1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright 2016 6WIND S.A.
3 Copyright 2016 Mellanox Technologies, Ltd
5 Generic flow API (rte_flow)
6 ===========================
11 This API provides a generic means to configure hardware to match specific
12 ingress or egress traffic, alter its fate and query related counters
13 according to any number of user-defined rules.
15 It is named *rte_flow* after the prefix used for all its symbols, and is
16 defined in ``rte_flow.h``.
18 - Matching can be performed on packet data (protocol headers, payload) and
19 properties (e.g. associated physical port, virtual device function ID).
21 - Possible operations include dropping traffic, diverting it to specific
22 queues, to virtual/physical device functions or ports, performing tunnel
23 offloads, adding marks and so on.
25 It is slightly higher-level than the legacy filtering framework which it
26 encompasses and supersedes (including all functions and filter types) in
27 order to expose a single interface with an unambiguous behavior that is
28 common to all poll-mode drivers (PMDs).
36 A flow rule is the combination of attributes with a matching pattern and a
37 list of actions. Flow rules form the basis of this API.
39 Flow rules can have several distinct actions (such as counting,
40 encapsulating, decapsulating before redirecting packets to a particular
41 queue, etc.), instead of relying on several rules to achieve this and having
42 applications deal with hardware implementation details regarding their
45 Support for different priority levels on a rule basis is provided, for
46 example in order to force a more specific rule to come before a more generic
47 one for packets matched by both. However hardware support for more than a
48 single priority level cannot be guaranteed. When supported, the number of
49 available priority levels is usually low, which is why they can also be
50 implemented in software by PMDs (e.g. missing priority levels may be
51 emulated by reordering rules).
53 In order to remain as hardware-agnostic as possible, by default all rules
54 are considered to have the same priority, which means that the order between
55 overlapping rules (when a packet is matched by several filters) is
58 PMDs may refuse to create overlapping rules at a given priority level when
59 they can be detected (e.g. if a pattern matches an existing filter).
61 Thus predictable results for a given priority level can only be achieved
62 with non-overlapping rules, using perfect matching on all protocol layers.
64 Flow rules can also be grouped, the flow rule priority is specific to the
65 group they belong to. All flow rules in a given group are thus processed within
66 the context of that group. Groups are not linked by default, so the logical
67 hierarchy of groups must be explicitly defined by flow rules themselves in each
68 group using the JUMP action to define the next group to redirect too. Only flow
69 rules defined in the default group 0 are guarantee to be matched against, this
70 makes group 0 the origin of any group hierarchy defined by an application.
72 Support for multiple actions per rule may be implemented internally on top
73 of non-default hardware priorities, as a result both features may not be
74 simultaneously available to applications.
76 Considering that allowed pattern/actions combinations cannot be known in
77 advance and would result in an impractically large number of capabilities to
78 expose, a method is provided to validate a given rule from the current
79 device configuration state.
81 This enables applications to check if the rule types they need is supported
82 at initialization time, before starting their data path. This method can be
83 used anytime, its only requirement being that the resources needed by a rule
84 should exist (e.g. a target RX queue should be configured first).
86 Each defined rule is associated with an opaque handle managed by the PMD,
87 applications are responsible for keeping it. These can be used for queries
88 and rules management, such as retrieving counters or other data and
91 To avoid resource leaks on the PMD side, handles must be explicitly
92 destroyed by the application before releasing associated resources such as
95 The following sections cover:
97 - **Attributes** (represented by ``struct rte_flow_attr``): properties of a
98 flow rule such as its direction (ingress or egress) and priority.
100 - **Pattern item** (represented by ``struct rte_flow_item``): part of a
101 matching pattern that either matches specific packet data or traffic
102 properties. It can also describe properties of the pattern itself, such as
105 - **Matching pattern**: traffic properties to look for, a combination of any
108 - **Actions** (represented by ``struct rte_flow_action``): operations to
109 perform whenever a packet is matched by a pattern.
117 Flow rules can be grouped by assigning them a common group number. Groups
118 allow a logical hierarchy of flow rule groups (tables) to be defined. These
119 groups can be supported virtually in the PMD or in the physical device.
120 Group 0 is the default group and this is the only group which flows are
121 guarantee to matched against, all subsequent groups can only be reached by
122 way of the JUMP action from a matched flow rule.
