2 Copyright(c) 2010-2016 Intel Corporation. All rights reserved.
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 Intel Corporation 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.
37 Vector PMD uses IntelĀ® SIMD instructions to optimize packet I/O.
38 It improves load/store bandwidth efficiency of L1 data cache by using a wider SSE/AVX register 1 (1).
39 The wider register gives space to hold multiple packet buffers so as to save instruction number when processing bulk of packets.
41 There is no change to PMD API. The RX/TX handler are the only two entries for vPMD packet I/O.
42 They are transparently registered at runtime RX/TX execution if all condition checks pass.
44 1. To date, only an SSE version of IX GBE vPMD is available.
45 To ensure that vPMD is in the binary code, ensure that the option CONFIG_RTE_IXGBE_INC_VECTOR=y is in the configure file.
47 Some constraints apply as pre-conditions for specific optimizations on bulk packet transfers.
48 The following sections explain RX and TX constraints in the vPMD.
53 Prerequisites and Pre-conditions
54 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
56 The following prerequisites apply:
58 * To enable vPMD to work for RX, bulk allocation for Rx must be allowed.
60 Ensure that the following pre-conditions are satisfied:
62 * rxq->rx_free_thresh >= RTE_PMD_IXGBE_RX_MAX_BURST
64 * rxq->rx_free_thresh < rxq->nb_rx_desc
66 * (rxq->nb_rx_desc % rxq->rx_free_thresh) == 0
68 * rxq->nb_rx_desc < (IXGBE_MAX_RING_DESC - RTE_PMD_IXGBE_RX_MAX_BURST)
70 These conditions are checked in the code.
72 Scattered packets are not supported in this mode.
73 If an incoming packet is greater than the maximum acceptable length of one "mbuf" data size (by default, the size is 2 KB),
74 vPMD for RX would be disabled.
76 By default, IXGBE_MAX_RING_DESC is set to 4096 and RTE_PMD_IXGBE_RX_MAX_BURST is set to 32.
78 Feature not Supported by RX Vector PMD
79 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
81 Some features are not supported when trying to increase the throughput in vPMD.
90 * RX checksum off load
92 Other features are supported using optional MACRO configuration. They include:
98 To guarantee the constraint, configuration flags in dev_conf.rxmode will be checked:
110 fdir_conf->mode will also be checked.
115 As vPMD is focused on high throughput, it assumes that the RX burst size is equal to or greater than 32 per burst.
116 It returns zero if using nb_pkt < 32 as the expected packet number in the receive handler.
124 The only prerequisite is related to tx_rs_thresh.
125 The tx_rs_thresh value must be greater than or equal to RTE_PMD_IXGBE_TX_MAX_BURST,
126 but less or equal to RTE_IXGBE_TX_MAX_FREE_BUF_SZ.
127 Consequently, by default the tx_rs_thresh value is in the range 32 to 64.
129 Feature not Supported by TX Vector PMD
130 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
132 TX vPMD only works when txq_flags is set to IXGBE_SIMPLE_FLAGS.
134 This means that it does not support TX multi-segment, VLAN offload and TX csum offload.
135 The following MACROs are used for these three features:
137 * ETH_TXQ_FLAGS_NOMULTSEGS
139 * ETH_TXQ_FLAGS_NOVLANOFFL
141 * ETH_TXQ_FLAGS_NOXSUMSCTP
143 * ETH_TXQ_FLAGS_NOXSUMUDP
145 * ETH_TXQ_FLAGS_NOXSUMTCP
147 Application Programming Interface
148 ---------------------------------
150 In DPDK release v16.11 an API for ixgbe specific functions has been added to the ixgbe PMD.
151 The declarations for the API functions are in the header ``rte_pmd_ixgbe.h``.
153 Sample Application Notes
154 ------------------------
159 When running l3fwd with vPMD, there is one thing to note.
160 In the configuration, ensure that port_conf.rxmode.hw_ip_checksum=0.
161 Otherwise, by default, RX vPMD is disabled.
166 As in the case of l3fwd, set configure port_conf.rxmode.hw_ip_checksum=0 to enable vPMD.
167 In addition, for improved performance, use -bsz "(32,32),(64,64),(32,32)" in load_balancer to avoid using the default burst size of 144.
