linux/net/ipv4
Neal Cardwell ed66dfaf23 tcp: when scheduling TLP, time of RTO should account for current ACK
Fix the TLP scheduling logic so that when scheduling a TLP probe, we
ensure that the estimated time at which an RTO would fire accounts for
the fact that ACKs indicating forward progress should push back RTO
times.

After the following fix:

df92c8394e ("tcp: fix xmit timer to only be reset if data ACKed/SACKed")

we had an unintentional behavior change in the following kind of
scenario: suppose the RTT variance has been very low recently. Then
suppose we send out a flight of N packets and our RTT is 100ms:

t=0: send a flight of N packets
t=100ms: receive an ACK for N-1 packets

The response before df92c8394e that was:
  -> schedule a TLP for now + RTO_interval

The response after df92c8394e is:
  -> schedule a TLP for t=0 + RTO_interval

Since RTO_interval = srtt + RTT_variance, this means that we have
scheduled a TLP timer at a point in the future that only accounts for
RTT_variance. If the RTT_variance term is small, this means that the
timer fires soon.

Before df92c8394e this would not happen, because in that code, when
we receive an ACK for a prefix of flight, we did:

    1) Near the top of tcp_ack(), switch from TLP timer to RTO
       at write_queue_head->paket_tx_time + RTO_interval:
            if (icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)
                   tcp_rearm_rto(sk);

    2) In tcp_clean_rtx_queue(), update the RTO to now + RTO_interval:
            if (flag & FLAG_ACKED) {
                   tcp_rearm_rto(sk);

    3) In tcp_ack() after tcp_fastretrans_alert() switch from RTO
       to TLP at now + RTO_interval:
            if (icsk->icsk_pending == ICSK_TIME_RETRANS)
                   tcp_schedule_loss_probe(sk);

In df92c8394e we removed that 3-phase dance, and instead directly
set the TLP timer once: we set the TLP timer in cases like this to
write_queue_head->packet_tx_time + RTO_interval. So if the RTT
variance is small, then this means that this is setting the TLP timer
to fire quite soon. This means if the ACK for the tail of the flight
takes longer than an RTT to arrive (often due to delayed ACKs), then
the TLP timer fires too quickly.

