCPOPT_TIMESTAMP << 8) |
 				   TCPOLEN_TIMESTAMP);
-		rep.opt[1] = htonl(tcp_time_stamp);
+		rep.opt[1] = htonl(tcp_time_stamp(tp));
 		rep.opt[2] = htonl(ts);
 		arg.iov[0].iov_len += TCPOLEN_TSTAMP_ALIGNED;
 	}
@@ -799,7 +801,7 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
 	struct inet_timewait_sock *tw = inet_twsk(sk);
 	struct tcp_timewait_sock *tcptw = tcp_twsk(sk);
 
-	tcp_v4_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
+	tcp_v4_send_ack(sk, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
 			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
 			tcptw->tw_ts_recent,
 			tw->tw_bound_dev_if,
@@ -817,7 +819,7 @@ static void tcp_v4_reqsk_send_ack(struct sock *sk, struct sk_buff *skb,
 	/* sk->sk_state == TCP_LISTEN -> for regular TCP_SYN_RECV
 	 * sk->sk_state == TCP_SYN_RECV -> for Fast Open.
 	 */
-	tcp_v4_send_ack(skb, (sk->sk_state == TCP_LISTEN) ?
+	tcp_v4_send_ack(sk, skb, (sk->sk_state == TCP_LISTEN) ?
 			tcp_rsk(req)->snt_isn + 1 : tcp_sk(sk)->snd_nxt,
 			tcp_rsk(req)->rcv_nxt, req->rcv_wnd,
 			req->ts_recent,
@@ -839,6 +841,7 @@ static int tcp_v4_send_synack(struct sock *sk, struct dst_entry *dst,
 			      u16 queue_mapping,
 			      bool nocache)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	const struct inet_request_sock *ireq = inet_rsk(req);
 	struct flowi4 fl4;
 	int err = -1;
@@ -859,7 +862,7 @@ static int tcp_v4_send_synack(struct sock *sk, struct dst_entry *dst,
 					    ireq->opt);
 		err = net_xmit_eval(err);
 		if (!tcp_rsk(req)->snt_synack && !err)
-			tcp_rsk(req)->snt_synack = tcp_time_stamp;
+			tcp_rsk(req)->snt_synack = tcp_time_stamp(tp);
 	}
 
 	return err;
@@ -1393,7 +1396,7 @@ static int tcp_v4_conn_req_fastopen(struct sock *sk,
 				    ireq->rmt_addr, ireq->opt);
 	err = net_xmit_eval(err);
 	if (!err)
-		tcp_rsk(req)->snt_synack = tcp_time_stamp;
+		tcp_rsk(req)->snt_synack = tcp_time_stamp(tp);
 	/* XXX (TFO) - is it ok to ignore error and continue? */
 
 	spin_lock(&queue->fastopenq->lock);
@@ -1649,7 +1652,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 		if (err || want_cookie)
 			goto drop_and_free;
 
-		tcp_rsk(req)->snt_synack = tcp_time_stamp;
+		tcp_rsk(req)->snt_synack = tcp_time_stamp(tp);
 		tcp_rsk(req)->listener = NULL;
 		/* Add the request_sock to the SYN table */
 		inet_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT);
diff --git a/net/ipv4/tcp_lp.c b/net/ipv4/tcp_lp.c
index 72f7218..a86cc91 100644
--- a/net/ipv4/tcp_lp.c
+++ b/net/ipv4/tcp_lp.c
@@ -269,11 +269,11 @@ static void tcp_lp_pkts_acked(struct sock *sk, u32 num_acked, s32 rtt_us)
 		tcp_lp_rtt_sample(sk, rtt_us);
 
 	/* calc inference */
-	if (tcp_time_stamp > tp->rx_opt.rcv_tsecr)
-		lp->inference = 3 * (tcp_time_stamp - tp->rx_opt.rcv_tsecr);
+	if (tcp_time_stamp(tp) > tp->rx_opt.rcv_tsecr)
+		lp->inference = 3 * (tcp_time_stamp(tp) - tp->rx_opt.rcv_tsecr);
 
 	/* test if within inference */
-	if (lp->last_drop && (tcp_time_stamp - lp->last_drop < lp->inference))
+	if (lp->last_drop && (tcp_time_stamp(tp) - lp->last_drop < lp->inference))
 		lp->flag |= LP_WITHIN_INF;
 	else
 		lp->flag &= ~LP_WITHIN_INF;
@@ -310,7 +310,7 @@ static void tcp_lp_pkts_acked(struct sock *sk, u32 num_acked, s32 rtt_us)
 		tp->snd_cwnd = max(tp->snd_cwnd >> 1U, 1U);
 
