2 Copyright 2012 6WIND S.A.
3 Copyright 2015 Mellanox
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 6WIND S.A. 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 MLX4 poll mode driver library
32 =============================
34 The MLX4 poll mode driver library (**librte_pmd_mlx4**) implements support
35 for **Mellanox ConnectX-3** and **Mellanox ConnectX-3 Pro** 10/40 Gbps adapters
36 as well as their virtual functions (VF) in SR-IOV context.
38 Information and documentation about this family of adapters can be found on
39 the `Mellanox website <http://www.mellanox.com>`_. Help is also provided by
40 the `Mellanox community <http://community.mellanox.com/welcome>`_.
42 There is also a `section dedicated to this poll mode driver
43 <http://www.mellanox.com/page/products_dyn?product_family=209&mtag=pmd_for_dpdk>`_.
47 Due to external dependencies, this driver is disabled by default. It must
48 be enabled manually by setting ``CONFIG_RTE_LIBRTE_MLX4_PMD=y`` and
51 Implementation details
52 ----------------------
54 Most Mellanox ConnectX-3 devices provide two ports but expose a single PCI
55 bus address, thus unlike most drivers, librte_pmd_mlx4 registers itself as a
56 PCI driver that allocates one Ethernet device per detected port.
58 For this reason, one cannot white/blacklist a single port without also
59 white/blacklisting the others on the same device.
61 Besides its dependency on libibverbs (that implies libmlx4 and associated
62 kernel support), librte_pmd_mlx4 relies heavily on system calls for control
63 operations such as querying/updating the MTU and flow control parameters.
65 For security reasons and robustness, this driver only deals with virtual
66 memory addresses. The way resources allocations are handled by the kernel
67 combined with hardware specifications that allow it to handle virtual memory
68 addresses directly ensure that DPDK applications cannot access random
69 physical memory (or memory that does not belong to the current process).
71 This capability allows the PMD to coexist with kernel network interfaces
72 which remain functional, although they stop receiving unicast packets as
73 long as they share the same MAC address.
75 Compiling librte_pmd_mlx4 causes DPDK to be linked against libibverbs.
80 - Multi arch support: x86_64 and POWER8.
81 - Link state information is provided.
90 These options can be modified in the ``.config`` file.
92 - ``CONFIG_RTE_LIBRTE_MLX4_PMD`` (default **n**)
94 Toggle compilation of librte_pmd_mlx4 itself.
96 - ``CONFIG_RTE_LIBRTE_MLX4_DEBUG`` (default **n**)
98 Toggle debugging code and stricter compilation flags. Enabling this option
99 adds additional run-time checks and debugging messages at the cost of
102 - ``CONFIG_RTE_LIBRTE_MLX4_DEBUG_BROKEN_VERBS`` (default **n**)
104 Mellanox OFED versions earlier than 4.2 may return false errors from
105 Verbs object destruction APIs after the device is plugged out.
106 Enabling this option replaces assertion checks that cause the program
107 to abort with harmless debugging messages as a workaround.
108 Relevant only when CONFIG_RTE_LIBRTE_MLX4_DEBUG is enabled.
110 - ``CONFIG_RTE_LIBRTE_MLX4_TX_MP_CACHE`` (default **8**)
112 Maximum number of cached memory pools (MPs) per TX queue. Each MP from
113 which buffers are to be transmitted must be associated to memory regions
114 (MRs). This is a slow operation that must be cached.
116 This value is always 1 for RX queues since they use a single MP.
118 Run-time configuration
119 ~~~~~~~~~~~~~~~~~~~~~~
121 - librte_pmd_mlx4 brings kernel network interfaces up during initialization
122 because it is affected by their state. Forcing them down prevents packets
125 - **ethtool** operations on related kernel interfaces also affect the PMD.
127 - ``port`` parameter [int]
129 This parameter provides a physical port to probe and can be specified multiple
130 times for additional ports. All ports are probed by default if left
133 Kernel module parameters
134 ~~~~~~~~~~~~~~~~~~~~~~~~
136 The **mlx4_core** kernel module has several parameters that affect the
137 behavior and/or the performance of librte_pmd_mlx4. Some of them are described
140 - **num_vfs** (integer or triplet, optionally prefixed by device address
143 Create the given number of VFs on the specified devices.
145 - **log_num_mgm_entry_size** (integer)
147 Device-managed flow steering (DMFS) is required by DPDK applications. It is
148 enabled by using a negative value, the last four bits of which have a
151 - **-1**: force device-managed flow steering (DMFS).
152 - **-7**: configure optimized steering mode to improve performance with the
153 following limitation: VLAN filtering is not supported with this mode.
