draft-ietf-quic-recovery-16.txt   draft-ietf-quic-recovery-latest.txt 
QUIC Working Group J. Iyengar, Ed. QUIC Working Group J. Iyengar, Ed.
Internet-Draft Fastly Internet-Draft Fastly
Intended status: Standards Track I. Swett, Ed. Intended status: Standards Track I. Swett, Ed.
Expires: April 26, 2019 Google Expires: May 21, 2019 Google
October 23, 2018 November 17, 2018
QUIC Loss Detection and Congestion Control QUIC Loss Detection and Congestion Control
draft-ietf-quic-recovery-16 draft-ietf-quic-recovery-latest
Abstract Abstract
This document describes loss detection and congestion control This document describes loss detection and congestion control
mechanisms for QUIC. mechanisms for QUIC.
Note to Readers Note to Readers
Discussion of this draft takes place on the QUIC working group Discussion of this draft takes place on the QUIC working group
mailing list (quic@ietf.org), which is archived at mailing list (quic@ietf.org), which is archived at
skipping to change at page 1, line 42 skipping to change at page 1, line 42
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet- working documents as Internet-Drafts. The list of current Internet-
Drafts is at https://datatracker.ietf.org/drafts/current/. Drafts is at https://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress." material or to cite them other than as "work in progress."
This Internet-Draft will expire on April 26, 2019. This Internet-Draft will expire on May 21, 2019.
Copyright Notice Copyright Notice
Copyright (c) 2018 IETF Trust and the persons identified as the Copyright (c) 2018 IETF Trust and the persons identified as the
document authors. All rights reserved. document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents Provisions Relating to IETF Documents
(https://trustee.ietf.org/license-info) in effect on the date of (https://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents publication of this document. Please review these documents
skipping to change at page 2, line 29 skipping to change at page 2, line 29
3. Design of the QUIC Transmission Machinery . . . . . . . . . . 4 3. Design of the QUIC Transmission Machinery . . . . . . . . . . 4
3.1. Relevant Differences Between QUIC and TCP . . . . . . . . 5 3.1. Relevant Differences Between QUIC and TCP . . . . . . . . 5
3.1.1. Separate Packet Number Spaces . . . . . . . . . . . . 5 3.1.1. Separate Packet Number Spaces . . . . . . . . . . . . 5
3.1.2. Monotonically Increasing Packet Numbers . . . . . . . 6 3.1.2. Monotonically Increasing Packet Numbers . . . . . . . 6
3.1.3. No Reneging . . . . . . . . . . . . . . . . . . . . . 6 3.1.3. No Reneging . . . . . . . . . . . . . . . . . . . . . 6
3.1.4. More ACK Ranges . . . . . . . . . . . . . . . . . . . 6 3.1.4. More ACK Ranges . . . . . . . . . . . . . . . . . . . 6
3.1.5. Explicit Correction For Delayed ACKs . . . . . . . . 6 3.1.5. Explicit Correction For Delayed ACKs . . . . . . . . 6
4. Loss Detection . . . . . . . . . . . . . . . . . . . . . . . 7 4. Loss Detection . . . . . . . . . . . . . . . . . . . . . . . 7
4.1. Computing the RTT estimate . . . . . . . . . . . . . . . 7 4.1. Computing the RTT estimate . . . . . . . . . . . . . . . 7
4.2. Ack-based Detection . . . . . . . . . . . . . . . . . . . 7 4.2. Ack-based Detection . . . . . . . . . . . . . . . . . . . 7
4.2.1. Fast Retransmit . . . . . . . . . . . . . . . . . . . 7 4.2.1. Fast Retransmit . . . . . . . . . . . . . . . . . . . 8
4.2.2. Early Retransmit . . . . . . . . . . . . . . . . . . 8 4.2.2. Early Retransmit . . . . . . . . . . . . . . . . . . 8
4.3. Timer-based Detection . . . . . . . . . . . . . . . . . . 9 4.3. Timeout Loss Detection . . . . . . . . . . . . . . . . . 9
4.3.1. Crypto Retransmission Timeout . . . . . . . . . . . . 9 4.3.1. Crypto Retransmission Timeout . . . . . . . . . . . . 9
4.3.2. Tail Loss Probe . . . . . . . . . . . . . . . . . . . 10 4.3.2. Tail Loss Probe . . . . . . . . . . . . . . . . . . . 11
4.3.3. Retransmission Timeout . . . . . . . . . . . . . . . 11 4.3.3. Retransmission Timeout . . . . . . . . . . . . . . . 12
4.4. Generating Acknowledgements . . . . . . . . . . . . . . . 12 4.4. Generating Acknowledgements . . . . . . . . . . . . . . . 13
4.4.1. Crypto Handshake Data . . . . . . . . . . . . . . . . 13 4.4.1. Crypto Handshake Data . . . . . . . . . . . . . . . . 13
4.4.2. ACK Ranges . . . . . . . . . . . . . . . . . . . . . 13 4.4.2. ACK Ranges . . . . . . . . . . . . . . . . . . . . . 14
4.4.3. Receiver Tracking of ACK Frames . . . . . . . . . . . 13 4.4.3. Receiver Tracking of ACK Frames . . . . . . . . . . . 14
4.5. Pseudocode . . . . . . . . . . . . . . . . . . . . . . . 14 4.5. Tracking Sent Packets . . . . . . . . . . . . . . . . . . 14
4.5.1. Constants of interest . . . . . . . . . . . . . . . . 14 4.5.1. Sent Packet Fields . . . . . . . . . . . . . . . . . 15
4.5.2. Variables of interest . . . . . . . . . . . . . . . . 14 4.6. Pseudocode . . . . . . . . . . . . . . . . . . . . . . . 15
4.5.3. Initialization . . . . . . . . . . . . . . . . . . . 16 4.6.1. Constants of interest . . . . . . . . . . . . . . . . 15
4.5.4. On Sending a Packet . . . . . . . . . . . . . . . . . 16 4.6.2. Variables of interest . . . . . . . . . . . . . . . . 16
4.5.5. On Receiving an Acknowledgment . . . . . . . . . . . 17 4.6.3. Initialization . . . . . . . . . . . . . . . . . . . 17
4.5.6. On Packet Acknowledgment . . . . . . . . . . . . . . 19 4.6.4. On Sending a Packet . . . . . . . . . . . . . . . . . 18
4.5.7. Setting the Loss Detection Timer . . . . . . . . . . 19 4.6.5. On Receiving an Acknowledgment . . . . . . . . . . . 18
4.5.8. On Timeout . . . . . . . . . . . . . . . . . . . . . 20 4.6.6. On Packet Acknowledgment . . . . . . . . . . . . . . 20
4.5.9. Detecting Lost Packets . . . . . . . . . . . . . . . 21 4.6.7. Setting the Loss Detection Timer . . . . . . . . . . 20
4.6. Discussion . . . . . . . . . . . . . . . . . . . . . . . 22 4.6.8. On Timeout . . . . . . . . . . . . . . . . . . . . . 21
5. Congestion Control . . . . . . . . . . . . . . . . . . . . . 22 4.6.9. Detecting Lost Packets . . . . . . . . . . . . . . . 22
5.1. Explicit Congestion Notification . . . . . . . . . . . . 23 4.7. Discussion . . . . . . . . . . . . . . . . . . . . . . . 23
5.2. Slow Start . . . . . . . . . . . . . . . . . . . . . . . 23 5. Congestion Control . . . . . . . . . . . . . . . . . . . . . 23
5.3. Congestion Avoidance . . . . . . . . . . . . . . . . . . 23 5.1. Explicit Congestion Notification . . . . . . . . . . . . 24
5.4. Recovery Period . . . . . . . . . . . . . . . . . . . . . 23 5.2. Slow Start . . . . . . . . . . . . . . . . . . . . . . . 24
5.5. Tail Loss Probe . . . . . . . . . . . . . . . . . . . . . 24 5.3. Congestion Avoidance . . . . . . . . . . . . . . . . . . 24
5.6. Retransmission Timeout . . . . . . . . . . . . . . . . . 24 5.4. Recovery Period . . . . . . . . . . . . . . . . . . . . . 24
5.7. Pacing . . . . . . . . . . . . . . . . . . . . . . . . . 24 5.5. Tail Loss Probe . . . . . . . . . . . . . . . . . . . . . 25
5.8. Pseudocode . . . . . . . . . . . . . . . . . . . . . . . 25 5.6. Retransmission Timeout . . . . . . . . . . . . . . . . . 25
5.8.1. Constants of interest . . . . . . . . . . . . . . . . 25 5.7. Pacing . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.8.2. Variables of interest . . . . . . . . . . . . . . . . 25 5.8. Pseudocode . . . . . . . . . . . . . . . . . . . . . . . 26
5.8.3. Initialization . . . . . . . . . . . . . . . . . . . 26 5.8.1. Constants of interest . . . . . . . . . . . . . . . . 26
5.8.4. On Packet Sent . . . . . . . . . . . . . . . . . . . 26 5.8.2. Variables of interest . . . . . . . . . . . . . . . . 26
5.8.5. On Packet Acknowledgement . . . . . . . . . . . . . . 26 5.8.3. Initialization . . . . . . . . . . . . . . . . . . . 27
5.8.6. On New Congestion Event . . . . . . . . . . . . . . . 27 5.8.4. On Packet Sent . . . . . . . . . . . . . . . . . . . 27
5.8.7. Process ECN Information . . . . . . . . . . . . . . . 27 5.8.5. On Packet Acknowledgement . . . . . . . . . . . . . . 27
5.8.8. On Packets Lost . . . . . . . . . . . . . . . . . . . 27 5.8.6. On New Congestion Event . . . . . . . . . . . . . . . 28
5.8.9. On Retransmission Timeout Verified . . . . . . . . . 28 5.8.7. Process ECN Information . . . . . . . . . . . . . . . 28
6. Security Considerations . . . . . . . . . . . . . . . . . . . 28 5.8.8. On Packets Lost . . . . . . . . . . . . . . . . . . . 28
6.1. Congestion Signals . . . . . . . . . . . . . . . . . . . 28 5.8.9. On Retransmission Timeout Verified . . . . . . . . . 29
6.2. Traffic Analysis . . . . . . . . . . . . . . . . . . . . 28 6. Security Considerations . . . . . . . . . . . . . . . . . . . 29
6.3. Misreporting ECN Markings . . . . . . . . . . . . . . . . 28 6.1. Congestion Signals . . . . . . . . . . . . . . . . . . . 29
7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 29 6.2. Traffic Analysis . . . . . . . . . . . . . . . . . . . . 29