 	/* record this drop time */
-	lp->last_drop = tcp_time_stamp;
+	lp->last_drop = tcp_time_stamp(tp);
 }
 
 static struct tcp_congestion_ops tcp_lp __read_mostly = {
diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c
index f696d7c..1732e3e 100644
--- a/net/ipv4/tcp_metrics.c
+++ b/net/ipv4/tcp_metrics.c
@@ -527,7 +527,7 @@ reset:
 		tp->snd_cwnd = 1;
 	else
 		tp->snd_cwnd = tcp_init_cwnd(tp, dst);
-	tp->snd_cwnd_stamp = tcp_time_stamp;
+	tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 }
 
 bool tcp_peer_is_proven(struct request_sock *req, struct dst_entry *dst, bool paws_check)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 5d45159..0ac097c 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -142,7 +142,7 @@ static __u16 tcp_advertise_mss(struct sock *sk)
 static void tcp_cwnd_restart(struct sock *sk, const struct dst_entry *dst)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
-	s32 delta = tcp_time_stamp - tp->lsndtime;
+	s32 delta = tcp_time_stamp(tp) - tp->lsndtime;
 	u32 restart_cwnd = tcp_init_cwnd(tp, dst);
 	u32 cwnd = tp->snd_cwnd;
 
@@ -154,7 +154,7 @@ static void tcp_cwnd_restart(struct sock *sk, const struct dst_entry *dst)
 	while ((delta -= inet_csk(sk)->icsk_rto) > 0 && cwnd > restart_cwnd)
 		cwnd >>= 1;
 	tp->snd_cwnd = max(cwnd, restart_cwnd);
-	tp->snd_cwnd_stamp = tcp_time_stamp;
+	tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 	tp->snd_cwnd_used = 0;
 }
 
@@ -163,7 +163,7 @@ static void tcp_event_data_sent(struct tcp_sock *tp,
 				struct sock *sk)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
-	const u32 now = tcp_time_stamp;
+	const u32 now = tcp_time_stamp(tp);
 
 	if (sysctl_tcp_slow_start_after_idle &&
 	    (!tp->packets_out && (s32)(now - tp->lsndtime) > icsk->icsk_rto))
@@ -1510,14 +1510,14 @@ static void tcp_cwnd_validate(struct sock *sk)
 	if (tp->packets_out >= tp->snd_cwnd) {
 		/* Network is feed fully. */
 		tp->snd_cwnd_used = 0;
-		tp->snd_cwnd_stamp = tcp_time_stamp;
+		tp->snd_cwnd_stamp = tcp_time_stamp(tp);
 	} else {
 		/* Network starves. */
 		if (tp->packets_out > tp->snd_cwnd_used)
 			tp->snd_cwnd_used = tp->packets_out;
 
 		if (sysctl_tcp_slow_start_after_idle &&
-		    (s32)(tcp_time_stamp - tp->snd_cwnd_stamp) >= inet_csk(sk)->icsk_rto)
+		    (s32)(tcp_time_stamp(tp) - tp->snd_cwnd_stamp) >= inet_csk(sk)->icsk_rto)
 			tcp_cwnd_application_limited(sk);
 	}
 }
@@ -1930,7 +1930,7 @@ static int tcp_mtu_probe(struct sock *sk)
 
 	/* We're ready to send.  If this fails, the probe will
 	 * be resegmented into mss-sized pieces by tcp_write_xmit(). */
-	TCP_SKB_CB(nskb)->when = tcp_time_stamp;
+	TCP_SKB_CB(nskb)->when = tcp_time_stamp(tp);
 	if (!tcp_transmit_skb(sk, nskb, 1, GFP_ATOMIC)) {
 		/* Decrement cwnd here because we are sending
 		 * effectively two packets. */
@@ -2024,7 +2024,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
 		    unlikely(tso_fragment(sk, skb, limit, mss_now, gfp)))
 			break;
 
-		TCP_SKB_CB(skb)->when = tcp_time_stamp;
+		TCP_SKB_CB(skb)->when = tcp_time_stamp(tp);
 
 		if (unlikely(tcp_transmit_skb(sk, skb, 1, gfp)))
 			break;
@@ -2380,7 +2380,7 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb)
 	/* Make a copy, if the first transmission SKB clone we made
 	 * is still in somebody's hands, else make a clone.
 	 */
-	TCP_SKB_CB(skb)->when = tcp_time_stamp;
+	TCP_SKB_CB(skb)->when = tcp_time_stamp(tp);
 
