<?xml version='1.0' encoding='utf-8'?>
<!DOCTYPE rfc [
  <!ENTITY nbsp    "&#160;">
  <!ENTITY zwsp   "&#8203;">
  <!ENTITY nbhy   "&#8209;">
  <!ENTITY wj     "&#8288;">
]>
<?xml-stylesheet type="text/xsl" href="rfc2629.xslt" ?>
<!-- generated by https://github.com/cabo/kramdown-rfc version 1.7.43 (Ruby 3.4.12) -->
<rfc xmlns:xi="http://www.w3.org/2001/XInclude" ipr="trust200902" docName="draft-sporeba-tsvwg-mobile-l4s-01" category="bcp" consensus="true" submissionType="IETF" tocInclude="true" sortRefs="true" symRefs="true" version="3">
  <!-- xml2rfc v2v3 conversion 3.35.0 -->
  <front>
    <title abbrev="Mobile L4S">Best Practices for L4S Implementation for Mobile Devices</title>
    <seriesInfo name="Internet-Draft" value="draft-sporeba-tsvwg-mobile-l4s-01"/>
    <author fullname="Sebastian Poreba">
      <organization>Google LLC</organization>
      <address>
        <email>sporeba@google.com</email>
      </address>
    </author>
    <author fullname="Lorenzo Colitti">
      <organization>Google LLC</organization>
      <address>
        <email>lorenzo@google.com</email>
      </address>
    </author>
    <author fullname="Sandeep Irlanki">
      <organization>Samsung R&amp;D</organization>
      <address>
        <email>irlanki.s@samsung.com</email>
      </address>
    </author>
    <date year="2026" month="November" day="20"/>
    <area>Transport</area>
    <workgroup>Transport Area Working Group</workgroup>
    <keyword>L4S</keyword>
    <keyword>Mobile</keyword>
    <keyword>Cellular</keyword>
    <keyword>ECN</keyword>
    <keyword>AccECN</keyword>
    <keyword>NQB</keyword>
    <abstract>
      <?line 96?>

<t>This document describes best practices for deployment of Low Latency, Low Loss, and Scalable Throughput (L4S) in mobile devices. It defines the responsibilities of the host operating system and the link-layer (modem and Wi-Fi) subsystems to ensure successful end-to-end low-latency communication.</t>
    </abstract>
    <note removeInRFC="true">
      <name>About This Document</name>
      <t>
        The latest revision of this draft can be found at <eref target="https://fridek.github.io/draft-sporeba-tsvwg-mobile-l4s/draft-sporeba-tsvwg-mobile-l4s.html"/>.
        Status information for this document may be found at <eref target="https://datatracker.ietf.org/doc/draft-sporeba-tsvwg-mobile-l4s/"/>.
      </t>
      <t>Source for this draft and an issue tracker can be found at
        <eref target="https://github.com/fridek/draft-sporeba-tsvwg-mobile-l4s"/>.</t>
    </note>
  </front>
  <middle>
    <?line 100?>

<section anchor="introduction">
      <name>Introduction</name>
      <t>L4S (Low Latency, Low Loss, Scalable Throughput) <xref target="RFC9330"/> offers a framework to significantly reduce queueing delay and loss while maintaining consistent throughput. Mobile devices often have to react to quickly changing connectivity conditions and may be subject to variable throughput and connection quality. This can cause large variations in user-perceived latency and greater bufferbloat than in other devices. Deploying L4S in a mobile ecosystem requires co-operation across multiple layers: the network stack, the host operating system (OS), link-layer drivers and firmware (e.g., Wi-Fi and cellular modem), and the network. This document outlines best current practices for the mobile device subsystems to achieve reliable, low-latency performance in the field.</t>
      <section anchor="conventions-and-definitions">
        <name>Conventions and Definitions</name>
        <t>The key words "<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL
NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>",
"<bcp14>MAY</bcp14>", and "<bcp14>OPTIONAL</bcp14>" in this document are to be interpreted as
described in BCP 14 <xref target="RFC2119"/> <xref target="RFC8174"/> when, and only when, they
appear in all capitals, as shown here.</t>
        <?line -18?>

<t>This document uses the four codepoint names of the 2-bit IP-ECN field defined in <xref target="RFC3168"/> and Section 1.3 of <xref target="RFC9332"/>: <tt>Not-ECT</tt>, <tt>ECT(0)</tt>, <tt>ECT(1)</tt>, and <tt>CE</tt>. In addition, the following collective terms are used:</t>
        <dl>
          <dt>ECN-capable packet:</dt>
          <dd>
            <t>An IP packet carrying <tt>ECT(0)</tt>, <tt>ECT(1)</tt>, or <tt>CE</tt> in the IP-ECN field (that is, any codepoint other than <tt>Not-ECT</tt>).</t>
          </dd>
          <dt>L4S identifier (and L4S packet):</dt>
          <dd>
            <t>An IP-ECN field set to <tt>ECT(1)</tt> or <tt>CE</tt>, which link-layer and network nodes use to classify packets as L4S packets without per-flow state (Section 5.1 of <xref target="RFC9331"/>, Section 2.3 of <xref target="RFC9332"/>). Note that sending endpoints mark outgoing L4S packets as <tt>ECT(1)</tt>, whereas <tt>CE</tt> indicates congestion experienced along the path.</t>
          </dd>
          <dt>NQB packet (and NQB traffic):</dt>
          <dd>
            <t>Non-queue building IP packet(s) with DSCP set to value <tt>45</tt> as defined by Section 6.1 of <xref target="RFC9956"/>.</t>
          </dd>
        </dl>
      </section>
    </section>
    <section anchor="mobile-device-architecture">
      <name>Mobile Device Architecture</name>
      <t>Modern mobile devices separate the host operating system from link-layer subsystems such as cellular modems and Wi-Fi (<xref target="fig-arch"/>). The host operating system runs applications and manages socket APIs, transport protocols, and the IP stack. Link-layer subsystems manage wireless channel acquisition and air-interface retransmissions, and maintain their own internal packet buffers in both directions. Both the host operating system and link-layer subsystems need to correctly implement behaviours on the downlink and uplink paths relevant to L4S to achieve low-latency benefits.</t>
      <figure anchor="fig-arch">
        <name>Mobile Device Data Path for Downlink and Uplink L4S</name>
        <artwork type="ascii-art"><![CDATA[
