2 Copyright (C) Cavium networks Ltd. 2016.
5 Redistribution and use in source and binary forms, with or without
6 modification, are permitted provided that the following conditions
9 * Redistributions of source code must retain the above copyright
10 notice, this list of conditions and the following disclaimer.
11 * Redistributions in binary form must reproduce the above copyright
12 notice, this list of conditions and the following disclaimer in
13 the documentation and/or other materials provided with the
15 * Neither the name of Cavium networks nor the names of its
16 contributors may be used to endorse or promote products derived
17 from this software without specific prior written permission.
19 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
20 "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
21 LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
22 A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
23 OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
24 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
25 LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
29 OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
31 ThunderX NICVF Poll Mode Driver
32 ===============================
34 The ThunderX NICVF PMD (**librte_pmd_thunderx_nicvf**) provides poll mode driver
35 support for the inbuilt NIC found in the **Cavium ThunderX** SoC family
36 as well as their virtual functions (VF) in SR-IOV context.
38 More information can be found at `Cavium Networks Official Website
39 <http://www.cavium.com/ThunderX_ARM_Processors.html>`_.
44 Features of the ThunderX PMD are:
46 - Multiple queues for TX and RX
47 - Receive Side Scaling (RSS)
48 - Packet type information
52 - Port hardware statistics
54 - Link state information
55 - Scattered and gather for TX and RX
59 - Multi queue set support (up to 96 queues (12 queue sets)) per port
61 Supported ThunderX SoCs
62 -----------------------
69 - Follow the DPDK :ref:`Getting Started Guide for Linux <linux_gsg>` to setup the basic DPDK environment.
71 Pre-Installation Configuration
72 ------------------------------
77 The following options can be modified in the ``config`` file.
78 Please note that enabling debugging options may affect system performance.
80 - ``CONFIG_RTE_LIBRTE_THUNDERX_NICVF_PMD`` (default ``y``)
82 Toggle compilation of the ``librte_pmd_thunderx_nicvf`` driver.
84 - ``CONFIG_RTE_LIBRTE_THUNDERX_NICVF_DEBUG_INIT`` (default ``n``)
86 Toggle display of initialization related messages.
88 - ``CONFIG_RTE_LIBRTE_THUNDERX_NICVF_DEBUG_RX`` (default ``n``)
90 Toggle display of receive fast path run-time message
92 - ``CONFIG_RTE_LIBRTE_THUNDERX_NICVF_DEBUG_TX`` (default ``n``)
94 Toggle display of transmit fast path run-time message
96 - ``CONFIG_RTE_LIBRTE_THUNDERX_NICVF_DEBUG_DRIVER`` (default ``n``)
98 Toggle display of generic debugging messages
100 - ``CONFIG_RTE_LIBRTE_THUNDERX_NICVF_DEBUG_MBOX`` (default ``n``)
102 Toggle display of PF mailbox related run-time check messages
107 To compile the ThunderX NICVF PMD for Linux arm64 gcc target, run the
108 following “make” command:
110 .. code-block:: console
112 cd <DPDK-source-directory>
113 make config T=arm64-thunderx-linuxapp-gcc install
118 .. _thunderx_testpmd_example:
123 This section demonstrates how to launch ``testpmd`` with ThunderX NIC VF device
124 managed by ``librte_pmd_thunderx_nicvf`` in the Linux operating system.
126 #. Load ``vfio-pci`` driver:
128 .. code-block:: console
132 .. _thunderx_vfio_noiommu:
134 #. Enable **VFIO-NOIOMMU** mode (optional):
136 .. code-block:: console
138 echo 1 > /sys/module/vfio/parameters/enable_unsafe_noiommu_mode
142 **VFIO-NOIOMMU** is required only when running in VM context and should not be enabled otherwise.
143 See also :ref:`SR-IOV: Prerequisites and sample Application Notes <thunderx_sriov_example>`.
145 #. Bind the ThunderX NIC VF device to ``vfio-pci`` loaded in the previous step:
147 Setup VFIO permissions for regular users and then bind to ``vfio-pci``:
149 .. code-block:: console
151 ./usertools/dpdk-devbind.py --bind vfio-pci 0002:01:00.2
153 #. Start ``testpmd`` with basic parameters:
