1 .. SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
3 =================================================
4 Mellanox ConnectX(R) mlx5 core VPI Network Driver
5 =================================================
7 Copyright (c) 2019, Mellanox Technologies LTD.
12 - `Enabling the driver and kconfig options`_
14 - `Devlink parameters`_
15 - `Devlink health reporters`_
18 Enabling the driver and kconfig options
19 ================================================
21 | mlx5 core is modular and most of the major mlx5 core driver features can be selected (compiled in/out)
22 | at build time via kernel Kconfig flags.
23 | Basic features, ethernet net device rx/tx offloads and XDP, are available with the most basic flags
24 | CONFIG_MLX5_CORE=y/m and CONFIG_MLX5_CORE_EN=y.
25 | For the list of advanced features please see below.
27 **CONFIG_MLX5_CORE=(y/m/n)** (module mlx5_core.ko)
29 | The driver can be enabled by choosing CONFIG_MLX5_CORE=y/m in kernel config.
30 | This will provide mlx5 core driver for mlx5 ulps to interface with (mlx5e, mlx5_ib).
33 **CONFIG_MLX5_CORE_EN=(y/n)**
35 | Choosing this option will allow basic ethernet netdevice support with all of the standard rx/tx offloads.
36 | mlx5e is the mlx5 ulp driver which provides netdevice kernel interface, when chosen, mlx5e will be
37 | built-in into mlx5_core.ko.
40 **CONFIG_MLX5_EN_ARFS=(y/n)**
42 | Enables Hardware-accelerated receive flow steering (arfs) support, and ntuple filtering.
43 | https://community.mellanox.com/s/article/howto-configure-arfs-on-connectx-4
46 **CONFIG_MLX5_EN_RXNFC=(y/n)**
48 | Enables ethtool receive network flow classification, which allows user defined
49 | flow rules to direct traffic into arbitrary rx queue via ethtool set/get_rxnfc API.
52 **CONFIG_MLX5_CORE_EN_DCB=(y/n)**:
54 | Enables `Data Center Bridging (DCB) Support <https://community.mellanox.com/s/article/howto-auto-config-pfc-and-ets-on-connectx-4-via-lldp-dcbx>`_.
57 **CONFIG_MLX5_MPFS=(y/n)**
59 | Ethernet Multi-Physical Function Switch (MPFS) support in ConnectX NIC.
60 | MPFs is required for when `Multi-Host <http://www.mellanox.com/page/multihost>`_ configuration is enabled to allow passing
61 | user configured unicast MAC addresses to the requesting PF.
64 **CONFIG_MLX5_ESWITCH=(y/n)**
66 | Ethernet SRIOV E-Switch support in ConnectX NIC. E-Switch provides internal SRIOV packet steering
67 | and switching for the enabled VFs and PF in two available modes:
68 | 1) `Legacy SRIOV mode (L2 mac vlan steering based) <https://community.mellanox.com/s/article/howto-configure-sr-iov-for-connectx-4-connectx-5-with-kvm--ethernet-x>`_.
69 | 2) `Switchdev mode (eswitch offloads) <https://www.mellanox.com/related-docs/prod_software/ASAP2_Hardware_Offloading_for_vSwitches_User_Manual_v4.4.pdf>`_.
72 **CONFIG_MLX5_CORE_IPOIB=(y/n)**
74 | IPoIB offloads & acceleration support.
75 | Requires CONFIG_MLX5_CORE_EN to provide an accelerated interface for the rdma
76 | IPoIB ulp netdevice.
79 **CONFIG_MLX5_FPGA=(y/n)**
81 | Build support for the Innova family of network cards by Mellanox Technologies.
82 | Innova network cards are comprised of a ConnectX chip and an FPGA chip on one board.
83 | If you select this option, the mlx5_core driver will include the Innova FPGA core and allow
84 | building sandbox-specific client drivers.