+---------------- Mobile Device ----------------+
|                                               |
| +--------------+   +------------------------+ |
| |   Host OS    |   |  Link-Layer Subsystem  | |
| |              |   |   (Wi-Fi / Cellular)   | |
| | Apps, Socket |   |                        | |
| | APIs & TCP/  |   | +--------------------+ | |   Downlink
| | UDP/IP Stack |<----| Inbound Reordering |<----------------
| |              |   | |       Buffer       | | |
| |              |   | +--------------------+ | |
| |              |   | +--------------------+ | |    Uplink
| |              |---->|   Uplink Queues    |---------------->
| |              |   | |  & Channel Access  | | |
| |              |   | +--------------------+ | |
| +--------------+   +------------------------+ |
+-----------------------------------------------+
]]></artwork>
      </figure>
    </section>
    <section anchor="downlink-requirements">
      <name>Downlink Requirements</name>
      <t>On the downlink, remote senders transmit L4S traffic toward the mobile device while bottleneck active queue management is performed in the network. The mobile device is responsible for delivering received packets without link-layer reordering delays, preserving congestion markings, and reporting accurate ECN feedback back to the sender.</t>
      <section anchor="downlink-host-os">
        <name>Host Operating System</name>
        <section anchor="udp-socket-apis-for-reading-ecn">
          <name>UDP Socket APIs for Reading ECN</name>
          <t>To enable UDP-based transport stacks (such as QUIC or real-time media over RTP) to receive L4S traffic and report congestion feedback to the sender, the OS <bcp14>MUST</bcp14> provide APIs that allow applications to read the ECN codepoints of incoming packets on a per-packet basis. This capability <bcp14>MUST</bcp14> be exposed via standard socket APIs (e.g., <tt>IP_RECVTOS</tt> and <tt>IPV6_RECVTCLASS</tt> via ancillary data) and <bcp14>MUST NOT</bcp14> be restricted by default security policies for standard application sockets. Configuring UDP sockets for ECN on common platforms is detailed in <xref target="RFC10055"/>.</t>
        </section>
        <section anchor="tcp-support-detection">
          <name>TCP Support Detection and Bootstrapping</name>
          <t>To allow L4S-capable senders (e.g., Internet servers) to take advantage of L4S, host operating systems <bcp14>MUST</bcp14> negotiate Accurate ECN (AccECN) support and verify path integrity as described in Section 3 of <xref target="RFC9768"/>.</t>
          <section anchor="per-network-detection">
            <name>Per-Network Detection and Latency Mitigation</name>
            <t>If misconfigured networks or servers drop ECN negotiation packets, and the client does not retry without ECN negotiation, TCP connections will fail. Therefore, clients <bcp14>MUST</bcp14> implement retry strategies that disable ECN negotiation, as described in Section 3.1.4.1 of <xref target="RFC9768"/>. Note that a SYN-ACK declining AccECN or ECN in the TCP header flags (falling back to Classic ECN or Non-ECN as per Section 3.1.2 of <xref target="RFC9768"/>) does not require any retries and is not considered a failure.</t>
            <t>Latency and loss are critical to mobile applications, and retry strategies dependent on retransmissions and timeouts can lead to a degraded user experience. To ensure that L4S can be safely enabled without degrading the user experience in the presence of networks or servers that drop ECN negotiations, Accurate ECN TCP options, or <tt>ECT(1)</tt>-marked packets, the client <bcp14>SHOULD</bcp14> balance implementing the retry strategies in Section 3 of <xref target="RFC9768"/> with mechanisms to reduce the latency impact of retransmissions. A possible strategy is to only attempt ECN negotiation once on the initial SYN (disabling it immediately upon retransmission). Additionally, the host <bcp14>MAY</bcp14> cache ECN negotiation or <tt>ECT(1)</tt> timeouts on a per-host, per-IP-address, or other basis so subsequent connections to servers known not to respond skip ECN negotiation.</t>
            <t>If a host is in a network that blocks all ECN negotiation regardless of destination, all TCP connections will suffer latency impact. This will degrade the user experience even if failures are cached, because any connection to a server that is not in the cache will suffer an additional round-trip delay.</t>
            <t>A host system that wants to be resilient to this <bcp14>MAY</bcp14> attempt a connectivity check to a known, L4S-supporting service. In case of check failure, the result can be used to turn off ECN negotiation attempts for a given network, represented by PLMN/APN (in carrier networks) or BSSID (in Wi-Fi networks).</t>
            <t>When maintaining per-destination or per-network caches, entries <bcp14>SHOULD</bcp14> be retired after a reasonable TTL (e.g., 7 days).</t>
          </section>
        </section>
      </section>
      <section anchor="downlink-link-layer">
        <name>Link-Layer Subsystems</name>
        <section anchor="inbound-packet-reordering">
          <name>Inbound Packet Reordering</name>
          <t>Some link layers provide strong ordering guarantees for inbound packets by assigning a link-layer sequence number to each frame and buffering incoming packets so they can be presented to the host operating system in order <xref target="I-D.ietf-intarea-reordering"/>. In cellular networks, physical-layer retransmissions frequently deliver packets out of order, while the modem buffers received packets up to a network-configurable timeout to deliver them in order. Similarly, in Wi-Fi <xref target="IEEE-802.11-2024"/>, when an individual frame within an aggregated A-MPDU burst is lost, BlockAck resequencing holds subsequent frames in that traffic identifier (including frames belonging to unrelated flows) until MAC contention and retransmission complete. Both mechanisms cause head-of-line blocking and latency spikes when air-interface transmission errors occur.</t>