8. References . . . . . . . . . . . . . . . . . . . . . . . . . 29 6.3. Misreporting ECN Markings . . . . . . . . . . . . . . . . 29
8.1. Normative References . . . . . . . . . . . . . . . . . . 29 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 30
8.2. Informative References . . . . . . . . . . . . . . . . . 29 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 30
8.3. URIs . . . . . . . . . . . . . . . . . . . . . . . . . . 30 8.1. Normative References . . . . . . . . . . . . . . . . . . 30
Appendix A. Change Log . . . . . . . . . . . . . . . . . . . . . 31 8.2. Informative References . . . . . . . . . . . . . . . . . 30
A.1. Since draft-ietf-quic-recovery-14 . . . . . . . . . . . . 31 8.3. URIs . . . . . . . . . . . . . . . . . . . . . . . . . . 31
A.2. Since draft-ietf-quic-recovery-13 . . . . . . . . . . . . 31 Appendix A. Change Log . . . . . . . . . . . . . . . . . . . . . 32
A.3. Since draft-ietf-quic-recovery-12 . . . . . . . . . . . . 31 A.1. Since draft-ietf-quic-recovery-14 . . . . . . . . . . . . 32
A.4. Since draft-ietf-quic-recovery-11 . . . . . . . . . . . . 31 A.2. Since draft-ietf-quic-recovery-13 . . . . . . . . . . . . 32
A.5. Since draft-ietf-quic-recovery-10 . . . . . . . . . . . . 31 A.3. Since draft-ietf-quic-recovery-12 . . . . . . . . . . . . 32
A.6. Since draft-ietf-quic-recovery-09 . . . . . . . . . . . . 32 A.4. Since draft-ietf-quic-recovery-11 . . . . . . . . . . . . 32
A.7. Since draft-ietf-quic-recovery-08 . . . . . . . . . . . . 32 A.5. Since draft-ietf-quic-recovery-10 . . . . . . . . . . . . 32
A.8. Since draft-ietf-quic-recovery-07 . . . . . . . . . . . . 32 A.6. Since draft-ietf-quic-recovery-09 . . . . . . . . . . . . 33
A.9. Since draft-ietf-quic-recovery-06 . . . . . . . . . . . . 32 A.7. Since draft-ietf-quic-recovery-08 . . . . . . . . . . . . 33
A.10. Since draft-ietf-quic-recovery-05 . . . . . . . . . . . . 32 A.8. Since draft-ietf-quic-recovery-07 . . . . . . . . . . . . 33
A.11. Since draft-ietf-quic-recovery-04 . . . . . . . . . . . . 32 A.9. Since draft-ietf-quic-recovery-06 . . . . . . . . . . . . 33
A.12. Since draft-ietf-quic-recovery-03 . . . . . . . . . . . . 32 A.10. Since draft-ietf-quic-recovery-05 . . . . . . . . . . . . 33
A.13. Since draft-ietf-quic-recovery-02 . . . . . . . . . . . . 32 A.11. Since draft-ietf-quic-recovery-04 . . . . . . . . . . . . 33
A.14. Since draft-ietf-quic-recovery-01 . . . . . . . . . . . . 33 A.12. Since draft-ietf-quic-recovery-03 . . . . . . . . . . . . 33
A.15. Since draft-ietf-quic-recovery-00 . . . . . . . . . . . . 33 A.13. Since draft-ietf-quic-recovery-02 . . . . . . . . . . . . 33
A.16. Since draft-iyengar-quic-loss-recovery-01 . . . . . . . . 33 A.14. Since draft-ietf-quic-recovery-01 . . . . . . . . . . . . 34
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 33 A.15. Since draft-ietf-quic-recovery-00 . . . . . . . . . . . . 34
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 33 A.16. Since draft-iyengar-quic-loss-recovery-01 . . . . . . . . 34
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 34
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 34
1. Introduction 1. Introduction
QUIC is a new multiplexed and secure transport atop UDP. QUIC builds QUIC is a new multiplexed and secure transport atop UDP. QUIC builds
on decades of transport and security experience, and implements on decades of transport and security experience, and implements
mechanisms that make it attractive as a modern general-purpose mechanisms that make it attractive as a modern general-purpose
transport. The QUIC protocol is described in [QUIC-TRANSPORT]. transport. The QUIC protocol is described in [QUIC-TRANSPORT].
QUIC implements the spirit of known TCP loss recovery mechanisms, QUIC implements the spirit of known TCP loss recovery mechanisms,
described in RFCs, various Internet-drafts, and also those prevalent described in RFCs, various Internet-drafts, and also those prevalent
skipping to change at page 4, line 29 skipping to change at page 4, line 29
2. Conventions and Definitions 2. Conventions and Definitions
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
"OPTIONAL" in this document are to be interpreted as described in BCP "OPTIONAL" in this document are to be interpreted as described in BCP
14 [RFC2119] [RFC8174] when, and only when, they appear in all 14 [RFC2119] [RFC8174] when, and only when, they appear in all
capitals, as shown here. capitals, as shown here.
Definitions of terms that are used in this document: Definitions of terms that are used in this document:
ACK-only: Any packet containing only an ACK frame. ACK-only: Any packet containing only one or more ACK frame(s).
In-flight: Packets are considered in-flight when they have been sent In-flight: Packets are considered in-flight when they have been sent
and neither acknowledged nor declared lost, and they are not ACK- and neither acknowledged nor declared lost, and they are not ACK-
only. only.
Retransmittable Frames: All frames besides ACK or PADDING are Retransmittable Frames: All frames besides ACK or PADDING are
considered retransmittable. considered retransmittable.
Retransmittable Packets: Packets that contain retransmittable frames Retransmittable Packets: Packets that contain retransmittable frames
elicit an ACK from the receiver and are called retransmittable elicit an ACK from the receiver within the maximum ack delay and
packets. are called retransmittable packets.
Crypto Packets: Packets containing CRYPTO data sent in Initial or Crypto Packets: Packets containing CRYPTO data sent in Initial or
Handshake packets. Handshake packets.
3. Design of the QUIC Transmission Machinery 3. Design of the QUIC Transmission Machinery
All transmissions in QUIC are sent with a packet-level header, which All transmissions in QUIC are sent with a packet-level header, which
indicates the encryption level and includes a packet sequence number indicates the encryption level and includes a packet sequence number
(referred to below as a packet number). The encryption level (referred to below as a packet number). The encryption level
indicates the packet number space, as described in [QUIC-TRANSPORT]. indicates the packet number space, as described in [QUIC-TRANSPORT].
skipping to change at page 5, line 18 skipping to change at page 5, line 18
transmissions and retransmissions and eliminates significant transmissions and retransmissions and eliminates significant
complexity from QUIC's interpretation of TCP loss detection complexity from QUIC's interpretation of TCP loss detection
mechanisms. mechanisms.
QUIC packets can contain multiple frames of different types. The QUIC packets can contain multiple frames of different types. The
recovery mechanisms ensure that data and frames that need reliable recovery mechanisms ensure that data and frames that need reliable
delivery are acknowledged or declared lost and sent in new packets as delivery are acknowledged or declared lost and sent in new packets as
necessary. The types of frames contained in a packet affect recovery necessary. The types of frames contained in a packet affect recovery
and congestion control logic: and congestion control logic:
o All packets are acknowledged, though packets that contain only ACK o All packets are acknowledged, though packets that contain no
and PADDING frames are not acknowledged immediately. retransmittable frames are only acknowledged along with
retransmittable packets.
o Long header packets that contain CRYPTO frames are critical to the o Long header packets that contain CRYPTO frames are critical to the
performance of the QUIC handshake and use shorter timers for performance of the QUIC handshake and use shorter timers for
acknowledgement and retransmission. acknowledgement and retransmission.
o Packets that contain only ACK frames do not count toward o Packets that contain only ACK frames do not count toward
congestion control limits and are not considered in-flight. Note congestion control limits and are not considered in-flight. Note
that this means PADDING frames cause packets to contribute toward that this means PADDING frames cause packets to contribute toward
bytes in flight without directly causing an acknowledgment to be bytes in flight without directly causing an acknowledgment to be
sent. sent.
skipping to change at page 6, line 7 skipping to change at page 6, line 7
QUIC uses separate packet number spaces for each encryption level, QUIC uses separate packet number spaces for each encryption level,
except 0-RTT and all generations of 1-RTT keys use the same packet except 0-RTT and all generations of 1-RTT keys use the same packet
number space. Separate packet number spaces ensures acknowledgement number space. Separate packet number spaces ensures acknowledgement
of packets sent with one level of encryption will not cause spurious of packets sent with one level of encryption will not cause spurious
retransmission of packets sent with a different encryption level. retransmission of packets sent with a different encryption level.
