mlx4.rst revision 97f17497
1..  BSD LICENSE
2    Copyright 2012-2015 6WIND S.A.
3
4    Redistribution and use in source and binary forms, with or without
5    modification, are permitted provided that the following conditions
6    are met:
7
8    * Redistributions of source code must retain the above copyright
9    notice, this list of conditions and the following disclaimer.
10    * Redistributions in binary form must reproduce the above copyright
11    notice, this list of conditions and the following disclaimer in
12    the documentation and/or other materials provided with the
13    distribution.
14    * Neither the name of 6WIND S.A. nor the names of its
15    contributors may be used to endorse or promote products derived
16    from this software without specific prior written permission.
17
18    THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
19    "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
20    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
21    A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
22    OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
23    SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
24    LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
25    DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
26    THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
27    (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
28    OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
29
30MLX4 poll mode driver library
31=============================
32
33The MLX4 poll mode driver library (**librte_pmd_mlx4**) implements support
34for **Mellanox ConnectX-3** and **Mellanox ConnectX-3 Pro** 10/40 Gbps adapters
35as well as their virtual functions (VF) in SR-IOV context.
36
37Information and documentation about this family of adapters can be found on
38the `Mellanox website <http://www.mellanox.com>`_. Help is also provided by
39the `Mellanox community <http://community.mellanox.com/welcome>`_.
40
41There is also a `section dedicated to this poll mode driver
42<http://www.mellanox.com/page/products_dyn?product_family=209&mtag=pmd_for_dpdk>`_.
43
44.. note::
45
46   Due to external dependencies, this driver is disabled by default. It must
47   be enabled manually by setting ``CONFIG_RTE_LIBRTE_MLX4_PMD=y`` and
48   recompiling DPDK.
49
50Implementation details
51----------------------
52
53Most Mellanox ConnectX-3 devices provide two ports but expose a single PCI
54bus address, thus unlike most drivers, librte_pmd_mlx4 registers itself as a
55PCI driver that allocates one Ethernet device per detected port.
56
57For this reason, one cannot white/blacklist a single port without also
58white/blacklisting the others on the same device.
59
60Besides its dependency on libibverbs (that implies libmlx4 and associated
61kernel support), librte_pmd_mlx4 relies heavily on system calls for control
62operations such as querying/updating the MTU and flow control parameters.
63
64For security reasons and robustness, this driver only deals with virtual
65memory addresses. The way resources allocations are handled by the kernel
66combined with hardware specifications that allow it to handle virtual memory
67addresses directly ensure that DPDK applications cannot access random
68physical memory (or memory that does not belong to the current process).
69
70This capability allows the PMD to coexist with kernel network interfaces
71which remain functional, although they stop receiving unicast packets as
72long as they share the same MAC address.
73
74Compiling librte_pmd_mlx4 causes DPDK to be linked against libibverbs.
75
76Features
77--------
78
79- RSS, also known as RCA, is supported. In this mode the number of
80  configured RX queues must be a power of two.
81- VLAN filtering is supported.
82- Link state information is provided.
83- Promiscuous mode is supported.
84- All multicast mode is supported.
85- Multiple MAC addresses (unicast, multicast) can be configured.
86- Scattered packets are supported for TX and RX.
87- Inner L3/L4 (IP, TCP and UDP) TX/RX checksum offloading and validation.
88- Outer L3 (IP) TX/RX checksum offloading and validation for VXLAN frames.
89- Secondary process TX is supported.
90
91Limitations
92-----------
93
94- RSS hash key cannot be modified.
95- RSS RETA cannot be configured
96- RSS always includes L3 (IPv4/IPv6) and L4 (UDP/TCP). They cannot be
97  dissociated.
98- Hardware counters are not implemented (they are software counters).
99- Secondary process RX is not supported.
100
101Configuration
102-------------
103
104Compilation options
105~~~~~~~~~~~~~~~~~~~
106
107These options can be modified in the ``.config`` file.
108
109- ``CONFIG_RTE_LIBRTE_MLX4_PMD`` (default **n**)
110
111  Toggle compilation of librte_pmd_mlx4 itself.
112
113- ``CONFIG_RTE_LIBRTE_MLX4_DEBUG`` (default **n**)
114
115  Toggle debugging code and stricter compilation flags. Enabling this option
116  adds additional run-time checks and debugging messages at the cost of
117  lower performance.
118
119- ``CONFIG_RTE_LIBRTE_MLX4_SGE_WR_N`` (default **4**)
120
121  Number of scatter/gather elements (SGEs) per work request (WR). Lowering
122  this number improves performance but also limits the ability to receive
123  scattered packets (packets that do not fit a single mbuf). The default
124  value is a safe tradeoff.
125
126- ``CONFIG_RTE_LIBRTE_MLX4_MAX_INLINE`` (default **0**)
127
128  Amount of data to be inlined during TX operations. Improves latency but
129  lowers throughput.
