redonkable
/
remarkable-linux
1
0
Fork 0
Code Releases Activity
remarkable-linux/net/ipv4
History
Eric Dumazet 45fa661525 tcp_bbr: better deal with suboptimal GSO 
[ Upstream commit 350c9f484b ]

BBR uses tcp_tso_autosize() in an attempt to probe what would be the
burst sizes and to adjust cwnd in bbr_target_cwnd() with following
gold formula :

/* Allow enough full-sized skbs in flight to utilize end systems. */
cwnd += 3 * bbr->tso_segs_goal;

But GSO can be lacking or be constrained to very small
units (ip link set dev ... gso_max_segs 2)

What we really want is to have enough packets in flight so that both
GSO and GRO are efficient.

So in the case GSO is off or downgraded, we still want to have the same
number of packets in flight as if GSO/TSO was fully operational, so
that GRO can hopefully be working efficiently.

To fix this issue, we make tcp_tso_autosize() unaware of
sk->sk_gso_max_segs

Only tcp_tso_segs() has to enforce the gso_max_segs limit.

Tested:

ethtool -K eth0 tso off gso off
tc qd replace dev eth0 root pfifo_fast

Before patch:
for f in {1..5}; do ./super_netperf 1 -H lpaa24 -- -K bbr; done
    691  (ss -temoi shows cwnd is stuck around 6 )
    667
    651
    631
    517

After patch :
# for f in {1..5}; do ./super_netperf 1 -H lpaa24 -- -K bbr; done
   1733 (ss -temoi shows cwnd is around 386 )
   1778
   1746
   1781
   1718