          <t>Delaying received packets increases latency, which is contrary to the low-latency goals of L4S. Also, by artificially introducing delays that were not imposed by the network, it reduces the accuracy of protocol rate estimation. The additional delays do not benefit protocol stacks which are equipped to handle out-of-order packets.</t>
          <t>Link-layer subsystems <bcp14>MUST NOT</bcp14> buffer inbound ECN-capable packets in a way that imposes measurable latency to the protocol stack. Correspondingly, cellular modems and Wi-Fi subsystems <bcp14>MUST</bcp14> ensure that ECN-capable packets can be delivered out of order rather than delayed to preserve link-layer ordering, so that air-interface retransmissions do not cause head-of-line blocking stalls.</t>
          <t>Section 4.3 of <xref target="RFC9331"/> specifies that loss detection <bcp14>SHOULD</bcp14> be resilient to packet reordering. However, to enable the link-layer behaviour described above, L4S-aware protocol stacks on mobile devices <bcp14>MUST</bcp14> be prepared to receive out-of-order packets.</t>
        </section>
        <section anchor="downlink-transparency-and-bleach-prevention">
          <name>Downlink Transparency and Bleach Prevention</name>
          <t>The link layer <bcp14>MUST NOT</bcp14> modify the ECN field, DSCP values, or AccECN TCP options (172 and 174) on inbound traffic or on outbound acknowledgements.</t>
        </section>
      </section>
    </section>
    <section anchor="uplink-requirements">
      <name>Uplink Requirements</name>
      <t>On the uplink, the mobile device originates traffic and isolates low-latency packets from queue-building traffic while acquiring wireless channel resources, relying on on-path network nodes for congestion marking.</t>
      <section anchor="uplink-host-os">
        <name>Host Operating System</name>
        <section anchor="udp-socket-apis-and-packet-marking">
          <name>UDP Socket APIs and Packet Marking</name>
          <t>To enable UDP-based transport stacks (such as QUIC or real-time media over RTP) to send L4S traffic, the OS <bcp14>MUST</bcp14> provide APIs that allow applications to set the ECN codepoint <xref target="RFC9331"/> on outgoing packets on either a per-socket or per-packet basis. This capability <bcp14>MUST</bcp14> be exposed via standard socket APIs (e.g., <tt>IP_TOS</tt> and <tt>IPV6_TCLASS</tt>, as detailed in <xref target="RFC10055"/>) and <bcp14>MUST NOT</bcp14> be restricted by default security policies for standard application sockets. On some platforms, setting ECN codepoints shares calls with DSCP values, and applications might need to take steps to avoid overwriting existing DSCP or ECN markings.</t>
          <t>UDP-based transport stacks <bcp14>SHOULD</bcp14> mark L4S-capable traffic as <tt>ECT(1)</tt> and <bcp14>MUST NOT</bcp14> mark non-L4S traffic as <tt>ECT(1)</tt>. NQB traffic is considered compatible with L4S and can share the low-latency queue (Section 3.3 of <xref target="RFC9956"/>), and <bcp14>MUST NOT</bcp14> be marked as <tt>ECT(1)</tt> unless it also satisfies the Prague L4S requirements <xref target="RFC9331"/>. Link-layer handling of misbehaving traffic is described in <xref target="defense-against-misbehaving-traffic"/>.</t>
          <t>Because some network paths drop or bleach ECN-marked UDP packets, UDP-based transport stacks <bcp14>SHOULD</bcp14> validate ECN path traversal and fall back to <tt>Not-ECT</tt> if <tt>ECT(1)</tt> packets are persistently lost or bleached (for example, as specified for QUIC in Section 13.4.2 of <xref target="RFC9000"/>).</t>
        </section>
        <section anchor="tcp-sender-requirements">
          <name>TCP Sender Requirements</name>
          <t>Where the host OS TCP stack sends L4S traffic on the uplink it <bcp14>MUST</bcp14> use a compliant L4S congestion control algorithm <xref target="RFC9331"/>.</t>
        </section>
      </section>
      <section anchor="uplink-link-layer">
        <name>Link-Layer Subsystems</name>
        <t>Link-layer subsystems, such as the cellular modem and Wi-Fi, schedule and manage transmission over the physical medium. Measurement studies on mobile devices have shown that bufferbloat forms inside link-layer driver and baseband firmware buffers rather than in the host operating system networking stack <xref target="IMC16-QCUT"/>. Moreover, because grant-based cellular links rely on buffer status reports derived from modem-resident queues to request uplink radio resources, queueing cannot simply be pushed up into the host operating system without degrading throughput. Consequently, uplink queue isolation <bcp14>MUST</bcp14> be implemented within the link-layer subsystem. The host operating system <bcp14>MAY</bcp14> also maintain a separate low-latency queue where link-layer buffering is kept shallow.</t>
        <section anchor="uplink-ce-marking">
          <name>Avoiding On-Device CE Marking</name>
          <t>The host operating system and link-layer subsystems <bcp14>MUST NOT</bcp14> apply <tt>CE</tt> congestion markings to outbound uplink packets. Instead, uplink congestion marking is left to on-path network nodes, while the mobile device isolates low-latency packets from local queue-building traffic (<xref target="low-latency-queueing"/>).</t>
          <t>In 5G cellular networks supporting L4S <xref target="TS23.501"/> <xref target="TS38.300"/>, uplink congestion detection and <tt>CE</tt> marking are performed in the radio access network (base station), which schedules uplink grants across radio bearers and across devices using real-time radio and cell-load metrics. Performing <tt>CE</tt> marking on the mobile device is redundant on such networks and risks premature <tt>CE</tt> marks during normal grant-scheduling turnaround delays, while host-level active queue management is ineffective because packets drain rapidly into modem buffers to populate buffer status reports <xref target="IMC16-QCUT"/>.</t>