Congestion control and RTT measurement are unified across packet Congestion control and RTT measurement are unified across packet
number spaces. number spaces.
3.1.2. Monotonically Increasing Packet Numbers 3.1.2. Monotonically Increasing Packet Numbers
TCP conflates transmission sequence number at the sender with TCP conflates transmission order at the sender with delivery order at
delivery sequence number at the receiver, which results in the receiver, which results in retransmissions of the same data
retransmissions of the same data carrying the same sequence number, carrying the same sequence number, and consequently leads to
and consequently to problems caused by "retransmission ambiguity". "retransmission ambiguity". QUIC separates the two: QUIC uses a
QUIC separates the two: QUIC uses a packet number for transmissions, packet number to indicate transmission order, and any application
and any application data is sent in one or more streams, with data is sent in one or more streams, with delivery order determined
delivery order determined by stream offsets encoded within STREAM by stream offsets encoded within STREAM frames.
frames.
QUIC's packet number is strictly increasing, and directly encodes QUIC's packet number is strictly increasing within a packet number
transmission order. A higher QUIC packet number signifies that the space, and directly encodes transmission order. A higher packet
packet was sent later, and a lower QUIC packet number signifies that number signifies that the packet was sent later, and a lower packet
the packet was sent earlier. When a packet containing frames is number signifies that the packet was sent earlier. When a packet
deemed lost, QUIC rebundles necessary frames in a new packet with a containing retransmittable frames is detected lost, QUIC rebundles
new packet number, removing ambiguity about which packet is necessary frames in a new packet with a new packet number, removing
acknowledged when an ACK is received. Consequently, more accurate ambiguity about which packet is acknowledged when an ACK is received.
RTT measurements can be made, spurious retransmissions are trivially Consequently, more accurate RTT measurements can be made, spurious
detected, and mechanisms such as Fast Retransmit can be applied retransmissions are trivially detected, and mechanisms such as Fast
universally, based only on packet number. Retransmit can be applied universally, based only on packet number.
This design point significantly simplifies loss detection mechanisms This design point significantly simplifies loss detection mechanisms
for QUIC. Most TCP mechanisms implicitly attempt to infer for QUIC. Most TCP mechanisms implicitly attempt to infer
transmission ordering based on TCP sequence numbers - a non-trivial transmission ordering based on TCP sequence numbers - a non-trivial
task, especially when TCP timestamps are not available. task, especially when TCP timestamps are not available.
3.1.3. No Reneging 3.1.3. No Reneging
QUIC ACKs contain information that is similar to TCP SACK, but QUIC QUIC ACKs contain information that is similar to TCP SACK, but QUIC
does not allow any acked packet to be reneged, greatly simplifying does not allow any acked packet to be reneged, greatly simplifying
skipping to change at page 7, line 15 skipping to change at page 7, line 14
useful in receivers which may incur delays such as context-switch useful in receivers which may incur delays such as context-switch
latency before a userspace QUIC receiver processes a received packet. latency before a userspace QUIC receiver processes a received packet.
4. Loss Detection 4. Loss Detection
QUIC senders use both ack information and timeouts to detect lost QUIC senders use both ack information and timeouts to detect lost
packets, and this section provides a description of these algorithms. packets, and this section provides a description of these algorithms.
Estimating the network round-trip time (RTT) is critical to these Estimating the network round-trip time (RTT) is critical to these
algorithms and is described first. algorithms and is described first.
If a packet is lost, the QUIC transport needs to recover from that
loss, such as by retransmitting the data, sending an updated frame,
or abandoning the frame. For more information, see Section 13.2 of
[QUIC-TRANSPORT].
4.1. Computing the RTT estimate 4.1. Computing the RTT estimate
RTT is calculated when an ACK frame arrives by computing the RTT is calculated when an ACK frame arrives by computing the
difference between the current time and the time the largest newly difference between the current time and the time the largest newly
acked packet was sent. If no packets are newly acknowledged, RTT acked packet was sent. If no packets are newly acknowledged, RTT
cannot be calculated. When RTT is calculated, the ack delay field cannot be calculated. When RTT is calculated, the ack delay field
from the ACK frame SHOULD be subtracted from the RTT as long as the from the ACK frame SHOULD be subtracted from the RTT as long as the
result is larger than the Min RTT. If the result is smaller than the result is larger than the Min RTT. If the result is smaller than the
min_rtt, the RTT should be used, but the ack delay field should be min_rtt, the RTT should be used, but the ack delay field should be
ignored. ignored.
skipping to change at page 7, line 39 skipping to change at page 7, line 43
Min RTT is the minimum RTT measured over the connection, prior to Min RTT is the minimum RTT measured over the connection, prior to
adjusting by ack delay. Ignoring ack delay for min RTT prevents adjusting by ack delay. Ignoring ack delay for min RTT prevents
intentional or unintentional underestimation of min RTT, which in intentional or unintentional underestimation of min RTT, which in
turn prevents underestimating smoothed RTT. turn prevents underestimating smoothed RTT.
4.2. Ack-based Detection 4.2. Ack-based Detection
Ack-based loss detection implements the spirit of TCP's Fast Ack-based loss detection implements the spirit of TCP's Fast
Retransmit [RFC5681], Early Retransmit [RFC5827], FACK, and SACK loss Retransmit [RFC5681], Early Retransmit [RFC5827], FACK, and SACK loss
recovery [RFC6675]. This section provides an overview of how these recovery [RFC6675]. This section provides an overview of how these
algorithms are implemented in QUIC. algorithms are implemented in QUIC. Though both time-based loss
detection and early retransmit use a timer, they are part of ack-
based detection because they do not use a timer to send probes, but
rather to declare packets lost.
4.2.1. Fast Retransmit 4.2.1. Fast Retransmit
An unacknowledged packet is marked as lost when an acknowledgment is An unacknowledged packet is marked as lost when an acknowledgment is
received for a packet that was sent a threshold number of packets received for a packet that was sent a threshold number of packets
(kReorderingThreshold) and/or a threshold amount of time after the (kReorderingThreshold) and/or a threshold amount of time after the
unacknowledged packet. Receipt of the acknowledgement indicates that unacknowledged packet. Receipt of the acknowledgement indicates that
a later packet was received, while the reordering threshold provides a later packet was received, while the reordering threshold provides
some tolerance for reordering of packets in the network. some tolerance for reordering of packets in the network.
Spuriously declaring packets lost leads to unnecessary
retransmissions and may result in degraded performance due to the
actions of the congestion controller upon detecting loss.
Implementations that detect spurious retransmissions and increase the
reordering threshold in packets or time MAY choose to start with
smaller initial reordering thresholds to minimize recovery latency.
4.2.1.1. Packet Threshold
The RECOMMENDED initial value for kReorderingThreshold is 3, based on The RECOMMENDED initial value for kReorderingThreshold is 3, based on
TCP loss recovery [RFC5681] [RFC6675]. Some networks may exhibit TCP loss recovery [RFC5681] [RFC6675]. Some networks may exhibit
higher degrees of reordering, causing a sender to detect spurious higher degrees of reordering, causing a sender to detect spurious
losses. Spuriously declaring packets lost leads to unnecessary losses. Implementers MAY use algorithms developed for TCP, such as
retransmissions and may result in degraded performance due to the TCP-NCR [RFC4653], to improve QUIC's reordering resilience.
actions of the congestion controller upon detecting loss.
Implementers MAY use algorithms developed for TCP, such as TCP-NCR
[RFC4653], to improve QUIC's reordering resilience.
QUIC implementations can use time-based loss detection to handle 4.2.1.2. Time Threshold
reordering based on time elapsed since the packet was sent. This may
be used either as a replacement for a packet reordering threshold or Time threshold loss detection uses a time threshold to determine how
in addition to it. The RECOMMENDED time threshold, expressed as a much reordering to tolerate. In this document, the threshold is
fraction of the round-trip time (kTimeReorderingFraction), is 1/8. expressed as a fraction of an RTT, but implemenantations MAY
experiment with absolute thresholds. This may be used either as a
replacement for a packet reordering threshold or in addition to it.
When a larger packet is acknowledged, if it was sent more than the
threshold after any in flight packets, those packets are immediately
declared lost. Otherwise, a timer is set for the the reordering
threshold minus the time difference between the earliest in flight
packet and the largest newly acknowledged packet. Note that in some
cases the timer could become longer when packets are acknowleged out
of order. The RECOMMENDED time threshold, expressed as a fraction of
the round-trip time (kTimeReorderingFraction), is 1/8.
4.2.2. Early Retransmit 4.2.2. Early Retransmit
Unacknowledged packets close to the tail may have fewer than Unacknowledged packets close to the tail may have fewer than
kReorderingThreshold retransmittable packets sent after them. Loss kReorderingThreshold retransmittable packets sent after them. Loss
of such packets cannot be detected via Fast Retransmit. To enable of such packets cannot be detected via Packet Threshold Fast
ack-based loss detection of such packets, receipt of an Retransmit. To enable ack-based loss detection of such packets,
acknowledgment for the last outstanding retransmittable packet receipt of an acknowledgment for the last outstanding retransmittable
triggers the Early Retransmit process, as follows. packet triggers the Early Retransmit process, as follows.
If there are unacknowledged in-flight packets still pending, they If there are unacknowledged in-flight packets still pending, they
should be marked as lost. To compensate for the reduced reordering should be marked as lost. To compensate for the reduced reordering
resilience, the sender SHOULD set a timer for a small period of time. resilience, the sender SHOULD set a timer for a small period of time.
