1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2018 Intel Corporation.
7 The ice PMD (**librte_net_ice**) provides poll mode driver support for
8 10/25/50/100 Gbps IntelĀ® Ethernet 810 Series Network Adapters based on
9 the Intel Ethernet Controller E810.
15 - Follow the DPDK :ref:`Getting Started Guide for Linux <linux_gsg>` to setup the basic DPDK environment.
17 - To get better performance on Intel platforms, please follow the "How to get best performance with NICs on Intel platforms"
18 section of the :ref:`Getting Started Guide for Linux <linux_gsg>`.
20 Pre-Installation Configuration
21 ------------------------------
24 Runtime Config Options
25 ~~~~~~~~~~~~~~~~~~~~~~
27 - ``Safe Mode Support`` (default ``0``)
29 If driver failed to load OS package, by default driver's initialization failed.
30 But if user intend to use the device without OS package, user can take ``devargs``
31 parameter ``safe-mode-support``, for example::
33 -w 80:00.0,safe-mode-support=1
35 Then the driver will be initialized successfully and the device will enter Safe Mode.
36 NOTE: In Safe mode, only very limited features are available, features like RSS,
37 checksum, fdir, tunneling ... are all disabled.
39 - ``Generic Flow Pipeline Mode Support`` (default ``0``)
41 In pipeline mode, a flow can be set at one specific stage by setting parameter
42 ``priority``. Currently, we support two stages: priority = 0 or !0. Flows with
43 priority 0 located at the first pipeline stage which typically be used as a firewall
44 to drop the packet on a blacklist(we called it permission stage). At this stage,
45 flow rules are created for the device's exact match engine: switch. Flows with priority
46 !0 located at the second stage, typically packets are classified here and be steered to
47 specific queue or queue group (we called it distribution stage), At this stage, flow
48 rules are created for device's flow director engine.
49 For none-pipeline mode, ``priority`` is ignored, a flow rule can be created as a flow director
50 rule or a switch rule depends on its pattern/action and the resource allocation situation,
51 all flows are virtually at the same pipeline stage.
52 By default, generic flow API is enabled in none-pipeline mode, user can choose to
53 use pipeline mode by setting ``devargs`` parameter ``pipeline-mode-support``,
56 -w 80:00.0,pipeline-mode-support=1
58 - ``Protocol extraction for per queue``
60 Configure the RX queues to do protocol extraction into mbuf for protocol
61 handling acceleration, like checking the TCP SYN packets quickly.
63 The argument format is::
65 -w 18:00.0,proto_xtr=<queues:protocol>[<queues:protocol>...]
66 -w 18:00.0,proto_xtr=<protocol>
68 Queues are grouped by ``(`` and ``)`` within the group. The ``-`` character
69 is used as a range separator and ``,`` is used as a single number separator.
70 The grouping ``()`` can be omitted for single element group. If no queues are
71 specified, PMD will use this protocol extraction type for all queues.
73 Protocol is : ``vlan, ipv4, ipv6, ipv6_flow, tcp, ip_offset``.
75 .. code-block:: console
77 dpdk-testpmd -w 18:00.0,proto_xtr='[(1,2-3,8-9):tcp,10-13:vlan]'
79 This setting means queues 1, 2-3, 8-9 are TCP extraction, queues 10-13 are
80 VLAN extraction, other queues run with no protocol extraction.
82 .. code-block:: console
84 dpdk-testpmd -w 18:00.0,proto_xtr=vlan,proto_xtr='[(1,2-3,8-9):tcp,10-23:ipv6]'
86 This setting means queues 1, 2-3, 8-9 are TCP extraction, queues 10-23 are
87 IPv6 extraction, other queues use the default VLAN extraction.
89 The extraction metadata is copied into the registered dynamic mbuf field, and
90 the related dynamic mbuf flags is set.
