tcp: tsq: restore minimal amount of queueing

After commit c9eeec26e32e ("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: c9eeec26e32e ("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>

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 8b8a0578764112c1dacc4a719f454c676c09100c..3c12d9a7ed00391d5c3f49ef80d7b1ae9abc40fe 100644 (file)
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -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

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 8e8529d3c8c92a4473a3ed9fd8db876ca83533bd..3dc0c6cf02a896e66071cd5f66b0a93f0dd3fc06 100644 (file)
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -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 */

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 672854664ff5c1d36783ac1bfb72fea481648ca1..c5231d9b06d700b9a4678d6cb6232fcb0f5ab8d3 100644 (file)
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -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);