 	/* make sure skb->data is aligned on arches that require it */
 	if (unlikely(NET_IP_ALIGN && ((unsigned long)skb->data & 3))) {
@@ -2607,6 +2607,7 @@ void tcp_send_fin(struct sock *sk)
  */
 void tcp_send_active_reset(struct sock *sk, gfp_t priority)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb;
 
 	/* NOTE: No TCP options attached and we never retransmit this. */
@@ -2621,7 +2622,7 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority)
 	tcp_init_nondata_skb(skb, tcp_acceptable_seq(sk),
 			     TCPHDR_ACK | TCPHDR_RST);
 	/* Send it off. */
-	TCP_SKB_CB(skb)->when = tcp_time_stamp;
+	TCP_SKB_CB(skb)->when = tcp_time_stamp(tp);
 	if (tcp_transmit_skb(sk, skb, 0, priority))
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTFAILED);
 
@@ -2636,6 +2637,7 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority)
  */
 int tcp_send_synack(struct sock *sk)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb;
 
 	skb = tcp_write_queue_head(sk);
@@ -2660,7 +2662,7 @@ int tcp_send_synack(struct sock *sk)
 		TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_ACK;
 		TCP_ECN_send_synack(tcp_sk(sk), skb);
 	}
-	TCP_SKB_CB(skb)->when = tcp_time_stamp;
+	TCP_SKB_CB(skb)->when = tcp_time_stamp(tp);
 	return tcp_transmit_skb(sk, skb, 1, GFP_ATOMIC);
 }
 
@@ -2732,10 +2734,10 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst,
 	memset(&opts, 0, sizeof(opts));
 #ifdef CONFIG_SYN_COOKIES
 	if (unlikely(req->cookie_ts))
-		TCP_SKB_CB(skb)->when = cookie_init_timestamp(req);
+		TCP_SKB_CB(skb)->when = cookie_init_timestamp(tp, req);
 	else
 #endif
-	TCP_SKB_CB(skb)->when = tcp_time_stamp;
+	TCP_SKB_CB(skb)->when = tcp_time_stamp(tp);
 	tcp_header_size = tcp_synack_options(sk, req, mss,
 					     skb, &opts, &md5, xvp, foc)
 			+ sizeof(*th);
@@ -3005,7 +3007,7 @@ int tcp_connect(struct sock *sk)
 	skb_reserve(buff, MAX_TCP_HEADER);
 
 	tcp_init_nondata_skb(buff, tp->write_seq++, TCPHDR_SYN);
-	tp->retrans_stamp = TCP_SKB_CB(buff)->when = tcp_time_stamp;
+	tp->retrans_stamp = TCP_SKB_CB(buff)->when = tcp_time_stamp(tp);
 	tcp_connect_queue_skb(sk, buff);
 	TCP_ECN_send_syn(sk, buff);
 
@@ -3088,6 +3090,7 @@ void tcp_send_delayed_ack(struct sock *sk)
 /* This routine sends an ack and also updates the window. */
 void tcp_send_ack(struct sock *sk)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *buff;
 
 	/* If we have been reset, we may not send again. */
@@ -3112,7 +3115,7 @@ void tcp_send_ack(struct sock *sk)
 	tcp_init_nondata_skb(buff, tcp_acceptable_seq(sk), TCPHDR_ACK);
 
 	/* Send it off, this clears delayed acks for us. */
-	TCP_SKB_CB(buff)->when = tcp_time_stamp;
+	TCP_SKB_CB(buff)->when = tcp_time_stamp(tp);
 	tcp_transmit_skb(sk, buff, 0, sk_gfp_atomic(sk, GFP_ATOMIC));
 }
 
@@ -3144,7 +3147,7 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent)
 	 * send it.
 	 */
 	tcp_init_nondata_skb(skb, tp->snd_una - !urgent, TCPHDR_ACK);
-	TCP_SKB_CB(skb)->when = tcp_time_stamp;
+	TCP_SKB_CB(skb)->when = tcp_time_stamp(tp);
 	return tcp_transmit_skb(sk, skb, 0, GFP_ATOMIC);
 }
 
@@ -3189,7 +3192,7 @@ int tcp_write_wakeup(struct sock *sk)
 			tcp_set_skb_tso_segs(sk, skb, mss);
 