130
131- ``CONFIG_RTE_LIBRTE_MLX4_TX_MP_CACHE`` (default **8**)
132
133  Maximum number of cached memory pools (MPs) per TX queue. Each MP from
134  which buffers are to be transmitted must be associated to memory regions
135  (MRs). This is a slow operation that must be cached.
136
137  This value is always 1 for RX queues since they use a single MP.
138
139- ``CONFIG_RTE_LIBRTE_MLX4_SOFT_COUNTERS`` (default **1**)
140
141  Toggle software counters. No counters are available if this option is
142  disabled since hardware counters are not supported.
143
144Environment variables
145~~~~~~~~~~~~~~~~~~~~~
146
147- ``MLX4_INLINE_RECV_SIZE``
148
149  A nonzero value enables inline receive for packets up to that size. May
150  significantly improve performance in some cases but lower it in
151  others. Requires careful testing.
152
153Run-time configuration
154~~~~~~~~~~~~~~~~~~~~~~
155
156- The only constraint when RSS mode is requested is to make sure the number
157  of RX queues is a power of two. This is a hardware requirement.
158
159- librte_pmd_mlx4 brings kernel network interfaces up during initialization
160  because it is affected by their state. Forcing them down prevents packets
161  reception.
162
163- **ethtool** operations on related kernel interfaces also affect the PMD.
164
165Kernel module parameters
166~~~~~~~~~~~~~~~~~~~~~~~~
167
168The **mlx4_core** kernel module has several parameters that affect the
169behavior and/or the performance of librte_pmd_mlx4. Some of them are described
170below.
171
172- **num_vfs** (integer or triplet, optionally prefixed by device address
173  strings)
174
175  Create the given number of VFs on the specified devices.
176
177- **log_num_mgm_entry_size** (integer)
178
179  Device-managed flow steering (DMFS) is required by DPDK applications. It is
180  enabled by using a negative value, the last four bits of which have a
181  special meaning.
182
183  - **-1**: force device-managed flow steering (DMFS).
184  - **-7**: configure optimized steering mode to improve performance with the
185    following limitation: VLAN filtering is not supported with this mode.
186    This is the recommended mode in case VLAN filter is not needed.
187
188Prerequisites
189-------------
190
191This driver relies on external libraries and kernel drivers for resources
192allocations and initialization. The following dependencies are not part of
193DPDK and must be installed separately:
194
195- **libibverbs**
196
197  User space verbs framework used by librte_pmd_mlx4. This library provides
198  a generic interface between the kernel and low-level user space drivers
199  such as libmlx4.
200
201  It allows slow and privileged operations (context initialization, hardware
202  resources allocations) to be managed by the kernel and fast operations to
203  never leave user space.
204
205- **libmlx4**
206
207  Low-level user space driver library for Mellanox ConnectX-3 devices,
208  it is automatically loaded by libibverbs.
209
210  This library basically implements send/receive calls to the hardware
211  queues.
212
213- **Kernel modules** (mlnx-ofed-kernel)
214
215  They provide the kernel-side verbs API and low level device drivers that
216  manage actual hardware initialization and resources sharing with user
217  space processes.
218
219  Unlike most other PMDs, these modules must remain loaded and bound to
220  their devices:
221
222  - mlx4_core: hardware driver managing Mellanox ConnectX-3 devices.
223  - mlx4_en: Ethernet device driver that provides kernel network interfaces.
224  - mlx4_ib: InifiniBand device driver.
225  - ib_uverbs: user space driver for verbs (entry point for libibverbs).
226
227- **Firmware update**
228
229  Mellanox OFED releases include firmware updates for ConnectX-3 adapters.
230
231  Because each release provides new features, these updates must be applied to
232  match the kernel modules and libraries they come with.
233
234.. note::
235
236   Both libraries are BSD and GPL licensed. Linux kernel modules are GPL
237   licensed.
238
239Currently supported by DPDK:
240
241- Mellanox OFED **3.1**.
242- Firmware version **2.35.5100** and higher.
243- Supported architectures:  **x86_64** and **POWER8**.
244
245Getting Mellanox OFED
246~~~~~~~~~~~~~~~~~~~~~
247
248While these libraries and kernel modules are available on OpenFabrics
249Alliance's `website <https://www.openfabrics.org/>`_ and provided by package
250managers on most distributions, this PMD requires Ethernet extensions that
251may not be supported at the moment (this is a work in progress).
252
253`Mellanox OFED
254<http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux_sw_drivers>`_
255includes the necessary support and should be used in the meantime. For DPDK,
256only libibverbs, libmlx4, mlnx-ofed-kernel packages and firmware updates are
257required from that distribution.
258
259.. note::
260
261   Several versions of Mellanox OFED are available. Installing the version
262   this DPDK release was developed and tested against is strongly
263   recommended. Please check the `prerequisites`_.
