175 lines · plain
1.. SPDX-License-Identifier: GPL-2.02.. include:: <isonum.txt>3 4===============5Multi-PF Netdev6===============7 8Contents9========10 11- `Background`_12- `Overview`_13- `mlx5 implementation`_14- `Channels distribution`_15- `Observability`_16- `Steering`_17- `Mutually exclusive features`_18 19Background20==========21 22The Multi-PF NIC technology enables several CPUs within a multi-socket server to connect directly to23the network, each through its own dedicated PCIe interface. Through either a connection harness that24splits the PCIe lanes between two cards or by bifurcating a PCIe slot for a single card. This25results in eliminating the network traffic traversing over the internal bus between the sockets,26significantly reducing overhead and latency, in addition to reducing CPU utilization and increasing27network throughput.28 29Overview30========31 32The feature adds support for combining multiple PFs of the same port in a Multi-PF environment under33one netdev instance. It is implemented in the netdev layer. Lower-layer instances like pci func,34sysfs entry, and devlink are kept separate.35Passing traffic through different devices belonging to different NUMA sockets saves cross-NUMA36traffic and allows apps running on the same netdev from different NUMAs to still feel a sense of37proximity to the device and achieve improved performance.38 39mlx5 implementation40===================41 42Multi-PF or Socket-direct in mlx5 is achieved by grouping PFs together which belong to the same43NIC and has the socket-direct property enabled, once all PFs are probed, we create a single netdev44to represent all of them, symmetrically, we destroy the netdev whenever any of the PFs is removed.45 46The netdev network channels are distributed between all devices, a proper configuration would utilize47the correct close NUMA node when working on a certain app/CPU.48 49We pick one PF to be a primary (leader), and it fills a special role. The other devices50(secondaries) are disconnected from the network at the chip level (set to silent mode). In silent51mode, no south <-> north traffic flowing directly through a secondary PF. It needs the assistance of52the leader PF (east <-> west traffic) to function. All Rx/Tx traffic is steered through the primary53to/from the secondaries.54 55Currently, we limit the support to PFs only, and up to two PFs (sockets).56 57Channels distribution58=====================59 60We distribute the channels between the different PFs to achieve local NUMA node performance61on multiple NUMA nodes.62 63Each combined channel works against one specific PF, creating all its datapath queues against it. We64distribute channels to PFs in a round-robin policy.65 66::67 68 Example for 2 PFs and 5 channels:69 +--------+--------+70 | ch idx | PF idx |71 +--------+--------+72 | 0 | 0 |73 | 1 | 1 |74 | 2 | 0 |75 | 3 | 1 |76 | 4 | 0 |77 +--------+--------+78 79 80The reason we prefer round-robin is, it is less influenced by changes in the number of channels. The81mapping between a channel index and a PF is fixed, no matter how many channels the user configures.82As the channel stats are persistent across channel's closure, changing the mapping every single time83would turn the accumulative stats less representing of the channel's history.84 85This is achieved by using the correct core device instance (mdev) in each channel, instead of them86all using the same instance under "priv->mdev".87 88Observability89=============90The relation between PF, irq, napi, and queue can be observed via netlink spec::91 92 $ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/netdev.yaml --dump queue-get --json='{"ifindex": 13}'93 [{'id': 0, 'ifindex': 13, 'napi-id': 539, 'type': 'rx'},94 {'id': 1, 'ifindex': 13, 'napi-id': 540, 'type': 'rx'},95 {'id': 2, 'ifindex': 13, 'napi-id': 541, 'type': 'rx'},96 {'id': 3, 'ifindex': 13, 'napi-id': 542, 'type': 'rx'},97 {'id': 4, 'ifindex': 13, 'napi-id': 543, 'type': 'rx'},98 {'id': 0, 'ifindex': 13, 'napi-id': 539, 'type': 'tx'},99 {'id': 1, 'ifindex': 13, 'napi-id': 540, 'type': 'tx'},100 {'id': 2, 'ifindex': 13, 'napi-id': 541, 'type': 'tx'},101 {'id': 3, 'ifindex': 13, 'napi-id': 542, 'type': 'tx'},102 {'id': 4, 'ifindex': 13, 'napi-id': 543, 'type': 'tx'}]103 104 $ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/netdev.yaml --dump napi-get --json='{"ifindex": 13}'105 [{'id': 543, 'ifindex': 13, 'irq': 42},106 {'id': 542, 'ifindex': 13, 'irq': 41},107 {'id': 541, 'ifindex': 13, 'irq': 40},108 {'id': 540, 'ifindex': 13, 'irq': 39},109 {'id': 539, 'ifindex': 13, 'irq': 36}]110 111Here you can clearly observe our channels distribution policy::112 113 $ ls /proc/irq/{36,39,40,41,42}/mlx5* -d -1114 /proc/irq/36/mlx5_comp0@pci:0000:08:00.0115 /proc/irq/39/mlx5_comp0@pci:0000:09:00.0116 /proc/irq/40/mlx5_comp1@pci:0000:08:00.0117 /proc/irq/41/mlx5_comp1@pci:0000:09:00.0118 /proc/irq/42/mlx5_comp2@pci:0000:08:00.0119 120Steering121========122Secondary PFs are set to "silent" mode, meaning they are disconnected from the network.123 124In Rx, the steering tables belong to the primary PF only, and it is its role to distribute incoming125traffic to other PFs, via cross-vhca steering capabilities. Still maintain a single default RSS table,126that is capable of pointing to the receive queues of a different PF.127 128In Tx, the primary PF creates a new Tx flow table, which is aliased by the secondaries, so they can129go out to the network through it.130 131In addition, we set default XPS configuration that, based on the CPU, selects an SQ belonging to the132PF on the same node as the CPU.133 134XPS default config example:135 136NUMA node(s): 2137NUMA node0 CPU(s): 0-11138NUMA node1 CPU(s): 12-23139 140PF0 on node0, PF1 on node1.141 142- /sys/class/net/eth2/queues/tx-0/xps_cpus:000001143- /sys/class/net/eth2/queues/tx-1/xps_cpus:001000144- /sys/class/net/eth2/queues/tx-2/xps_cpus:000002145- /sys/class/net/eth2/queues/tx-3/xps_cpus:002000146- /sys/class/net/eth2/queues/tx-4/xps_cpus:000004147- /sys/class/net/eth2/queues/tx-5/xps_cpus:004000148- /sys/class/net/eth2/queues/tx-6/xps_cpus:000008149- /sys/class/net/eth2/queues/tx-7/xps_cpus:008000150- /sys/class/net/eth2/queues/tx-8/xps_cpus:000010151- /sys/class/net/eth2/queues/tx-9/xps_cpus:010000152- /sys/class/net/eth2/queues/tx-10/xps_cpus:000020153- /sys/class/net/eth2/queues/tx-11/xps_cpus:020000154- /sys/class/net/eth2/queues/tx-12/xps_cpus:000040155- /sys/class/net/eth2/queues/tx-13/xps_cpus:040000156- /sys/class/net/eth2/queues/tx-14/xps_cpus:000080157- /sys/class/net/eth2/queues/tx-15/xps_cpus:080000158- /sys/class/net/eth2/queues/tx-16/xps_cpus:000100159- /sys/class/net/eth2/queues/tx-17/xps_cpus:100000160- /sys/class/net/eth2/queues/tx-18/xps_cpus:000200161- /sys/class/net/eth2/queues/tx-19/xps_cpus:200000162- /sys/class/net/eth2/queues/tx-20/xps_cpus:000400163- /sys/class/net/eth2/queues/tx-21/xps_cpus:400000164- /sys/class/net/eth2/queues/tx-22/xps_cpus:000800165- /sys/class/net/eth2/queues/tx-23/xps_cpus:800000166 167Mutually exclusive features168===========================169 170The nature of Multi-PF, where different channels work with different PFs, conflicts with171stateful features where the state is maintained in one of the PFs.172For example, in the TLS device-offload feature, special context objects are created per connection173and maintained in the PF. Transitioning between different RQs/SQs would break the feature. Hence,174we disable this combination for now.175