 		TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_PSH;
-		TCP_SKB_CB(skb)->when = tcp_time_stamp;
+		TCP_SKB_CB(skb)->when = tcp_time_stamp(tp);
 		err = tcp_transmit_skb(sk, skb, 1, GFP_ATOMIC);
 		if (!err)
 			tcp_event_new_data_sent(sk, skb);
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index b78aac3..6be8aa0 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -59,7 +59,7 @@ static int tcp_out_of_resources(struct sock *sk, int do_reset)
 
 	/* If peer does not open window for long time, or did not transmit
 	 * anything for long time, penalize it. */
-	if ((s32)(tcp_time_stamp - tp->lsndtime) > 2*TCP_RTO_MAX || !do_reset)
+	if ((s32)(tcp_time_stamp(tp) - tp->lsndtime) > 2*TCP_RTO_MAX || !do_reset)
 		shift++;
 
 	/* If some dubious ICMP arrived, penalize even more. */
@@ -69,7 +69,7 @@ static int tcp_out_of_resources(struct sock *sk, int do_reset)
 	if (tcp_check_oom(sk, shift)) {
 		/* Catch exceptional cases, when connection requires reset.
 		 *      1. Last segment was sent recently. */
-		if ((s32)(tcp_time_stamp - tp->lsndtime) <= TCP_TIMEWAIT_LEN ||
+		if ((s32)(tcp_time_stamp(tp) - tp->lsndtime) <= TCP_TIMEWAIT_LEN ||
 		    /*  2. Window is closed. */
 		    (!tp->snd_wnd && !tp->packets_out))
 			do_reset = 1;
@@ -129,6 +129,7 @@ static bool retransmits_timed_out(struct sock *sk,
 				  unsigned int timeout,
 				  bool syn_set)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	unsigned int linear_backoff_thresh, start_ts;
 	unsigned int rto_base = syn_set ? TCP_TIMEOUT_INIT : TCP_RTO_MIN;
 
@@ -149,7 +150,7 @@ static bool retransmits_timed_out(struct sock *sk,
 			timeout = ((2 << linear_backoff_thresh) - 1) * rto_base +
 				(boundary - linear_backoff_thresh) * TCP_RTO_MAX;
 	}
-	return (tcp_time_stamp - start_ts) >= timeout;
+	return (tcp_time_stamp(tp) - start_ts) >= timeout;
 }
 
 /* A write timeout has occurred. Process the after effects. */
@@ -383,7 +384,7 @@ void tcp_retransmit_timer(struct sock *sk)
 				       tp->snd_una, tp->snd_nxt);
 		}
 #endif
-		if (tcp_time_stamp - tp->rcv_tstamp > TCP_RTO_MAX) {
+		if (tcp_time_stamp(tp) - tp->rcv_tstamp > TCP_RTO_MAX) {
 			tcp_write_err(sk);
 			goto out;
 		}
diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp_westwood.c
index 1b91bf4..348db8d 100644
--- a/net/ipv4/tcp_westwood.c
+++ b/net/ipv4/tcp_westwood.c
@@ -60,6 +60,7 @@ struct westwood {
  */
 static void tcp_westwood_init(struct sock *sk)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	struct westwood *w = inet_csk_ca(sk);
 
 	w->bk = 0;
@@ -69,7 +70,7 @@ static void tcp_westwood_init(struct sock *sk)
 	w->cumul_ack = 0;
 	w->reset_rtt_min = 1;
 	w->rtt_min = w->rtt = TCP_WESTWOOD_INIT_RTT;
-	w->rtt_win_sx = tcp_time_stamp;
+	w->rtt_win_sx = tcp_time_stamp(tp);
 	w->snd_una = tcp_sk(sk)->snd_una;
 	w->first_ack = 1;
 }
@@ -115,8 +116,9 @@ static void tcp_westwood_pkts_acked(struct sock *sk, u32 cnt, s32 rtt)
  */
 static void westwood_update_window(struct sock *sk)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	struct westwood *w = inet_csk_ca(sk);
-	s32 delta = tcp_time_stamp - w->rtt_win_sx;
+	s32 delta = tcp_time_stamp(tp) - w->rtt_win_sx;
 
 	/* Initialize w->snd_una with the first acked sequence number in order
 	 * to fix mismatch between tp->snd_una and w->snd_una for the first
@@ -140,7 +142,7 @@ static void westwood_update_window(struct sock *sk)
 		westwood_filter(w, delta);
 
 		w->bk = 0;
-		w->rtt_win_sx = tcp_time_stamp;
+		w->rtt_win_sx = tcp_time_stamp(tp);
 	}
 }
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 93825dd..3f7a74d 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -711,9 +711,11 @@ static const struct tcp_request_sock_ops tcp_request_sock_ipv6_ops = {
 };
 #endif
 