If the unacknowledged in-flight packets are not acknowledged during If the unacknowledged in-flight packets are not acknowledged during
this time, then these packets MUST be marked as lost. this time, then these packets MUST be marked as lost.
An endpoint SHOULD set the timer such that a packet is marked as lost An endpoint SHOULD set the timer such that a packet is marked as lost
no earlier than 1.125 * max(SRTT, latest_RTT) since when it was sent. no earlier than 1.125 * max(SRTT, latest_RTT) since when it was sent.
skipping to change at page 9, line 12 skipping to change at page 9, line 39
retransmissions, and a higher multiplier increases loss recovery retransmissions, and a higher multiplier increases loss recovery
delay. delay.
This mechanism is based on Early Retransmit for TCP [RFC5827]. This mechanism is based on Early Retransmit for TCP [RFC5827].
However, [RFC5827] does not include the timer described above. Early However, [RFC5827] does not include the timer described above. Early
Retransmit is prone to spurious retransmissions due to its reduced Retransmit is prone to spurious retransmissions due to its reduced
reordering resilence without the timer. This observation led Linux reordering resilence without the timer. This observation led Linux
TCP implementers to implement a timer for TCP as well, and this TCP implementers to implement a timer for TCP as well, and this
document incorporates this advancement. document incorporates this advancement.
4.3. Timer-based Detection 4.3. Timeout Loss Detection
Timer-based loss detection recovers from losses that cannot be Timeout loss detection recovers from losses that cannot be handled by
handled by ack-based loss detection. It uses a single timer which ack-based loss detection. It uses a single timer which switches
switches between a crypto retransmission timer, a Tail Loss Probe between a crypto retransmission timer, a Tail Loss Probe timer and
timer and Retransmission Timeout mechanisms. Retransmission Timeout mechanisms.
4.3.1. Crypto Retransmission Timeout 4.3.1. Crypto Retransmission Timeout
Data in CRYPTO frames is critical to QUIC transport and crypto Data in CRYPTO frames is critical to QUIC transport and crypto
negotiation, so a more aggressive timeout is used to retransmit it. negotiation, so a more aggressive timeout is used to retransmit it.
The initial crypto retransmission timeout SHOULD be set to twice the The initial crypto retransmission timeout SHOULD be set to twice the
initial RTT. initial RTT.
At the beginning, there are no prior RTT samples within a connection. At the beginning, there are no prior RTT samples within a connection.
skipping to change at page 9, line 40 skipping to change at page 10, line 21
initial RTT. If no previous RTT is available, or if the network initial RTT. If no previous RTT is available, or if the network
changes, the initial RTT SHOULD be set to 100ms. When an changes, the initial RTT SHOULD be set to 100ms. When an
acknowledgement is received, a new RTT is computed and the timer acknowledgement is received, a new RTT is computed and the timer
SHOULD be set for twice the newly computed smoothed RTT. SHOULD be set for twice the newly computed smoothed RTT.
When crypto packets are sent, the sender MUST set a timer for the When crypto packets are sent, the sender MUST set a timer for the
crypto timeout period. Upon timeout, the sender MUST retransmit all crypto timeout period. Upon timeout, the sender MUST retransmit all
unacknowledged CRYPTO data if possible. unacknowledged CRYPTO data if possible.
Until the server has validated the client's address on the path, the Until the server has validated the client's address on the path, the
number of bytes it can send is limited, as specified in amount of data it can send is limited, as specified in
[QUIC-TRANSPORT]. If not all unacknowledged CRYPTO data can be sent, [QUIC-TRANSPORT]. If not all unacknowledged CRYPTO data can be sent,
then all unacknowledged CRYPTO data sent in Initial packets should be then all unacknowledged CRYPTO data sent in Initial packets should be
retransmitted. If no bytes can be sent, then no alarm should be retransmitted. If no data can be sent, then no alarm should be armed
armed until bytes have been received from the client. until data has been received from the client.
Because the server could be blocked until more packets are received, Because the server could be blocked until more packets are received,
the client MUST start the crypto retransmission timer even if there the client MUST start the crypto retransmission timer even if there
is no unacknowledged CRYPTO data. If the timer expires and the is no unacknowledged CRYPTO data. If the timer expires and the
client has no CRYPTO data to retransmit and does not have Handshake client has no CRYPTO data to retransmit and does not have Handshake
keys, it SHOULD send an Initial packet in a UDP datagram of at least keys, it SHOULD send an Initial packet in a UDP datagram of at least
1200 octets. If the client has Handshake keys, it SHOULD send a 1200 bytes. If the client has Handshake keys, it SHOULD send a
Handshake packet. Handshake packet.
On each consecutive expiration of the crypto timer without receiving On each consecutive expiration of the crypto timer without receiving
an acknowledgement for a new packet, the sender SHOULD double the an acknowledgement for a new packet, the sender SHOULD double the
crypto retransmission timeout and set a timer for this period. crypto retransmission timeout and set a timer for this period.
When crypto packets are outstanding, the TLP and RTO timers are not When crypto packets are outstanding, the TLP and RTO timers are not
active. active.
4.3.1.1. Retry and Version Negotiation 4.3.1.1. Retry and Version Negotiation
skipping to change at page 12, line 5 skipping to change at page 12, line 37
event therefore makes the connection very sensitive to single packet event therefore makes the connection very sensitive to single packet
loss. Sending two packets instead of one significantly increases loss. Sending two packets instead of one significantly increases
resilience to packet drop in both directions, thus reducing the resilience to packet drop in both directions, thus reducing the
probability of consecutive RTO events. probability of consecutive RTO events.
QUIC's RTO algorithm differs from TCP in that the firing of an RTO QUIC's RTO algorithm differs from TCP in that the firing of an RTO
timer is not considered a strong enough signal of packet loss, so timer is not considered a strong enough signal of packet loss, so
does not result in an immediate change to congestion window or does not result in an immediate change to congestion window or
recovery state. An RTO timer expires only when there's a prolonged recovery state. An RTO timer expires only when there's a prolonged
period of network silence, which could be caused by a change in the period of network silence, which could be caused by a change in the
underlying network RTT. network RTT.
QUIC also diverges from TCP by including MaxAckDelay in the RTO QUIC also diverges from TCP by including MaxAckDelay in the RTO
period. Since QUIC corrects for this delay in its SRTT and RTTVAR period. Since QUIC corrects for this delay in its SRTT and RTTVAR
computations, it is necessary to add this delay explicitly in the TLP computations, it is necessary to add this delay explicitly in the TLP
and RTO computation. and RTO computation.
When an acknowledgment is received for a packet sent on an RTO event, When an ACK is received that acknowledges only one or more packets
any unacknowledged packets with lower packet numbers than those sent on an RTO event, all unacknowledged packets with lower packet
acknowledged MUST be marked as lost. If an acknowledgement for a numbers MUST be marked as lost. If packets sent prior to the first
packet sent on an RTO is received at the same time packets sent prior RTO are acknowledged in the same ACK, the RTO is considered spurious
to the first RTO are acknowledged, the RTO is considered spurious and and standard loss detection rules apply.
standard loss detection rules apply.
A packet sent when an RTO timer expires MAY carry new data if A packet sent when an RTO timer expires MAY carry new data if
available or unacknowledged data to potentially reduce recovery time. available or unacknowledged data to potentially reduce recovery time.
Since this packet is sent as a probe into the network prior to Since this packet is sent as a probe into the network prior to
establishing any packet loss, prior unacknowledged packets SHOULD NOT establishing any packet loss, prior unacknowledged packets SHOULD NOT
be marked as lost. be marked as lost when the timer expires.
A packet sent on an RTO timer MUST NOT be blocked by the sender's A packet sent on an RTO timer MUST NOT be blocked by the sender's
congestion controller. A sender MUST however count these bytes as congestion controller. A sender MUST however count these packets as
additional bytes in flight, since this packet adds network load being in flight, since this packet adds network load without
without establishing packet loss. establishing packet loss.
4.4. Generating Acknowledgements 4.4. Generating Acknowledgements
QUIC SHOULD delay sending acknowledgements in response to packets, QUIC SHOULD delay sending acknowledgements in response to packets,
but MUST NOT excessively delay acknowledgements of packets containing but MUST NOT excessively delay acknowledgements of retransmittable
frames other than ACK. Specifically, implementations MUST attempt to packets. Specifically, implementations MUST attempt to enforce a
enforce a maximum ack delay to avoid causing the peer spurious maximum ack delay to avoid causing the peer spurious timeouts. The
timeouts. The maximum ack delay is communicated in the maximum ack delay is communicated in the "max_ack_delay" transport
"max_ack_delay" transport parameter and the default value is 25ms. parameter and the default value is 25ms.
An acknowledgement SHOULD be sent immediately upon receipt of a An acknowledgement SHOULD be sent immediately upon receipt of a
second packet but the delay SHOULD NOT exceed the maximum ack delay. second packet but the delay SHOULD NOT exceed the maximum ack delay.
QUIC recovery algorithms do not assume the peer generates an QUIC recovery algorithms do not assume the peer generates an
acknowledgement immediately when receiving a second full-packet. acknowledgement immediately when receiving a second full-packet.
Out-of-order packets SHOULD be acknowledged more quickly, in order to Out-of-order packets SHOULD be acknowledged more quickly, in order to
accelerate loss recovery. The receiver SHOULD send an immediate ACK accelerate loss recovery. The receiver SHOULD send an immediate ACK
when it receives a new packet which is not one greater than the when it receives a new packet which is not one greater than the
largest received packet number. largest received packet number.
skipping to change at page 13, line 43 skipping to change at page 14, line 30
4.4.3. Receiver Tracking of ACK Frames 4.4.3. Receiver Tracking of ACK Frames
When a packet containing an ACK frame is sent, the largest When a packet containing an ACK frame is sent, the largest
acknowledged in that frame may be saved. When a packet containing an acknowledged in that frame may be saved. When a packet containing an
ACK frame is acknowledged, the receiver can stop acknowledging ACK frame is acknowledged, the receiver can stop acknowledging
packets less than or equal to the largest acknowledged in the sent packets less than or equal to the largest acknowledged in the sent
ACK frame. ACK frame.
