1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2018 Intel Corporation.
7 The ice PMD (librte_pmd_ice) provides poll mode driver support for
8 10/25 Gbps IntelĀ® Ethernet 810 Series Network Adapters based on
9 the Intel Ethernet Controller E810.
15 - Identifying your adapter using `Intel Support
16 <http://www.intel.com/support>`_ and get the latest NVM/FW images.
18 - Follow the DPDK :ref:`Getting Started Guide for Linux <linux_gsg>` to setup the basic DPDK environment.
20 - To get better performance on Intel platforms, please follow the "How to get best performance with NICs on Intel platforms"
21 section of the :ref:`Getting Started Guide for Linux <linux_gsg>`.
24 Pre-Installation Configuration
25 ------------------------------
30 The following options can be modified in the ``config`` file.
31 Please note that enabling debugging options may affect system performance.
33 - ``CONFIG_RTE_LIBRTE_ICE_PMD`` (default ``y``)
35 Toggle compilation of the ``librte_pmd_ice`` driver.
37 - ``CONFIG_RTE_LIBRTE_ICE_DEBUG_*`` (default ``n``)
39 Toggle display of generic debugging messages.
41 - ``CONFIG_RTE_LIBRTE_ICE_RX_ALLOW_BULK_ALLOC`` (default ``y``)
43 Toggle bulk allocation for RX.
45 - ``CONFIG_RTE_LIBRTE_ICE_16BYTE_RX_DESC`` (default ``n``)
47 Toggle to use a 16-byte RX descriptor, by default the RX descriptor is 32 byte.
49 Runtime Config Options
50 ~~~~~~~~~~~~~~~~~~~~~~
52 - ``Safe Mode Support`` (default ``0``)
54 If driver failed to load OS package, by default driver's initialization failed.
55 But if user intend to use the device without OS package, user can take ``devargs``
56 parameter ``safe-mode-support``, for example::
58 -w 80:00.0,safe-mode-support=1
60 Then the driver will be initialized successfully and the device will enter Safe Mode.
61 NOTE: In Safe mode, only very limited features are available, features like RSS,
62 checksum, fdir, tunneling ... are all disabled.
64 - ``Generic Flow Pipeline Mode Support`` (default ``0``)
66 In pipeline mode, a flow can be set at one specific stage by setting parameter
67 ``priority``. Currently, we support two stages: priority = 0 or !0. Flows with
68 priority 0 located at the first pipeline stage which typically be used as a firewall
69 to drop the packet on a blacklist(we called it permission stage). At this stage,
70 flow rules are created for the device's exact match engine: switch. Flows with priority
71 !0 located at the second stage, typically packets are classified here and be steered to
72 specific queue or queue group (we called it distribution stage), At this stage, flow
73 rules are created for device's flow director engine.
74 For none-pipeline mode, ``priority`` is ignored, a flow rule can be created as a flow director
75 rule or a switch rule depends on its pattern/action and the resource allocation situation,
76 all flows are virtually at the same pipeline stage.
77 By default, generic flow API is enabled in none-pipeline mode, user can choose to
78 use pipeline mode by setting ``devargs`` parameter ``pipeline-mode-support``,
81 -w 80:00.0,pipeline-mode-support=1
83 - ``Protocol extraction for per queue``
85 Configure the RX queues to do protocol extraction into ``rte_mbuf::udata64``
86 for protocol handling acceleration, like checking the TCP SYN packets quickly.
88 The argument format is::
90 -w 18:00.0,proto_xtr=<queues:protocol>[<queues:protocol>...]
91 -w 18:00.0,proto_xtr=<protocol>
93 Queues are grouped by ``(`` and ``)`` within the group. The ``-`` character
94 is used as a range separator and ``,`` is used as a single number separator.
95 The grouping ``()`` can be omitted for single element group. If no queues are
96 specified, PMD will use this protocol extraction type for all queues.
98 Protocol is : ``vlan, ipv4, ipv6, ipv6_flow, tcp``.
100 .. code-block:: console
102 testpmd -w 18:00.0,proto_xtr='[(1,2-3,8-9):tcp,10-13:vlan]'
104 This setting means queues 1, 2-3, 8-9 are TCP extraction, queues 10-13 are
105 VLAN extraction, other queues run with no protocol extraction.
107 .. code-block:: console
109 testpmd -w 18:00.0,proto_xtr=vlan,proto_xtr='[(1,2-3,8-9):tcp,10-23:ipv6]'