87 **CONFIG_MLX5_EN_IPSEC=(y/n)**
89 | Enables `IPSec XFRM cryptography-offload accelaration <http://www.mellanox.com/related-docs/prod_software/Mellanox_Innova_IPsec_Ethernet_Adapter_Card_User_Manual.pdf>`_.
91 **CONFIG_MLX5_EN_TLS=(y/n)**
93 | TLS cryptography-offload accelaration.
96 **CONFIG_MLX5_INFINIBAND=(y/n/m)** (module mlx5_ib.ko)
98 | Provides low-level InfiniBand/RDMA and `RoCE <https://community.mellanox.com/s/article/recommended-network-configuration-examples-for-roce-deployment>`_ support.
101 **External options** ( Choose if the corresponding mlx5 feature is required )
103 - CONFIG_PTP_1588_CLOCK: When chosen, mlx5 ptp support will be enabled
104 - CONFIG_VXLAN: When chosen, mlx5 vxaln support will be enabled.
105 - CONFIG_MLXFW: When chosen, mlx5 firmware flashing support will be enabled (via devlink and ethtool).
110 The devlink info reports the running and stored firmware versions on device.
111 It also prints the device PSID which represents the HCA board type ID.
113 User command example::
115 $ devlink dev info pci/0000:00:06.0
120 fw.psid MT_0000000009
122 fw.version 16.26.0100
124 fw.version 16.26.0100
129 flow_steering_mode: Device flow steering mode
130 ---------------------------------------------
131 The flow steering mode parameter controls the flow steering mode of the driver.
132 Two modes are supported:
133 1. 'dmfs' - Device managed flow steering.
134 2. 'smfs - Software/Driver managed flow steering.
136 In DMFS mode, the HW steering entities are created and managed through the
138 In SMFS mode, the HW steering entities are created and managed though by
139 the driver directly into Hardware without firmware intervention.
141 SMFS mode is faster and provides better rule inserstion rate compared to default DMFS mode.
143 User command examples:
145 - Set SMFS flow steering mode::
147 $ devlink dev param set pci/0000:06:00.0 name flow_steering_mode value "smfs" cmode runtime
149 - Read device flow steering mode::
151 $ devlink dev param show pci/0000:06:00.0 name flow_steering_mode
153 name flow_steering_mode type driver-specific
155 cmode runtime value smfs
158 Devlink health reporters
159 ========================
163 The tx reporter is responsible for reporting and recovering of the following two error scenarios:
166 Report on kernel tx timeout detection.
167 Recover by searching lost interrupts.
168 - TX error completion
169 Report on error tx completion.
170 Recover by flushing the TX queue and reset it.
172 TX reporter also support on demand diagnose callback, on which it provides
173 real time information of its send queues status.
175 User commands examples:
177 - Diagnose send queues status::
179 $ devlink health diagnose pci/0000:82:00.0 reporter tx
181 NOTE: This command has valid output only when interface is up, otherwise the command has empty output.
183 - Show number of tx errors indicated, number of recover flows ended successfully,
184 is autorecover enabled and graceful period from last recover::
186 $ devlink health show pci/0000:82:00.0 reporter tx
190 The rx reporter is responsible for reporting and recovering of the following two error scenarios:
192 - RX queues initialization (population) timeout
193 RX queues descriptors population on ring initialization is done in
194 napi context via triggering an irq, in case of a failure to get
195 the minimum amount of descriptors, a timeout would occur and it
196 could be recoverable by polling the EQ (Event Queue).
197 - RX completions with errors (reported by HW on interrupt context)
198 Report on rx completion error.
199 Recover (if needed) by flushing the related queue and reset it.
201 RX reporter also supports on demand diagnose callback, on which it
202 provides real time information of its receive queues status.
204 - Diagnose rx queues status, and corresponding completion queue::
206 $ devlink health diagnose pci/0000:82:00.0 reporter rx
208 NOTE: This command has valid output only when interface is up, otherwise the command has empty output.
210 - Show number of rx errors indicated, number of recover flows ended successfully,
211 is autorecover enabled and graceful period from last recover::
213 $ devlink health show pci/0000:82:00.0 reporter rx
217 The fw reporter implements diagnose and dump callbacks.
218 It follows symptoms of fw error such as fw syndrome by triggering
219 fw core dump and storing it into the dump buffer.