154 This is the recommended mode in case VLAN filter is not needed.
159 This driver relies on external libraries and kernel drivers for resources
160 allocations and initialization. The following dependencies are not part of
161 DPDK and must be installed separately:
165 User space verbs framework used by librte_pmd_mlx4. This library provides
166 a generic interface between the kernel and low-level user space drivers
169 It allows slow and privileged operations (context initialization, hardware
170 resources allocations) to be managed by the kernel and fast operations to
171 never leave user space.
175 Low-level user space driver library for Mellanox ConnectX-3 devices,
176 it is automatically loaded by libibverbs.
178 This library basically implements send/receive calls to the hardware
181 - **Kernel modules** (mlnx-ofed-kernel)
183 They provide the kernel-side verbs API and low level device drivers that
184 manage actual hardware initialization and resources sharing with user
187 Unlike most other PMDs, these modules must remain loaded and bound to
190 - mlx4_core: hardware driver managing Mellanox ConnectX-3 devices.
191 - mlx4_en: Ethernet device driver that provides kernel network interfaces.
192 - mlx4_ib: InifiniBand device driver.
193 - ib_uverbs: user space driver for verbs (entry point for libibverbs).
195 - **Firmware update**
197 Mellanox OFED releases include firmware updates for ConnectX-3 adapters.
199 Because each release provides new features, these updates must be applied to
200 match the kernel modules and libraries they come with.
204 Both libraries are BSD and GPL licensed. Linux kernel modules are GPL
207 Currently supported by DPDK:
209 - Mellanox OFED **4.1**.
210 - Firmware version **2.36.5000** and above.
212 Getting Mellanox OFED
213 ~~~~~~~~~~~~~~~~~~~~~
215 While these libraries and kernel modules are available on OpenFabrics
216 Alliance's `website <https://www.openfabrics.org/>`_ and provided by package
217 managers on most distributions, this PMD requires Ethernet extensions that
218 may not be supported at the moment (this is a work in progress).
221 <http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux_sw_drivers>`_
222 includes the necessary support and should be used in the meantime. For DPDK,
223 only libibverbs, libmlx4, mlnx-ofed-kernel packages and firmware updates are
224 required from that distribution.
228 Several versions of Mellanox OFED are available. Installing the version
229 this DPDK release was developed and tested against is strongly
230 recommended. Please check the `prerequisites`_.
235 * Mellanox(R) ConnectX(R)-3 Pro 40G MCX354A-FCC_Ax (2*40G)
240 1. Download latest Mellanox OFED. For more info check the `prerequisites`_.
242 2. Install the required libraries and kernel modules either by installing
243 only the required set, or by installing the entire Mellanox OFED:
247 .. code-block:: console
251 For SR-IOV hypervisors use:
253 .. code-block:: console
255 ./mlnxofedinstall --enable-sriov -hypervisor
257 For SR-IOV virtual machine use:
259 .. code-block:: console
261 ./mlnxofedinstall --guest
263 3. Verify the firmware is the correct one:
265 .. code-block:: console
269 4. Set all ports links to Ethernet, follow instructions on the screen:
271 .. code-block:: console
275 Or in the manual way:
277 .. code-block:: console
279 PCI=<NIC PCI address>
280 echo eth > "/sys/bus/pci/devices/$PCI/mlx4_port0"
281 echo eth > "/sys/bus/pci/devices/$PCI/mlx4_port1"
283 5. In case of bare metal or hypervisor, configure optimized steering mode
284 by adding the following line to ``/etc/modprobe.d/mlx4_core.conf``:
286 .. code-block:: console
288 options mlx4_core log_num_mgm_entry_size=-7
292 If VLAN filtering is used, set log_num_mgm_entry_size=-1.
293 Performance degradation can occur on this case.
295 6. Restart the driver:
297 .. code-block:: console
299 /etc/init.d/openibd restart
303 .. code-block:: console
305 service openibd restart
307 7. Compile DPDK and you are ready to go. See instructions on
308 :ref:`Development Kit Build System <Development_Kit_Build_System>`
313 1. Verify the optimized steering mode is configured:
315 .. code-block:: console
317 cat /sys/module/mlx4_core/parameters/log_num_mgm_entry_size
319 2. Use the CPU near local NUMA node to which the PCIe adapter is connected,
320 for better performance. For VMs, verify that the right CPU
321 and NUMA node are pinned according to the above. Run:
323 .. code-block:: console
327 to identify the NUMA node to which the PCIe adapter is connected.
329 3. If more than one adapter is used, and root complex capabilities allow
330 to put both adapters on the same NUMA node without PCI bandwidth degradation,
331 it is recommended to locate both adapters on the same NUMA node.
332 This in order to forward packets from one to the other without
333 NUMA performance penalty.
335 4. Disable pause frames:
337 .. code-block:: console
339 ethtool -A <netdev> rx off tx off
341 5. Verify IO non-posted prefetch is disabled by default. This can be checked
342 via the BIOS configuration. Please contact you server provider for more
343 information about the settings.