124 Although optional, applications are encouraged to group similar rules as
125 much as possible to fully take advantage of hardware capabilities
126 (e.g. optimized matching) and work around limitations (e.g. a single pattern
127 type possibly allowed in a given group), while being aware that the groups
128 hierarchies must be programmed explicitly.
130 Note that support for more than a single group is not guaranteed.
135 A priority level can be assigned to a flow rule, lower values
136 denote higher priority, with 0 as the maximum.
138 Priority levels are arbitrary and up to the application, they do
139 not need to be contiguous nor start from 0, however the maximum number
140 varies between devices and may be affected by existing flow rules.
142 A flow which matches multiple rules in the same group will always matched by
143 the rule with the highest priority in that group.
145 If a packet is matched by several rules of a given group for a given
146 priority level, the outcome is undefined. It can take any path, may be
147 duplicated or even cause unrecoverable errors.
149 Note that support for more than a single priority level is not guaranteed.
151 Attribute: Traffic direction
152 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
154 Flow rule patterns apply to inbound and/or outbound traffic.
156 In the context of this API, **ingress** and **egress** respectively stand
157 for **inbound** and **outbound** based on the standpoint of the application
158 creating a flow rule.
160 There are no exceptions to this definition.
162 Several pattern items and actions are valid and can be used in both
163 directions. At least one direction must be specified.
165 Specifying both directions at once for a given rule is not recommended but
166 may be valid in a few cases (e.g. shared counters).
171 Instead of simply matching the properties of traffic as it would appear on a
172 given DPDK port ID, enabling this attribute transfers a flow rule to the
173 lowest possible level of any device endpoints found in the pattern.
175 When supported, this effectively enables an application to reroute traffic
176 not necessarily intended for it (e.g. coming from or addressed to different
177 physical ports, VFs or applications) at the device level.
179 It complements the behavior of some pattern items such as `Item: PHY_PORT`_
180 and is meaningless without them.
182 When transferring flow rules, **ingress** and **egress** attributes
183 (`Attribute: Traffic direction`_) keep their original meaning, as if
184 processing traffic emitted or received by the application.
189 Pattern items fall in two categories:
191 - Matching protocol headers and packet data, usually associated with a
192 specification structure. These must be stacked in the same order as the
193 protocol layers to match inside packets, starting from the lowest.
195 - Matching meta-data or affecting pattern processing, often without a
196 specification structure. Since they do not match packet contents, their
197 position in the list is usually not relevant.
199 Item specification structures are used to match specific values among
200 protocol fields (or item properties). Documentation describes for each item
201 whether they are associated with one and their type name if so.
203 Up to three structures of the same type can be set for a given item:
205 - ``spec``: values to match (e.g. a given IPv4 address).
207 - ``last``: upper bound for an inclusive range with corresponding fields in
210 - ``mask``: bit-mask applied to both ``spec`` and ``last`` whose purpose is
211 to distinguish the values to take into account and/or partially mask them
212 out (e.g. in order to match an IPv4 address prefix).
214 Usage restrictions and expected behavior:
216 - Setting either ``mask`` or ``last`` without ``spec`` is an error.
218 - Field values in ``last`` which are either 0 or equal to the corresponding
219 values in ``spec`` are ignored; they do not generate a range. Nonzero
220 values lower than those in ``spec`` are not supported.
222 - Setting ``spec`` and optionally ``last`` without ``mask`` causes the PMD
223 to use the default mask defined for that item (defined as
224 ``rte_flow_item_{name}_mask`` constants).
226 - Not setting any of them (assuming item type allows it) is equivalent to
227 providing an empty (zeroed) ``mask`` for broad (nonspecific) matching.
229 - ``mask`` is a simple bit-mask applied before interpreting the contents of
230 ``spec`` and ``last``, which may yield unexpected results if not used
231 carefully. For example, if for an IPv4 address field, ``spec`` provides
232 *10.1.2.3*, ``last`` provides *10.3.4.5* and ``mask`` provides
233 *255.255.0.0*, the effective range becomes *10.1.0.0* to *10.3.255.255*.