111 This setting means queues 1, 2-3, 8-9 are TCP extraction, queues 10-23 are
112 IPv6 extraction, other queues use the default VLAN extraction.
114 The extraction will be copied into the lower 32 bit of ``rte_mbuf::udata64``.
116 .. table:: Protocol extraction : ``vlan``
118 +----------------------------+----------------------------+
120 +======+===+=================+======+===+=================+
121 | PCP | D | VID | PCP | D | VID |
122 +------+---+-----------------+------+---+-----------------+
124 VLAN1 - single or EVLAN (first for QinQ).
126 VLAN2 - C-VLAN (second for QinQ).
128 .. table:: Protocol extraction : ``ipv4``
130 +----------------------------+----------------------------+
132 +======+=======+=============+==============+=============+
133 | Ver |Hdr Len| ToS | TTL | Protocol |
134 +------+-------+-------------+--------------+-------------+
136 IPHDR1 - IPv4 header word 4, "TTL" and "Protocol" fields.
138 IPHDR2 - IPv4 header word 0, "Ver", "Hdr Len" and "Type of Service" fields.
140 .. table:: Protocol extraction : ``ipv6``
142 +----------------------------+----------------------------+
144 +=====+=============+========+=============+==============+
145 | Ver |Traffic class| Flow | Next Header | Hop Limit |
146 +-----+-------------+--------+-------------+--------------+
148 IPHDR1 - IPv6 header word 3, "Next Header" and "Hop Limit" fields.
150 IPHDR2 - IPv6 header word 0, "Ver", "Traffic class" and high 4 bits of
153 .. table:: Protocol extraction : ``ipv6_flow``
155 +----------------------------+----------------------------+
157 +=====+=============+========+============================+
158 | Ver |Traffic class| Flow Label |
159 +-----+-------------+-------------------------------------+
161 IPHDR1 - IPv6 header word 1, 16 low bits of the "Flow Label" field.
163 IPHDR2 - IPv6 header word 0, "Ver", "Traffic class" and high 4 bits of
166 .. table:: Protocol extraction : ``tcp``
168 +----------------------------+----------------------------+
169 | TCPHDR2 | TCPHDR1 |
170 +============================+======+======+==============+
171 | Reserved |Offset| RSV | Flags |
172 +----------------------------+------+------+--------------+
174 TCPHDR1 - TCP header word 6, "Data Offset" and "Flags" fields.
178 Use ``get_proto_xtr_flds(struct rte_mbuf *mb)`` to access the protocol
179 extraction, do not use ``rte_mbuf::udata64`` directly.
181 The ``dump_proto_xtr_flds(struct rte_mbuf *mb)`` routine shows how to
182 access the protocol extraction result in ``struct rte_mbuf``.
184 Driver compilation and testing
185 ------------------------------
187 Refer to the document :ref:`compiling and testing a PMD for a NIC <pmd_build_and_test>`
196 Vector PMD for RX and TX path are selected automatically. The paths
197 are chosen based on 2 conditions.
200 On the X86 platform, the driver checks if the CPU supports AVX2.
201 If it's supported, AVX2 paths will be chosen. If not, SSE is chosen.
203 - ``Offload features``
204 The supported HW offload features are described in the document ice_vec.ini.
205 If any not supported features are used, ICE vector PMD is disabled and the
206 normal paths are chosen.
208 Malicious driver detection (MDD)
209 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
211 It's not appropriate to send a packet, if this packet's destination MAC address
212 is just this port's MAC address. If SW tries to send such packets, HW will
213 report a MDD event and drop the packets.
215 The APPs based on DPDK should avoid providing such packets.
217 Sample Application Notes
218 ------------------------
223 Vlan filter only works when Promiscuous mode is off.
225 To start ``testpmd``, and add vlan 10 to port 0:
227 .. code-block:: console
229 ./app/testpmd -l 0-15 -n 4 -- -i
232 testpmd> rx_vlan add 10 0
234 Limitations or Known issues
235 ---------------------------
237 The Intel E810 requires a programmable pipeline package be downloaded
238 by the driver to support normal operations. The E810 has a limited
239 functionality built in to allow PXE boot and other use cases, but the
240 driver must download a package file during the driver initialization
243 The default DDP package file name is ice.pkg. For a specific NIC, the
244 DDP package supposed to be loaded can have a filename: ice-xxxxxx.pkg,
245 where 'xxxxxx' is the 64-bit PCIe Device Serial Number of the NIC. For
246 example, if the NIC's device serial number is 00-CC-BB-FF-FF-AA-05-68,
247 the device-specific DDP package filename is ice-00ccbbffffaa0568.pkg
248 (in hex and all low case). During initialization, the driver searches
249 in the following paths in order: /lib/firmware/updates/intel/ice/ddp
250 and /lib/firmware/intel/ice/ddp. The corresponding device-specific DDP
251 package will be downloaded first if the file exists. If not, then the
252 driver tries to load the default package. The type of loaded package
253 is stored in ``ice_adapter->active_pkg_type``.
255 A symbolic link to the DDP package file is also ok. The same package
256 file is used by both the kernel driver and the DPDK PMD.
261 Ice code released in 19.02 is for evaluation only.