Fixes: df92c8394e ("tcp: fix xmit timer to only be reset if data ACKed/SACKed")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-19 12:25:26 +09:00
..
netfilter Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next 2017-11-08 14:22:50 +09:00
af_inet.c ipv4: mark expected switch fall-throughs 2017-10-18 14:10:29 +01:00
ah4.c net: use -ENOSPC for transient busy indication 2017-11-03 22:11:17 +08:00
arp.c ipv4: mark expected switch fall-throughs 2017-10-18 14:10:29 +01:00
cipso_ipv4.c tcp/dccp: fix ireq->opt races 2017-10-21 01:33:19 +01:00
datagram.c
devinet.c net: Add extack to validator_info structs used for address notifier 2017-10-20 13:15:07 +01:00
esp4_offload.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-09-01 17:42:05 -07:00
esp4.c net: use -ENOSPC for transient busy indication 2017-11-03 22:11:17 +08:00
fib_frontend.c ipv4: fix validate_source for VRF setup 2017-11-01 21:25:15 +09:00
fib_lookup.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
fib_notifier.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
fib_rules.c net: fib_rules: Implement notification logic in core 2017-08-03 15:35:59 -07:00
fib_semantics.c tcp: Namespace-ify sysctl_tcp_default_congestion_control 2017-11-15 14:09:52 +09:00
fib_trie.c net: Add extack to fib_notifier_info 2017-11-01 11:50:43 +09:00
fou.c gue: fix remcsum when GRO on and CHECKSUM_PARTIAL boundary is outer UDP 2017-08-01 16:09:14 -07:00
gre_demux.c
gre_offload.c gso: fix payload length when gso_size is zero 2017-10-08 10:12:15 -07:00
icmp.c ipv4: icmp: use BUG_ON instead of if condition followed by BUG 2017-10-24 18:44:42 +09:00
igmp.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-08-21 17:06:42 -07:00
inet_connection_sock.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-10-30 21:09:24 +09:00
inet_diag.c inet_diag: allow protocols to provide additional data 2017-09-01 18:38:09 -07:00
inet_fragment.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2017-11-15 11:56:19 -08:00
inet_hashtables.c soreuseport: fix initialization race 2017-10-22 02:03:51 +01:00
inet_timewait_sock.c Merge branch 'akpm' (patches from Andrew) 2017-11-15 19:42:40 -08:00
inetpeer.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-10-05 18:19:22 -07:00
ip_forward.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ip_fragment.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-11-04 09:26:51 +09:00
ip_gre.c ip_gre: add the support for i/o_flags update via ioctl 2017-11-10 14:39:53 +09:00
ip_input.c IPv4: early demux can return an error code 2017-10-01 03:55:47 +01:00
ip_options.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ip_output.c udp: remove unreachable ufo branches 2017-08-22 14:27:18 -07:00
ip_sockglue.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ip_tunnel_core.c net: store port/representator id in metadata_dst 2017-06-25 11:42:01 -04:00
ip_tunnel.c ipv4: speedup ipv6 tunnels dismantle 2017-09-19 16:32:24 -07:00
ip_vti.c ip_vti: remove the useless err_count check in vti_xmit 2017-11-01 11:57:32 +09:00
ipcomp.c
ipconfig.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ipip.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-10-30 21:09:24 +09:00
ipmr.c ipv4: mark expected switch fall-throughs 2017-10-18 14:10:29 +01:00
Kconfig ip: update policy routing config help 2017-10-12 22:57:11 -07:00
Makefile License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
netfilter.c netfilter: use skb_to_full_sk in ip_route_me_harder 2017-02-28 12:49:36 +01:00
ping.c net: convert sock.sk_refcnt from atomic_t to refcount_t 2017-07-01 07:39:08 -07:00
proc.c tcp: retire FACK loss detection 2017-11-11 18:53:16 +09:00
protocol.c net: Add sysctl to toggle early demux for tcp and udp 2017-03-24 13:17:07 -07:00
raw_diag.c net: ipv6: add second dif to raw socket lookups 2017-08-07 11:39:22 -07:00
raw.c net: ipv4: add second dif to multicast source filter 2017-08-07 11:39:22 -07:00
route.c route: also update fnhe_genid when updating a route cache 2017-11-18 10:32:41 +09:00
syncookies.c tcp: Namespace-ify sysctl_tcp_workaround_signed_windows 2017-10-28 19:24:38 +09:00