92 .. table:: Protocol extraction : ``vlan``
94 +----------------------------+----------------------------+
96 +======+===+=================+======+===+=================+
97 | PCP | D | VID | PCP | D | VID |
98 +------+---+-----------------+------+---+-----------------+
100 VLAN1 - single or EVLAN (first for QinQ).
102 VLAN2 - C-VLAN (second for QinQ).
104 .. table:: Protocol extraction : ``ipv4``
106 +----------------------------+----------------------------+
108 +======+=======+=============+==============+=============+
109 | Ver |Hdr Len| ToS | TTL | Protocol |
110 +------+-------+-------------+--------------+-------------+
112 IPHDR1 - IPv4 header word 4, "TTL" and "Protocol" fields.
114 IPHDR2 - IPv4 header word 0, "Ver", "Hdr Len" and "Type of Service" fields.
116 .. table:: Protocol extraction : ``ipv6``
118 +----------------------------+----------------------------+
120 +=====+=============+========+=============+==============+
121 | Ver |Traffic class| Flow | Next Header | Hop Limit |
122 +-----+-------------+--------+-------------+--------------+
124 IPHDR1 - IPv6 header word 3, "Next Header" and "Hop Limit" fields.
126 IPHDR2 - IPv6 header word 0, "Ver", "Traffic class" and high 4 bits of
129 .. table:: Protocol extraction : ``ipv6_flow``
131 +----------------------------+----------------------------+
133 +=====+=============+========+============================+
134 | Ver |Traffic class| Flow Label |
135 +-----+-------------+-------------------------------------+
137 IPHDR1 - IPv6 header word 1, 16 low bits of the "Flow Label" field.
139 IPHDR2 - IPv6 header word 0, "Ver", "Traffic class" and high 4 bits of
142 .. table:: Protocol extraction : ``tcp``
144 +----------------------------+----------------------------+
145 | TCPHDR2 | TCPHDR1 |
146 +============================+======+======+==============+
147 | Reserved |Offset| RSV | Flags |
148 +----------------------------+------+------+--------------+
150 TCPHDR1 - TCP header word 6, "Data Offset" and "Flags" fields.
154 .. table:: Protocol extraction : ``ip_offset``
156 +----------------------------+----------------------------+
158 +============================+============================+
159 | IPv6 HDR Offset | IPv4 HDR Offset |
160 +----------------------------+----------------------------+
162 IPHDR1 - Outer/Single IPv4 Header offset.
164 IPHDR2 - Outer/Single IPv6 Header offset.
166 Use ``rte_net_ice_dynf_proto_xtr_metadata_get`` to access the protocol
167 extraction metadata, and use ``RTE_PKT_RX_DYNF_PROTO_XTR_*`` to get the
168 metadata type of ``struct rte_mbuf::ol_flags``.
170 The ``rte_net_ice_dump_proto_xtr_metadata`` routine shows how to
171 access the protocol extraction result in ``struct rte_mbuf``.
173 Driver compilation and testing
174 ------------------------------
176 Refer to the document :ref:`compiling and testing a PMD for a NIC <pmd_build_and_test>`
185 Vector PMD for RX and TX path are selected automatically. The paths
186 are chosen based on 2 conditions.
189 On the X86 platform, the driver checks if the CPU supports AVX2.
190 If it's supported, AVX2 paths will be chosen. If not, SSE is chosen.
192 - ``Offload features``
193 The supported HW offload features are described in the document ice_vec.ini.
194 If any not supported features are used, ICE vector PMD is disabled and the
195 normal paths are chosen.
197 Malicious driver detection (MDD)
198 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
200 It's not appropriate to send a packet, if this packet's destination MAC address
201 is just this port's MAC address. If SW tries to send such packets, HW will
202 report a MDD event and drop the packets.
204 The APPs based on DPDK should avoid providing such packets.
206 Device Config Function (DCF)
207 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
209 This section demonstrates ICE DCF PMD, which shares the core module with ICE
212 A DCF (Device Config Function) PMD bounds to the device's trusted VF with ID 0,
213 it can act as a sole controlling entity to exercise advance functionality (such
214 as switch, ACL) for the rest VFs.
216 The DCF PMD needs to advertise and acquire DCF capability which allows DCF to
217 send AdminQ commands that it would like to execute over to the PF and receive
218 responses for the same from PF.
222 .. figure:: img/ice_dcf.*
224 DCF Communication flow.
228 echo 4 > /sys/bus/pci/devices/0000\:18\:00.0/sriov_numvfs
230 #. Enable the VF0 trust on::
232 ip link set dev enp24s0f0 vf 0 trust on
234 #. Bind the VF0, and run testpmd with 'cap=dcf' devarg::
236 dpdk-testpmd -l 22-25 -n 4 -w 18:01.0,cap=dcf -- -i
238 #. Monitor the VF2 interface network traffic::
240 tcpdump -e -nn -i enp24s1f2
242 #. Create one flow to redirect the traffic to VF2 by DCF::
244 flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.2 \
245 dst is 192.168.0.3 / end actions vf id 2 / end
247 #. Send the packet, and it should be displayed on tcpdump::
249 sendp(Ether(src='3c:fd:fe:aa:bb:78', dst='00:00:00:01:02:03')/IP(src=' \
250 192.168.0.2', dst="192.168.0.3")/TCP(flags='S')/Raw(load='XXXXXXXXXX'), \
251 iface="enp24s0f0", count=10)
253 Sample Application Notes
254 ------------------------
259 Vlan filter only works when Promiscuous mode is off.
261 To start ``testpmd``, and add vlan 10 to port 0:
263 .. code-block:: console
265 ./app/dpdk-testpmd -l 0-15 -n 4 -- -i
268 testpmd> rx_vlan add 10 0
270 Limitations or Known issues
271 ---------------------------
273 The Intel E810 requires a programmable pipeline package be downloaded
274 by the driver to support normal operations. The E810 has a limited
275 functionality built in to allow PXE boot and other use cases, but the
276 driver must download a package file during the driver initialization
279 The default DDP package file name is ice.pkg. For a specific NIC, the
280 DDP package supposed to be loaded can have a filename: ice-xxxxxx.pkg,
281 where 'xxxxxx' is the 64-bit PCIe Device Serial Number of the NIC. For
282 example, if the NIC's device serial number is 00-CC-BB-FF-FF-AA-05-68,
283 the device-specific DDP package filename is ice-00ccbbffffaa0568.pkg
284 (in hex and all low case). During initialization, the driver searches
285 in the following paths in order: /lib/firmware/updates/intel/ice/ddp
286 and /lib/firmware/intel/ice/ddp. The corresponding device-specific DDP
287 package will be downloaded first if the file exists. If not, then the
288 driver tries to load the default package. The type of loaded package
289 is stored in ``ice_adapter->active_pkg_type``.
291 A symbolic link to the DDP package file is also ok. The same package
292 file is used by both the kernel driver and the DPDK PMD.