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 Environment variables
119 ~~~~~~~~~~~~~~~~~~~~~
121 - ``MLX4_INLINE_RECV_SIZE``
123 A nonzero value enables inline receive for packets up to that size. May
124 significantly improve performance in some cases but lower it in
125 others. Requires careful testing.
127 Run-time configuration
128 ~~~~~~~~~~~~~~~~~~~~~~
130 - librte_pmd_mlx4 brings kernel network interfaces up during initialization
131 because it is affected by their state. Forcing them down prevents packets
134 - **ethtool** operations on related kernel interfaces also affect the PMD.
136 - ``port`` parameter [int]
138 This parameter provides a physical port to probe and can be specified multiple
139 times for additional ports. All ports are probed by default if left
142 Kernel module parameters
143 ~~~~~~~~~~~~~~~~~~~~~~~~
145 The **mlx4_core** kernel module has several parameters that affect the
146 behavior and/or the performance of librte_pmd_mlx4. Some of them are described
149 - **num_vfs** (integer or triplet, optionally prefixed by device address
152 Create the given number of VFs on the specified devices.
154 - **log_num_mgm_entry_size** (integer)
156 Device-managed flow steering (DMFS) is required by DPDK applications. It is
157 enabled by using a negative value, the last four bits of which have a
160 - **-1**: force device-managed flow steering (DMFS).
161 - **-7**: configure optimized steering mode to improve performance with the
162 following limitation: VLAN filtering is not supported with this mode.
163 This is the recommended mode in case VLAN filter is not needed.
168 This driver relies on external libraries and kernel drivers for resources
169 allocations and initialization. The following dependencies are not part of
170 DPDK and must be installed separately:
174 User space verbs framework used by librte_pmd_mlx4. This library provides
175 a generic interface between the kernel and low-level user space drivers
178 It allows slow and privileged operations (context initialization, hardware
179 resources allocations) to be managed by the kernel and fast operations to
180 never leave user space.
184 Low-level user space driver library for Mellanox ConnectX-3 devices,
185 it is automatically loaded by libibverbs.
187 This library basically implements send/receive calls to the hardware
190 - **Kernel modules** (mlnx-ofed-kernel)
192 They provide the kernel-side verbs API and low level device drivers that
193 manage actual hardware initialization and resources sharing with user
196 Unlike most other PMDs, these modules must remain loaded and bound to
199 - mlx4_core: hardware driver managing Mellanox ConnectX-3 devices.
200 - mlx4_en: Ethernet device driver that provides kernel network interfaces.
201 - mlx4_ib: InifiniBand device driver.
202 - ib_uverbs: user space driver for verbs (entry point for libibverbs).
204 - **Firmware update**
206 Mellanox OFED releases include firmware updates for ConnectX-3 adapters.
208 Because each release provides new features, these updates must be applied to
209 match the kernel modules and libraries they come with.
213 Both libraries are BSD and GPL licensed. Linux kernel modules are GPL
216 Currently supported by DPDK:
218 - Mellanox OFED **4.1**.
219 - Firmware version **2.36.5000** and above.
221 Getting Mellanox OFED
222 ~~~~~~~~~~~~~~~~~~~~~
224 While these libraries and kernel modules are available on OpenFabrics
225 Alliance's `website <https://www.openfabrics.org/>`_ and provided by package
226 managers on most distributions, this PMD requires Ethernet extensions that
227 may not be supported at the moment (this is a work in progress).
230 <http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux_sw_drivers>`_
231 includes the necessary support and should be used in the meantime. For DPDK,
232 only libibverbs, libmlx4, mlnx-ofed-kernel packages and firmware updates are
233 required from that distribution.
237 Several versions of Mellanox OFED are available. Installing the version
238 this DPDK release was developed and tested against is strongly
239 recommended. Please check the `prerequisites`_.
244 * Mellanox(R) ConnectX(R)-3 Pro 40G MCX354A-FCC_Ax (2*40G)
249 1. Download latest Mellanox OFED. For more info check the `prerequisites`_.
251 2. Install the required libraries and kernel modules either by installing
252 only the required set, or by installing the entire Mellanox OFED:
256 .. code-block:: console
260 For SR-IOV hypervisors use:
262 .. code-block:: console
264 ./mlnxofedinstall --enable-sriov -hypervisor
266 For SR-IOV virtual machine use:
268 .. code-block:: console
270 ./mlnxofedinstall --guest
272 3. Verify the firmware is the correct one:
274 .. code-block:: console
278 4. Set all ports links to Ethernet, follow instructions on the screen:
280 .. code-block:: console
284 Or in the manual way:
286 .. code-block:: console
288 PCI=<NIC PCI address>
289 echo eth > "/sys/bus/pci/devices/$PCI/mlx4_port0"
290 echo eth > "/sys/bus/pci/devices/$PCI/mlx4_port1"
292 5. In case of bare metal or hypervisor, configure optimized steering mode
293 by adding the following line to ``/etc/modprobe.d/mlx4_core.conf``:
295 .. code-block:: console
297 options mlx4_core log_num_mgm_entry_size=-7
301 If VLAN filtering is used, set log_num_mgm_entry_size=-1.
302 Performance degradation can occur on this case.
304 6. Restart the driver:
306 .. code-block:: console
308 /etc/init.d/openibd restart
312 .. code-block:: console
314 service openibd restart
316 7. Compile DPDK and you are ready to go. See instructions on
317 :ref:`Development Kit Build System <Development_Kit_Build_System>`
322 1. Verify the optimized steering mode is configured:
324 .. code-block:: console
326 cat /sys/module/mlx4_core/parameters/log_num_mgm_entry_size
328 2. Use environment variable MLX4_INLINE_RECV_SIZE=64 to get maximum
329 performance for 64B messages.