347 On some machines, depends on the machine integrator, it is beneficial
348 to set the PCI max read request parameter to 1K. This can be
349 done in the following way:
351 To query the read request size use:
353 .. code-block:: console
355 setpci -s <NIC PCI address> 68.w
357 If the output is different than 3XXX, set it by:
359 .. code-block:: console
361 setpci -s <NIC PCI address> 68.w=3XXX
363 The XXX can be different on different systems. Make sure to configure
364 according to the setpci output.
369 This section demonstrates how to launch **testpmd** with Mellanox ConnectX-3
370 devices managed by librte_pmd_mlx4.
372 #. Load the kernel modules:
374 .. code-block:: console
376 modprobe -a ib_uverbs mlx4_en mlx4_core mlx4_ib
378 Alternatively if MLNX_OFED is fully installed, the following script can
381 .. code-block:: console
383 /etc/init.d/openibd restart
387 User space I/O kernel modules (uio and igb_uio) are not used and do
388 not have to be loaded.
390 #. Make sure Ethernet interfaces are in working order and linked to kernel
391 verbs. Related sysfs entries should be present:
393 .. code-block:: console
395 ls -d /sys/class/net/*/device/infiniband_verbs/uverbs* | cut -d / -f 5
399 .. code-block:: console
406 #. Optionally, retrieve their PCI bus addresses for whitelisting:
408 .. code-block:: console
411 for intf in eth2 eth3 eth4 eth5;
413 (cd "/sys/class/net/${intf}/device/" && pwd -P);
416 sed -n 's,.*/\(.*\),-w \1,p'
420 .. code-block:: console
429 There are only two distinct PCI bus addresses because the Mellanox
430 ConnectX-3 adapters installed on this system are dual port.
432 #. Request huge pages:
434 .. code-block:: console
436 echo 1024 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages/nr_hugepages
438 #. Start testpmd with basic parameters:
440 .. code-block:: console
442 testpmd -l 8-15 -n 4 -w 0000:83:00.0 -w 0000:84:00.0 -- --rxq=2 --txq=2 -i
446 .. code-block:: console
449 EAL: PCI device 0000:83:00.0 on NUMA socket 1
450 EAL: probe driver: 15b3:1007 librte_pmd_mlx4
451 PMD: librte_pmd_mlx4: PCI information matches, using device "mlx4_0" (VF: false)
452 PMD: librte_pmd_mlx4: 2 port(s) detected
453 PMD: librte_pmd_mlx4: port 1 MAC address is 00:02:c9:b5:b7:50
454 PMD: librte_pmd_mlx4: port 2 MAC address is 00:02:c9:b5:b7:51
455 EAL: PCI device 0000:84:00.0 on NUMA socket 1
456 EAL: probe driver: 15b3:1007 librte_pmd_mlx4
457 PMD: librte_pmd_mlx4: PCI information matches, using device "mlx4_1" (VF: false)
458 PMD: librte_pmd_mlx4: 2 port(s) detected
459 PMD: librte_pmd_mlx4: port 1 MAC address is 00:02:c9:b5:ba:b0
460 PMD: librte_pmd_mlx4: port 2 MAC address is 00:02:c9:b5:ba:b1
461 Interactive-mode selected
462 Configuring Port 0 (socket 0)
463 PMD: librte_pmd_mlx4: 0x867d60: TX queues number update: 0 -> 2
464 PMD: librte_pmd_mlx4: 0x867d60: RX queues number update: 0 -> 2
465 Port 0: 00:02:C9:B5:B7:50
466 Configuring Port 1 (socket 0)
467 PMD: librte_pmd_mlx4: 0x867da0: TX queues number update: 0 -> 2
468 PMD: librte_pmd_mlx4: 0x867da0: RX queues number update: 0 -> 2
469 Port 1: 00:02:C9:B5:B7:51
470 Configuring Port 2 (socket 0)
471 PMD: librte_pmd_mlx4: 0x867de0: TX queues number update: 0 -> 2
472 PMD: librte_pmd_mlx4: 0x867de0: RX queues number update: 0 -> 2
473 Port 2: 00:02:C9:B5:BA:B0
474 Configuring Port 3 (socket 0)
475 PMD: librte_pmd_mlx4: 0x867e20: TX queues number update: 0 -> 2
476 PMD: librte_pmd_mlx4: 0x867e20: RX queues number update: 0 -> 2
477 Port 3: 00:02:C9:B5:BA:B1
478 Checking link statuses...
479 Port 0 Link Up - speed 10000 Mbps - full-duplex
480 Port 1 Link Up - speed 40000 Mbps - full-duplex
481 Port 2 Link Up - speed 10000 Mbps - full-duplex
482 Port 3 Link Up - speed 40000 Mbps - full-duplex