[mirror_ubuntu-bionic-kernel.git] / Documentation / networking / tcp.txt

TCP protocol
============

Last updated: 3 June 2017

Contents
========

- Congestion control
- How the new TCP output machine [nyi] works

Congestion control
==================

The following variables are used in the tcp_sock for congestion control:
snd_cwnd		The size of the congestion window
snd_ssthresh		Slow start threshold. We are in slow start if
			snd_cwnd is less than this.
snd_cwnd_cnt		A counter used to slow down the rate of increase
			once we exceed slow start threshold.
snd_cwnd_clamp		This is the maximum size that snd_cwnd can grow to.
snd_cwnd_stamp		Timestamp for when congestion window last validated.
snd_cwnd_used		Used as a highwater mark for how much of the
			congestion window is in use. It is used to adjust
			snd_cwnd down when the link is limited by the
			application rather than the network.

As of 2.6.13, Linux supports pluggable congestion control algorithms.
A congestion control mechanism can be registered through functions in
tcp_cong.c. The functions used by the congestion control mechanism are
registered via passing a tcp_congestion_ops struct to
tcp_register_congestion_control. As a minimum, the congestion control
mechanism must provide a valid name and must implement either ssthresh,
cong_avoid and undo_cwnd hooks or the "omnipotent" cong_control hook.

Private data for a congestion control mechanism is stored in tp->ca_priv.
tcp_ca(tp) returns a pointer to this space.  This is preallocated space - it
is important to check the size of your private data will fit this space, or
alternatively, space could be allocated elsewhere and a pointer to it could
be stored here.

There are three kinds of congestion control algorithms currently: The
simplest ones are derived from TCP reno (highspeed, scalable) and just
provide an alternative congestion window calculation. More complex
ones like BIC try to look at other events to provide better
heuristics.  There are also round trip time based algorithms like
Vegas and Westwood+.

Good TCP congestion control is a complex problem because the algorithm
needs to maintain fairness and performance. Please review current
research and RFC's before developing new modules.

The default congestion control mechanism is chosen based on the
DEFAULT_TCP_CONG Kconfig parameter. If you really want a particular default
value then you can set it using sysctl net.ipv4.tcp_congestion_control. The
module will be autoloaded if needed and you will get the expected protocol. If
you ask for an unknown congestion method, then the sysctl attempt will fail.

If you remove a TCP congestion control module, then you will get the next
available one. Since reno cannot be built as a module, and cannot be
removed, it will always be available.

How the new TCP output machine [nyi] works.
===========================================

Data is kept on a single queue. The skb->users flag tells us if the frame is
one that has been queued already. To add a frame we throw it on the end. Ack
walks down the list from the start.

We keep a set of control flags


	sk->tcp_pend_event

		TCP_PEND_ACK			Ack needed
		TCP_ACK_NOW			Needed now
		TCP_WINDOW			Window update check
		TCP_WINZERO			Zero probing


	sk->transmit_queue		The transmission frame begin
	sk->transmit_new		First new frame pointer
	sk->transmit_end		Where to add frames

	sk->tcp_last_tx_ack		Last ack seen
	sk->tcp_dup_ack			Dup ack count for fast retransmit


Frames are queued for output by tcp_write. We do our best to send the frames
off immediately if possible, but otherwise queue and compute the body
checksum in the copy. 

When a write is done we try to clear any pending events and piggy back them.
If the window is full we queue full sized frames. On the first timeout in
zero window we split this.