sysctl_net_ipv4.c tcp: Namespace-ify sysctl_tcp_default_congestion_control 2017-11-15 14:09:52 +09:00
tcp_bbr.c tcp_bbr: init pacing rate on first RTT sample 2017-07-15 14:43:29 -07:00
tcp_bic.c tcp: consolidate congestion control undo functions 2017-08-06 21:25:10 -07:00
tcp_cdg.c tcp: cdg: make struct tcp_cdg static 2017-10-16 21:24:25 +01:00
tcp_cong.c tcp: Namespace-ify sysctl_tcp_default_congestion_control 2017-11-15 14:09:52 +09:00
tcp_cubic.c tcp: consolidate congestion control undo functions 2017-08-06 21:25:10 -07:00
tcp_dctcp.c Revert "dctcp: update cwnd on congestion event" 2016-12-06 11:34:24 -05:00
tcp_diag.c tcp_diag: report TCP MD5 signing keys and addresses 2017-09-01 18:38:09 -07:00
tcp_fastopen.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2017-11-04 09:26:51 +09:00
tcp_highspeed.c tcp: consolidate congestion control undo functions 2017-08-06 21:25:10 -07:00
tcp_htcp.c tcp: fix cwnd undo in Reno and HTCP congestion controls 2017-08-06 21:25:10 -07:00
tcp_hybla.c tcp: make undo_cwnd mandatory for congestion modules 2016-11-21 13:20:17 -05:00
tcp_illinois.c tcp: consolidate congestion control undo functions 2017-08-06 21:25:10 -07:00
tcp_input.c tcp: when scheduling TLP, time of RTO should account for current ACK 2017-11-19 12:25:26 +09:00
tcp_ipv4.c tcp: Namespace-ify sysctl_tcp_default_congestion_control 2017-11-15 14:09:52 +09:00
tcp_lp.c tcp: switch TCP TS option (RFC 7323) to 1ms clock 2017-05-17 16:06:01 -04:00
tcp_metrics.c tcp: retire FACK loss detection 2017-11-11 18:53:16 +09:00
tcp_minisocks.c tcp: use sequence distance to detect reordering 2017-11-11 18:53:16 +09:00
tcp_nv.c tcp_nv: use do_div() instead of expensive div64_u64() 2017-11-04 22:34:52 +09:00
tcp_offload.c tcp: gso: avoid refcount_t warning from tcp_gso_segment() 2017-11-10 18:07:15 +09:00
tcp_output.c tcp: when scheduling TLP, time of RTO should account for current ACK 2017-11-19 12:25:26 +09:00
tcp_probe.c tcp: remove redundant argument from tcp_rcv_established() 2017-07-24 17:28:12 -07:00
tcp_rate.c tcp: export do_tcp_sendpages and tcp_rate_check_app_limited functions 2017-06-15 12:12:40 -04:00
tcp_recovery.c tcp: higher throughput under reordering with adaptive RACK reordering wnd 2017-11-05 23:15:42 +09:00
tcp_scalable.c tcp: consolidate congestion control undo functions 2017-08-06 21:25:10 -07:00
tcp_timer.c tcp: tcp_mtu_probing() cleanup 2017-11-05 22:14:23 +09:00
tcp_ulp.c tcp: ulp: avoid module refcnt leak in tcp_set_ulp 2017-08-14 22:17:05 -07:00
tcp_vegas.c tcp: fix under-evaluated ssthresh in TCP Vegas 2017-09-29 06:07:00 +01:00
tcp_vegas.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
tcp_veno.c tcp: consolidate congestion control undo functions 2017-08-06 21:25:10 -07:00
tcp_westwood.c tcp: Revert "tcp: remove CA_ACK_SLOWPATH" 2017-08-30 11:20:08 -07:00
tcp_yeah.c tcp: consolidate congestion control undo functions 2017-08-06 21:25:10 -07:00
tcp.c tcp: use sequence distance to detect reordering 2017-11-11 18:53:16 +09:00
tunnel4.c
udp_diag.c net: ipv6: add second dif to udp socket lookups 2017-08-07 11:39:22 -07:00
udp_impl.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
udp_offload.c gso: fix payload length when gso_size is zero 2017-10-08 10:12:15 -07:00
udp_tunnel.c net: add infrastructure to un-offload UDP tunnel port 2017-07-24 13:52:59 -07:00
udp.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2017-11-15 11:56:19 -08:00
udplite.c udplite: call proper backlog handlers 2016-11-24 15:32:14 -05:00
xfrm4_input.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
xfrm4_mode_beet.c networking: make skb_pull & friends return void pointers 2017-06-16 11:48:39 -04:00
xfrm4_mode_transport.c xfrm: Add encapsulation header offsets while SKB is not encrypted 2017-04-14 10:07:39 +02:00
xfrm4_mode_tunnel.c xfrm: Add encapsulation header offsets while SKB is not encrypted 2017-04-14 10:07:39 +02:00
xfrm4_output.c xfrm: Add an IPsec hardware offloading API 2017-04-14 10:06:10 +02:00
xfrm4_policy.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
xfrm4_protocol.c xfrm: input: constify xfrm_input_afinfo 2017-02-09 10:22:17 +01:00
xfrm4_state.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
xfrm4_tunnel.c