170 Limitations or Known issues
171 ---------------------------
173 Malicious Driver Detection not Supported
174 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
176 The Intel x550 series NICs support a feature called MDD (Malicious
177 Driver Detection) which checks the behavior of the VF driver.
178 If this feature is enabled, the VF must use the advanced context descriptor
179 correctly and set the CC (Check Context) bit.
180 DPDK PF doesn't support MDD, but kernel PF does. We may hit problem in this
181 scenario kernel PF + DPDK VF. If user enables MDD in kernel PF, DPDK VF will
182 not work. Because kernel PF thinks the VF is malicious. But actually it's not.
183 The only reason is the VF doesn't act as MDD required.
184 There's significant performance impact to support MDD. DPDK should check if
185 the advanced context descriptor should be set and set it. And DPDK has to ask
186 the info about the header length from the upper layer, because parsing the
187 packet itself is not acceptable. So, it's too expensive to support MDD.
188 When using kernel PF + DPDK VF on x550, please make sure to use a kernel
189 PF driver that disables MDD or can disable MDD.
191 Some kernel drivers already disable MDD by default while some kernels can use
192 the command ``insmod ixgbe.ko MDD=0,0`` to disable MDD. Each "0" in the
193 command refers to a port. For example, if there are 6 ixgbe ports, the command
194 should be changed to ``insmod ixgbe.ko MDD=0,0,0,0,0,0``.
200 The statistics of ixgbe hardware must be polled regularly in order for it to
201 remain consistent. Running a DPDK application without polling the statistics will
202 cause registers on hardware to count to the maximum value, and "stick" at
205 In order to avoid statistic registers every reaching the maximum value,
206 read the statistics from the hardware using ``rte_eth_stats_get()`` or
207 ``rte_eth_xstats_get()``.
209 The maximum time between statistics polls that ensures consistent results can
210 be calculated as follows:
214 max_read_interval = UINT_MAX / max_packets_per_second
215 max_read_interval = 4294967295 / 14880952
216 max_read_interval = 288.6218096127183 (seconds)
217 max_read_interval = ~4 mins 48 sec.
219 In order to ensure valid results, it is recommended to poll every 4 minutes.
224 Although the user can set the MTU separately on PF and VF ports, the ixgbe NIC
225 only supports one global MTU per physical port.
226 So when the user sets different MTUs on PF and VF ports in one physical port,
227 the real MTU for all these PF and VF ports is the largest value set.
228 This behavior is based on the kernel driver behavior.
231 Supported Chipsets and NICs
232 ---------------------------
234 - Intel 82599EB 10 Gigabit Ethernet Controller
235 - Intel 82598EB 10 Gigabit Ethernet Controller
236 - Intel 82599ES 10 Gigabit Ethernet Controller
237 - Intel 82599EN 10 Gigabit Ethernet Controller
238 - Intel Ethernet Controller X540-AT2
239 - Intel Ethernet Controller X550-BT2
240 - Intel Ethernet Controller X550-AT2
241 - Intel Ethernet Controller X550-AT
242 - Intel Ethernet Converged Network Adapter X520-SR1
243 - Intel Ethernet Converged Network Adapter X520-SR2
244 - Intel Ethernet Converged Network Adapter X520-LR1
245 - Intel Ethernet Converged Network Adapter X520-DA1
246 - Intel Ethernet Converged Network Adapter X520-DA2
247 - Intel Ethernet Converged Network Adapter X520-DA4
248 - Intel Ethernet Converged Network Adapter X520-QDA1
249 - Intel Ethernet Converged Network Adapter X520-T2
250 - Intel 10 Gigabit AF DA Dual Port Server Adapter
251 - Intel 10 Gigabit AT Server Adapter
252 - Intel 10 Gigabit AT2 Server Adapter
253 - Intel 10 Gigabit CX4 Dual Port Server Adapter
254 - Intel 10 Gigabit XF LR Server Adapter
255 - Intel 10 Gigabit XF SR Dual Port Server Adapter
256 - Intel 10 Gigabit XF SR Server Adapter
257 - Intel Ethernet Converged Network Adapter X540-T1
258 - Intel Ethernet Converged Network Adapter X540-T2
259 - Intel Ethernet Converged Network Adapter X550-T1
260 - Intel Ethernet Converged Network Adapter X550-T2