Fixes: 0f8782ea14 ("tcp_bbr: add BBR congestion control")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
 		2018-03-11 16:21:33 +01:00
..
netfilter netfilter: on sockopt() acquire sock lock only in the required scope 2018-02-25 11:05:43 +01:00
Kconfig tcp: Set DEFAULT_TCP_CONG to bbr if DEFAULT_BBR is set 2016-11-28 12:15:00 -05:00
Makefile tcp_bbr: add BBR congestion control 2016-09-21 00:23:01 -04:00
af_inet.c igmp: Fix regression caused by igmp sysctl namespace code. 2017-08-12 19:31:22 -07:00
ah4.c IPsec: do not ignore crypto err in ah4 input 2017-11-15 15:53:15 +01:00
arp.c ipv4: Make neigh lookup keys for loopback/point-to-point devices be INADDR_ANY 2018-01-31 12:55:55 +01:00
cipso_ipv4.c tcp/dccp: fix ireq->opt races 2017-11-18 11:22:22 +01:00
datagram.c net: Set sk_txhash from a random number 2015-07-29 22:44:04 -07:00
devinet.c ipv4: igmp: guard against silly MTU values 2018-01-02 20:35:10 +01:00
esp4.c esp4: Fix integrity verification when ESN are used 2016-11-30 11:09:39 +01:00
fib_frontend.c ipv4: Fix use-after-free when flushing FIB tables 2018-01-02 20:35:12 +01:00
fib_lookup.h ipv4: consider TOS in fib_select_default 2015-07-24 22:46:11 -07:00
fib_rules.c switchdev: remove FIB offload infrastructure 2016-09-28 04:48:00 -04:00
fib_semantics.c fib_semantics: Don't match route with mismatching tclassid 2018-03-11 16:21:31 +01:00
fib_trie.c net: Improve handling of failures on link and route dumps 2017-06-07 12:07:44 +02:00
fou.c net: add recursion limit to GRO 2016-10-20 14:32:22 -04:00
gre_demux.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2016-06-30 05:03:36 -04:00
gre_offload.c gso: fix payload length when gso_size is zero 2017-11-18 11:22:20 +01:00
icmp.c icmp: don't fail on fragment reassembly time exceeded 2017-12-20 10:07:33 +01:00
igmp.c net: igmp: add a missing rcu locking section 2018-02-13 12:35:56 +01:00
inet_connection_sock.c tcp/dccp: fix other lockdep splats accessing ireq_opt 2017-11-18 11:22:22 +01:00
inet_diag.c net: inet: diag: expose the socket mark to privileged processes. 2016-09-08 16:13:09 -07:00
inet_fragment.c Revert "net: use lib/percpu_counter API for fragmentation mem accounting" 2017-09-20 08:19:55 +02:00
inet_hashtables.c soreuseport: fix initialization race 2017-11-18 11:22:22 +01:00
inet_timewait_sock.c timers, net/ipv4/inet: Initialize connection request timers as pinned 2016-07-07 10:35:06 +02:00
inetpeer.c net: Add helper function to compare inetpeer addresses 2015-08-28 13:32:36 -07:00
ip_forward.c ipv4: allow local fragmentation in ip_finish_output_gso() 2016-11-03 16:10:26 -04:00
ip_fragment.c inet: frag: release spinlock before calling icmp_send() 2017-12-25 14:23:39 +01:00
ip_gre.c net/ip_tunnels: Introduce tunnel_id_to_key32() and key32_to_tunnel_id() 2016-09-10 20:53:55 -07:00
ip_input.c net: VRF: Pass original iif to ip_route_input() 2016-09-16 04:24:07 -04:00
ip_options.c net: ipv4: Convert IP network timestamps to be y2038 safe 2016-03-01 17:18:44 -05:00
ip_output.c udp: consistently apply ufo or fragmentation 2017-08-12 19:31:22 -07:00
ip_sockglue.c netfilter: drop outermost socket lock in getsockopt() 2018-02-28 10:18:30 +01:00
ip_tunnel.c ipv4: igmp: guard against silly MTU values 2018-01-02 20:35:10 +01:00
ip_tunnel_core.c net: Specify the owning module for lwtunnel ops 2017-02-04 09:47:11 +01:00
ip_vti.c vti: fix use after free in vti_tunnel_xmit/vti6_tnl_xmit 2017-10-12 11:51:22 +02:00
ipcomp.c ipv4: coding style: comparison for equality with NULL 2015-04-03 12:11:15 -04:00
ipconfig.c net: ipconfig: fix ic_close_devs() use-after-free 2017-12-25 14:23:42 +01:00
ipip.c ipip: only increase err_count for some certain type icmp in ipip_err 2017-11-18 11:22:23 +01:00
ipmr.c ipv4: allow local fragmentation in ip_finish_output_gso() 2016-11-03 16:10:26 -04:00
netfilter.c netfilter: use skb_to_full_sk in ip_route_me_harder 2017-07-05 14:40:28 +02:00
ping.c ping: implement proper locking 2017-05-03 08:36:34 -07:00
proc.c net: Suppress the "Comparison to NULL could be written" warnings 2016-09-30 01:50:45 -04:00
protocol.c net: Export inet_offloads and inet6_offloads 2014-09-19 17:15:31 -04:00