155 .. code-block:: console
157 ./arm64-thunderx-linuxapp-gcc/app/testpmd -l 0-3 -n 4 -w 0002:01:00.2 \
158 -- -i --disable-hw-vlan-filter --crc-strip --no-flush-rx \
163 .. code-block:: console
167 PMD: rte_nicvf_pmd_init(): librte_pmd_thunderx nicvf version 1.0
170 EAL: probe driver: 177d:11 rte_nicvf_pmd
171 EAL: using IOMMU type 1 (Type 1)
172 EAL: PCI memory mapped at 0x3ffade50000
173 EAL: Trying to map BAR 4 that contains the MSI-X table.
174 Trying offsets: 0x40000000000:0x0000, 0x10000:0x1f0000
175 EAL: PCI memory mapped at 0x3ffadc60000
176 PMD: nicvf_eth_dev_init(): nicvf: device (177d:11) 2:1:0:2
177 PMD: nicvf_eth_dev_init(): node=0 vf=1 mode=tns-bypass sqs=false
178 loopback_supported=true
179 PMD: nicvf_eth_dev_init(): Port 0 (177d:11) mac=a6:c6:d9:17:78:01
180 Interactive-mode selected
181 Configuring Port 0 (socket 0)
184 PMD: nicvf_dev_configure(): Configured ethdev port0 hwcap=0x0
185 Port 0: A6:C6:D9:17:78:01
186 Checking link statuses...
187 Port 0 Link Up - speed 10000 Mbps - full-duplex
191 .. _thunderx_sriov_example:
193 SR-IOV: Prerequisites and sample Application Notes
194 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
196 Current ThunderX NIC PF/VF kernel modules maps each physical Ethernet port
197 automatically to virtual function (VF) and presented them as PCIe-like SR-IOV device.
198 This section provides instructions to configure SR-IOV with Linux OS.
200 #. Verify PF devices capabilities using ``lspci``:
202 .. code-block:: console
208 .. code-block:: console
210 0002:01:00.0 Ethernet controller: Cavium Networks Device a01e (rev 01)
212 Capabilities: [100 v1] Alternative Routing-ID Interpretation (ARI)
214 Capabilities: [180 v1] Single Root I/O Virtualization (SR-IOV)
216 Kernel driver in use: thunder-nic
221 Unless ``thunder-nic`` driver is in use make sure your kernel config includes ``CONFIG_THUNDER_NIC_PF`` setting.
223 #. Verify VF devices capabilities and drivers using ``lspci``:
225 .. code-block:: console
231 .. code-block:: console
233 0002:01:00.1 Ethernet controller: Cavium Networks Device 0011 (rev 01)
235 Capabilities: [100 v1] Alternative Routing-ID Interpretation (ARI)
237 Kernel driver in use: thunder-nicvf
240 0002:01:00.2 Ethernet controller: Cavium Networks Device 0011 (rev 01)
242 Capabilities: [100 v1] Alternative Routing-ID Interpretation (ARI)
244 Kernel driver in use: thunder-nicvf
249 Unless ``thunder-nicvf`` driver is in use make sure your kernel config includes ``CONFIG_THUNDER_NIC_VF`` setting.
251 #. Verify PF/VF bind using ``dpdk-devbind.py``:
253 .. code-block:: console
255 ./usertools/dpdk-devbind.py --status
259 .. code-block:: console
262 0002:01:00.0 'Device a01e' if= drv=thunder-nic unused=vfio-pci
263 0002:01:00.1 'Device 0011' if=eth0 drv=thunder-nicvf unused=vfio-pci
264 0002:01:00.2 'Device 0011' if=eth1 drv=thunder-nicvf unused=vfio-pci
267 #. Load ``vfio-pci`` driver:
269 .. code-block:: console
273 #. Bind VF devices to ``vfio-pci`` using ``dpdk-devbind.py``:
275 .. code-block:: console
277 ./usertools/dpdk-devbind.py --bind vfio-pci 0002:01:00.1
278 ./usertools/dpdk-devbind.py --bind vfio-pci 0002:01:00.2
280 #. Verify VF bind using ``dpdk-devbind.py``:
282 .. code-block:: console
284 ./usertools/dpdk-devbind.py --status
288 .. code-block:: console
291 0002:01:00.1 'Device 0011' drv=vfio-pci unused=
292 0002:01:00.2 'Device 0011' drv=vfio-pci unused=
294 0002:01:00.0 'Device a01e' if= drv=thunder-nic unused=vfio-pci
297 #. Pass VF device to VM context (PCIe Passthrough):
299 The VF devices may be passed through to the guest VM using qemu or
300 virt-manager or virsh etc.
301 ``librte_pmd_thunderx_nicvf`` or ``thunder-nicvf`` should be used to bind
302 the VF devices in the guest VM in :ref:`VFIO-NOIOMMU <thunderx_vfio_noiommu>` mode.
304 Example qemu guest launch command:
306 .. code-block:: console
308 sudo qemu-system-aarch64 -name vm1 \
309 -machine virt,gic_version=3,accel=kvm,usb=off \
311 -smp 4,sockets=1,cores=8,threads=1 \
312 -nographic -nodefaults \
313 -kernel <kernel image> \
314 -append "root=/dev/vda console=ttyAMA0 rw hugepagesz=512M hugepages=3" \
315 -device vfio-pci,host=0002:01:00.1 \
316 -drive file=<rootfs.ext3>,if=none,id=disk1,format=raw \
317 -device virtio-blk-device,scsi=off,drive=disk1,id=virtio-disk1,bootindex=1 \
318 -netdev tap,id=net0,ifname=tap0,script=/etc/qemu-ifup_thunder \
319 -device virtio-net-device,netdev=net0 \
323 #. Refer to section :ref:`Running testpmd <thunderx_testpmd_example>` for instruction
324 how to launch ``testpmd`` application.