-static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win,
-				 u32 ts, struct tcp_md5sig_key *key, int rst, u8 tclass)
+static void tcp_v6_send_response(struct sock *sk, struct sk_buff *skb,
+				 u32 seq, u32 ack, u32 win, u32 ts,
+				 struct tcp_md5sig_key *key, int rst, u8 tclass)
 {
+	const struct tcp_sock *tp = tcp_sk(sk);
 	const struct tcphdr *th = tcp_hdr(skb);
 	struct tcphdr *t1;
 	struct sk_buff *buff;
@@ -757,7 +759,7 @@ static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win,
 	if (ts) {
 		*topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
 				(TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP);
-		*topt++ = htonl(tcp_time_stamp);
+		*topt++ = htonl(tcp_time_stamp(tp));
 		*topt++ = htonl(ts);
 	}
 
@@ -858,7 +860,7 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb)
 		ack_seq = ntohl(th->seq) + th->syn + th->fin + skb->len -
 			  (th->doff << 2);
 
-	tcp_v6_send_response(skb, seq, ack_seq, 0, 0, key, 1, 0);
+	tcp_v6_send_response(sk, skb, seq, ack_seq, 0, 0, key, 1, 0);
 
 #ifdef CONFIG_TCP_MD5SIG
 release_sk1:
@@ -869,10 +871,11 @@ release_sk1:
 #endif
 }
 
-static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts,
+static void tcp_v6_send_ack(struct sock *sk, struct sk_buff *skb,
+			    u32 seq, u32 ack, u32 win, u32 ts,
 			    struct tcp_md5sig_key *key, u8 tclass)
 {
-	tcp_v6_send_response(skb, seq, ack, win, ts, key, 0, tclass);
+	tcp_v6_send_response(sk, skb, seq, ack, win, ts, key, 0, tclass);
 }
 
 static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb)
@@ -880,7 +883,7 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb)
 	struct inet_timewait_sock *tw = inet_twsk(sk);
 	struct tcp_timewait_sock *tcptw = tcp_twsk(sk);
 
-	tcp_v6_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
+	tcp_v6_send_ack(sk, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
 			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
 			tcptw->tw_ts_recent, tcp_twsk_md5_key(tcptw),
 			tw->tw_tclass);
@@ -891,7 +894,8 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb)
 static void tcp_v6_reqsk_send_ack(struct sock *sk, struct sk_buff *skb,
 				  struct request_sock *req)
 {
-	tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent,
+	tcp_v6_send_ack(sk, skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1,
+			req->rcv_wnd, req->ts_recent,
 			tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->daddr), 0);
 }
 
@@ -1098,7 +1102,7 @@ have_isn:
 	    want_cookie)
 		goto drop_and_free;
 
-	tcp_rsk(req)->snt_synack = tcp_time_stamp;
+	tcp_rsk(req)->snt_synack = tcp_time_stamp(tp);
 	tcp_rsk(req)->listener = NULL;
 	inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT);
 	return 0;
-- 
1.7.11.7

Re: [PATCH 1/2] tcp: make tcp_timestamp dependent on tcp_sock

[Eric Dumazet]

