1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2016 QLogic Corporation
3 Copyright(c) 2017 Cavium, Inc
8 The QEDE poll mode driver library (**librte_pmd_qede**) implements support
9 for **QLogic FastLinQ QL4xxxx 10G/25G/40G/50G/100G Intelligent Ethernet Adapters (IEA) and Converged Network Adapters (CNA)** family of adapters as well as SR-IOV virtual functions (VF). It is supported on
10 several standard Linux distros like RHEL, SLES, Ubuntu etc.
11 It is compile-tested under FreeBSD OS.
13 More information can be found at `QLogic Corporation's Website
14 <http://www.qlogic.com>`_.
19 - Unicast/Multicast filtering
22 - Port hardware statistics
24 - Multiple MAC address
26 - Default pause flow control
29 - Multiple Rx/Tx queues
30 - RSS (with RETA/hash table/key)
32 - Stateless checksum offloads (IPv4/IPv6/TCP/UDP)
34 - VLAN offload - Filtering and stripping
35 - N-tuple filter and flow director (limited support)
36 - NPAR (NIC Partitioning)
38 - GRE Tunneling offload
39 - GENEVE Tunneling offload
40 - VXLAN Tunneling offload
41 - MPLSoUDP Tx Tunneling offload
44 Non-supported Features
45 ----------------------
49 Co-existence considerations
50 ---------------------------
52 - QLogic FastLinQ QL4xxxx CNAs support Ethernet, RDMA, iSCSI and FCoE
53 functionalities. These functionalities are supported using
54 QLogic Linux kernel drivers qed, qede, qedr, qedi and qedf. DPDK is
55 supported on these adapters using qede PMD.
57 - When SR-IOV is not enabled on the adapter,
58 QLogic Linux kernel drivers (qed, qede, qedr, qedi and qedf) and qede
59 PMD can’t be attached to different PFs on a given QLogic FastLinQ
61 A given adapter needs to be completely used by DPDK or Linux drivers
62 Before binding DPDK driver to one or more PFs on the adapter,
63 please make sure to unbind Linux drivers from all PFs of the adapter.
64 If there are multiple adapters on the system, one or more adapters
65 can be used by DPDK driver completely and other adapters can be used
66 by Linux drivers completely.
68 - When SR-IOV is enabled on the adapter,
69 Linux kernel drivers (qed, qede, qedr, qedi and qedf) can be bound
70 to the PFs of a given adapter and either qede PMD or Linux drivers
71 (qed and qede) can be bound to the VFs of the adapter.
73 - For sharing an adapter between DPDK and Linux drivers, SRIOV needs
74 to be enabled. Bind all the PFs to Linux Drivers(qed/qede). Create
75 a VF on PFs where DPDK is desired and bind these VFs to qede_pmd.
76 Binding of PFs simultaneously to DPDK and Linux drivers on a given
77 adapter is not supported.
79 Supported QLogic Adapters
80 -------------------------
82 - QLogic FastLinQ QL4xxxx 10G/25G/40G/50G/100G Intelligent Ethernet Adapters (IEA) and Converged Network Adapters (CNA)
87 - Requires storm firmware version **8.40.33.0**. Firmware may be available
88 inbox in certain newer Linux distros under the standard directory
89 ``E.g. /lib/firmware/qed/qed_init_values-8.40.33.0.bin``.
90 If the required firmware files are not available then download it from
91 `linux-firmware git repository <http://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/tree/qed>`_.
93 - Requires the NIC be updated minimally with **8.30.x.x** Management firmware(MFW) version supported for that NIC.
94 It is highly recommended that the NIC be updated with the latest available management firmware version to get latest feature set.
95 Management Firmware and Firmware Upgrade Utility for Cavium FastLinQ(r) branded adapters can be downloaded from
96 `Driver Download Center <http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI/DefaultNewSearch.aspx>`_.
97 For downloading Firmware Upgrade Utility, select NIC category, model and Linux distro.
98 To update the management firmware, refer to the instructions in the Firmware Upgrade Utility Readme document.
99 For OEM branded adapters please follow the instruction provided by the OEM to update the Management Firmware on the NIC.