235 Example of an item specification matching an Ethernet header:
237 .. _table_rte_flow_pattern_item_example:
239 .. table:: Ethernet item
241 +----------+----------+-----------------------+
242 | Field | Subfield | Value |
243 +==========+==========+=======================+
244 | ``spec`` | ``src`` | ``00:00:01:02:03:04`` |
245 | +----------+-----------------------+
246 | | ``dst`` | ``00:00:2a:66:00:01`` |
247 | +----------+-----------------------+
248 | | ``type`` | ``0x22aa`` |
249 +----------+----------+-----------------------+
250 | ``last`` | unspecified |
251 +----------+----------+-----------------------+
252 | ``mask`` | ``src`` | ``00:00:ff:ff:ff:00`` |
253 | +----------+-----------------------+
254 | | ``dst`` | ``00:00:00:00:00:ff`` |
255 | +----------+-----------------------+
256 | | ``type`` | ``0x0000`` |
257 +----------+----------+-----------------------+
259 Non-masked bits stand for any value (shown as ``?`` below), Ethernet headers
260 with the following properties are thus matched:
262 - ``src``: ``??:??:01:02:03:??``
263 - ``dst``: ``??:??:??:??:??:01``
264 - ``type``: ``0x????``
269 A pattern is formed by stacking items starting from the lowest protocol
270 layer to match. This stacking restriction does not apply to meta items which
271 can be placed anywhere in the stack without affecting the meaning of the
274 Patterns are terminated by END items.
278 .. _table_rte_flow_tcpv4_as_l4:
280 .. table:: TCPv4 as L4
296 .. _table_rte_flow_tcpv6_in_vxlan:
298 .. table:: TCPv6 in VXLAN
300 +-------+------------+
302 +=======+============+
304 +-------+------------+
306 +-------+------------+
308 +-------+------------+
310 +-------+------------+
312 +-------+------------+
314 +-------+------------+
316 +-------+------------+
318 +-------+------------+
322 .. _table_rte_flow_tcpv4_as_l4_meta:
324 .. table:: TCPv4 as L4 with meta items
346 The above example shows how meta items do not affect packet data matching
347 items, as long as those remain stacked properly. The resulting matching
348 pattern is identical to "TCPv4 as L4".
350 .. _table_rte_flow_udpv6_anywhere:
352 .. table:: UDPv6 anywhere
364 If supported by the PMD, omitting one or several protocol layers at the
365 bottom of the stack as in the above example (missing an Ethernet
366 specification) enables looking up anywhere in packets.
368 It is unspecified whether the payload of supported encapsulations
369 (e.g. VXLAN payload) is matched by such a pattern, which may apply to inner,
370 outer or both packets.
372 .. _table_rte_flow_invalid_l3:
374 .. table:: Invalid, missing L3
386 The above pattern is invalid due to a missing L3 specification between L2
387 (Ethernet) and L4 (UDP). Doing so is only allowed at the bottom and at the
393 They match meta-data or affect pattern processing instead of matching packet
394 data directly, most of them do not need a specification structure. This
395 particularity allows them to be specified anywhere in the stack without
396 causing any side effect.
401 End marker for item lists. Prevents further processing of items, thereby
404 - Its numeric value is 0 for convenience.
405 - PMD support is mandatory.
406 - ``spec``, ``last`` and ``mask`` are ignored.
408 .. _table_rte_flow_item_end:
412 +----------+---------+
414 +==========+=========+
415 | ``spec`` | ignored |
416 +----------+---------+
417 | ``last`` | ignored |
418 +----------+---------+
419 | ``mask`` | ignored |
420 +----------+---------+
425 Used as a placeholder for convenience. It is ignored and simply discarded by
428 - PMD support is mandatory.
429 - ``spec``, ``last`` and ``mask`` are ignored.
431 .. _table_rte_flow_item_void:
435 +----------+---------+
437 +==========+=========+
438 | ``spec`` | ignored |
439 +----------+---------+
440 | ``last`` | ignored |
441 +----------+---------+
442 | ``mask`` | ignored |
443 +----------+---------+
445 One usage example for this type is generating rules that share a common
446 prefix quickly without reallocating memory, only by updating item types:
448 .. _table_rte_flow_item_void_example:
450 .. table:: TCP, UDP or ICMP as L4
452 +-------+--------------------+
454 +=======+====================+
456 +-------+--------------------+
458 +-------+------+------+------+
459 | 2 | UDP | VOID | VOID |
460 +-------+------+------+------+
461 | 3 | VOID | TCP | VOID |
462 +-------+------+------+------+
463 | 4 | VOID | VOID | ICMP |
464 +-------+------+------+------+
466 +-------+--------------------+
471 Inverted matching, i.e. process packets that do not match the pattern.
473 - ``spec``, ``last`` and ``mask`` are ignored.