326 Multiple Queue Set per DPDK port configuration
327 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
329 There are two types of VFs:
334 Each port consists of a primary VF and n secondary VF(s). Each VF provides 8 Tx/Rx queues to a port.
335 When a given port is configured to use more than 8 queues, it requires one (or more) secondary VF.
336 Each secondary VF adds 8 additional queues to the queue set.
338 During PMD driver initialization, the primary VF's are enumerated by checking the
339 specific flag (see sqs message in DPDK boot log - sqs indicates secondary queue set).
340 They are at the beginning of VF list (the remain ones are secondary VF's).
342 The primary VFs are used as master queue sets. Secondary VFs provide
343 additional queue sets for primary ones. If a port is configured for more then
344 8 queues than it will request for additional queues from secondary VFs.
346 Secondary VFs cannot be shared between primary VFs.
348 Primary VFs are present on the beginning of the 'Network devices using kernel
349 driver' list, secondary VFs are on the remaining on the remaining part of the list.
353 The VNIC driver in the multiqueue setup works differently than other drivers like `ixgbe`.
354 We need to bind separately each specific queue set device with the ``usertools/dpdk-devbind.py`` utility.
358 Depending on the hardware used, the kernel driver sets a threshold ``vf_id``. VFs that try to attached with an id below or equal to
359 this boundary are considered primary VFs. VFs that try to attach with an id above this boundary are considered secondary VFs.
362 Example device binding
363 ~~~~~~~~~~~~~~~~~~~~~~
365 If a system has three interfaces, a total of 18 VF devices will be created
366 on a non-NUMA machine.
370 NUMA systems have 12 VFs per port and non-NUMA 6 VFs per port.
372 .. code-block:: console
374 # usertools/dpdk-devbind.py --status
376 Network devices using DPDK-compatible driver
377 ============================================
380 Network devices using kernel driver
381 ===================================
382 0000:01:10.0 'Device a026' if= drv=thunder-BGX unused=vfio-pci,uio_pci_generic
383 0000:01:10.1 'Device a026' if= drv=thunder-BGX unused=vfio-pci,uio_pci_generic
384 0002:01:00.0 'Device a01e' if= drv=thunder-nic unused=vfio-pci,uio_pci_generic
385 0002:01:00.1 'Device 0011' if=eth0 drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
386 0002:01:00.2 'Device 0011' if=eth1 drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
387 0002:01:00.3 'Device 0011' if=eth2 drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
388 0002:01:00.4 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
389 0002:01:00.5 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
390 0002:01:00.6 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
391 0002:01:00.7 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
392 0002:01:01.0 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
393 0002:01:01.1 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
394 0002:01:01.2 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
395 0002:01:01.3 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
396 0002:01:01.4 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
397 0002:01:01.5 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
398 0002:01:01.6 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
399 0002:01:01.7 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
400 0002:01:02.0 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
401 0002:01:02.1 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
402 0002:01:02.2 'Device 0011' if= drv=thunder-nicvf unused=vfio-pci,uio_pci_generic
404 Other network devices
405 =====================
406 0002:00:03.0 'Device a01f' unused=vfio-pci,uio_pci_generic
409 We want to bind two physical interfaces with 24 queues each device, we attach two primary VFs
410 and four secondary queues. In our example we choose two 10G interfaces eth1 (0002:01:00.2) and eth2 (0002:01:00.3).
411 We will choose four secondary queue sets from the ending of the list (0002:01:01.7-0002:01:02.2).
414 #. Bind two primary VFs to the ``vfio-pci`` driver:
416 .. code-block:: console
418 usertools/dpdk-devbind.py -b vfio-pci 0002:01:00.2
419 usertools/dpdk-devbind.py -b vfio-pci 0002:01:00.3
421 #. Bind four primary VFs to the ``vfio-pci`` driver:
423 .. code-block:: console
425 usertools/dpdk-devbind.py -b vfio-pci 0002:01:01.7
426 usertools/dpdk-devbind.py -b vfio-pci 0002:01:02.0
427 usertools/dpdk-devbind.py -b vfio-pci 0002:01:02.1
428 usertools/dpdk-devbind.py -b vfio-pci 0002:01:02.2
430 The nicvf thunderx driver will make use of attached secondary VFs automatically during the interface configuration stage.
438 The ThunderX SoC family NICs strip the CRC for every packets coming into the
439 host interface. So, CRC will be stripped even when the
440 ``rxmode.hw_strip_crc`` member is set to 0 in ``struct rte_eth_conf``.
442 Maximum packet length
443 ~~~~~~~~~~~~~~~~~~~~~
445 The ThunderX SoC family NICs support a maximum of a 9K jumbo frame. The value
446 is fixed and cannot be changed. So, even when the ``rxmode.max_rx_pkt_len``
447 member of ``struct rte_eth_conf`` is set to a value lower than 9200, frames
448 up to 9200 bytes can still reach the host interface.
450 Maximum packet segments
451 ~~~~~~~~~~~~~~~~~~~~~~~
453 The ThunderX SoC family NICs support up to 12 segments per packet when working
454 in scatter/gather mode. So, setting MTU will result with ``EINVAL`` when the
455 frame size does not fit in the maximum number of segments.