264
265Usage example
266-------------
267
268This section demonstrates how to launch **testpmd** with Mellanox ConnectX-3
269devices managed by librte_pmd_mlx4.
270
271#. Load the kernel modules:
272
273   .. code-block:: console
274
275      modprobe -a ib_uverbs mlx4_en mlx4_core mlx4_ib
276
277   Alternatively if MLNX_OFED is fully installed, the following script can
278   be run:
279
280   .. code-block:: console
281
282      /etc/init.d/openibd restart
283
284   .. note::
285
286      User space I/O kernel modules (uio and igb_uio) are not used and do
287      not have to be loaded.
288
289#. Make sure Ethernet interfaces are in working order and linked to kernel
290   verbs. Related sysfs entries should be present:
291
292   .. code-block:: console
293
294      ls -d /sys/class/net/*/device/infiniband_verbs/uverbs* | cut -d / -f 5
295
296   Example output:
297
298   .. code-block:: console
299
300      eth2
301      eth3
302      eth4
303      eth5
304
305#. Optionally, retrieve their PCI bus addresses for whitelisting:
306
307   .. code-block:: console
308
309      {
310          for intf in eth2 eth3 eth4 eth5;
311          do
312              (cd "/sys/class/net/${intf}/device/" && pwd -P);
313          done;
314      } |
315      sed -n 's,.*/\(.*\),-w \1,p'
316
317   Example output:
318
319   .. code-block:: console
320
321      -w 0000:83:00.0
322      -w 0000:83:00.0
323      -w 0000:84:00.0
324      -w 0000:84:00.0
325
326   .. note::
327
328      There are only two distinct PCI bus addresses because the Mellanox
329      ConnectX-3 adapters installed on this system are dual port.
330
331#. Request huge pages:
332
333   .. code-block:: console
334
335      echo 1024 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages/nr_hugepages
336
337#. Start testpmd with basic parameters:
338
339   .. code-block:: console
340
341      testpmd -c 0xff00 -n 4 -w 0000:83:00.0 -w 0000:84:00.0 -- --rxq=2 --txq=2 -i
342
343   Example output:
344
345   .. code-block:: console
346
347      [...]
348      EAL: PCI device 0000:83:00.0 on NUMA socket 1
349      EAL:   probe driver: 15b3:1007 librte_pmd_mlx4
350      PMD: librte_pmd_mlx4: PCI information matches, using device "mlx4_0" (VF: false)
351      PMD: librte_pmd_mlx4: 2 port(s) detected
352      PMD: librte_pmd_mlx4: port 1 MAC address is 00:02:c9:b5:b7:50
353      PMD: librte_pmd_mlx4: port 2 MAC address is 00:02:c9:b5:b7:51
354      EAL: PCI device 0000:84:00.0 on NUMA socket 1
355      EAL:   probe driver: 15b3:1007 librte_pmd_mlx4
356      PMD: librte_pmd_mlx4: PCI information matches, using device "mlx4_1" (VF: false)
357      PMD: librte_pmd_mlx4: 2 port(s) detected
358      PMD: librte_pmd_mlx4: port 1 MAC address is 00:02:c9:b5:ba:b0
359      PMD: librte_pmd_mlx4: port 2 MAC address is 00:02:c9:b5:ba:b1
360      Interactive-mode selected
361      Configuring Port 0 (socket 0)
362      PMD: librte_pmd_mlx4: 0x867d60: TX queues number update: 0 -> 2
363      PMD: librte_pmd_mlx4: 0x867d60: RX queues number update: 0 -> 2
364      Port 0: 00:02:C9:B5:B7:50
365      Configuring Port 1 (socket 0)
366      PMD: librte_pmd_mlx4: 0x867da0: TX queues number update: 0 -> 2
367      PMD: librte_pmd_mlx4: 0x867da0: RX queues number update: 0 -> 2
368      Port 1: 00:02:C9:B5:B7:51
369      Configuring Port 2 (socket 0)
370      PMD: librte_pmd_mlx4: 0x867de0: TX queues number update: 0 -> 2
371      PMD: librte_pmd_mlx4: 0x867de0: RX queues number update: 0 -> 2
372      Port 2: 00:02:C9:B5:BA:B0
373      Configuring Port 3 (socket 0)
374      PMD: librte_pmd_mlx4: 0x867e20: TX queues number update: 0 -> 2
375      PMD: librte_pmd_mlx4: 0x867e20: RX queues number update: 0 -> 2
376      Port 3: 00:02:C9:B5:BA:B1
377      Checking link statuses...
378      Port 0 Link Up - speed 10000 Mbps - full-duplex
379      Port 1 Link Up - speed 40000 Mbps - full-duplex
380      Port 2 Link Up - speed 10000 Mbps - full-duplex
381      Port 3 Link Up - speed 40000 Mbps - full-duplex
382      Done
383      testpmd>
384