In cases without ACK frame loss, this algorithm allows for a minimum In cases without ACK frame loss, this algorithm allows for a minimum
of 1 RTT of reordering. In cases with ACK frame loss, this approach of 1 RTT of reordering. In cases with ACK frame loss and reordering,
does not guarantee that every acknowledgement is seen by the sender this approach does not guarantee that every acknowledgement is seen
before it is no longer included in the ACK frame. Packets could be by the sender before it is no longer included in the ACK frame.
received out of order and all subsequent ACK frames containing them Packets could be received out of order and all subsequent ACK frames
could be lost. In this case, the loss recovery algorithm may cause containing them could be lost. In this case, the loss recovery
spurious retransmits, but the sender will continue making forward algorithm may cause spurious retransmits, but the sender will
progress. continue making forward progress.
4.5. Pseudocode 4.5. Tracking Sent Packets
4.5.1. Constants of interest To correctly implement congestion control, a QUIC sender tracks every
retransmittable packet until the packet is acknowledged or lost. It
is expected that implementations will be able to access this
information by packet number and crypto context and store the per-
packet fields (Section 4.5.1) for loss recovery and congestion
control.
After a packet is declared lost, it SHOULD be tracked for an amount
of time comparable to the maximum expected packet reordering, such as
1 RTT. This allows for detection of spurious retransmissions.
Sent packets are tracked for each packet number space, and ACK
processing only applies to a single space.
4.5.1. Sent Packet Fields
packet_number: The packet number of the sent packet.
retransmittable: A boolean that indicates whether a packet is
retransmittable. If true, it is expected that an acknowledgement
will be received, though the peer could delay sending the ACK
frame containing it by up to the MaxAckDelay.
in_flight: A boolean that indicates whether the packet counts
towards bytes in flight.
is_crypto_packet: A boolean that indicates whether the packet
contains cryptographic handshake messages critical to the
completion of the QUIC handshake. In this version of QUIC, this
includes any packet with the long header that includes a CRYPTO
frame.
sent_bytes: The number of bytes sent in the packet, not including
UDP or IP overhead, but including QUIC framing overhead.
time: The time the packet was sent.
4.6. Pseudocode
4.6.1. Constants of interest
Constants used in loss recovery are based on a combination of RFCs, Constants used in loss recovery are based on a combination of RFCs,
papers, and common practice. Some may need to be changed or papers, and common practice. Some may need to be changed or
negotiated in order to better suit a variety of environments. negotiated in order to better suit a variety of environments.
kMaxTLPs: Maximum number of tail loss probes before an RTO expires. kMaxTLPs: Maximum number of tail loss probes before an RTO expires.
The RECOMMENDED value is 2. The RECOMMENDED value is 2.
kReorderingThreshold: Maximum reordering in packet number space kReorderingThreshold: Maximum reordering in packet number space
before FACK style loss detection considers a packet lost. The before FACK style loss detection considers a packet lost. The
RECOMMENDED value is 3. RECOMMENDED value is 3.
kTimeReorderingFraction: Maximum reordering in time space before kTimeReorderingFraction: Maximum reordering in time space before
time based loss detection considers a packet lost. In fraction of time threshold loss detection considers a packet lost. In
an RTT. The RECOMMENDED value is 1/8. fraction of an RTT. The RECOMMENDED value is 1/8.
kUsingTimeLossDetection: Whether time based loss detection is in kUsingTimeLossDetection: Whether time threshold loss detection is in
use. If false, uses FACK style loss detection. The RECOMMENDED use. If false, uses only packet threshold loss detection. The
value is false. RECOMMENDED value is false.
kMinTLPTimeout: Minimum time in the future a tail loss probe timer kMinTLPTimeout: Minimum time in the future a tail loss probe timer
may be set for. The RECOMMENDED value is 10ms. may be set for. The RECOMMENDED value is 10ms.
kMinRTOTimeout: Minimum time in the future an RTO timer may be set kMinRTOTimeout: Minimum time in the future an RTO timer may be set
for. The RECOMMENDED value is 200ms. for. The RECOMMENDED value is 200ms.
kDelayedAckTimeout: The length of the peer's delayed ack timer. The kDelayedAckTimeout: The length of the peer's delayed ack timer. The
RECOMMENDED value is 25ms. RECOMMENDED value is 25ms.
kInitialRtt: The RTT used before an RTT sample is taken. The kInitialRtt: The RTT used before an RTT sample is taken. The
RECOMMENDED value is 100ms. RECOMMENDED value is 100ms.
4.5.2. Variables of interest 4.6.2. Variables of interest
Variables required to implement the congestion control mechanisms are Variables required to implement the congestion control mechanisms are
described in this section. described in this section.
loss_detection_timer: Multi-modal timer used for loss detection. loss_detection_timer: Multi-modal timer used for loss detection.
crypto_count: The number of times all unacknowledged CRYPTO data has crypto_count: The number of times all unacknowledged CRYPTO data has
been retransmitted without receiving an ack. been retransmitted without receiving an ack.
tlp_count: The number of times a tail loss probe has been sent tlp_count: The number of times a tail loss probe has been sent
skipping to change at page 15, line 30 skipping to change at page 17, line 4
largest_acked_packet: The largest packet number acknowledged in an largest_acked_packet: The largest packet number acknowledged in an
ACK frame. ACK frame.
latest_rtt: The most recent RTT measurement made when receiving an latest_rtt: The most recent RTT measurement made when receiving an
ack for a previously unacked packet. ack for a previously unacked packet.
smoothed_rtt: The smoothed RTT of the connection, computed as smoothed_rtt: The smoothed RTT of the connection, computed as
described in [RFC6298] described in [RFC6298]
rttvar: The RTT variance, computed as described in [RFC6298] rttvar: The RTT variance, computed as described in [RFC6298]
min_rtt: The minimum RTT seen in the connection, ignoring ack delay. min_rtt: The minimum RTT seen in the connection, ignoring ack delay.
max_ack_delay: The maximum amount of time by which the receiver max_ack_delay: The maximum amount of time by which the receiver
intends to delay acknowledgments, in milliseconds. The actual intends to delay acknowledgments, in milliseconds. The actual
ack_delay in a received ACK frame may be larger due to late ack_delay in a received ACK frame may be larger due to late
timers, reordering, or lost ACKs. timers, reordering, or lost ACKs.
reordering_threshold: The largest packet number gap between the reordering_threshold: The largest packet number gap between the
largest acknowledged retransmittable packet and an unacknowledged largest acknowledged retransmittable packet and an unacknowledged
retransmittable packet before it is declared lost. retransmittable packet before it is declared lost.
time_reordering_fraction: The reordering window as a fraction of time_reordering_fraction: The reordering window as a fraction of
max(smoothed_rtt, latest_rtt). max(smoothed_rtt, latest_rtt).
loss_time: The time at which the next packet will be considered lost loss_time: The time at which the next packet will be considered lost
based on early transmit or exceeding the reordering window in based on early transmit or exceeding the reordering window in
time. time.
sent_packets: An association of packet numbers to information about sent_packets: An association of packet numbers to information about
them, including a number field indicating the packet number, a them. Described in detail above in Section 4.5.
time field indicating the time a packet was sent, a boolean
indicating whether the packet is ack-only, a boolean indicating
whether it counts towards bytes in flight, and a bytes field
indicating the packet's size. sent_packets is ordered by packet
number, and packets remain in sent_packets until acknowledged or
lost. A sent_packets data structure is maintained per packet
number space, and ACK processing only applies to a single space.
4.5.3. Initialization 4.6.3. Initialization
At the beginning of the connection, initialize the loss detection At the beginning of the connection, initialize the loss detection
variables as follows: variables as follows:
loss_detection_timer.reset() loss_detection_timer.reset()
crypto_count = 0 crypto_count = 0
tlp_count = 0 tlp_count = 0
rto_count = 0 rto_count = 0
if (kUsingTimeLossDetection) if (kUsingTimeLossDetection)
reordering_threshold = infinite reordering_threshold = infinite
skipping to change at page 16, line 34 skipping to change at page 18, line 5
time_reordering_fraction = infinite time_reordering_fraction = infinite
loss_time = 0 loss_time = 0
smoothed_rtt = 0 smoothed_rtt = 0
rttvar = 0 rttvar = 0
min_rtt = infinite min_rtt = infinite
largest_sent_before_rto = 0 largest_sent_before_rto = 0
time_of_last_sent_retransmittable_packet = 0 time_of_last_sent_retransmittable_packet = 0
time_of_last_sent_crypto_packet = 0 time_of_last_sent_crypto_packet = 0
largest_sent_packet = 0 largest_sent_packet = 0
4.5.4. On Sending a Packet 4.6.4. On Sending a Packet
After any packet is sent, be it a new transmission or a rebundled
transmission, the following OnPacketSent function is called. The
parameters to OnPacketSent are as follows:
o packet_number: The packet number of the sent packet.
o ack_only: A boolean that indicates whether a packet contains only
ACK or PADDING frame(s). If true, it is still expected an ack
will be received for this packet, but it is not retransmittable.
o in_flight: A boolean that indicates whether the packet counts
towards bytes in flight.
o is_crypto_packet: A boolean that indicates whether the packet
contains cryptographic handshake messages critical to the
completion of the QUIC handshake. In this version of QUIC, this
includes any packet with the long header that includes a CRYPTO
frame.
o sent_bytes: The number of bytes sent in the packet, not including After a packet is sent, information about the packet is stored. The
UDP or IP overhead, but including QUIC framing overhead. parameters to OnPacketSent are described in detail above in
Section 4.5.1.