          <t>In shared contention-based wireless media such as Wi-Fi <xref target="IEEE-802.11-2024"/>, transmission opportunities depend on the channel acquisition across contending stations. When delay is caused by channel contention rather than an local sending rate, marking <tt>CE</tt> on the device is counterproductive. In this scenario backing off does not improve transmission window opportunities and places the mobile device at a throughput disadvantage relative to other contending stations.</t>
        </section>
        <section anchor="low-latency-queueing">
          <name>Low-Latency Queueing and Bearer Isolation</name>
          <t>On the uplink, the device needs to prevent Classic queue-building flows from delaying L4S and NQB packets.</t>
          <t>In cellular networks supporting L4S (such as 5G <xref target="TS23.501"/> <xref target="TS38.300"/>), the network can assign L4S traffic to a dedicated data radio bearer - for example, dynamically upon detecting L4S packets, or via session policy - and provision classification rules on the device so that matching uplink packets are steered onto that bearer. Because each data radio bearer maintains its own link-layer queue under network-controlled scheduling, this bearer separation isolates L4S traffic from Classic queue buildup without requiring additional local queue separation in the modem. Cellular modems <bcp14>MUST</bcp14> support matching on the 2-bit IP-ECN field in network-signalled packet filters, and the host-to-modem data path <bcp14>MUST</bcp14> ensure that backpressure from a congested Classic bearer does not cause head-of-line blocking for packets destined for a dedicated L4S bearer.</t>
          <t>Where low-latency traffic is not isolated onto a dedicated radio bearer by the network (such as on Wi-Fi links, or on shared cellular bearers where the network does not support L4S), a link-layer subsystem that supports L4S <bcp14>MUST</bcp14> provide a dedicated low-latency queue for L4S and NQB traffic. Some link-layer systems already support high-priority and low-priority queues within a shared bearer. Because such high-priority queues are typically not latency-bounded, any local low-latency queue <bcp14>MUST</bcp14> remain distinct from them:</t>
          <ul spacing="normal">
            <li>
              <t><strong>Low-Latency Queue:</strong> For L4S and NQB traffic.</t>
            </li>
            <li>
              <t><strong>High-Priority Queue:</strong> For high-priority but potentially queue-building traffic.</t>
            </li>
            <li>
              <t><strong>Low-Priority Queue:</strong> For queue-building traffic (e.g., CUBIC/Reno) and bulk data.</t>
            </li>
          </ul>
          <t>Traffic in the high-priority queue <bcp14>MUST NOT</bcp14> take unbounded precedence over the low-latency queue in a manner that introduces queueing delay to low-latency packets.</t>
          <t>The host operating system <bcp14>MAY</bcp14> also provide a low-latency queue where link-layer driver and firmware buffering is bounded (for example, to one or two transmission opportunities).</t>
          <t>Because the device does not perform <tt>CE</tt> marking (<xref target="uplink-ce-marking"/>), all low-latency queues on device <bcp14>MUST</bcp14> follow the shallow-buffer and scheduling principles in Section 5.1 of <xref target="RFC9956"/>: they <bcp14>MUST</bcp14> enforce a shallow buffer (for example, 10 ms or less at the egress rate) and <bcp14>MUST</bcp14> use a work-conserving scheduler with a bounded weight (such as Deficit Round Robin) rather than strict priority, so that un-marked traffic in the low-latency queue cannot starve Classic traffic.</t>
          <t>In Wi-Fi standards that support multi-Traffic-Identifier (multi-TID) frame aggregation (such as IEEE 802.11be), the link-layer subsystem <bcp14>SHOULD NOT</bcp14> bundle frames from the low-latency queue with traffic using longer transmission windows.</t>
        </section>
        <section anchor="packet-classification">
          <name>Packet Classification</name>
          <t>Where a local low-latency queue is used, the subsystem <bcp14>MUST</bcp14> steer uplink packets carrying the L4S identifier in the IP header to the low-latency queue, and <bcp14>SHOULD</bcp14> also support mapping based on the NQB DSCP value (45) <xref target="RFC9956"/> as an alternative or supplementary local classifier. Because DSCP markings are frequently bleached or remarked at network boundaries, ECN mapping remains the most reliable end-to-end classifier across mobile networks.</t>
          <t>Where a device uses local host metadata (such as per-application or per-user priority markings) to steer traffic into a high-priority queue, classification based on network-signalled bearer rules, the L4S identifier, and NQB markings <bcp14>MUST</bcp14> take precedence so that low-latency packets from prioritised applications are not diverted into a queue-building queue.</t>
          <t>Link-layer subsystems <bcp14>MUST NOT</bcp14> attempt to dynamically classify packets for a low-latency queue using heuristic traffic inference or Deep Packet Inspection (DPI). Apart from network-signalled bearer mapping rules (<xref target="low-latency-queueing"/>), classification <bcp14>MUST</bcp14> rely solely on explicit packet markings set by the application endpoints. This ensures compatibility with fully encrypted payloads and aligns with the end-to-end principle and permissionless innovation, as discussed in the ISP deployment observations in <xref target="I-D.livingood-low-latency-deployment"/> (which also contains details on Wi-Fi link-layer queueing considerations).</t>
        </section>
        <section anchor="defense-against-misbehaving-traffic">
          <name>Defence Against Misbehaving Traffic</name>
          <t>Applications may erroneously or maliciously mark queue-building or unresponsive traffic with the L4S identifier or NQB DSCP without implementing responsive congestion control. Because the device's low-latency queue does not apply <tt>CE</tt> marking (<xref target="uplink-ce-marking"/>), such traffic can overrun the shallow low-latency buffer and induce queueing delay or packet loss for well-behaved low-latency flows sharing the queue.</t>
          <t>To bound queueing delay and prevent starvation of Classic traffic, the low-latency queue <bcp14>MUST</bcp14> enforce a shallow buffer limit (dropping or redirecting excess packets upon overflow) and a bounded scheduling weight as described in <xref target="low-latency-queueing"/> (Section 5.1 of <xref target="RFC9956"/>). In addition, link-layer subsystems (or the host operating system) <bcp14>SHOULD</bcp14> implement traffic protection (Section 5.2 of <xref target="RFC9956"/>) to detect queue-building flows in the low-latency queue and reclassify (demote) them to a queue-building queue. Where maintaining per-flow state in modem or Wi-Fi subsystem is impractical due to hardware limitations, or simply ineffective (e.g. across encrypted VPN tunnels as described in Section 4.1 of <xref target="RFC9332"/>), implementations <bcp14>MAY</bcp14> rely on the fallback mechanisms described in <xref target="no-ce-marking-path"/>.</t>
        </section>
        <section anchor="uplink-transparency-and-bleach-prevention">
          <name>Uplink Transparency and Bleach Prevention</name>
          <t>The link layer <bcp14>MUST NOT</bcp14> modify the ECN field, DSCP values, or AccECN TCP options (172 and 174) on outbound traffic.</t>
        </section>
      </section>
      <section anchor="no-ce-marking-path">
        <name>Behaviour on a Path That Never Marks CE</name>
        <t>Sender on a mobile device might traverse a paths where no nodes leading to and including the node that is the transmission bottleneck apply <tt>CE</tt> markings.</t>
        <t>Where <tt>ECT(1)</tt> traffic shares a standard queue with Classic traffic (such as on a default cellular radio bearer without a local low-latency queue), compliant L4S congestion controllers fall back to Classic congestion control behaviour upon packet loss (Section 4.3 of <xref target="RFC9331"/>), achieving throughput and fairness comparable to Classic flows without further intervention.</t>
        <t>However, where the device steers <tt>ECT(1)</tt> packets into a shallow local low-latency queue (<xref target="low-latency-queueing"/>) and no downstream node marks <tt>CE</tt>, a capacity-seeking L4S flow will build a queue locally and suffer premature shallow-buffer drops, degrading both its own throughput and the latency of other flows sharing that queue. To mitigate this:</t>
        <ol spacing="normal" type="1"><li>
            <t>Where per-flow traffic protection is supported (Section 5.2 of <xref target="RFC9956"/>), the link-layer subsystem (or host operating system) <bcp14>SHOULD</bcp14> selectively demote queue-building flows from the low-latency queue to a Classic queue. Similarly, when a sending transport on the host operating system observes repeated loss episodes on a connection while receiving zero <tt>CE</tt> marks, it <bcp14>SHOULD</bcp14> fall back to Classic congestion control and stop marking outgoing packets as <tt>ECT(1)</tt> for that flow (Section 4.3 of <xref target="RFC9331"/>), which steers subsequent packets into the Classic queue without requiring per-flow state in link-layer hardware.</t>
          </li>
          <li>
            <t>If shallow low-latency queue overflows persist across multiple flows, or if per-flow isolation is not possible, the device <bcp14>SHOULD</bcp14> disable <tt>ECT(1)</tt> marking on a per-destination or per-network basis. To prevent overly eager disabling of L4S in networks experiencing temporary issues, this decision should be ephemeral (instead of cached as in <xref target="per-network-detection"/>).</t>
          </li>
        </ol>
      </section>
    </section>
    <section anchor="security-considerations">
      <name>Security Considerations</name>
      <t>L4S introduces potential abuse vectors where applications mark queue-building traffic as low-latency. This risk is amplified if networks without low-latency queues instead prioritise <tt>ECT(1)</tt>-marked traffic using alternative heuristics (such as bandwidth allocation multipliers) without an active end-to-end congestion-marking feedback loop, incentivising endpoints to mark <tt>ECT(1)</tt> without implementing L4S congestion control. Regardless of network behaviour, host stacks and applications <bcp14>MUST NOT</bcp14> mark traffic as <tt>ECT(1)</tt> unless using compliant L4S congestion control. As described in <xref target="defense-against-misbehaving-traffic"/>, the link-layer subsystem <bcp14>MUST</bcp14> implement queueing and scheduling safeguards to defend against starvation and unbounded queueing delay.</t>
    </section>
    <section anchor="iana-considerations">
      <name>IANA Considerations</name>
      <t>This document has no IANA actions.</t>
    </section>
  </middle>
  <back>
    <references anchor="sec-combined-references">
      <name>References</name>
      <references anchor="sec-normative-references">
        <name>Normative References</name>
        <reference anchor="RFC9331">
          <front>
            <title>The Explicit Congestion Notification (ECN) Protocol for Low Latency, Low Loss, and Scalable Throughput (L4S)</title>
            <author fullname="K. De Schepper" initials="K." surname="De Schepper"/>
            <author fullname="B. Briscoe" initials="B." role="editor" surname="Briscoe"/>
            <date month="January" year="2023"/>
            <abstract>
              <t>This specification defines the protocol to be used for a new network service called Low Latency, Low Loss, and Scalable throughput (L4S). L4S uses an Explicit Congestion Notification (ECN) scheme at the IP layer that is similar to the original (or 'Classic') ECN approach, except as specified within. L4S uses 'Scalable' congestion control, which induces much more frequent control signals from the network, and it responds to them with much more fine-grained adjustments so that very low (typically sub-millisecond on average) and consistently low queuing delay becomes possible for L4S traffic without compromising link utilization. Thus, even capacity-seeking (TCP-like) traffic can have high bandwidth and very low delay at the same time, even during periods of high traffic load.</t>