On Wed, 2013-01-23 at 19:01 +0400, Andrey Vagin wrote:
> This will be used for setting timestamp offset.
> 
> This patch converts the macros tcp_timestamp to a function with one
> argument "struct tcp_sock *tp" and nothing else.
> 
> Cc: "David S. Miller" <davem@davemloft.net>
> Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
> Cc: James Morris <jmorris@namei.org>
> Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
> Cc: Patrick McHardy <kaber@trash.net>
> Cc: Eric Dumazet <edumazet@google.com>
> Cc: Yuchung Cheng <ycheng@google.com>
> Cc: Neal Cardwell <ncardwell@google.com>
> Cc: Pavel Emelyanov <xemul@parallels.com>
> Cc: Dave Jones <davej@redhat.com>
> Cc: Michael Kerrisk <mtk.manpages@gmail.com>
> Signed-off-by: Andrey Vagin <avagin@openvz.org>
> ---
>  include/net/tcp.h       | 15 +++++++++-----
>  net/ipv4/syncookies.c   |  4 ++--
>  net/ipv4/tcp.c          |  2 +-
>  net/ipv4/tcp_bic.c      | 10 +++++-----
>  net/ipv4/tcp_cubic.c    | 14 ++++++-------
>  net/ipv4/tcp_htcp.c     |  2 +-
>  net/ipv4/tcp_input.c    | 53 ++++++++++++++++++++++++++-----------------------
>  net/ipv4/tcp_ipv4.c     | 19 ++++++++++--------
>  net/ipv4/tcp_lp.c       |  8 ++++----
>  net/ipv4/tcp_metrics.c  |  2 +-
>  net/ipv4/tcp_output.c   | 35 +++++++++++++++++---------------
>  net/ipv4/tcp_timer.c    |  9 +++++----
>  net/ipv4/tcp_westwood.c |  8 +++++---
>  net/ipv6/tcp_ipv6.c     | 22 +++++++++++---------
>  14 files changed, 112 insertions(+), 91 deletions(-)
> 
> diff --git a/include/net/tcp.h b/include/net/tcp.h
> index aed42c7..3e242ba 100644
> --- a/include/net/tcp.h
> +++ b/include/net/tcp.h
> @@ -503,7 +503,7 @@ static inline __u32 cookie_v4_init_sequence(struct sock *sk,
>  }
>  #endif
>  
> -extern __u32 cookie_init_timestamp(struct request_sock *req);
> +extern __u32 cookie_init_timestamp(struct tcp_sock *tp, struct request_sock *req);
>  extern bool cookie_check_timestamp(struct tcp_options_received *opt, bool *);
>  
>  /* From net/ipv6/syncookies.c */
> @@ -675,7 +675,10 @@ void tcp_send_window_probe(struct sock *sk);
>   * to use only the low 32-bits of jiffies and hide the ugly
>   * casts with the following macro.
>   */
> -#define tcp_time_stamp		((__u32)(jiffies))
> +static inline u32 tcp_time_stamp(const struct tcp_sock *tp)
> +{
> +	return (__u32)jiffies;
> +}
>  
>  #define tcp_flag_byte(th) (((u_int8_t *)th)[13])
>  
> @@ -1142,9 +1145,11 @@ static inline void tcp_openreq_init(struct request_sock *req,
>  static inline void tcp_synack_rtt_meas(struct sock *sk,
>  				       struct request_sock *req)
>  {
> +	const struct tcp_sock *tp = tcp_sk(sk);
> +
>  	if (tcp_rsk(req)->snt_synack)
>  		tcp_valid_rtt_meas(sk,
> -		    tcp_time_stamp - tcp_rsk(req)->snt_synack);
> +		    tcp_time_stamp(tp) - tcp_rsk(req)->snt_synack);
>  }
>  

This first chunk looks not needed.

Can SYN_RECV sockets be live-migrated ? Probably not.

Quite frankly I am sure there is an issue for timewait (not real
sockets).

Have you really tested this patch, or is it an RFC ?

Re: [PATCH 1/2] tcp: make tcp_timestamp dependent on tcp_sock

[Andrew Vagin]

On Wed, Jan 23, 2013 at 08:25:17AM -0800, Eric Dumazet wrote:
> > @@ -1142,9 +1145,11 @@ static inline void tcp_openreq_init(struct request_sock *req,
> >  static inline void tcp_synack_rtt_meas(struct sock *sk,
> >  				       struct request_sock *req)
> >  {
> > +	const struct tcp_sock *tp = tcp_sk(sk);
> > +
> >  	if (tcp_rsk(req)->snt_synack)
> >  		tcp_valid_rtt_meas(sk,
> > -		    tcp_time_stamp - tcp_rsk(req)->snt_synack);
> > +		    tcp_time_stamp(tp) - tcp_rsk(req)->snt_synack);
> >  }
> >  
> 
> This first chunk looks not needed.
> 
> Can SYN_RECV sockets be live-migrated ? Probably not.
> 
> Quite frankly I am sure there is an issue for timewait (not real
> sockets).
> 
> Have you really tested this patch, or is it an RFC ?

I set RFC, when I'm not sure, that this idea is right and want to get
opinions and comments before spend a lot of time.

I have tested this patch. I have a virtuall machine, where I test
kernels. SSH is used for accessing this VM. My test dumps an outside
TCP connection, then reboots the VM and restores the TCP connection.
Without this patch the tcp connection are not restored correctly, but
with this patch it works properly.

A timestamp offset is zero by default, so I did not catch any bugs.
I will test my patches more carefully.

