OpenCloudOS-Kernel/net
Soheil Hassas Yeganeh d7722e8570 tcp: track application-limited rate samples
This commit adds code to track whether the delivery rate represented
by each rate_sample was limited by the application.

Upon each transmit, we store in the is_app_limited field in the skb a
boolean bit indicating whether there is a known "bubble in the pipe":
a point in the rate sample interval where the sender was
application-limited, and did not transmit even though the cwnd and
pacing rate allowed it.

This logic marks the flow app-limited on a write if *all* of the
following are true:

  1) There is less than 1 MSS of unsent data in the write queue
     available to transmit.

  2) There is no packet in the sender's queues (e.g. in fq or the NIC
     tx queue).

  3) The connection is not limited by cwnd.

  4) There are no lost packets to retransmit.

The tcp_rate_check_app_limited() code in tcp_rate.c determines whether
the connection is application-limited at the moment. If the flow is
application-limited, it sets the tp->app_limited field. If the flow is
application-limited then that means there is effectively a "bubble" of
silence in the pipe now, and this silence will be reflected in a lower
bandwidth sample for any rate samples from now until we get an ACK
indicating this bubble has exited the pipe: specifically, until we get
an ACK for the next packet we transmit.

When we send every skb we record in scb->tx.is_app_limited whether the
resulting rate sample will be application-limited.

The code in tcp_rate_gen() checks to see when it is safe to mark all
known application-limited bubbles of silence as having exited the
pipe. It does this by checking to see when the delivered count moves
past the tp->app_limited marker. At this point it zeroes the
tp->app_limited marker, as all known bubbles are out of the pipe.

We make room for the tx.is_app_limited bit in the skb by borrowing a
bit from the in_flight field used by NV to record the number of bytes
in flight. The receive window in the TCP header is 16 bits, and the
max receive window scaling shift factor is 14 (RFC 1323). So the max
receive window offered by the TCP protocol is 2^(16+14) = 2^30. So we
only need 30 bits for the tx.in_flight used by NV.

Signed-off-by: Van Jacobson <vanj@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Nandita Dukkipati <nanditad@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-09-21 00:23:00 -04:00
..
6lowpan 6lowpan: ndisc: no overreact if no short address is available 2016-09-19 20:19:34 +02:00
9p 9p/trans_virtio: use kvfree() for iov_iter_get_pages_alloc() 2016-08-09 13:42:36 +03:00
802
8021q net: remove type_check from dev_get_nest_level() 2016-08-13 15:15:54 -07:00
appletalk appletalk: use IS_ENABLED() instead of checking for built-in or module 2016-09-10 21:19:10 -07:00
atm lec: use IS_ENABLED() instead of checking for built-in or module 2016-09-10 21:19:10 -07:00
ax25 AX.25: Close socket connection on session completion 2016-06-18 20:55:34 -07:00
batman-adv batman: make netlink attributes const 2016-09-01 14:09:00 -07:00
bluetooth Bluetooth: Set appearance only for LE capable controllers 2016-09-19 21:48:22 +03:00
bridge net: bridge: add helper to call /sbin/bridge-stp 2016-09-13 11:21:31 -04:00
caif caif: Remove unneeded header file 2016-06-28 05:26:14 -04:00
can can: only call can_stat_update with procfs 2016-06-23 11:23:49 +02:00
ceph libceph: using kfree_rcu() to simplify the code 2016-08-08 21:41:42 +02:00
core bpf: direct packet write and access for helpers for clsact progs 2016-09-20 23:32:11 -04:00
dcb
dccp Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security 2016-07-29 17:38:46 -07:00
decnet net: fix decnet rtnexthop parsing 2016-07-05 14:08:47 -07:00
dns_resolver KEYS: Add a facility to restrict new links into a keyring 2016-04-11 22:37:37 +01:00
dsa net-next: dsa: make the set_addr() operation optional 2016-09-20 04:47:44 -04:00
ethernet eth: Pull header from first fragment via eth_get_headlen 2016-02-24 13:58:05 -05:00
hsr net/hsr: Use setup_timer and mod_timer. 2016-05-16 14:00:43 -04:00
ieee802154 ieee802154: 6lowpan: fix intra pan id check 2016-07-08 13:23:12 +02:00
ipv4 tcp: track application-limited rate samples 2016-09-21 00:23:00 -04:00
ipv6 gso: Support partial splitting at the frag_list pointer 2016-09-19 20:59:34 -04:00
ipx
irda net/irda: remove pointless assignment/check 2016-08-19 18:07:24 -07:00
iucv Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security 2016-07-29 17:38:46 -07:00
kcm Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2016-09-12 15:52:44 -07:00
key
l2tp l2tp: constify net_device_ops structures 2016-09-17 10:07:23 -04:00
l3mdev net: ipv6: Remove l3mdev_get_saddr6 2016-09-10 23:12:53 -07:00
lapb net/lapb: tuse %*ph to dump buffers 2016-05-29 22:33:25 -07:00
llc llc: switch type to bool as the timeout is only tested versus 0 2016-09-17 10:05:05 -04:00
mac80211 mac80211: Use rhltable instead of rhashtable 2016-09-20 04:43:36 -04:00
mac802154 mac802154: use rate limited warnings for malformed frames 2016-09-19 20:19:34 +02:00
mpls mpls: get rid of trivial returns 2016-09-01 10:13:15 -07:00
ncsi net/ncsi: avoid maybe-uninitialized warning 2016-07-25 10:32:59 -07:00
netfilter net: Add _nf_(un)register_hooks symbols 2016-09-19 01:25:22 -04:00
netlabel netlabel: Implement CALIPSO config functions for SMACK. 2016-06-27 15:06:18 -04:00
netlink netlink: don't forget to release a rhashtable_iter structure 2016-09-07 17:29:38 -07:00
netrom
nfc NFC: digital: Fix RTOX supervisor PDU handling 2016-07-11 02:02:03 +02:00
openvswitch openvswitch: avoid resetting flow key while installing new flow. 2016-09-20 22:54:35 -04:00
packet Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2016-07-24 00:53:32 -04:00
phonet
qrtr Merge tag 'qcom-soc-for-4.7-2' into net-next 2016-05-17 14:11:19 -04:00
rds RDS: add __printf format attribute to error reporting functions 2016-08-08 16:16:21 -07:00
rfkill rfkill: Use switch to demux userspace operations 2016-04-05 10:48:53 +02:00
rose rose: limit sk_filter trim to payload 2016-07-13 11:53:40 -07:00
rxrpc rxrpc: Add config to inject packet loss 2016-09-17 11:24:04 +01:00
sched net_sched: sch_fq: add low_rate_threshold parameter 2016-09-21 00:23:00 -04:00
sctp sctp: Remove some redundant code 2016-09-19 01:34:01 -04:00
strparser kcm: Remove TCP specific references from kcm and strparser 2016-08-28 23:32:41 -04:00
sunrpc SUNRPC: Silence WARN_ON when NFSv4.1 over RDMA is in use 2016-08-24 22:32:55 -04:00
switchdev rtnetlink: fdb dump: optimize by saving last interface markers 2016-09-01 16:56:15 -07:00
tipc tipc: fix possible memory leak in tipc_udp_enable() 2016-09-13 11:28:32 -04:00
unix af_unix: split 'u->readlock' into two: 'iolock' and 'bindlock' 2016-09-04 13:29:29 -07:00
vmw_vsock vhost/vsock: drop space available check for TX vq 2016-08-15 05:05:21 +03:00
wimax
wireless This time we have various things - all across the board: 2016-09-18 22:29:08 -04:00
x25 net: x25: remove null checks on arrays calling_ae and called_ae 2016-09-09 18:13:30 -07:00
xfrm Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2016-09-12 15:52:44 -07:00
Kconfig strparser: Stream parser for messages 2016-08-17 19:36:23 -04:00
Makefile strparser: Stream parser for messages 2016-08-17 19:36:23 -04:00
compat.c packet: compat support for sock_fprog 2016-06-09 23:41:03 -07:00
socket.c fs: poll/select/recvmmsg: use timespec64 for timeout events 2016-05-19 19:12:14 -07:00
sysctl_net.c net: make net namespace sysctls belong to container's owner 2016-08-14 21:08:58 -07:00