101 - SR-IOV requires Linux PF driver version **8.20.x.x** or higher.
102 If the required PF driver is not available then download it from
103 `QLogic Driver Download Center <http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI/DefaultNewSearch.aspx>`_.
104 For downloading PF driver, select adapter category, model and Linux distro.
109 - For better performance, it is recommended to use 4K or higher RX/TX rings.
114 The following options can be modified in the ``.config`` file. Please note that
115 enabling debugging options may affect system performance.
117 - ``CONFIG_RTE_LIBRTE_QEDE_PMD`` (default **y**)
119 Toggle compilation of QEDE PMD driver.
121 - ``CONFIG_RTE_LIBRTE_QEDE_DEBUG_TX`` (default **n**)
123 Toggle display of transmit fast path run-time messages.
125 - ``CONFIG_RTE_LIBRTE_QEDE_DEBUG_RX`` (default **n**)
127 Toggle display of receive fast path run-time messages.
129 - ``CONFIG_RTE_LIBRTE_QEDE_FW`` (default **""**)
131 Gives absolute path of firmware file.
132 ``Eg: "/lib/firmware/qed/qed_init_values-8.40.33.0.bin"``
133 Empty string indicates driver will pick up the firmware file
134 from the default location /lib/firmware/qed.
135 CAUTION this option is more for custom firmware, it is not
136 recommended for use under normal condition.
141 When there are multiple adapters and/or large number of Rx/Tx queues
142 configured on the adapters, the default (2560) number of memzone
143 descriptors may not be enough. Please increase the number of memzone
144 descriptors to a higher number as needed. When sufficient number of
145 memzone descriptors are not configured, user can potentially run into
148 .. code-block:: console
150 EAL: memzone_reserve_aligned_thread_unsafe(): No more room in config
152 Driver compilation and testing
153 ------------------------------
155 Refer to the document :ref:`compiling and testing a PMD for a NIC <pmd_build_and_test>`
161 QLogic FastLinQ QL4xxxx NICs has support for the following patterns and
166 .. _table_qede_supported_flow_item_types:
168 .. table:: Item types
170 +----+--------------------------------+
172 +====+================================+
173 | 1 | RTE_FLOW_ITEM_TYPE_IPV4 |
174 +----+--------------------------------+
175 | 2 | RTE_FLOW_ITEM_TYPE_IPV6 |
176 +----+--------------------------------+
177 | 3 | RTE_FLOW_ITEM_TYPE_UDP |
178 +----+--------------------------------+
179 | 4 | RTE_FLOW_ITEM_TYPE_TCP |
180 +----+--------------------------------+
184 .. _table_qede_supported_ingress_action_types:
186 .. table:: Ingress action types
188 +----+--------------------------------+
190 +====+================================+
191 | 1 | RTE_FLOW_ACTION_TYPE_QUEUE |
192 +----+--------------------------------+
193 | 2 | RTE_FLOW_ACTION_TYPE_DROP |
194 +----+--------------------------------+
196 SR-IOV: Prerequisites and Sample Application Notes
197 --------------------------------------------------
199 This section provides instructions to configure SR-IOV with Linux OS.
201 **Note**: librte_pmd_qede will be used to bind to SR-IOV VF device and Linux native kernel driver (qede) will function as SR-IOV PF driver. Requires PF driver to be 8.20.x.x or higher.