331 3. Use the CPU near local NUMA node to which the PCIe adapter is connected,
332 for better performance. For VMs, verify that the right CPU
333 and NUMA node are pinned according to the above. Run:
335 .. code-block:: console
339 to identify the NUMA node to which the PCIe adapter is connected.
341 4. If more than one adapter is used, and root complex capabilities allow
342 to put both adapters on the same NUMA node without PCI bandwidth degradation,
343 it is recommended to locate both adapters on the same NUMA node.
344 This in order to forward packets from one to the other without
345 NUMA performance penalty.
347 5. Disable pause frames:
349 .. code-block:: console
351 ethtool -A <netdev> rx off tx off
353 6. Verify IO non-posted prefetch is disabled by default. This can be checked
354 via the BIOS configuration. Please contact you server provider for more
355 information about the settings.
359 On some machines, depends on the machine integrator, it is beneficial
360 to set the PCI max read request parameter to 1K. This can be
361 done in the following way:
363 To query the read request size use:
365 .. code-block:: console
367 setpci -s <NIC PCI address> 68.w
369 If the output is different than 3XXX, set it by:
371 .. code-block:: console
373 setpci -s <NIC PCI address> 68.w=3XXX
375 The XXX can be different on different systems. Make sure to configure
376 according to the setpci output.
381 This section demonstrates how to launch **testpmd** with Mellanox ConnectX-3
382 devices managed by librte_pmd_mlx4.
384 #. Load the kernel modules:
386 .. code-block:: console
388 modprobe -a ib_uverbs mlx4_en mlx4_core mlx4_ib
390 Alternatively if MLNX_OFED is fully installed, the following script can
393 .. code-block:: console
395 /etc/init.d/openibd restart
399 User space I/O kernel modules (uio and igb_uio) are not used and do
400 not have to be loaded.
402 #. Make sure Ethernet interfaces are in working order and linked to kernel
403 verbs. Related sysfs entries should be present:
405 .. code-block:: console
407 ls -d /sys/class/net/*/device/infiniband_verbs/uverbs* | cut -d / -f 5
411 .. code-block:: console
418 #. Optionally, retrieve their PCI bus addresses for whitelisting:
420 .. code-block:: console
423 for intf in eth2 eth3 eth4 eth5;
425 (cd "/sys/class/net/${intf}/device/" && pwd -P);
428 sed -n 's,.*/\(.*\),-w \1,p'
432 .. code-block:: console
441 There are only two distinct PCI bus addresses because the Mellanox
442 ConnectX-3 adapters installed on this system are dual port.
444 #. Request huge pages:
446 .. code-block:: console
448 echo 1024 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages/nr_hugepages
450 #. Start testpmd with basic parameters:
452 .. code-block:: console
454 testpmd -l 8-15 -n 4 -w 0000:83:00.0 -w 0000:84:00.0 -- --rxq=2 --txq=2 -i
458 .. code-block:: console
461 EAL: PCI device 0000:83:00.0 on NUMA socket 1
462 EAL: probe driver: 15b3:1007 librte_pmd_mlx4
463 PMD: librte_pmd_mlx4: PCI information matches, using device "mlx4_0" (VF: false)
464 PMD: librte_pmd_mlx4: 2 port(s) detected
465 PMD: librte_pmd_mlx4: port 1 MAC address is 00:02:c9:b5:b7:50
466 PMD: librte_pmd_mlx4: port 2 MAC address is 00:02:c9:b5:b7:51
467 EAL: PCI device 0000:84:00.0 on NUMA socket 1
468 EAL: probe driver: 15b3:1007 librte_pmd_mlx4
469 PMD: librte_pmd_mlx4: PCI information matches, using device "mlx4_1" (VF: false)
470 PMD: librte_pmd_mlx4: 2 port(s) detected
471 PMD: librte_pmd_mlx4: port 1 MAC address is 00:02:c9:b5:ba:b0
472 PMD: librte_pmd_mlx4: port 2 MAC address is 00:02:c9:b5:ba:b1
473 Interactive-mode selected
474 Configuring Port 0 (socket 0)
475 PMD: librte_pmd_mlx4: 0x867d60: TX queues number update: 0 -> 2
476 PMD: librte_pmd_mlx4: 0x867d60: RX queues number update: 0 -> 2
477 Port 0: 00:02:C9:B5:B7:50
478 Configuring Port 1 (socket 0)
479 PMD: librte_pmd_mlx4: 0x867da0: TX queues number update: 0 -> 2
480 PMD: librte_pmd_mlx4: 0x867da0: RX queues number update: 0 -> 2
481 Port 1: 00:02:C9:B5:B7:51
482 Configuring Port 2 (socket 0)
483 PMD: librte_pmd_mlx4: 0x867de0: TX queues number update: 0 -> 2
484 PMD: librte_pmd_mlx4: 0x867de0: RX queues number update: 0 -> 2
485 Port 2: 00:02:C9:B5:BA:B0
486 Configuring Port 3 (socket 0)
487 PMD: librte_pmd_mlx4: 0x867e20: TX queues number update: 0 -> 2
488 PMD: librte_pmd_mlx4: 0x867e20: RX queues number update: 0 -> 2
489 Port 3: 00:02:C9:B5:BA:B1
490 Checking link statuses...
491 Port 0 Link Up - speed 10000 Mbps - full-duplex
492 Port 1 Link Up - speed 40000 Mbps - full-duplex
493 Port 2 Link Up - speed 10000 Mbps - full-duplex
494 Port 3 Link Up - speed 40000 Mbps - full-duplex