475 .. _table_rte_flow_item_invert:
479 +----------+---------+
481 +==========+=========+
482 | ``spec`` | ignored |
483 +----------+---------+
484 | ``last`` | ignored |
485 +----------+---------+
486 | ``mask`` | ignored |
487 +----------+---------+
489 Usage example, matching non-TCPv4 packets only:
491 .. _table_rte_flow_item_invert_example:
493 .. table:: Anything but TCPv4
512 Matches traffic originating from (ingress) or going to (egress) the physical
513 function of the current device.
515 If supported, should work even if the physical function is not managed by
516 the application and thus not associated with a DPDK port ID.
518 - Can be combined with any number of `Item: VF`_ to match both PF and VF
520 - ``spec``, ``last`` and ``mask`` must not be set.
522 .. _table_rte_flow_item_pf:
539 Matches traffic originating from (ingress) or going to (egress) a given
540 virtual function of the current device.
542 If supported, should work even if the virtual function is not managed by the
543 application and thus not associated with a DPDK port ID.
545 Note this pattern item does not match VF representors traffic which, as
546 separate entities, should be addressed through their own DPDK port IDs.
548 - Can be specified multiple times to match traffic addressed to several VF
550 - Can be combined with a PF item to match both PF and VF traffic.
551 - Default ``mask`` matches any VF ID.
553 .. _table_rte_flow_item_vf:
557 +----------+----------+---------------------------+
558 | Field | Subfield | Value |
559 +==========+==========+===========================+
560 | ``spec`` | ``id`` | destination VF ID |
561 +----------+----------+---------------------------+
562 | ``last`` | ``id`` | upper range value |
563 +----------+----------+---------------------------+
564 | ``mask`` | ``id`` | zeroed to match any VF ID |
565 +----------+----------+---------------------------+
570 Matches traffic originating from (ingress) or going to (egress) a physical
571 port of the underlying device.
573 The first PHY_PORT item overrides the physical port normally associated with
574 the specified DPDK input port (port_id). This item can be provided several
575 times to match additional physical ports.
577 Note that physical ports are not necessarily tied to DPDK input ports
578 (port_id) when those are not under DPDK control. Possible values are
579 specific to each device, they are not necessarily indexed from zero and may
582 As a device property, the list of allowed values as well as the value
583 associated with a port_id should be retrieved by other means.
585 - Default ``mask`` matches any port index.
587 .. _table_rte_flow_item_phy_port:
591 +----------+-----------+--------------------------------+
592 | Field | Subfield | Value |
593 +==========+===========+================================+
594 | ``spec`` | ``index`` | physical port index |
595 +----------+-----------+--------------------------------+
596 | ``last`` | ``index`` | upper range value |
597 +----------+-----------+--------------------------------+
598 | ``mask`` | ``index`` | zeroed to match any port index |
599 +----------+-----------+--------------------------------+
604 Matches traffic originating from (ingress) or going to (egress) a given DPDK
607 Normally only supported if the port ID in question is known by the
608 underlying PMD and related to the device the flow rule is created against.
610 This must not be confused with `Item: PHY_PORT`_ which refers to the
611 physical port of a device, whereas `Item: PORT_ID`_ refers to a ``struct
612 rte_eth_dev`` object on the application side (also known as "port
613 representor" depending on the kind of underlying device).
615 - Default ``mask`` matches the specified DPDK port ID.
617 .. _table_rte_flow_item_port_id:
621 +----------+----------+-----------------------------+
622 | Field | Subfield | Value |
623 +==========+==========+=============================+
624 | ``spec`` | ``id`` | DPDK port ID |
625 +----------+----------+-----------------------------+
626 | ``last`` | ``id`` | upper range value |
627 +----------+----------+-----------------------------+
628 | ``mask`` | ``id`` | zeroed to match any port ID |
629 +----------+----------+-----------------------------+
634 Matches an arbitrary integer value which was set using the ``MARK`` action in
635 a previously matched rule.
637 This item can only specified once as a match criteria as the ``MARK`` action can
638 only be specified once in a flow action.
