tcp: fix tcp_cwnd_validate() to not forget is_cwnd_limited

This commit fixes a bug in the tracking of max_packets_out and
is_cwnd_limited. This bug can cause the connection to fail to remember
that is_cwnd_limited is true, causing the connection to fail to grow
cwnd when it should, causing throughput to be lower than it should be.

The following event sequence is an example that triggers the bug:

 (a) The connection is cwnd_limited, but packets_out is not at its
     peak due to TSO deferral deciding not to send another skb yet.
     In such cases the connection can advance max_packets_seq and set
     tp->is_cwnd_limited to true and max_packets_out to a small
     number.

(b) Then later in the round trip the connection is pacing-limited (not
     cwnd-limited), and packets_out is larger. In such cases the
     connection would raise max_packets_out to a bigger number but
     (unexpectedly) flip tp->is_cwnd_limited from true to false.

This commit fixes that bug.

One straightforward fix would be to separately track (a) the next
window after max_packets_out reaches a maximum, and (b) the next
window after tp->is_cwnd_limited is set to true. But this would
require consuming an extra u32 sequence number.

Instead, to save space we track only the most important
information. Specifically, we track the strongest available signal of
the degree to which the cwnd is fully utilized:

(1) If the connection is cwnd-limited then we remember that fact for
the current window.

(2) If the connection not cwnd-limited then we track the maximum
number of outstanding packets in the current window.

In particular, note that the new logic cannot trigger the buggy
(a)/(b) sequence above because with the new logic a condition where
tp->packets_out > tp->max_packets_out can only trigger an update of
tp->is_cwnd_limited if tp->is_cwnd_limited is false.

This first showed up in a testing of a BBRv2 dev branch, but this
buggy behavior highlighted a general issue with the
tcp_cwnd_validate() logic that can cause cwnd to fail to increase at
the proper rate for any TCP congestion control, including Reno or
CUBIC.

Fixes: ca8a22634381 ("tcp: make cwnd-limited checks measurement-based, and gentler")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Kevin(Yudong) Yang <yyd@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index a9fbe22732c3ba67584df13344955b584f242b54..4791fd80194541f0ef4b1966c4fc6ab297dfe37f 100644 (file)
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -295,7 +295,7 @@ struct tcp_sock {
        u32     packets_out;    /* Packets which are "in flight"        */
        u32     retrans_out;    /* Retransmitted packets out            */
        u32     max_packets_out;  /* max packets_out in last window */
-       u32     max_packets_seq;  /* right edge of max_packets_out flight */
+       u32     cwnd_usage_seq;  /* right edge of cwnd usage tracking flight */
 
        u16     urg_data;       /* Saved octet of OOB data and control flags */
        u8      ecn_flags;      /* ECN status bits.                     */

diff --git a/include/net/tcp.h b/include/net/tcp.h
index d10962b9f0d0d25f46331b6be3fd7f9c2a136a80..95c1d51393ac40ebcde1ab13817bdaad6c5ce51a 100644 (file)
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1295,11 +1295,14 @@ static inline bool tcp_is_cwnd_limited(const struct sock *sk)
 {
        const struct tcp_sock *tp = tcp_sk(sk);
 
+       if (tp->is_cwnd_limited)
+               return true;
+
        /* If in slow start, ensure cwnd grows to twice what was ACKed. */
        if (tcp_in_slow_start(tp))
                return tcp_snd_cwnd(tp) < 2 * tp->max_packets_out;
 
-       return tp->is_cwnd_limited;
+       return false;
 }
 
 /* BBR congestion control needs pacing.

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index e373dde1f46f7b286c01aef022117cb10ba64e45..997a80ce1e139ea5782c37bc0e9b582ad460c7c3 100644 (file)
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3137,6 +3137,8 @@ int tcp_disconnect(struct sock *sk, int flags)
        tp->snd_ssthresh = TCP_INFINITE_SSTHRESH;
        tcp_snd_cwnd_set(tp, TCP_INIT_CWND);
        tp->snd_cwnd_cnt = 0;
+       tp->is_cwnd_limited = 0;
+       tp->max_packets_out = 0;
        tp->window_clamp = 0;
        tp->delivered = 0;
        tp->delivered_ce = 0;

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 290019de766dc2b8dae3bbfbcd5011d6d03f0f45..c69f4d966024ce2c37dae63f7be94fe79c7dd648 100644 (file)
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1875,15 +1875,20 @@ static void tcp_cwnd_validate(struct sock *sk, bool is_cwnd_limited)
        const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops;
        struct tcp_sock *tp = tcp_sk(sk);
 
-       /* Track the maximum number of outstanding packets in each
-        * window, and remember whether we were cwnd-limited then.
+       /* Track the strongest available signal of the degree to which the cwnd
+        * is fully utilized. If cwnd-limited then remember that fact for the
+        * current window. If not cwnd-limited then track the maximum number of
+        * outstanding packets in the current window. (If cwnd-limited then we
+        * chose to not update tp->max_packets_out to avoid an extra else
+        * clause with no functional impact.)
         */
-       if (!before(tp->snd_una, tp->max_packets_seq) ||
-           tp->packets_out > tp->max_packets_out ||
-           is_cwnd_limited) {
-               tp->max_packets_out = tp->packets_out;
-               tp->max_packets_seq = tp->snd_nxt;
+       if (!before(tp->snd_una, tp->cwnd_usage_seq) ||
+           is_cwnd_limited ||
+           (!tp->is_cwnd_limited &&
+            tp->packets_out > tp->max_packets_out)) {
                tp->is_cwnd_limited = is_cwnd_limited;
+               tp->max_packets_out = tp->packets_out;
+               tp->cwnd_usage_seq = tp->snd_nxt;
        }
 
        if (tcp_is_cwnd_limited(sk)) {