Pseudocode for OnPacketSent follows: Pseudocode for OnPacketSent follows:
OnPacketSent(packet_number, ack_only, in_flight, OnPacketSent(packet_number, retransmittable, in_flight,
is_crypto_packet, sent_bytes): is_crypto_packet, sent_bytes):
largest_sent_packet = packet_number largest_sent_packet = packet_number
sent_packets[packet_number].packet_number = packet_number sent_packets[packet_number].packet_number = packet_number
sent_packets[packet_number].time = now sent_packets[packet_number].time = now
sent_packets[packet_number].ack_only = ack_only sent_packets[packet_number].retransmittable = retransmittable
sent_packets[packet_number].in_flight = in_flight sent_packets[packet_number].in_flight = in_flight
if !ack_only: if retransmittable:
if is_crypto_packet: if is_crypto_packet:
time_of_last_sent_crypto_packet = now time_of_last_sent_crypto_packet = now
time_of_last_sent_retransmittable_packet = now time_of_last_sent_retransmittable_packet = now
OnPacketSentCC(sent_bytes) OnPacketSentCC(sent_bytes)
sent_packets[packet_number].bytes = sent_bytes sent_packets[packet_number].size = sent_bytes
SetLossDetectionTimer() SetLossDetectionTimer()
4.5.5. On Receiving an Acknowledgment 4.6.5. On Receiving an Acknowledgment
When an ACK frame is received, it may newly acknowledge any number of When an ACK frame is received, it may newly acknowledge any number of
packets. packets.
Pseudocode for OnAckReceived and UpdateRtt follow: Pseudocode for OnAckReceived and UpdateRtt follow:
OnAckReceived(ack): OnAckReceived(ack):
largest_acked_packet = ack.largest_acked
// If the largest acknowledged is newly acked, // If the largest acknowledged is newly acked,
// update the RTT. // update the RTT.
if (sent_packets[ack.largest_acked]): if (sent_packets[ack.largest_acked]):
latest_rtt = now - sent_packets[ack.largest_acked].time latest_rtt = now - sent_packets[ack.largest_acked].time
UpdateRtt(latest_rtt, ack.ack_delay) UpdateRtt(latest_rtt, ack.ack_delay)
// Find all newly acked packets in this ACK frame // Find all newly acked packets in this ACK frame
newly_acked_packets = DetermineNewlyAckedPackets(ack) newly_acked_packets = DetermineNewlyAckedPackets(ack)
for acked_packet in newly_acked_packets: for acked_packet in newly_acked_packets:
OnPacketAcked(acked_packet.packet_number) OnPacketAcked(acked_packet.packet_number)
skipping to change at page 18, line 31 skipping to change at page 19, line 30
FindSmallestNewlyAcked(newly_acked_packets) FindSmallestNewlyAcked(newly_acked_packets)
// If any packets sent prior to RTO were acked, then the // If any packets sent prior to RTO were acked, then the
// RTO was spurious. Otherwise, inform congestion control. // RTO was spurious. Otherwise, inform congestion control.
if (rto_count > 0 && if (rto_count > 0 &&
smallest_newly_acked > largest_sent_before_rto): smallest_newly_acked > largest_sent_before_rto):
OnRetransmissionTimeoutVerified(smallest_newly_acked) OnRetransmissionTimeoutVerified(smallest_newly_acked)
crypto_count = 0 crypto_count = 0
tlp_count = 0 tlp_count = 0
rto_count = 0 rto_count = 0
DetectLostPackets(ack.largest_acked_packet) DetectLostPackets(ack.acked_packet)
SetLossDetectionTimer() SetLossDetectionTimer()
// Process ECN information if present. // Process ECN information if present.
if (ACK frame contains ECN information): if (ACK frame contains ECN information):
ProcessECN(ack) ProcessECN(ack)
UpdateRtt(latest_rtt, ack_delay): UpdateRtt(latest_rtt, ack_delay):
// min_rtt ignores ack delay. // min_rtt ignores ack delay.
min_rtt = min(min_rtt, latest_rtt) min_rtt = min(min_rtt, latest_rtt)
// Adjust for ack delay if it's plausible. // Adjust for ack delay if it's plausible.
skipping to change at page 19, line 5 skipping to change at page 20, line 5
latest_rtt -= ack_delay latest_rtt -= ack_delay
// Based on {{RFC6298}}. // Based on {{RFC6298}}.
if (smoothed_rtt == 0): if (smoothed_rtt == 0):
smoothed_rtt = latest_rtt smoothed_rtt = latest_rtt
rttvar = latest_rtt / 2 rttvar = latest_rtt / 2
else: else:
rttvar_sample = abs(smoothed_rtt - latest_rtt) rttvar_sample = abs(smoothed_rtt - latest_rtt)
rttvar = 3/4 * rttvar + 1/4 * rttvar_sample rttvar = 3/4 * rttvar + 1/4 * rttvar_sample
smoothed_rtt = 7/8 * smoothed_rtt + 1/8 * latest_rtt smoothed_rtt = 7/8 * smoothed_rtt + 1/8 * latest_rtt
4.5.6. On Packet Acknowledgment 4.6.6. On Packet Acknowledgment
When a packet is acked for the first time, the following When a packet is acknowledged for the first time, the following
OnPacketAcked function is called. Note that a single ACK frame may OnPacketAcked function is called. Note that a single ACK frame may
newly acknowledge several packets. OnPacketAcked must be called once newly acknowledge several packets. OnPacketAcked must be called once
for each of these newly acked packets. for each of these newly acknowledged packets.
OnPacketAcked takes one parameter, acked_packet, which is the struct OnPacketAcked takes one parameter, acked_packet, which is the struct
of the newly acked packet. detailed in Section 4.5.1.
If this is the first acknowledgement following RTO, check if the
smallest newly acknowledged packet is one sent by the RTO, and if so,
inform congestion control of a verified RTO, similar to F-RTO
[RFC5682].
Pseudocode for OnPacketAcked follows: Pseudocode for OnPacketAcked follows:
OnPacketAcked(acked_packet): OnPacketAcked(acked_packet):
if (!acked_packet.is_ack_only): if (acked_packet.retransmittable):
OnPacketAckedCC(acked_packet) OnPacketAckedCC(acked_packet)
sent_packets.remove(acked_packet.packet_number) sent_packets.remove(acked_packet.packet_number)
4.5.7. Setting the Loss Detection Timer 4.6.7. Setting the Loss Detection Timer
QUIC loss detection uses a single timer for all timer-based loss QUIC loss detection uses a single timer for all timeout loss
detection. The duration of the timer is based on the timer's mode, detection. The duration of the timer is based on the timer's mode,
which is set in the packet and timer events further below. The which is set in the packet and timer events further below. The
function SetLossDetectionTimer defined below shows how the single function SetLossDetectionTimer defined below shows how the single
timer is set. timer is set.
Pseudocode for SetLossDetectionTimer follows: Pseudocode for SetLossDetectionTimer follows:
SetLossDetectionTimer(): SetLossDetectionTimer():
// Don't arm timer if there are no retransmittable packets // Don't arm timer if there are no retransmittable packets
// in flight. // in flight.
skipping to change at page 20, line 43 skipping to change at page 21, line 43
timeout = timeout * (2 ^ rto_count) timeout = timeout * (2 ^ rto_count)
if (tlp_count < kMaxTLPs): if (tlp_count < kMaxTLPs):
// Tail Loss Probe // Tail Loss Probe
tlp_timeout = max(1.5 * smoothed_rtt tlp_timeout = max(1.5 * smoothed_rtt
+ max_ack_delay, kMinTLPTimeout) + max_ack_delay, kMinTLPTimeout)
timeout = min(tlp_timeout, timeout) timeout = min(tlp_timeout, timeout)
loss_detection_timer.set( loss_detection_timer.set(
time_of_last_sent_retransmittable_packet + timeout) time_of_last_sent_retransmittable_packet + timeout)
4.5.8. On Timeout 4.6.8. On Timeout
When the loss detection timer expires, the timer's mode determines When the loss detection timer expires, the timer's mode determines
the action to be performed. the action to be performed.
Pseudocode for OnLossDetectionTimeout follows: Pseudocode for OnLossDetectionTimeout follows:
OnLossDetectionTimeout(): OnLossDetectionTimeout():
if (crypto packets are outstanding): if (crypto packets are outstanding):
// Crypto retransmission timeout. // Crypto retransmission timeout.
RetransmitUnackedCryptoData() RetransmitUnackedCryptoData()
skipping to change at page 21, line 26 skipping to change at page 22, line 26
tlp_count++ tlp_count++
else: else:
// RTO. // RTO.
if (rto_count == 0) if (rto_count == 0)
largest_sent_before_rto = largest_sent_packet largest_sent_before_rto = largest_sent_packet
SendTwoPackets() SendTwoPackets()
rto_count++ rto_count++
SetLossDetectionTimer() SetLossDetectionTimer()
4.5.9. Detecting Lost Packets 4.6.9. Detecting Lost Packets
Packets in QUIC are only considered lost once a larger packet number
in the same packet number space is acknowledged. DetectLostPackets
is called every time an ack is received and operates on the
sent_packets for that packet number space. If the loss detection
timer expires and the loss_time is set, the previous largest acked
packet is supplied.
