Network Function Representors

This document describes the semantics and usage of representor netdevices, as used to control internal switching on SmartNICs. For the closely-related port representors on physical (multi-port) switches, see Documentation/networking/switchdev.rst.

Motivation

Since the mid-2010s, network cards have started offering more complex virtualisation capabilities than the legacy SR-IOV approach (with its simple MAC/VLAN-based switching model) can support. This led to a desire to offload software-defined networks (such as OpenVSwitch) to these NICs to specify the network connectivity of each function. The resulting designs are variously called SmartNICs or DPUs.

Network function representors bring the standard Linux networking stack to virtual switches and IOV devices. Just as each physical port of a Linux- controlled switch has a separate netdev, so does each virtual port of a virtual switch. When the system boots, and before any offload is configured, all packets from the virtual functions appear in the networking stack of the PF via the representors. The PF can thus always communicate freely with the virtual functions. The PF can configure standard Linux forwarding between representors, the uplink or any other netdev (routing, bridging, TC classifiers).

Thus, a representor is both a control plane object (representing the function in administrative commands) and a data plane object (one end of a virtual pipe). As a virtual link endpoint, the representor can be configured like any other netdevice; in some cases (e.g. link state) the representee will follow the representor's configuration, while in others there are separate APIs to configure the representee.

Definitions

This document uses the term "switchdev function" to refer to the PCIe function which has administrative control over the virtual switch on the device. Typically, this will be a PF, but conceivably a NIC could be configured to grant these administrative privileges instead to a VF or SF (subfunction). Depending on NIC design, a multi-port NIC might have a single switchdev function for the whole device or might have a separate virtual switch, and hence switchdev function, for each physical network port. If the NIC supports nested switching, there might be separate switchdev functions for each nested switch, in which case each switchdev function should only create representors for the ports on the (sub-)switch it directly administers.

A "representee" is the object that a representor represents. So for example in the case of a VF representor, the representee is the corresponding VF.

What does a representor do?

A representor has three main roles.

  1. It is used to configure the network connection the representee sees, e.g. link up/down, MTU, etc. For instance, bringing the representor administratively UP should cause the representee to see a link up / carrier on event.

  2. It provides the slow path for traffic which does not hit any offloaded fast-path rules in the virtual switch. Packets transmitted on the representor netdevice should be delivered to the representee; packets transmitted by the representee which fail to match any switching rule should be received on the representor netdevice. (That is, there is a virtual pipe connecting the representor to the representee, similar in concept to a veth pair.) This allows software switch implementations (such as OpenVSwitch or a Linux bridge) to forward packets between representees and the rest of the network.

  3. It acts as a handle by which switching rules (such as TC filters) can refer to the representee, allowing these rules to be offloaded.

The combination of 2) and 3) means that the behaviour (apart from performance) should be the same whether a TC filter is offloaded or not. E.g. a TC rule on a VF representor applies in software to packets received on that representor netdevice, while in hardware offload it would apply to packets transmitted by the representee VF. Conversely, a mirred egress redirect to a VF representor corresponds in hardware to delivery directly to the representee VF.

What functions should have a representor?

Essentially, for each virtual port on the device's internal switch, there should be a representor. Some vendors have chosen to omit representors for the uplink and the physical network port, which can simplify usage (the uplink netdev becomes in effect the physical port's representor) but does not generalise to devices with multiple ports or uplinks.

Thus, the following should all have representors:

  • VFs belonging to the switchdev function.

  • Other PFs on the local PCIe controller, and any VFs belonging to them.

  • PFs and VFs on external PCIe controllers on the device (e.g. for any embedded System-on-Chip within the SmartNIC).

  • PFs and VFs with other personalities, including network block devices (such as a vDPA virtio-blk PF backed by remote/distributed storage), if (and only if) their network access is implemented through a virtual switch port. [1] Note that such functions can require a representor despite the representee not having a netdev.

  • Subfunctions (SFs) belonging to any of the above PFs or VFs, if they have their own port on the switch (as opposed to using their parent PF's port).

  • Any accelerators or plugins on the device whose interface to the network is through a virtual switch port, even if they do not have a corresponding PCIe PF or VF.

This allows the entire switching behaviour of the NIC to be controlled through representor TC rules.

It is a common misunderstanding to conflate virtual ports with PCIe virtual functions or their netdevs. While in simple cases there will be a 1:1 correspondence between VF netdevices and VF representors, more advanced device configurations may not follow this. A PCIe function which does not have network access through the internal switch (not even indirectly through the hardware implementation of whatever services the function provides) should not have a representor (even if it has a netdev). Such a function has no switch virtual port for the representor to configure or to be the other end of the virtual pipe. The representor represents the virtual port, not the PCIe function nor the 'end user' netdevice.

How are representors created?