              <t>The L4S identifier defined in this document distinguishes L4S from 'Classic' (e.g., TCP-Reno-friendly) traffic. Then, network bottlenecks can be incrementally modified to distinguish and isolate existing traffic that still follows the Classic behaviour, to prevent it from degrading the low queuing delay and low loss of L4S traffic. This Experimental specification defines the rules that L4S transports and network elements need to follow, with the intention that L4S flows neither harm each other's performance nor that of Classic traffic. It also suggests open questions to be investigated during experimentation. Examples of new Active Queue Management (AQM) marking algorithms and new transports (whether TCP-like or real time) are specified separately.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9331"/>
          <seriesInfo name="DOI" value="10.17487/RFC9331"/>
        </reference>
        <reference anchor="RFC9768">
          <front>
            <title>More Accurate Explicit Congestion Notification (AccECN) Feedback in TCP</title>
            <author fullname="B. Briscoe" initials="B." surname="Briscoe"/>
            <author fullname="M. Kühlewind" initials="M." surname="Kühlewind"/>
            <author fullname="R. Scheffenegger" initials="R." surname="Scheffenegger"/>
            <date month="April" year="2026"/>
            <abstract>
              <t>Explicit Congestion Notification (ECN) is a mechanism by which network nodes can mark IP packets instead of dropping them to indicate incipient congestion to the endpoints. Receivers with an ECN-capable transport protocol feed back this information to the sender. ECN was originally specified for TCP in such a way that only one feedback signal can be transmitted per Round-Trip Time (RTT). More recently defined mechanisms like Congestion Exposure (ConEx), Data Center TCP (DCTCP), or Low Latency, Low Loss, and Scalable Throughput (L4S) need more accurate ECN feedback information whenever more than one marking is received in one RTT. This document updates the original ECN specification defined in RFC 3168 by specifying a scheme that provides more than one feedback signal per RTT in the TCP header, called More Accurate ECN (AccECN). Given TCP header space is scarce, it allocates a reserved header bit previously assigned to the ECN-nonce. It also overloads the two existing ECN flags in the TCP header. The resulting extra space is additionally exploited to feed back the IP-ECN field received during the TCP connection establishment. Supplementary feedback information can optionally be provided in two new TCP Option alternatives, which are never used on the TCP SYN. The document also specifies the treatment of this updated TCP wire protocol by middleboxes.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9768"/>
          <seriesInfo name="DOI" value="10.17487/RFC9768"/>
        </reference>
        <reference anchor="RFC9956">
          <front>
            <title>A Non-Queue-Building Per-Hop Behavior (NQB PHB) for Differentiated Services</title>
            <author fullname="G. White" initials="G." surname="White"/>
            <author fullname="T. Fossati" initials="T." surname="Fossati"/>
            <author fullname="R. Geib" initials="R." surname="Geib"/>
            <date month="May" year="2026"/>
            <abstract>
              <t>This document specifies characteristics of a Non-Queue-Building Per-Hop Behavior (NQB PHB). The NQB PHB provides a shallow-buffered, best-effort service as a complement to a Default deep-buffered, best-effort service for Internet services. The purpose of this NQB PHB is to provide a separate queue that enables smooth (i.e., non-bursty), low-data-rate, application-limited traffic microflows, to avoid the delay, delay variation and loss that would ordinarily be caused by sharing a queue with bursty, capacity-seeking traffic. This PHB is implemented without prioritization and can be implemented without rate policing, making it suitable for environments where the use of these features is restricted. The NQB PHB has been developed primarily for use by access network segments, where queuing delay and queuing loss caused by Queue-Building (QB) protocols are manifested; however, its use is not limited to such segments. In particular, the application of NQB PHB to cable broadband links, Wi-Fi links, and mobile network radio/core segments are discussed in this document. This document recommends a specific Differentiated Services Code Point (DSCP) to identify NQB microflows and updates the guidance in RFC 8325 on mapping Differentiated Services (Diffserv or DS) to IEEE 802.11 for this codepoint.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9956"/>
          <seriesInfo name="DOI" value="10.17487/RFC9956"/>
        </reference>
        <reference anchor="RFC2119">
          <front>
            <title>Key words for use in RFCs to Indicate Requirement Levels</title>
            <author fullname="S. Bradner" initials="S." surname="Bradner"/>
            <date month="March" year="1997"/>
            <abstract>
              <t>In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements.</t>
            </abstract>
          </front>
          <seriesInfo name="BCP" value="14"/>
          <seriesInfo name="RFC" value="2119"/>
          <seriesInfo name="DOI" value="10.17487/RFC2119"/>
        </reference>
        <reference anchor="RFC8174">
          <front>
            <title>Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words</title>
            <author fullname="B. Leiba" initials="B." surname="Leiba"/>
            <date month="May" year="2017"/>
            <abstract>
              <t>RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings.</t>
            </abstract>
          </front>
          <seriesInfo name="BCP" value="14"/>
          <seriesInfo name="RFC" value="8174"/>
          <seriesInfo name="DOI" value="10.17487/RFC8174"/>
