tcp: tsq: no longer use limit_output_bytes for paced flows

FQ pacing guarantees that paced packets queued by one flow do not
add head-of-line blocking for other flows.

After TCP GSO conversion, increasing limit_output_bytes to 1 MB is safe,
since this maps to 16 skbs at most in qdisc or device queues.
(or slightly more if some drivers lower {gso_max_segs|size})

We still can queue at most 1 ms worth of traffic (this can be scaled
by wifi drivers if they need to)

Tested:

# ethtool -c eth0 | egrep "tx-usecs:|tx-frames:" # 40 Gbit mlx4 NIC
tx-usecs: 16
tx-frames: 16
# tc qdisc replace dev eth0 root fq
# for f in {1..10};do netperf -P0 -H lpaa24,6 -o THROUGHPUT;done

Before patch:
27711
26118
27107
27377
27712
27388
27340
27117
27278
27509

After patch:
37434
36949
36658
36998
37711
37291
37605
36659
36544
37349

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Documentation/networking/ip-sysctl.txt

@@ -759,7 +759,7 @@ tcp_limit_output_bytes - INTEGER
 	flows, for typical pfifo_fast qdiscs.  tcp_limit_output_bytes
 	limits the number of bytes on qdisc or device to reduce artificial
 	RTT/cwnd and reduce bufferbloat.
-	Default: 262144
+	Default: 1048576 (16 * 65536)
 
 tcp_challenge_ack_limit - INTEGER
 	Limits number of Challenge ACK sent per second, as recommended

net/ipv4/tcp_ipv4.c

@@ -2574,8 +2574,8 @@ static int __net_init tcp_sk_init(struct net *net)
 	 * which are too large can cause TCP streams to be bursty.
 	 */
 	net->ipv4.sysctl_tcp_tso_win_divisor = 3;
-	/* Default TSQ limit of four TSO segments */
-	net->ipv4.sysctl_tcp_limit_output_bytes = 262144;
+	/* Default TSQ limit of 16 TSO segments */
+	net->ipv4.sysctl_tcp_limit_output_bytes = 16 * 65536;
 	/* rfc5961 challenge ack rate limiting */
 	net->ipv4.sysctl_tcp_challenge_ack_limit = 1000;
 	net->ipv4.sysctl_tcp_min_tso_segs = 2;

net/ipv4/tcp_output.c

@@ -2220,8 +2220,9 @@ static bool tcp_small_queue_check(struct sock *sk, const struct sk_buff *skb,
 	limit = max_t(unsigned long,
 		      2 * skb->truesize,
 		      sk->sk_pacing_rate >> sk->sk_pacing_shift);
-	limit = min_t(unsigned long, limit,
-		      sock_net(sk)->ipv4.sysctl_tcp_limit_output_bytes);
+	if (sk->sk_pacing_status == SK_PACING_NONE)
+		limit = min_t(unsigned long, limit,
+			      sock_net(sk)->ipv4.sysctl_tcp_limit_output_bytes);
 	limit <<= factor;
 
 	if (refcount_read(&sk->sk_wmem_alloc) > limit) {