203 #. Verify SR-IOV and ARI capability is enabled on the adapter using ``lspci``:
205 .. code-block:: console
211 .. code-block:: console
214 Capabilities: [1b8 v1] Alternative Routing-ID Interpretation (ARI)
216 Capabilities: [1c0 v1] Single Root I/O Virtualization (SR-IOV)
218 Kernel driver in use: igb_uio
220 #. Load the kernel module:
222 .. code-block:: console
228 .. code-block:: console
230 systemd-udevd[4848]: renamed network interface eth0 to ens5f0
231 systemd-udevd[4848]: renamed network interface eth1 to ens5f1
233 #. Bring up the PF ports:
235 .. code-block:: console
240 #. Create VF device(s):
242 Echo the number of VFs to be created into ``"sriov_numvfs"`` sysfs entry
247 .. code-block:: console
249 echo 2 > /sys/devices/pci0000:00/0000:00:03.0/0000:81:00.0/sriov_numvfs
252 #. Assign VF MAC address:
254 Assign MAC address to the VF using iproute2 utility. The syntax is::
256 ip link set <PF iface> vf <VF id> mac <macaddr>
260 .. code-block:: console
262 ip link set ens5f0 vf 0 mac 52:54:00:2f:9d:e8
267 The VF devices may be passed through to the guest VM using ``virt-manager`` or
268 ``virsh``. QEDE PMD should be used to bind the VF devices in the guest VM
269 using the instructions from Driver compilation and testing section above.
273 (Supply ``--log-level="pmd.net.qede.driver:info`` to view informational messages):
275 Refer to the document
276 :ref:`compiling and testing a PMD for a NIC <pmd_build_and_test>` to run
277 ``testpmd`` application.
281 .. code-block:: console
283 testpmd -l 0,4-11 -n 4 -- -i --nb-cores=8 --portmask=0xf --rxd=4096 \
284 --txd=4096 --txfreet=4068 --enable-rx-cksum --rxq=4 --txq=4 \
289 EAL: PCI device 0000:84:00.0 on NUMA socket 1
290 EAL: probe driver: 1077:1634 rte_qede_pmd
291 EAL: Not managed by a supported kernel driver, skipped
292 EAL: PCI device 0000:84:00.1 on NUMA socket 1
293 EAL: probe driver: 1077:1634 rte_qede_pmd
294 EAL: Not managed by a supported kernel driver, skipped
295 EAL: PCI device 0000:88:00.0 on NUMA socket 1
296 EAL: probe driver: 1077:1656 rte_qede_pmd
297 EAL: PCI memory mapped at 0x7f738b200000
298 EAL: PCI memory mapped at 0x7f738b280000
299 EAL: PCI memory mapped at 0x7f738b300000
300 PMD: Chip details : BB1
301 PMD: Driver version : QEDE PMD 8.7.9.0_1.0.0
302 PMD: Firmware version : 8.7.7.0
303 PMD: Management firmware version : 8.7.8.0
304 PMD: Firmware file : /lib/firmware/qed/qed_init_values_zipped-8.7.7.0.bin
305 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_common_dev_init:macaddr \
308 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_tx_queue_setup:txq 0 num_desc 4096 \
309 tx_free_thresh 4068 socket 0
310 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_tx_queue_setup:txq 1 num_desc 4096 \
311 tx_free_thresh 4068 socket 0
312 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_tx_queue_setup:txq 2 num_desc 4096 \
313 tx_free_thresh 4068 socket 0
314 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_tx_queue_setup:txq 3 num_desc 4096 \
315 tx_free_thresh 4068 socket 0
316 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_rx_queue_setup:rxq 0 num_desc 4096 \
317 rx_buf_size=2148 socket 0
318 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_rx_queue_setup:rxq 1 num_desc 4096 \
319 rx_buf_size=2148 socket 0
320 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_rx_queue_setup:rxq 2 num_desc 4096 \
321 rx_buf_size=2148 socket 0
322 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_rx_queue_setup:rxq 3 num_desc 4096 \
323 rx_buf_size=2148 socket 0
324 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_dev_start:port 0
325 [QEDE PMD: (84:00.0:dpdk-port-0)]qede_dev_start:link status: down
327 Checking link statuses...
328 Port 0 Link Up - speed 25000 Mbps - full-duplex
329 Port 1 Link Up - speed 25000 Mbps - full-duplex
330 Port 2 Link Up - speed 25000 Mbps - full-duplex
331 Port 3 Link Up - speed 25000 Mbps - full-duplex