640 Note the value of MARK field is arbitrary and application defined.
642 Depending on the underlying implementation the MARK item may be supported on
643 the physical device, with virtual groups in the PMD or not at all.
645 - Default ``mask`` matches any integer value.
647 .. _table_rte_flow_item_mark:
651 +----------+----------+---------------------------+
652 | Field | Subfield | Value |
653 +==========+==========+===========================+
654 | ``spec`` | ``id`` | integer value |
655 +----------+--------------------------------------+
656 | ``last`` | ``id`` | upper range value |
657 +----------+----------+---------------------------+
658 | ``mask`` | ``id`` | zeroed to match any value |
659 +----------+----------+---------------------------+
661 Data matching item types
662 ~~~~~~~~~~~~~~~~~~~~~~~~
664 Most of these are basically protocol header definitions with associated
665 bit-masks. They must be specified (stacked) from lowest to highest protocol
666 layer to form a matching pattern.
668 The following list is not exhaustive, new protocols will be added in the
674 Matches any protocol in place of the current layer, a single ANY may also
675 stand for several protocol layers.
677 This is usually specified as the first pattern item when looking for a
678 protocol anywhere in a packet.
680 - Default ``mask`` stands for any number of layers.
682 .. _table_rte_flow_item_any:
686 +----------+----------+--------------------------------------+
687 | Field | Subfield | Value |
688 +==========+==========+======================================+
689 | ``spec`` | ``num`` | number of layers covered |
690 +----------+----------+--------------------------------------+
691 | ``last`` | ``num`` | upper range value |
692 +----------+----------+--------------------------------------+
693 | ``mask`` | ``num`` | zeroed to cover any number of layers |
694 +----------+----------+--------------------------------------+
696 Example for VXLAN TCP payload matching regardless of outer L3 (IPv4 or IPv6)
697 and L4 (UDP) both matched by the first ANY specification, and inner L3 (IPv4
698 or IPv6) matched by the second ANY specification:
700 .. _table_rte_flow_item_any_example:
702 .. table:: TCP in VXLAN with wildcards
704 +-------+------+----------+----------+-------+
705 | Index | Item | Field | Subfield | Value |
706 +=======+======+==========+==========+=======+
708 +-------+------+----------+----------+-------+
709 | 1 | ANY | ``spec`` | ``num`` | 2 |
710 +-------+------+----------+----------+-------+
712 +-------+------------------------------------+
714 +-------+------+----------+----------+-------+
715 | 4 | ANY | ``spec`` | ``num`` | 1 |
716 +-------+------+----------+----------+-------+
718 +-------+------------------------------------+
720 +-------+------------------------------------+
725 Matches a byte string of a given length at a given offset.
727 Offset is either absolute (using the start of the packet) or relative to the
728 end of the previous matched item in the stack, in which case negative values
731 If search is enabled, offset is used as the starting point. The search area
732 can be delimited by setting limit to a nonzero value, which is the maximum
733 number of bytes after offset where the pattern may start.
735 Matching a zero-length pattern is allowed, doing so resets the relative
736 offset for subsequent items.
738 - This type does not support ranges (``last`` field).
739 - Default ``mask`` matches all fields exactly.