raw.c net: ipv4: fix for a race condition in raw_sendmsg 2018-01-02 20:35:12 +01:00
route.c net: ipv4: don't allow setting net.ipv4.route.min_pmtu below 68 2018-03-11 16:21:31 +01:00
syncookies.c tcp/dccp: fix ireq->opt races 2017-11-18 11:22:22 +01:00
sysctl_net_ipv4.c ipv4: make tcp_notsent_lowat sysctl knob behave as true unsigned int 2017-08-11 08:49:35 -07:00
tcp.c tcp: release sk_frag.page in tcp_disconnect 2018-02-13 12:35:56 +01:00
tcp_bbr.c tcp_bbr: fix pacing_gain to always be unity when using lt_bw 2018-02-13 12:35:56 +01:00
tcp_bic.c tcp: replace cnt & rtt with struct in pkts_acked() 2016-05-11 14:43:19 -04:00
tcp_cdg.c tcp: cdg: rename struct minmax in tcp_cdg.c to avoid a naming conflict 2016-09-21 00:22:59 -04:00
tcp_cong.c tcp: disallow cwnd undo when switching congestion control 2017-06-14 15:05:52 +02:00
tcp_cubic.c tcp: replace cnt & rtt with struct in pkts_acked() 2016-05-11 14:43:19 -04:00
tcp_dctcp.c dctcp: avoid bogus doubling of cwnd after loss 2016-10-31 15:16:28 -04:00
tcp_diag.c net: diag: Fix refcnt leak in error path destroying socket 2016-08-23 23:11:36 -07:00
tcp_fastopen.c tcp: initialize max window for a new fastopen socket 2017-02-04 09:47:10 +01:00
tcp_highspeed.c tcp: add tcp_in_slow_start helper 2015-07-09 14:22:52 -07:00
tcp_htcp.c tcp: replace cnt & rtt with struct in pkts_acked() 2016-05-11 14:43:19 -04:00
tcp_hybla.c tcp: do not slow start when cwnd equals ssthresh 2015-07-09 14:22:52 -07:00
tcp_illinois.c tcp: replace cnt & rtt with struct in pkts_acked() 2016-05-11 14:43:19 -04:00
tcp_input.c tcp: invalidate rate samples during SACK reneging 2018-01-02 20:35:13 +01:00
tcp_ipv4.c tcp md5sig: Use skb's saddr when replying to an incoming segment 2018-01-02 20:35:11 +01:00
tcp_lp.c tcp: fix wraparound issue in tcp_lp 2017-05-14 14:00:21 +02:00
tcp_metrics.c tcp: make nla_policy const 2016-09-01 14:09:01 -07:00
tcp_minisocks.c tcp/dccp: block bh before arming time_wait timer 2017-12-16 16:25:46 +01:00
tcp_nv.c tcp_nv: fix division by zero in tcpnv_acked() 2017-11-24 08:33:40 +01:00
tcp_offload.c gso: validate gso_type in GSO handlers 2018-01-31 12:55:55 +01:00
tcp_output.c tcp_bbr: better deal with suboptimal GSO 2018-03-11 16:21:33 +01:00
tcp_probe.c tcp: tcp_probe: use spin_lock_bh() 2017-06-17 06:41:49 +02:00
tcp_rate.c tcp: invalidate rate samples during SACK reneging 2018-01-02 20:35:13 +01:00
tcp_recovery.c tcp: do not assume TCP code is non preemptible 2016-05-02 17:02:25 -04:00
tcp_scalable.c tcp: add tcp_in_slow_start helper 2015-07-09 14:22:52 -07:00
tcp_timer.c net: tcp: close sock if net namespace is exiting 2018-01-31 12:55:54 +01:00
tcp_vegas.c tcp: fix under-evaluated ssthresh in TCP Vegas 2017-12-25 14:23:45 +01:00
tcp_vegas.h tcp: replace cnt & rtt with struct in pkts_acked() 2016-05-11 14:43:19 -04:00
tcp_veno.c tcp: replace cnt & rtt with struct in pkts_acked() 2016-05-11 14:43:19 -04:00
tcp_westwood.c tcp: replace cnt & rtt with struct in pkts_acked() 2016-05-11 14:43:19 -04:00
tcp_yeah.c tcp: cwnd does not increase in TCP YeAH 2016-09-08 17:16:12 -07:00
tunnel4.c tunnels: correct conditional build of MPLS and IPv6 2016-07-11 13:27:06 -07:00
udp.c udplite: fix partial checksum initialization 2018-03-11 16:21:32 +01:00
udp_diag.c net: inet: diag: expose the socket mark to privileged processes. 2016-09-08 16:13:09 -07:00
udp_impl.h udplite: call proper backlog handlers 2016-11-24 15:32:14 -05:00
udp_offload.c gso: validate gso_type in GSO handlers 2018-01-31 12:55:55 +01:00
udp_tunnel.c net: Remove deprecated tunnel specific UDP offload functions 2016-06-17 20:23:32 -07:00
udplite.c udplite: call proper backlog handlers 2016-11-24 15:32:14 -05:00
xfrm4_input.c netfilter: Pass net into okfn 2015-09-17 17:18:37 -07:00
xfrm4_mode_beet.c
…
xfrm4_mode_transport.c
…
xfrm4_mode_tunnel.c ipv4: hash net ptr into fragmentation bucket selection 2015-03-25 14:07:04 -04:00
xfrm4_output.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2015-10-24 06:54:12 -07:00
xfrm4_policy.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2016-09-12 15:52:44 -07:00
xfrm4_protocol.c xfrm4: Remove duplicate semicolon 2014-06-30 07:49:47 +02:00
xfrm4_state.c
…
xfrm4_tunnel.c
…