tcp: tsq: restore minimal amount of queueing (98e09386) · Commits · 戴 / test

After commit ("tcp: TSQ can use a dynamic limit"), several users reported throughput regressions, notably on mvneta and wifi adapters. 802.11 AMPDU requires a fair amount of queueing to be effective. This patch partially reverts the change done in tcp_write_xmit() so that the minimal amount is sysctl_tcp_limit_output_bytes. It also remove the use of this sysctl while building skb stored in write queue, as TSO autosizing does the right thing anyway. Users with well behaving NICS and correct qdisc (like sch_fq), can then lower the default sysctl_tcp_limit_output_bytes value from 128KB to 8KB. This new usage of sysctl_tcp_limit_output_bytes permits each driver authors to check how their driver performs when/if the value is set to a minimum of 4KB. Normally, line rate for a single TCP flow should be possible, but some drivers rely on timers to perform TX completion and too long TX completion delays prevent reaching full throughput. Fixes: ("tcp: TSQ can use a dynamic limit") Signed-off-by:

Eric Dumazet <edumazet@google.com> Reported-by:

Sujith Manoharan <sujith@msujith.org> Reported-by:

Arnaud Ebalard <arno@natisbad.org> Tested-by:

Sujith Manoharan <sujith@msujith.org> Cc: Felix Fietkau <nbd@openwrt.org> Signed-off-by:

David S. Miller <davem@davemloft.net>

Documentation/networking/ip-sysctl.txt

+0 −3

Original line number	Diff line number	Diff line
		@@ -577,9 +577,6 @@ tcp_limit_output_bytes - INTEGER
		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.
		Note: For GSO/TSO enabled flows, we try to have at least two
		packets in flight. Reducing tcp_limit_output_bytes might also
		reduce the size of individual GSO packet (64KB being the max)
		Default: 131072

		tcp_challenge_ack_limit - INTEGER

net/ipv4/tcp.c

+0 −6

Original line number	Diff line number	Diff line
		@@ -808,12 +808,6 @@ static unsigned int tcp_xmit_size_goal(struct sock *sk, u32 mss_now,
		xmit_size_goal = min_t(u32, gso_size,
		sk->sk_gso_max_size - 1 - hlen);

		/* TSQ : try to have at least two segments in flight
		* (one in NIC TX ring, another in Qdisc)
		*/
		xmit_size_goal = min_t(u32, xmit_size_goal,
		sysctl_tcp_limit_output_bytes >> 1);

		xmit_size_goal = tcp_bound_to_half_wnd(tp, xmit_size_goal);

		/* We try hard to avoid divides here */

net/ipv4/tcp_output.c

+5 −1

Original line number	Diff line number	Diff line
		@@ -1875,8 +1875,12 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
		* - better RTT estimation and ACK scheduling
		* - faster recovery
		* - high rates
		* Alas, some drivers / subsystems require a fair amount
		* of queued bytes to ensure line rate.
		* One example is wifi aggregation (802.11 AMPDU)
		*/
		limit = max(skb->truesize, sk->sk_pacing_rate >> 10);
		limit = max_t(unsigned int, sysctl_tcp_limit_output_bytes,
		sk->sk_pacing_rate >> 10);

		if (atomic_read(&sk->sk_wmem_alloc) > limit) {
		set_bit(TSQ_THROTTLED, &tp->tsq_flags);

Admin message