[PATCH net 0/3] tcp_bbr: PROBE_RTT minor bug fixes

[PATCH net 0/3] tcp_bbr: PROBE_RTT minor bug fixes

[Kevin Yang]

From: "Kevin(Yudong) Yang" <yyd@google.com>

This series includes two minor bug fixes for the TCP BBR PROBE_RTT
mechanism, and one preparatory patch:

(1) A preparatory patch to reorganize the PROBE_RTT logic by refactoring
    (into its own function) the code to exit PROBE_RTT, since the next
    patch will be using that code in a new context.

(2) Fix: When BBR restarts from idle and if BBR is in PROBE_RTT mode,
    BBR should check if it's time to exit PROBE_RTT. If yes, then BBR
    should exit PROBE_RTT mode and restore the cwnd to its full value.

(3) Fix: Apply the PROBE_RTT cwnd cap even if the count of fully-ACKed
    packets is 0.

Kevin Yang (3):
  tcp_bbr: add bbr_check_probe_rtt_done() helper
  tcp_bbr: in restart from idle, see if we should exit PROBE_RTT
  tcp_bbr: apply PROBE_RTT cwnd cap even if acked==0

 net/ipv4/tcp_bbr.c | 42 ++++++++++++++++++++++++------------------
 1 file changed, 24 insertions(+), 18 deletions(-)

-- 
2.18.0.1017.ga543ac7ca45-goog

[PATCH net 1/3] tcp_bbr: add bbr_check_probe_rtt_done() helper

[Kevin Yang]

This patch add a helper function bbr_check_probe_rtt_done() to
  1. check the condition to see if bbr should exit probe_rtt mode;
  2. process the logic of exiting probe_rtt mode.

Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control")
Signed-off-by: Kevin Yang <yyd@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Soheil Hassas Yeganeh <soheil@google.com>
---
 net/ipv4/tcp_bbr.c | 34 ++++++++++++++++++----------------
 1 file changed, 18 insertions(+), 16 deletions(-)

diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c
index 13d34427ca3dd..fd7bccf36a263 100644
--- a/net/ipv4/tcp_bbr.c
+++ b/net/ipv4/tcp_bbr.c
@@ -95,11 +95,10 @@ struct bbr {
 	u32     mode:3,		     /* current bbr_mode in state machine */
 		prev_ca_state:3,     /* CA state on previous ACK */
 		packet_conservation:1,  /* use packet conservation? */
-		restore_cwnd:1,	     /* decided to revert cwnd to old value */
 		round_start:1,	     /* start of packet-timed tx->ack round? */
 		idle_restart:1,	     /* restarting after idle? */
 		probe_rtt_round_done:1,  /* a BBR_PROBE_RTT round at 4 pkts? */
-		unused:12,
+		unused:13,
 		lt_is_sampling:1,    /* taking long-term ("LT") samples now? */
 		lt_rtt_cnt:7,	     /* round trips in long-term interval */
 		lt_use_bw:1;	     /* use lt_bw as our bw estimate? */
@@ -396,17 +395,11 @@ static bool bbr_set_cwnd_to_recover_or_restore(
 		cwnd = tcp_packets_in_flight(tp) + acked;
 	} else if (prev_state >= TCP_CA_Recovery && state < TCP_CA_Recovery) {
 		/* Exiting loss recovery; restore cwnd saved before recovery. */
-		bbr->restore_cwnd = 1;
+		cwnd = max(cwnd, bbr->prior_cwnd);
 		bbr->packet_conservation = 0;
 	}
 	bbr->prev_ca_state = state;
 
-	if (bbr->restore_cwnd) {
-		/* Restore cwnd after exiting loss recovery or PROBE_RTT. */
-		cwnd = max(cwnd, bbr->prior_cwnd);
-		bbr->restore_cwnd = 0;
-	}
-
 	if (bbr->packet_conservation) {
 		*new_cwnd = max(cwnd, tcp_packets_in_flight(tp) + acked);
 		return true;	/* yes, using packet conservation */
@@ -748,6 +741,20 @@ static void bbr_check_drain(struct sock *sk, const struct rate_sample *rs)
 		bbr_reset_probe_bw_mode(sk);  /* we estimate queue is drained */
 }
 
+static void bbr_check_probe_rtt_done(struct sock *sk)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	struct bbr *bbr = inet_csk_ca(sk);
+
+	if (!(bbr->probe_rtt_done_stamp &&
+	      after(tcp_jiffies32, bbr->probe_rtt_done_stamp)))
+		return;
+
+	bbr->min_rtt_stamp = tcp_jiffies32;  /* wait a while until PROBE_RTT */
+	tp->snd_cwnd = max(tp->snd_cwnd, bbr->prior_cwnd);
+	bbr_reset_mode(sk);
+}
+
 /* The goal of PROBE_RTT mode is to have BBR flows cooperatively and
  * periodically drain the bottleneck queue, to converge to measure the true
  * min_rtt (unloaded propagation delay). This allows the flows to keep queues
@@ -806,12 +813,8 @@ static void bbr_update_min_rtt(struct sock *sk, const struct rate_sample *rs)
 		} else if (bbr->probe_rtt_done_stamp) {
 			if (bbr->round_start)
 				bbr->probe_rtt_round_done = 1;
-			if (bbr->probe_rtt_round_done &&
-			    after(tcp_jiffies32, bbr->probe_rtt_done_stamp)) {
-				bbr->min_rtt_stamp = tcp_jiffies32;
-				bbr->restore_cwnd = 1;  /* snap to prior_cwnd */
-				bbr_reset_mode(sk);
-			}
+			if (bbr->probe_rtt_round_done)
+				bbr_check_probe_rtt_done(sk);
 		}
 	}
 	/* Restart after idle ends only once we process a new S/ACK for data */
@@ -862,7 +865,6 @@ static void bbr_init(struct sock *sk)
 	bbr->has_seen_rtt = 0;
 	bbr_init_pacing_rate_from_rtt(sk);
 