        </reference>
      </references>
      <references anchor="sec-informative-references">
        <name>Informative References</name>
        <reference anchor="RFC3168">
          <front>
            <title>The Addition of Explicit Congestion Notification (ECN) to IP</title>
            <author fullname="K. Ramakrishnan" initials="K." surname="Ramakrishnan"/>
            <author fullname="S. Floyd" initials="S." surname="Floyd"/>
            <author fullname="D. Black" initials="D." surname="Black"/>
            <date month="September" year="2001"/>
            <abstract>
              <t>This memo specifies the incorporation of ECN (Explicit Congestion Notification) to TCP and IP, including ECN's use of two bits in the IP header. [STANDARDS-TRACK]</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="3168"/>
          <seriesInfo name="DOI" value="10.17487/RFC3168"/>
        </reference>
        <reference anchor="RFC9000">
          <front>
            <title>QUIC: A UDP-Based Multiplexed and Secure Transport</title>
            <author fullname="J. Iyengar" initials="J." role="editor" surname="Iyengar"/>
            <author fullname="M. Thomson" initials="M." role="editor" surname="Thomson"/>
            <date month="May" year="2021"/>
            <abstract>
              <t>This document defines the core of the QUIC transport protocol. QUIC provides applications with flow-controlled streams for structured communication, low-latency connection establishment, and network path migration. QUIC includes security measures that ensure confidentiality, integrity, and availability in a range of deployment circumstances. Accompanying documents describe the integration of TLS for key negotiation, loss detection, and an exemplary congestion control algorithm.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9000"/>
          <seriesInfo name="DOI" value="10.17487/RFC9000"/>
        </reference>
        <reference anchor="RFC9330">
          <front>
            <title>Low Latency, Low Loss, and Scalable Throughput (L4S) Internet Service: Architecture</title>
            <author fullname="B. Briscoe" initials="B." role="editor" surname="Briscoe"/>
            <author fullname="K. De Schepper" initials="K." surname="De Schepper"/>
            <author fullname="M. Bagnulo" initials="M." surname="Bagnulo"/>
            <author fullname="G. White" initials="G." surname="White"/>
            <date month="January" year="2023"/>
            <abstract>
              <t>This document describes the L4S architecture, which enables Internet applications to achieve low queuing latency, low congestion loss, and scalable throughput control. L4S is based on the insight that the root cause of queuing delay is in the capacity-seeking congestion controllers of senders, not in the queue itself. With the L4S architecture, all Internet applications could (but do not have to) transition away from congestion control algorithms that cause substantial queuing delay and instead adopt a new class of congestion controls that can seek capacity with very little queuing. These are aided by a modified form of Explicit Congestion Notification (ECN) from the network. With this new architecture, applications can have both low latency and high throughput.</t>
              <t>The architecture primarily concerns incremental deployment. It defines mechanisms that allow the new class of L4S congestion controls to coexist with 'Classic' congestion controls in a shared network. The aim is for L4S latency and throughput to be usually much better (and rarely worse) while typically not impacting Classic performance.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9330"/>
          <seriesInfo name="DOI" value="10.17487/RFC9330"/>
        </reference>
        <reference anchor="RFC9332">
          <front>
            <title>Dual-Queue Coupled Active Queue Management (AQM) for Low Latency, Low Loss, and Scalable Throughput (L4S)</title>
            <author fullname="K. De Schepper" initials="K." surname="De Schepper"/>
            <author fullname="B. Briscoe" initials="B." role="editor" surname="Briscoe"/>
            <author fullname="G. White" initials="G." surname="White"/>
            <date month="January" year="2023"/>
            <abstract>
              <t>This specification defines a framework for coupling the Active Queue Management (AQM) algorithms in two queues intended for flows with different responses to congestion. This provides a way for the Internet to transition from the scaling problems of standard TCP-Reno-friendly ('Classic') congestion controls to the family of 'Scalable' congestion controls. These are designed for consistently very low queuing latency, very low congestion loss, and scaling of per-flow throughput by using Explicit Congestion Notification (ECN) in a modified way. Until the Coupled Dual Queue (DualQ), these Scalable L4S congestion controls could only be deployed where a clean-slate environment could be arranged, such as in private data centres.</t>
              <t>This specification first explains how a Coupled DualQ works. It then gives the normative requirements that are necessary for it to work well. All this is independent of which two AQMs are used, but pseudocode examples of specific AQMs are given in appendices.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9332"/>
          <seriesInfo name="DOI" value="10.17487/RFC9332"/>
        </reference>
        <reference anchor="IEEE-802.11-2024" target="https://doi.org/10.1109/IEEESTD.2025.10979691">
          <front>
            <title>IEEE Standard for Information Technology--Telecommunications and Information Exchange between Systems - Local and Metropolitan Area Networks--Specific Requirements - Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications</title>
            <author>
              <organization>IEEE</organization>
            </author>
            <date year="2025" month="April"/>
          </front>
          <seriesInfo name="IEEE" value="Std 802.11-2024"/>