Thank you all for the comments and sorry for noise in the ML.

> 
> 

Re: [PATCH 1/2] tcp: make tcp_timestamp dependent on tcp_sock

[Christoph Paasch]

On Wednesday 23 January 2013 19:01:53 Andrey Vagin wrote:
> This will be used for setting timestamp offset.
> 
> This patch converts the macros tcp_timestamp to a function with one
> argument "struct tcp_sock *tp" and nothing else.
> 
> Cc: "David S. Miller" <davem@davemloft.net>
> Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
> Cc: James Morris <jmorris@namei.org>
> Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
> Cc: Patrick McHardy <kaber@trash.net>
> Cc: Eric Dumazet <edumazet@google.com>
> Cc: Yuchung Cheng <ycheng@google.com>
> Cc: Neal Cardwell <ncardwell@google.com>
> Cc: Pavel Emelyanov <xemul@parallels.com>
> Cc: Dave Jones <davej@redhat.com>
> Cc: Michael Kerrisk <mtk.manpages@gmail.com>
> Signed-off-by: Andrey Vagin <avagin@openvz.org>
> ---
>  include/net/tcp.h       | 15 +++++++++-----
>  net/ipv4/syncookies.c   |  4 ++--
>  net/ipv4/tcp.c          |  2 +-
>  net/ipv4/tcp_bic.c      | 10 +++++-----
>  net/ipv4/tcp_cubic.c    | 14 ++++++-------
>  net/ipv4/tcp_htcp.c     |  2 +-
>  net/ipv4/tcp_input.c    | 53
> ++++++++++++++++++++++++++----------------------- net/ipv4/tcp_ipv4.c     |
> 19 ++++++++++--------
>  net/ipv4/tcp_lp.c       |  8 ++++----
>  net/ipv4/tcp_metrics.c  |  2 +-
>  net/ipv4/tcp_output.c   | 35 +++++++++++++++++---------------
>  net/ipv4/tcp_timer.c    |  9 +++++----
>  net/ipv4/tcp_westwood.c |  8 +++++---
>  net/ipv6/tcp_ipv6.c     | 22 +++++++++++---------
>  14 files changed, 112 insertions(+), 91 deletions(-)
> 
> diff --git a/include/net/tcp.h b/include/net/tcp.h
> index aed42c7..3e242ba 100644
> --- a/include/net/tcp.h
> +++ b/include/net/tcp.h
> @@ -503,7 +503,7 @@ static inline __u32 cookie_v4_init_sequence(struct sock
> *sk, }
>  #endif
>  
> -extern __u32 cookie_init_timestamp(struct request_sock *req);
> +extern __u32 cookie_init_timestamp(struct tcp_sock *tp, struct request_sock
> *req); extern bool cookie_check_timestamp(struct tcp_options_received *opt,
> bool *); 
>  /* From net/ipv6/syncookies.c */
> @@ -675,7 +675,10 @@ void tcp_send_window_probe(struct sock *sk);
>   * to use only the low 32-bits of jiffies and hide the ugly
>   * casts with the following macro.
>   */
> -#define tcp_time_stamp         ((__u32)(jiffies))
> +static inline u32 tcp_time_stamp(const struct tcp_sock *tp)
> +{
> +       return (__u32)jiffies;
> +}

DCCP also accesses tcp_time_stamp. So, it won't compile with DCCP enabled.


Christoph


-- 
IP Networking Lab --- http://inl.info.ucl.ac.be
MultiPath TCP in the Linux Kernel --- http://mptcp.info.ucl.ac.be
UCLouvain
--

[PATCH 2/2] tcp: add ability to set a timestamp offset (v2)

[Andrey Vagin]

If a TCP socket will get live-migrated from one box to another the
timestamps (which are typically ON) will get screwed up -- the new
kernel will generate TS values that has nothing to do with what they
were on dump. The solution is to yet again fix the kernel and put a
"timestamp offset" on a socket.

This patch adds one more tcp socket option TCP_TIMESTAMP, which is used
for getting and setting a current value of tcp_timestamp.