-	bbr->restore_cwnd = 0;
 	bbr->round_start = 0;
 	bbr->idle_restart = 0;
 	bbr->full_bw_reached = 0;
-- 
2.18.0.1017.ga543ac7ca45-goog

[PATCH net 2/3] tcp_bbr: in restart from idle, see if we should exit PROBE_RTT

[Kevin Yang]

This patch fix the case where BBR does not exit PROBE_RTT mode when
it restarts from idle. When BBR restarts from idle and if BBR is in
PROBE_RTT mode, BBR should check if it's time to exit PROBE_RTT. If
yes, then BBR should exit PROBE_RTT mode and restore the cwnd to its
full value.

Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control")
Signed-off-by: Kevin Yang <yyd@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Soheil Hassas Yeganeh <soheil@google.com>
---
 net/ipv4/tcp_bbr.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c
index fd7bccf36a263..1d4bdd3b5e4d0 100644
--- a/net/ipv4/tcp_bbr.c
+++ b/net/ipv4/tcp_bbr.c
@@ -174,6 +174,8 @@ static const u32 bbr_lt_bw_diff = 4000 / 8;
 /* If we estimate we're policed, use lt_bw for this many round trips: */
 static const u32 bbr_lt_bw_max_rtts = 48;
 
+static void bbr_check_probe_rtt_done(struct sock *sk);
+
 /* Do we estimate that STARTUP filled the pipe? */
 static bool bbr_full_bw_reached(const struct sock *sk)
 {
@@ -308,6 +310,8 @@ static void bbr_cwnd_event(struct sock *sk, enum tcp_ca_event event)
 		 */
 		if (bbr->mode == BBR_PROBE_BW)
 			bbr_set_pacing_rate(sk, bbr_bw(sk), BBR_UNIT);
+		else if (bbr->mode == BBR_PROBE_RTT)
+			bbr_check_probe_rtt_done(sk);
 	}
 }
 
-- 
2.18.0.1017.ga543ac7ca45-goog

[PATCH net 3/3] tcp_bbr: apply PROBE_RTT cwnd cap even if acked==0

[Kevin Yang]

This commit fixes a corner case where TCP BBR would enter PROBE_RTT
mode but not reduce its cwnd. If a TCP receiver ACKed less than one
full segment, the number of delivered/acked packets was 0, so that
bbr_set_cwnd() would short-circuit and exit early, without cutting
cwnd to the value we want for PROBE_RTT.

The fix is to instead make sure that even when 0 full packets are
ACKed, we do apply all the appropriate caps, including the cap that
applies in PROBE_RTT mode.

Fixes: 0f8782ea1497 ("tcp_bbr: add BBR congestion control")
Signed-off-by: Kevin Yang <yyd@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Soheil Hassas Yeganeh <soheil@google.com>
---
 net/ipv4/tcp_bbr.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c
index 1d4bdd3b5e4d0..02ff2dde96094 100644
--- a/net/ipv4/tcp_bbr.c
+++ b/net/ipv4/tcp_bbr.c
@@ -420,10 +420,10 @@ static void bbr_set_cwnd(struct sock *sk, const struct rate_sample *rs,
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct bbr *bbr = inet_csk_ca(sk);
-	u32 cwnd = 0, target_cwnd = 0;
+	u32 cwnd = tp->snd_cwnd, target_cwnd = 0;
 
 	if (!acked)
-		return;
+		goto done;  /* no packet fully ACKed; just apply caps */
 
 	if (bbr_set_cwnd_to_recover_or_restore(sk, rs, acked, &cwnd))
 		goto done;
-- 
2.18.0.1017.ga543ac7ca45-goog

Re: [PATCH net 0/3] tcp_bbr: PROBE_RTT minor bug fixes

[David Miller]

From: Kevin Yang <yyd@google.com>
Date: Wed, 22 Aug 2018 17:43:13 -0400

> From: "Kevin(Yudong) Yang" <yyd@google.com>
> 
> This series includes two minor bug fixes for the TCP BBR PROBE_RTT
> mechanism, and one preparatory patch:
> 
> (1) A preparatory patch to reorganize the PROBE_RTT logic by refactoring
>     (into its own function) the code to exit PROBE_RTT, since the next
>     patch will be using that code in a new context.
> 
> (2) Fix: When BBR restarts from idle and if BBR is in PROBE_RTT mode,
>     BBR should check if it's time to exit PROBE_RTT. If yes, then BBR
>     should exit PROBE_RTT mode and restore the cwnd to its full value.
> 
> (3) Fix: Apply the PROBE_RTT cwnd cap even if the count of fully-ACKed
>     packets is 0.

Series applied, thank you.