The driver instance attached to the switchdev function should, for each virtual port on the switch, create a pure-software netdevice which has some form of in-kernel reference to the switchdev function's own netdevice or driver private data (netdev_priv()). This may be by enumerating ports at probe time, reacting dynamically to the creation and destruction of ports at run time, or a combination of the two.

The operations of the representor netdevice will generally involve acting through the switchdev function. For example, ndo_start_xmit() might send the packet through a hardware TX queue attached to the switchdev function, with either packet metadata or queue configuration marking it for delivery to the representee.

How are representors identified?

The representor netdevice should not directly refer to a PCIe device (e.g. through net_dev->dev.parent / SET_NETDEV_DEV()), either of the representee or of the switchdev function. Instead, the driver should use the SET_NETDEV_DEVLINK_PORT macro to assign a devlink port instance to the netdevice before registering the netdevice; the kernel uses the devlink port to provide the phys_switch_id and phys_port_name sysfs nodes. (Some legacy drivers implement ndo_get_port_parent_id() and ndo_get_phys_port_name() directly, but this is deprecated.) See Documentation/networking/devlink/devlink-port.rst for the details of this API.

It is expected that userland will use this information (e.g. through udev rules) to construct an appropriately informative name or alias for the netdevice. For instance if the switchdev function is eth4 then a representor with a phys_port_name of p0pf1vf2 might be renamed eth4pf1vf2rep.

There are as yet no established conventions for naming representors which do not correspond to PCIe functions (e.g. accelerators and plugins).

How do representors interact with TC rules?

Any TC rule on a representor applies (in software TC) to packets received by that representor netdevice. Thus, if the delivery part of the rule corresponds to another port on the virtual switch, the driver may choose to offload it to hardware, applying it to packets transmitted by the representee.

Similarly, since a TC mirred egress action targeting the representor would (in software) send the packet through the representor (and thus indirectly deliver it to the representee), hardware offload should interpret this as delivery to the representee.

As a simple example, if PORT_DEV is the physical port representor and REP_DEV is a VF representor, the following rules:

tc filter add dev $REP_DEV parent ffff: protocol ipv4 flower \
    action mirred egress redirect dev $PORT_DEV
tc filter add dev $PORT_DEV parent ffff: protocol ipv4 flower skip_sw \
    action mirred egress mirror dev $REP_DEV

would mean that all IPv4 packets from the VF are sent out the physical port, and all IPv4 packets received on the physical port are delivered to the VF in addition to PORT_DEV. (Note that without skip_sw on the second rule, the VF would get two copies, as the packet reception on PORT_DEV would trigger the TC rule again and mirror the packet to REP_DEV.)

On devices without separate port and uplink representors, PORT_DEV would instead be the switchdev function's own uplink netdevice.

Of course the rules can (if supported by the NIC) include packet-modifying actions (e.g. VLAN push/pop), which should be performed by the virtual switch.

Tunnel encapsulation and decapsulation are rather more complicated, as they involve a third netdevice (a tunnel netdev operating in metadata mode, such as a VxLAN device created with ip link add vxlan0 type vxlan external) and require an IP address to be bound to the underlay device (e.g. switchdev function uplink netdev or port representor). TC rules such as:

tc filter add dev $REP_DEV parent ffff: flower \
    action tunnel_key set id $VNI src_ip $LOCAL_IP dst_ip $REMOTE_IP \
                          dst_port 4789 \
    action mirred egress redirect dev vxlan0
tc filter add dev vxlan0 parent ffff: flower enc_src_ip $REMOTE_IP \
    enc_dst_ip $LOCAL_IP enc_key_id $VNI enc_dst_port 4789 \
    action tunnel_key unset action mirred egress redirect dev $REP_DEV

where LOCAL_IP is an IP address bound to PORT_DEV, and REMOTE_IP is another IP address on the same subnet, mean that packets sent by the VF should be VxLAN encapsulated and sent out the physical port (the driver has to deduce this by a route lookup of LOCAL_IP leading to PORT_DEV, and also perform an ARP/neighbour table lookup to find the MAC addresses to use in the outer Ethernet frame), while UDP packets received on the physical port with UDP port 4789 should be parsed as VxLAN and, if their VSID matches $VNI, decapsulated and forwarded to the VF.

If this all seems complicated, just remember the 'golden rule' of TC offload: the hardware should ensure the same final results as if the packets were processed through the slow path, traversed software TC (except ignoring any skip_hw rules and applying any skip_sw rules) and were transmitted or received through the representor netdevices.

Configuring the representee's MAC

The representee's link state is controlled through the representor. Setting the representor administratively UP or DOWN should cause carrier ON or OFF at the representee.

Setting an MTU on the representor should cause that same MTU to be reported to the representee. (On hardware that allows configuring separate and distinct MTU and MRU values, the representor MTU should correspond to the representee's MRU and vice-versa.)

Currently there is no way to use the representor to set the station permanent MAC address of the representee; other methods available to do this include: