1 .. SPDX-License-Identifier: GPL-2.0
2 .. include:: <isonum.txt>
3
4 ===============
5 Multi-PF Netdev
6 ===============
7
8 Contents
9 ========
10
11 - `Background`_
12 - `Overview`_
13 - `mlx5 implementation`_
14 - `Channels distribution`_
15 - `Observability`_
16 - `Steering`_
17 - `Mutually exclusive features`_
18
19 Background
20 ==========
21
22 The Multi-PF NIC technology enables several CPUs within a multi-socket server to connect directly to
23 the network, each through its own dedicated PCIe interface. Through either a connection harness that
24 splits the PCIe lanes between two cards or by bifurcating a PCIe slot for a single card. This
25 results in eliminating the network traffic traversing over the internal bus between the sockets,
26 significantly reducing overhead and latency, in addition to reducing CPU utilization and increasing
27 network throughput.
28
29 Overview
30 ========
31
32 The feature adds support for combining multiple PFs of the same port in a Multi-PF environment under
33 one netdev instance. It is implemented in the netdev layer. Lower-layer instances like pci func,
34 sysfs entry, and devlink are kept separate.
35 Passing traffic through different devices belonging to different NUMA sockets saves cross-NUMA
36 traffic and allows apps running on the same netdev from different NUMAs to still feel a sense of
37 proximity to the device and achieve improved performance.
38
39 mlx5 implementation
40 ===================
41
42 Multi-PF or Socket-direct in mlx5 is achieved by grouping PFs together which belong to the same
43 NIC and has the socket-direct property enabled, once all PFs are probed, we create a single netdev
44 to represent all of them, symmetrically, we destroy the netdev whenever any of the PFs is removed.
45
46 The netdev network channels are distributed between all devices, a proper configuration would utilize
47 the correct close NUMA node when working on a certain app/CPU.
48
49 We pick one PF to be a primary (leader), and it fills a special role. The other devices
50 (secondaries) are disconnected from the network at the chip level (set to silent mode). In silent
51 mode, no south <-> north traffic flowing directly through a secondary PF. It needs the assistance of
52 the leader PF (east <-> west traffic) to function. All Rx/Tx traffic is steered through the primary
53 to/from the secondaries.
54
55 Currently, we limit the support to PFs only, and up to two PFs (sockets).
56
57 Channels distribution
58 =====================
59
60 We distribute the channels between the different PFs to achieve local NUMA node performance
61 on multiple NUMA nodes.
62
63 Each combined channel works against one specific PF, creating all its datapath queues against it. We
64 distribute 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
80 The reason we prefer round-robin is, it is less influenced by changes in the number of channels. The
81 mapping between a channel index and a PF is fixed, no matter how many channels the user configures.
82 As the channel stats are persistent across channel's closure, changing the mapping every single time
83 would turn the accumulative stats less representing of the channel's history.
84
85 This is achieved by using the correct core device instance (mdev) in each channel, instead of them
86 all using the same instance under "priv->mdev".
87
88 Observability
89 =============
90 The 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
111 Here you can clearly observe our channels distribution policy::
112
113 $ ls /proc/irq/{36,39,40,41,42}/mlx5* -d -1
114 /proc/irq/36/mlx5_comp1@pci:0000:08:00.0
115 /proc/irq/39/mlx5_comp1@pci:0000:09:00.0
116 /proc/irq/40/mlx5_comp2@pci:0000:08:00.0
117 /proc/irq/41/mlx5_comp2@pci:0000:09:00.0
118 /proc/irq/42/mlx5_comp3@pci:0000:08:00.0
119
120 Steering
121 ========
122 Secondary PFs are set to "silent" mode, meaning they are disconnected from the network.
123
124 In Rx, the steering tables belong to the primary PF only, and it is its role to distribute incoming
125 traffic to other PFs, via cross-vhca steering capabilities. Still maintain a single default RSS table,
126 that is capable of pointing to the receive queues of a different PF.
127
128 In Tx, the primary PF creates a new Tx flow table, which is aliased by the secondaries, so they can
129 go out to the network through it.
130
131 In addition, we set default XPS configuration that, based on the CPU, selects an SQ belonging to the
132 PF on the same node as the CPU.
133
134 XPS default config example:
135
136 NUMA node(s): 2
137 NUMA node0 CPU(s): 0-11
138 NUMA node1 CPU(s): 12-23
139
140 PF0 on node0, PF1 on node1.
141
142 - /sys/class/net/eth2/queues/tx-0/xps_cpus:000001
143 - /sys/class/net/eth2/queues/tx-1/xps_cpus:001000
144 - /sys/class/net/eth2/queues/tx-2/xps_cpus:000002
145 - /sys/class/net/eth2/queues/tx-3/xps_cpus:002000
146 - /sys/class/net/eth2/queues/tx-4/xps_cpus:000004
147 - /sys/class/net/eth2/queues/tx-5/xps_cpus:004000
148 - /sys/class/net/eth2/queues/tx-6/xps_cpus:000008
149 - /sys/class/net/eth2/queues/tx-7/xps_cpus:008000
150 - /sys/class/net/eth2/queues/tx-8/xps_cpus:000010
151 - /sys/class/net/eth2/queues/tx-9/xps_cpus:010000
152 - /sys/class/net/eth2/queues/tx-10/xps_cpus:000020
153 - /sys/class/net/eth2/queues/tx-11/xps_cpus:020000
154 - /sys/class/net/eth2/queues/tx-12/xps_cpus:000040
155 - /sys/class/net/eth2/queues/tx-13/xps_cpus:040000
156 - /sys/class/net/eth2/queues/tx-14/xps_cpus:000080
157 - /sys/class/net/eth2/queues/tx-15/xps_cpus:080000
158 - /sys/class/net/eth2/queues/tx-16/xps_cpus:000100
159 - /sys/class/net/eth2/queues/tx-17/xps_cpus:100000
160 - /sys/class/net/eth2/queues/tx-18/xps_cpus:000200
161 - /sys/class/net/eth2/queues/tx-19/xps_cpus:200000
162 - /sys/class/net/eth2/queues/tx-20/xps_cpus:000400
163 - /sys/class/net/eth2/queues/tx-21/xps_cpus:400000
164 - /sys/class/net/eth2/queues/tx-22/xps_cpus:000800
165 - /sys/class/net/eth2/queues/tx-23/xps_cpus:800000
166
167 Mutually exclusive features
168 ===========================
169
170 The nature of Multi-PF, where different channels work with different PFs, conflicts with
171 stateful features where the state is maintained in one of the PFs.
172 For example, in the TLS device-offload feature, special context objects are created per connection
173 and maintained in the PF. Transitioning between different RQs/SQs would break the feature. Hence,
174 we disable this combination for now.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.