4.5.9.1. Pseudocode DetectLostPackets is called every time an ACK is received and
operates on the sent_packets for that packet number space. If the
loss detection timer expires and the loss_time is set, the previous
largest acknowledged packet is supplied.
DetectLostPackets takes one parameter, acked, which is the largest DetectLostPackets takes one parameter, largest_acked, which is the
acked packet. largest acked packet.
Pseudocode for DetectLostPackets follows: Pseudocode for DetectLostPackets follows:
DetectLostPackets(largest_acked): DetectLostPackets(largest_acked):
loss_time = 0 loss_time = 0
lost_packets = {} lost_packets = {}
delay_until_lost = infinite delay_until_lost = infinite
if (kUsingTimeLossDetection): if (kUsingTimeLossDetection):
delay_until_lost = delay_until_lost =
(1 + time_reordering_fraction) * (1 + time_reordering_fraction) *
max(latest_rtt, smoothed_rtt) max(latest_rtt, smoothed_rtt)
else if (largest_acked.packet_number == largest_sent_packet): else if (largest_acked.packet_number == largest_sent_packet):
// Early retransmit timer. // Early retransmit timer.
delay_until_lost = 9/8 * max(latest_rtt, smoothed_rtt) delay_until_lost = 9/8 * max(latest_rtt, smoothed_rtt)
foreach (unacked < largest_acked.packet_number): foreach (unacked < largest_acked.packet_number):
time_since_sent = now() - unacked.time_sent time_since_sent = now() - unacked.time_sent
delta = largest_acked.packet_number - unacked.packet_number delta = largest_acked.packet_number - unacked.packet_number
if (time_since_sent > delay_until_lost || if (time_since_sent > delay_until_lost ||
delta > reordering_threshold): delta > reordering_threshold):
sent_packets.remove(unacked.packet_number) sent_packets.remove(unacked.packet_number)
if (!unacked.is_ack_only): if (unacked.retransmittable):
lost_packets.insert(unacked) lost_packets.insert(unacked)
else if (loss_time == 0 && delay_until_lost != infinite): else if (loss_time == 0 && delay_until_lost != infinite):
loss_time = now() + delay_until_lost - time_since_sent loss_time = now() + delay_until_lost - time_since_sent
// Inform the congestion controller of lost packets and // Inform the congestion controller of lost packets and
// lets it decide whether to retransmit immediately. // lets it decide whether to retransmit immediately.
if (!lost_packets.empty()): if (!lost_packets.empty()):
OnPacketsLost(lost_packets) OnPacketsLost(lost_packets)
4.6. Discussion 4.7. Discussion
The majority of constants were derived from best common practices The majority of constants were derived from best common practices
among widely deployed TCP implementations on the internet. among widely deployed TCP implementations on the internet.
Exceptions follow. Exceptions follow.
A shorter delayed ack time of 25ms was chosen because longer delayed A shorter delayed ack time of 25ms was chosen because longer delayed
acks can delay loss recovery and for the small number of connections acks can delay loss recovery and for the small number of connections
where less than packet per 25ms is delivered, acking every packet is where less than packet per 25ms is delivered, acking every packet is
beneficial to congestion control and loss recovery. beneficial to congestion control and loss recovery.
skipping to change at page 23, line 11 skipping to change at page 24, line 11
is a congestion window based congestion control. QUIC specifies the is a congestion window based congestion control. QUIC specifies the
congestion window in bytes rather than packets due to finer control congestion window in bytes rather than packets due to finer control
and the ease of appropriate byte counting [RFC3465]. and the ease of appropriate byte counting [RFC3465].
QUIC hosts MUST NOT send packets if they would increase QUIC hosts MUST NOT send packets if they would increase
bytes_in_flight (defined in Section 5.8.2) beyond the available bytes_in_flight (defined in Section 5.8.2) beyond the available
congestion window, unless the packet is a probe packet sent after the congestion window, unless the packet is a probe packet sent after the
TLP or RTO timer expires, as described in Section 4.3.2 and TLP or RTO timer expires, as described in Section 4.3.2 and
Section 4.3.3. Section 4.3.3.
Implementations MAY use other congestion control algorithms, and Implementations MAY use other congestion control algorithms, such as
endpoints MAY use different algorithms from one another. The signals Cubic [RFC8312], and endpoints MAY use different algorithms from one
QUIC provides for congestion control are generic and are designed to another. The signals QUIC provides for congestion control are
support different algorithms. generic and are designed to support different algorithms.
5.1. Explicit Congestion Notification 5.1. Explicit Congestion Notification
If a path has been verified to support ECN, QUIC treats a Congestion If a path has been verified to support ECN, QUIC treats a Congestion
Experienced codepoint in the IP header as a signal of congestion. Experienced codepoint in the IP header as a signal of congestion.
This document specifies an endpoint's response when its peer receives This document specifies an endpoint's response when its peer receives
packets with the Congestion Experienced codepoint. As discussed in packets with the Congestion Experienced codepoint. As discussed in
[RFC8311], endpoints are permitted to experiment with other response [RFC8311], endpoints are permitted to experiment with other response
functions. functions.
5.2. Slow Start 5.2. Slow Start
QUIC begins every connection in slow start and exits slow start upon QUIC begins every connection in slow start and exits slow start upon
loss or upon increase in the ECN-CE counter. QUIC re-enters slow loss or upon increase in the ECN-CE counter. QUIC re-enters slow
start anytime the congestion window is less than ssthresh, which start anytime the congestion window is less than ssthresh, which
typically only occurs after an RTO. While in slow start, QUIC typically only occurs after an RTO. While in slow start, QUIC
increases the congestion window by the number of bytes acknowledged increases the congestion window by the number of bytes acknowledged
when each ack is processed. when each acknowledgment is processed.
5.3. Congestion Avoidance 5.3. Congestion Avoidance
Slow start exits to congestion avoidance. Congestion avoidance in Slow start exits to congestion avoidance. Congestion avoidance in
NewReno uses an additive increase multiplicative decrease (AIMD) NewReno uses an additive increase multiplicative decrease (AIMD)
approach that increases the congestion window by one maximum packet approach that increases the congestion window by one maximum packet
size per congestion window acknowledged. When a loss is detected, size per congestion window acknowledged. When a loss is detected,
NewReno halves the congestion window and sets the slow start NewReno halves the congestion window and sets the slow start
threshold to the new congestion window. threshold to the new congestion window.
5.4. Recovery Period 5.4. Recovery Period
Recovery is a period of time beginning with detection of a lost Recovery is a period of time beginning with detection of a lost
packet or an increase in the ECN-CE counter. Because QUIC packet or an increase in the ECN-CE counter. Because QUIC does not
retransmits stream data and control frames, not packets, it defines retransmit packets, it defines the end of recovery as a packet sent
the end of recovery as a packet sent after the start of recovery after the start of recovery being acknowledged. This is slightly
being acknowledged. This is slightly different from TCP's definition different from TCP's definition of recovery, which ends when the lost
of recovery, which ends when the lost packet that started recovery is packet that started recovery is acknowledged.
acknowledged.
The recovery period limits congestion window reduction to once per The recovery period limits congestion window reduction to once per
round trip. During recovery, the congestion window remains unchanged round trip. During recovery, the congestion window remains unchanged
irrespective of new losses or increases in the ECN-CE counter. irrespective of new losses or increases in the ECN-CE counter.
5.5. Tail Loss Probe 5.5. Tail Loss Probe
A TLP packet MUST NOT be blocked by the sender's congestion A TLP packet MUST NOT be blocked by the sender's congestion
controller. The sender MUST however count these bytes as additional controller. The sender MUST however count TLP packets against bytes
bytes-in-flight, since a TLP adds network load without establishing in flight, since a TLP adds network load without establishing packet
packet loss. loss.
Acknowledgement or loss of tail loss probes are treated like any Acknowledgement or loss of tail loss probes are treated like any
other packet. other packet.
5.6. Retransmission Timeout 5.6. Retransmission Timeout
When retransmissions are sent due to a retransmission timeout timer, When retransmissions are sent due to a retransmission timeout timer,
no change is made to the congestion window until the next no change is made to the congestion window until the next
acknowledgement arrives. The retransmission timeout is considered acknowledgement arrives. The retransmission timeout is considered
spurious when this acknowledgement acknowledges packets sent prior to spurious when this acknowledgement acknowledges packets sent prior to
skipping to change at page 25, line 49 skipping to change at page 26, line 49
that contain at least one retransmittable or PADDING frame, and that contain at least one retransmittable or PADDING frame, and
have not been acked or declared lost. The size does not include have not been acked or declared lost. The size does not include
IP or UDP overhead, but does include the QUIC header and AEAD IP or UDP overhead, but does include the QUIC header and AEAD
overhead. Packets only containing ACK frames do not count towards overhead. Packets only containing ACK frames do not count towards
bytes_in_flight to ensure congestion control does not impede bytes_in_flight to ensure congestion control does not impede
congestion feedback. congestion feedback.
congestion_window: Maximum number of bytes-in-flight that may be congestion_window: Maximum number of bytes-in-flight that may be
sent. sent.
end_of_recovery: The largest packet number sent when QUIC detects a recovery_start_time: The time when QUIC first detects a loss,
loss. When a larger packet is acknowledged, QUIC exits recovery. causing it to enter recovery. When a packet sent after this time
is acknowledged, QUIC exits recovery.
ssthresh: Slow start threshold in bytes. When the congestion window ssthresh: Slow start threshold in bytes. When the congestion window
is below ssthresh, the mode is slow start and the window grows by is below ssthresh, the mode is slow start and the window grows by
the number of bytes acknowledged. the number of bytes acknowledged.
5.8.3. Initialization 5.8.3. Initialization
At the beginning of the connection, initialize the congestion control At the beginning of the connection, initialize the congestion control
variables as follows: variables as follows:
congestion_window = kInitialWindow congestion_window = kInitialWindow
bytes_in_flight = 0 bytes_in_flight = 0
end_of_recovery = 0 recovery_start_time = 0
ssthresh = infinite ssthresh = infinite
ecn_ce_counter = 0 ecn_ce_counter = 0
5.8.4. On Packet Sent 5.8.4. On Packet Sent
Whenever a packet is sent, and it contains non-ACK frames, the packet Whenever a packet is sent, and it contains non-ACK frames, the packet
increases bytes_in_flight. increases bytes_in_flight.
OnPacketSentCC(bytes_sent): OnPacketSentCC(bytes_sent):
bytes_in_flight += bytes_sent bytes_in_flight += bytes_sent
5.8.5. On Packet Acknowledgement 5.8.5. On Packet Acknowledgement
Invoked from loss detection's OnPacketAcked and is supplied with Invoked from loss detection's OnPacketAcked and is supplied with the
acked_packet from sent_packets. acked_packet from sent_packets.
InRecovery(packet_number): InRecovery(sent_time):
return packet_number <= end_of_recovery return sent_time <= recovery_start_time
OnPacketAckedCC(acked_packet): OnPacketAckedCC(acked_packet):
// Remove from bytes_in_flight. // Remove from bytes_in_flight.
bytes_in_flight -= acked_packet.bytes bytes_in_flight -= acked_packet.size
if (InRecovery(acked_packet.packet_number)): if (InRecovery(acked_packet.time)):
// Do not increase congestion window in recovery period. // Do not increase congestion window in recovery period.
return return
if (congestion_window < ssthresh): if (congestion_window < ssthresh):
// Slow start. // Slow start.
congestion_window += acked_packet.bytes congestion_window += acked_packet.size
else: else:
// Congestion avoidance. // Congestion avoidance.
congestion_window += kMaxDatagramSize * acked_packet.bytes congestion_window += kMaxDatagramSize * acked_packet.size
/ congestion_window / congestion_window
5.8.6. On New Congestion Event 5.8.6. On New Congestion Event
Invoked from ProcessECN and OnPacketsLost when a new congestion event Invoked from ProcessECN and OnPacketsLost when a new congestion event
is detected. Starts a new recovery period and reduces the congestion is detected. May start a new recovery period and reduces the
window. congestion window.
CongestionEvent(packet_number): CongestionEvent(sent_time):
// Start a new congestion event if packet_number // Start a new congestion event if the sent time is larger
// is larger than the end of the previous recovery epoch. // than the start time of the previous recovery epoch.
if (!InRecovery(packet_number)): if (!InRecovery(sent_time)):
end_of_recovery = largest_sent_packet recovery_start_time = Now()
congestion_window *= kLossReductionFactor congestion_window *= kLossReductionFactor
congestion_window = max(congestion_window, kMinimumWindow) congestion_window = max(congestion_window, kMinimumWindow)
ssthresh = congestion_window ssthresh = congestion_window
5.8.7. Process ECN Information 5.8.7. Process ECN Information
Invoked when an ACK frame with an ECN section is received from the Invoked when an ACK frame with an ECN section is received from the
peer. peer.
ProcessECN(ack): ProcessECN(ack):
// If the ECN-CE counter reported by the peer has increased, // If the ECN-CE counter reported by the peer has increased,
// this could be a new congestion event. // this could be a new congestion event.
if (ack.ce_counter > ecn_ce_counter): if (ack.ce_counter > ecn_ce_counter):
ecn_ce_counter = ack.ce_counter ecn_ce_counter = ack.ce_counter
// Start a new congestion event if the last acknowledged // Start a new congestion event if the last acknowledged
// packet is past the end of the previous recovery epoch. // packet was sent after the start of the previous
CongestionEvent(ack.largest_acked_packet) // recovery epoch.
CongestionEvent(sent_packets[ack.largest_acked].time)
5.8.8. On Packets Lost 5.8.8. On Packets Lost
Invoked by loss detection from DetectLostPackets when new packets are Invoked by loss detection from DetectLostPackets when new packets are
detected lost. detected lost.
OnPacketsLost(lost_packets): OnPacketsLost(lost_packets):
// Remove lost packets from bytes_in_flight. // Remove lost packets from bytes_in_flight.
for (lost_packet : lost_packets): for (lost_packet : lost_packets):
bytes_in_flight -= lost_packet.bytes bytes_in_flight -= lost_packet.size
largest_lost_packet = lost_packets.last() largest_lost_packet = lost_packets.last()
// Start a new congestion epoch if the last lost packet // Start a new congestion epoch if the last lost packet
// is past the end of the previous recovery epoch. // is past the end of the previous recovery epoch.
CongestionEvent(largest_lost_packet.packet_number) CongestionEvent(largest_lost_packet.time)
5.8.9. On Retransmission Timeout Verified 5.8.9. On Retransmission Timeout Verified
QUIC decreases the congestion window to the minimum value once the QUIC decreases the congestion window to the minimum value once the
retransmission timeout has been verified and removes any packets sent retransmission timeout has been verified and removes any packets sent
before the newly acknowledged RTO packet. before the newly acknowledged RTO packet.
OnRetransmissionTimeoutVerified(packet_number) OnRetransmissionTimeoutVerified(packet_number)
congestion_window = kMinimumWindow congestion_window = kMinimumWindow
// Declare all packets prior to packet_number lost. // Declare all packets prior to packet_number lost.
for (sent_packet: sent_packets): for (sent_packet: sent_packets):
if (sent_packet.packet_number < packet_number): if (sent_packet.packet_number < packet_number):
bytes_in_flight -= sent_packet.bytes bytes_in_flight -= sent_packet.size
sent_packets.remove(sent_packet.packet_number) sent_packets.remove(sent_packet.packet_number)
6. Security Considerations 6. Security Considerations
6.1. Congestion Signals 6.1. Congestion Signals
Congestion control fundamentally involves the consumption of signals Congestion control fundamentally involves the consumption of signals
- both loss and ECN codepoints - from unauthenticated entities. On- - both loss and ECN codepoints - from unauthenticated entities. On-
path attackers can spoof or alter these signals. An attacker can path attackers can spoof or alter these signals. An attacker can
cause endpoints to reduce their sending rate by dropping packets, or cause endpoints to reduce their sending rate by dropping packets, or
skipping to change at page 29, line 22 skipping to change at page 30, line 22
This document has no IANA actions. Yet. This document has no IANA actions. Yet.
8. References 8. References
8.1. Normative References 8.1. Normative References
[QUIC-TRANSPORT] [QUIC-TRANSPORT]
Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based
Multiplexed and Secure Transport", draft-ietf-quic- Multiplexed and Secure Transport", draft-ietf-quic-
transport-16 (work in progress). transport-latest (work in progress).
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997, DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>. <https://www.rfc-editor.org/info/rfc2119>.
[RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC
2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
May 2017, <https://www.rfc-editor.org/info/rfc8174>. May 2017, <https://www.rfc-editor.org/info/rfc8174>.
skipping to change at page 30, line 38 skipping to change at page 31, line 38
and Y. Nishida, "A Conservative Loss Recovery Algorithm and Y. Nishida, "A Conservative Loss Recovery Algorithm
Based on Selective Acknowledgment (SACK) for TCP", Based on Selective Acknowledgment (SACK) for TCP",
RFC 6675, DOI 10.17487/RFC6675, August 2012, RFC 6675, DOI 10.17487/RFC6675, August 2012,
<https://www.rfc-editor.org/info/rfc6675>. <https://www.rfc-editor.org/info/rfc6675>.
[RFC6928] Chu, J., Dukkipati, N., Cheng, Y., and M. Mathis, [RFC6928] Chu, J., Dukkipati, N., Cheng, Y., and M. Mathis,
"Increasing TCP's Initial Window", RFC 6928, "Increasing TCP's Initial Window", RFC 6928,
DOI 10.17487/RFC6928, April 2013, DOI 10.17487/RFC6928, April 2013,
<https://www.rfc-editor.org/info/rfc6928>. <https://www.rfc-editor.org/info/rfc6928>.
[RFC8312] Rhee, I., Xu, L., Ha, S., Zimmermann, A., Eggert, L., and
R. Scheffenegger, "CUBIC for Fast Long-Distance Networks",
RFC 8312, DOI 10.17487/RFC8312, February 2018,
<https://www.rfc-editor.org/info/rfc8312>.
[TLP] Dukkipati, N., Cardwell, N., Cheng, Y., and M. Mathis, [TLP] Dukkipati, N., Cardwell, N., Cheng, Y., and M. Mathis,
"Tail Loss Probe (TLP): An Algorithm for Fast Recovery of "Tail Loss Probe (TLP): An Algorithm for Fast Recovery of
Tail Losses", draft-dukkipati-tcpm-tcp-loss-probe-01 (work Tail Losses", draft-dukkipati-tcpm-tcp-loss-probe-01 (work
in progress), February 2013. in progress), February 2013.
8.3. URIs 8.3. URIs
[1] https://mailarchive.ietf.org/arch/search/?email_list=quic [1] https://mailarchive.ietf.org/arch/search/?email_list=quic
[2] https://github.com/quicwg [2] https://github.com/quicwg
 End of changes. 78 change blocks. 
248 lines changed or deleted 283 lines changed or added

This html diff was produced by rfcdiff 1.44jr. The latest version is available from http://tools.ietf.org/tools/rfcdiff/