220 The fw reporter diagnose command can be triggered any time by the user to check
223 User commands examples:
225 - Check fw heath status::
227 $ devlink health diagnose pci/0000:82:00.0 reporter fw
229 - Read FW core dump if already stored or trigger new one::
231 $ devlink health dump show pci/0000:82:00.0 reporter fw
233 NOTE: This command can run only on the PF which has fw tracer ownership,
234 running it on other PF or any VF will return "Operation not permitted".
238 The fw fatal reporter implements dump and recover callbacks.
239 It follows fatal errors indications by CR-space dump and recover flow.
240 The CR-space dump uses vsc interface which is valid even if the FW command
241 interface is not functional, which is the case in most FW fatal errors.
242 The recover function runs recover flow which reloads the driver and triggers fw
245 User commands examples:
247 - Run fw recover flow manually::
249 $ devlink health recover pci/0000:82:00.0 reporter fw_fatal
251 - Read FW CR-space dump if already strored or trigger new one::
253 $ devlink health dump show pci/0000:82:00.1 reporter fw_fatal
255 NOTE: This command can run only on PF.
260 mlx5 driver provides internal trace points for tracking and debugging using
261 kernel tracepoints interfaces (refer to Documentation/trace/ftrase.rst).
263 For the list of support mlx5 events check /sys/kernel/debug/tracing/events/mlx5/
265 tc and eswitch offloads tracepoints:
267 - mlx5e_configure_flower: trace flower filter actions and cookies offloaded to mlx5::
269 $ echo mlx5:mlx5e_configure_flower >> /sys/kernel/debug/tracing/set_event
270 $ cat /sys/kernel/debug/tracing/trace
272 tc-6535 [019] ...1 2672.404466: mlx5e_configure_flower: cookie=0000000067874a55 actions= REDIRECT
274 - mlx5e_delete_flower: trace flower filter actions and cookies deleted from mlx5::
276 $ echo mlx5:mlx5e_delete_flower >> /sys/kernel/debug/tracing/set_event
277 $ cat /sys/kernel/debug/tracing/trace
279 tc-6569 [010] .N.1 2686.379075: mlx5e_delete_flower: cookie=0000000067874a55 actions= NULL
281 - mlx5e_stats_flower: trace flower stats request::
283 $ echo mlx5:mlx5e_stats_flower >> /sys/kernel/debug/tracing/set_event
284 $ cat /sys/kernel/debug/tracing/trace
286 tc-6546 [010] ...1 2679.704889: mlx5e_stats_flower: cookie=0000000060eb3d6a bytes=0 packets=0 lastused=4295560217
288 - mlx5e_tc_update_neigh_used_value: trace tunnel rule neigh update value offloaded to mlx5::
290 $ echo mlx5:mlx5e_tc_update_neigh_used_value >> /sys/kernel/debug/tracing/set_event
291 $ cat /sys/kernel/debug/tracing/trace
293 kworker/u48:4-8806 [009] ...1 55117.882428: mlx5e_tc_update_neigh_used_value: netdev: ens1f0 IPv4: 1.1.1.10 IPv6: ::ffff:1.1.1.10 neigh_used=1
295 - mlx5e_rep_neigh_update: trace neigh update tasks scheduled due to neigh state change events::
297 $ echo mlx5:mlx5e_rep_neigh_update >> /sys/kernel/debug/tracing/set_event
298 $ cat /sys/kernel/debug/tracing/trace
300 kworker/u48:7-2221 [009] ...1 1475.387435: mlx5e_rep_neigh_update: netdev: ens1f0 MAC: 24:8a:07:9a:17:9a IPv4: 1.1.1.10 IPv6: ::ffff:1.1.1.10 neigh_connected=1