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: Dave Jones <davej@redhat.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Andrey Vagin <avagin@openvz.org>
---
 include/linux/tcp.h      | 2 ++
 include/net/tcp.h        | 2 +-
 include/uapi/linux/tcp.h | 1 +
 net/ipv4/tcp.c           | 7 +++++++
 4 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 4e1d228..5ebc67c 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -162,6 +162,8 @@ struct tcp_sock {
 	u32	rcv_tstamp;	/* timestamp of last received ACK (for keepalives) */
 	u32	lsndtime;	/* timestamp of last sent data packet (for restart window) */
 
+	u32	tsoffset;	/* timestamp offset */
+
 	struct list_head tsq_node; /* anchor in tsq_tasklet.head list */
 	unsigned long	tsq_flags;
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 3e242ba..25119c0 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -677,7 +677,7 @@ void tcp_send_window_probe(struct sock *sk);
  */
 static inline u32 tcp_time_stamp(const struct tcp_sock *tp)
 {
-	return (__u32)jiffies;
+	return (__u32)(jiffies) + tp->tsoffset;
 }
 
 #define tcp_flag_byte(th) (((u_int8_t *)th)[13])
diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h
index e962faa..6b1ead0 100644
--- a/include/uapi/linux/tcp.h
+++ b/include/uapi/linux/tcp.h
@@ -111,6 +111,7 @@ enum {
 #define TCP_QUEUE_SEQ		21
 #define TCP_REPAIR_OPTIONS	22
 #define TCP_FASTOPEN		23	/* Enable FastOpen on listeners */
+#define TCP_TIMESTAMP		24
 
 struct tcp_repair_opt {
 	__u32	opt_code;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 086ceda..23f93d4 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2704,6 +2704,10 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
 		else
 			err = -EINVAL;
 		break;
+	case TCP_TIMESTAMP:
+		tp->tsoffset = 0;
+		tp->tsoffset = val - tcp_time_stamp(tp);
+		break;
 	default:
 		err = -ENOPROTOOPT;
 		break;
@@ -2952,6 +2956,9 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 	case TCP_USER_TIMEOUT:
 		val = jiffies_to_msecs(icsk->icsk_user_timeout);
 		break;
+	case TCP_TIMESTAMP:
+		val = tcp_time_stamp(tp);
+		break;
 	default:
 		return -ENOPROTOOPT;
 	}
-- 
1.7.11.7

Re: [PATCH 2/2] tcp: add ability to set a timestamp offset (v2)

[Pavel Emelyanov]

> diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
> index 086ceda..23f93d4 100644
> --- a/net/ipv4/tcp.c
> +++ b/net/ipv4/tcp.c
> @@ -2704,6 +2704,10 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
>  		else
>  			err = -EINVAL;
>  		break;
> +	case TCP_TIMESTAMP:
> +		tp->tsoffset = 0;
> +		tp->tsoffset = val - tcp_time_stamp(tp);

I believe we should check that the socket state is TCP_CLOSE.
Otherwise tcp_time_stamp() readers might be surprised while
reading the value.

> +		break;
>  	default:
>  		err = -ENOPROTOOPT;
>  		break;

Re: [PATCH] [RFC] tcp: add ability to set a timestamp offset (v2)

[Alexey Kuznetsov]

Hello!

On Wed, Jan 23, 2013 at 07:01:52PM +0400, Andrey Vagin wrote:
> -#define tcp_time_stamp         ((__u32)(jiffies))
> +#define tcp_time_stamp(tp)     ((__u32)(jiffies) + tp->tsoffset)

This implies that you always have some tp in hands. AFAIK this is not true,
so that I am puzzled how you were able to make something compilable with this definition.

At least in tcp_v4_send_ack() you obviously use some invalid tp, there is
no socket in hands there. The thing which is called "sk" is dummy.

With container migration I used per-container offset. When you have no container,
you have to do something a little bit smarter.

Alexey

Re: [PATCH] [RFC] tcp: add ability to set a timestamp offset (v2)

[David Miller]

From: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Date: Wed, 23 Jan 2013 20:20:25 +0400

> Hello!
> 
> On Wed, Jan 23, 2013 at 07:01:52PM +0400, Andrey Vagin wrote:
>> -#define tcp_time_stamp         ((__u32)(jiffies))
>> +#define tcp_time_stamp(tp)     ((__u32)(jiffies) + tp->tsoffset)
> 
> This implies that you always have some tp in hands. AFAIK this is not true,
> so that I am puzzled how you were able to make something compilable with this definition.
> 
> At least in tcp_v4_send_ack() you obviously use some invalid tp, there is
> no socket in hands there. The thing which is called "sk" is dummy.
> 
> With container migration I used per-container offset. When you have no container,
> you have to do something a little bit smarter.

I'm starting to have almost no confidence in this change, and even the
basic conceptual idea behind it.

Please go back to the drawing board and really analyze this thing
completely rather than just bouncing broken patches to the list every
day.

Thanks.