On a timer we walk the retransmit list to send any retransmits, update the
backoff timers etc. A change of route table stamp causes a change of header
and recompute. We add any new tcp level headers and refinish the checksum
before sending.
Commit	Line	Data
9d7bcfc6 SH	1	TCP protocol
	2	============
	3
1e0ce2a1	4	Last updated: 3 June 2017
9d7bcfc6 SH	5
	6	Contents
	7	========
	8
	9	- Congestion control
	10	- How the new TCP output machine [nyi] works
	11
	12	Congestion control
	13	==================
	14
	15	The following variables are used in the tcp_sock for congestion control:
	16	snd_cwnd The size of the congestion window
	17	snd_ssthresh Slow start threshold. We are in slow start if
	18	snd_cwnd is less than this.
	19	snd_cwnd_cnt A counter used to slow down the rate of increase
	20	once we exceed slow start threshold.
	21	snd_cwnd_clamp This is the maximum size that snd_cwnd can grow to.
	22	snd_cwnd_stamp Timestamp for when congestion window last validated.
	23	snd_cwnd_used Used as a highwater mark for how much of the
	24	congestion window is in use. It is used to adjust
	25	snd_cwnd down when the link is limited by the
	26	application rather than the network.
	27
	28	As of 2.6.13, Linux supports pluggable congestion control algorithms.
	29	A congestion control mechanism can be registered through functions in
	30	tcp_cong.c. The functions used by the congestion control mechanism are
	31	registered via passing a tcp_congestion_ops struct to
1e0ce2a1 AS	32	tcp_register_congestion_control. As a minimum, the congestion control
	33	mechanism must provide a valid name and must implement either ssthresh,
	34	cong_avoid and undo_cwnd hooks or the "omnipotent" cong_control hook.
1da177e4	35
9d7bcfc6 SH	36	Private data for a congestion control mechanism is stored in tp->ca_priv.
	37	tcp_ca(tp) returns a pointer to this space. This is preallocated space - it
	38	is important to check the size of your private data will fit this space, or
1e0ce2a1	39	alternatively, space could be allocated elsewhere and a pointer to it could
9d7bcfc6 SH	40	be stored here.
	41
	42	There are three kinds of congestion control algorithms currently: The
	43	simplest ones are derived from TCP reno (highspeed, scalable) and just
1e0ce2a1	44	provide an alternative congestion window calculation. More complex
9d7bcfc6 SH	45	ones like BIC try to look at other events to provide better
	46	heuristics. There are also round trip time based algorithms like
	47	Vegas and Westwood+.
	48
	49	Good TCP congestion control is a complex problem because the algorithm
	50	needs to maintain fairness and performance. Please review current
	51	research and RFC's before developing new modules.
	52
1e0ce2a1 AS	53	The default congestion control mechanism is chosen based on the
	54	DEFAULT_TCP_CONG Kconfig parameter. If you really want a particular default
	55	value then you can set it using sysctl net.ipv4.tcp_congestion_control. The
	56	module will be autoloaded if needed and you will get the expected protocol. If
	57	you ask for an unknown congestion method, then the sysctl attempt will fail.
9d7bcfc6	58
1e0ce2a1	59	If you remove a TCP congestion control module, then you will get the next
84eb8d06	60	available one. Since reno cannot be built as a module, and cannot be
1e0ce2a1	61	removed, it will always be available.
9d7bcfc6 SH	62
	63	How the new TCP output machine [nyi] works.
	64	===========================================
1da177e4 LT	65
	66	Data is kept on a single queue. The skb->users flag tells us if the frame is
	67	one that has been queued already. To add a frame we throw it on the end. Ack
	68	walks down the list from the start.
	69
	70	We keep a set of control flags
	71
	72
	73	sk->tcp_pend_event
	74
	75	TCP_PEND_ACK Ack needed
	76	TCP_ACK_NOW Needed now
	77	TCP_WINDOW Window update check
	78	TCP_WINZERO Zero probing
	79
	80
	81	sk->transmit_queue The transmission frame begin
	82	sk->transmit_new First new frame pointer
	83	sk->transmit_end Where to add frames
	84
	85	sk->tcp_last_tx_ack Last ack seen
	86	sk->tcp_dup_ack Dup ack count for fast retransmit
	87
	88
	89	Frames are queued for output by tcp_write. We do our best to send the frames
	90	off immediately if possible, but otherwise queue and compute the body
	91	checksum in the copy.
	92
	93	When a write is done we try to clear any pending events and piggy back them.
	94	If the window is full we queue full sized frames. On the first timeout in
	95	zero window we split this.
	96
	97	On a timer we walk the retransmit list to send any retransmits, update the
	98	backoff timers etc. A change of route table stamp causes a change of header
	99	and recompute. We add any new tcp level headers and refinish the checksum
	100	before sending.
	101