          <seriesInfo name="DOI" value="10.1109/IEEESTD.2025.10979691"/>
        </reference>
        <reference anchor="TS23.501" target="https://www.3gpp.org/dynareport/23501.htm">
          <front>
            <title>System architecture for the 5G System (5GS); Stage 2</title>
            <author>
              <organization>3GPP</organization>
            </author>
            <date>n.d.</date>
          </front>
          <seriesInfo name="3GPP" value="TS 23.501"/>
        </reference>
        <reference anchor="TS38.300" target="https://www.3gpp.org/dynareport/38300.htm">
          <front>
            <title>NR; NR and NG-RAN Overall description; Stage-2</title>
            <author>
              <organization>3GPP</organization>
            </author>
            <date>n.d.</date>
          </front>
          <seriesInfo name="3GPP" value="TS 38.300"/>
        </reference>
        <reference anchor="I-D.livingood-low-latency-deployment">
          <front>
            <title>ISP Dual Queue Networking Deployment Observations</title>
            <author fullname="Jason Livingood" initials="J." surname="Livingood">
              <organization>Comcast</organization>
            </author>
            <date day="17" month="April" year="2026"/>
            <abstract>
              <t>   The IETF's Transport and Services Working Group (TSVWG) has finalized
   experimental RFCs for Low Latency, Low Loss, Scalable Throughput
   (L4S) and new Non-Queue-Building (NQB) per hop behavior.  These
   documents describe a new architecture and protocol for deploying low
   latency networking.  Since deployment decisions are left to
   implementers, this document explores some of the implications of
   those decisions and makes suggestions that can help drive adoption
   and acceptance of L4S and NQB based on observations from the world's
   first large scale deployment.

              </t>
            </abstract>
          </front>
          <seriesInfo name="Internet-Draft" value="draft-livingood-low-latency-deployment-16"/>
        </reference>
        <reference anchor="I-D.ietf-intarea-reordering">
          <front>
            <title>Proposal for Updates to Guidance on Packet Reordering</title>
            <author fullname="Greg White" initials="G." surname="White">
              <organization>CableLabs</organization>
            </author>
            <author fullname="Ingemar Johansson" initials="I." surname="Johansson">
              <organization>Ericsson</organization>
            </author>
            <author fullname="Dibakar Das" initials="D." surname="Das">
              <organization>Intel</organization>
            </author>
            <author fullname="Chris Box" initials="C." surname="Box">
              <organization>BT Group</organization>
            </author>
            <date day="27" month="August" year="2026"/>
            <abstract>
              <t>   Several link technology standards mandate that equipment guarantee
   in-order delivery of layer 2 frames, apparently due to a belief that
   this is required by higher layer protocols.  To meet this requirement
   they implement a "resequencing" operation to restore the original
   packet order.  This can introduce delays that result in net
   degradation of performance.  Modern TCP and QUIC implementations
   support features that significantly improve their tolerance to out-
   of-order delivery.  This draft is intended to provide new information
   for layer 2 technology standards regarding the need to assure in-
   order delivery to support IETF protocols.

              </t>
            </abstract>
          </front>
          <seriesInfo name="Internet-Draft" value="draft-ietf-intarea-reordering-00"/>
        </reference>
        <reference anchor="RFC10055">
          <front>
            <title>Configuring UDP Sockets for Explicit Congestion Notification (ECN) for Common Platforms</title>
            <author fullname="M. Duke" initials="M." surname="Duke"/>
            <date month="October" year="2026"/>
            <abstract>
              <t>Explicit Congestion Notification (ECN) applies to all transport protocols in principle. However, it had limited deployment for UDP until QUIC became widely adopted. As a result, documentation of UDP socket APIs for ECN on various platforms is sparse. This document records the results of experimenting with these APIs in order to get ECN working on UDP for Chromium on Apple, Linux, and Windows platforms.</t>
              <t>This document provides a snapshot of the ECN state of affairs as supported by these platforms at the time of writing. Readers should refer to the documentation of the various platforms for more up-to-date information.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="10055"/>
          <seriesInfo name="DOI" value="10.17487/RFC10055"/>
        </reference>
        <reference anchor="IMC16-QCUT" target="https://doi.org/10.1145/2987443.2987490">
          <front>
            <title>Understanding On-device Bufferbloat for Cellular Upload</title>
            <author initials="Y." surname="Guo" fullname="Yihua Guo">
              <organization/>
            </author>
            <author initials="F." surname="Qian" fullname="Feng Qian">
              <organization/>
            </author>
            <author initials="Q. A." surname="Chen" fullname="Qi Alfred Chen">
              <organization/>
            </author>
            <author initials="Z. M." surname="Mao" fullname="Zhuoqing Morley Mao">
              <organization/>
            </author>
            <author initials="S." surname="Sen" fullname="Subhabrata Sen">
              <organization/>
            </author>
            <date year="2016" month="November"/>
          </front>
          <seriesInfo name="ACM" value="Proceedings of the 2016 Internet Measurement Conference (IMC '16)"/>
          <seriesInfo name="DOI" value="10.1145/2987443.2987490"/>
        </reference>
      </references>
    </references>
  </back>
  <!-- ##markdown-source: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-->

</rfc>