741 .. _table_rte_flow_item_raw:
745 +----------+--------------+-------------------------------------------------+
746 | Field | Subfield | Value |
747 +==========+==============+=================================================+
748 | ``spec`` | ``relative`` | look for pattern after the previous item |
749 | +--------------+-------------------------------------------------+
750 | | ``search`` | search pattern from offset (see also ``limit``) |
751 | +--------------+-------------------------------------------------+
752 | | ``reserved`` | reserved, must be set to zero |
753 | +--------------+-------------------------------------------------+
754 | | ``offset`` | absolute or relative offset for ``pattern`` |
755 | +--------------+-------------------------------------------------+
756 | | ``limit`` | search area limit for start of ``pattern`` |
757 | +--------------+-------------------------------------------------+
758 | | ``length`` | ``pattern`` length |
759 | +--------------+-------------------------------------------------+
760 | | ``pattern`` | byte string to look for |
761 +----------+--------------+-------------------------------------------------+
762 | ``last`` | if specified, either all 0 or with the same values as ``spec`` |
763 +----------+----------------------------------------------------------------+
764 | ``mask`` | bit-mask applied to ``spec`` values with usual behavior |
765 +----------+----------------------------------------------------------------+
767 Example pattern looking for several strings at various offsets of a UDP
768 payload, using combined RAW items:
770 .. _table_rte_flow_item_raw_example:
772 .. table:: UDP payload matching
774 +-------+------+----------+--------------+-------+
775 | Index | Item | Field | Subfield | Value |
776 +=======+======+==========+==============+=======+
778 +-------+----------------------------------------+
780 +-------+----------------------------------------+
782 +-------+------+----------+--------------+-------+
783 | 3 | RAW | ``spec`` | ``relative`` | 1 |
784 | | | +--------------+-------+
785 | | | | ``search`` | 1 |
786 | | | +--------------+-------+
787 | | | | ``offset`` | 10 |
788 | | | +--------------+-------+
789 | | | | ``limit`` | 0 |
790 | | | +--------------+-------+
791 | | | | ``length`` | 3 |
792 | | | +--------------+-------+
793 | | | | ``pattern`` | "foo" |
794 +-------+------+----------+--------------+-------+
795 | 4 | RAW | ``spec`` | ``relative`` | 1 |
796 | | | +--------------+-------+
797 | | | | ``search`` | 0 |
798 | | | +--------------+-------+
799 | | | | ``offset`` | 20 |
800 | | | +--------------+-------+
801 | | | | ``limit`` | 0 |
802 | | | +--------------+-------+
803 | | | | ``length`` | 3 |
804 | | | +--------------+-------+
805 | | | | ``pattern`` | "bar" |
806 +-------+------+----------+--------------+-------+
807 | 5 | RAW | ``spec`` | ``relative`` | 1 |
808 | | | +--------------+-------+
809 | | | | ``search`` | 0 |
810 | | | +--------------+-------+
811 | | | | ``offset`` | -29 |
812 | | | +--------------+-------+
813 | | | | ``limit`` | 0 |
814 | | | +--------------+-------+
815 | | | | ``length`` | 3 |
816 | | | +--------------+-------+
817 | | | | ``pattern`` | "baz" |
818 +-------+------+----------+--------------+-------+
820 +-------+----------------------------------------+
824 - Locate "foo" at least 10 bytes deep inside UDP payload.
825 - Locate "bar" after "foo" plus 20 bytes.
826 - Locate "baz" after "bar" minus 29 bytes.
828 Such a packet may be represented as follows (not to scale)::
831 | |<--------->| |<--------->|
833 |-----|------|-----|-----|-----|-----|-----------|-----|------|
834 | ETH | IPv4 | UDP | ... | baz | foo | ......... | bar | .... |
835 |-----|------|-----|-----|-----|-----|-----------|-----|------|
837 |<--------------------------->|
840 Note that matching subsequent pattern items would resume after "baz", not
841 "bar" since matching is always performed after the previous item of the
847 Matches an Ethernet header.
849 The ``type`` field either stands for "EtherType" or "TPID" when followed by
850 so-called layer 2.5 pattern items such as ``RTE_FLOW_ITEM_TYPE_VLAN``. In
851 the latter case, ``type`` refers to that of the outer header, with the inner
852 EtherType/TPID provided by the subsequent pattern item. This is the same
853 order as on the wire.
855 - ``dst``: destination MAC.
856 - ``src``: source MAC.
857 - ``type``: EtherType or TPID.
858 - Default ``mask`` matches destination and source addresses only.
863 Matches an 802.1Q/ad VLAN tag.
865 The corresponding standard outer EtherType (TPID) values are
866 ``RTE_ETHER_TYPE_VLAN`` or ``RTE_ETHER_TYPE_QINQ``. It can be overridden by the
867 preceding pattern item.
869 - ``tci``: tag control information.
870 - ``inner_type``: inner EtherType or TPID.
871 - Default ``mask`` matches the VID part of TCI only (lower 12 bits).
876 Matches an IPv4 header.
878 Note: IPv4 options are handled by dedicated pattern items.
880 - ``hdr``: IPv4 header definition (``rte_ip.h``).
881 - Default ``mask`` matches source and destination addresses only.
886 Matches an IPv6 header.
888 Note: IPv6 options are handled by dedicated pattern items, see `Item:
891 - ``hdr``: IPv6 header definition (``rte_ip.h``).
892 - Default ``mask`` matches source and destination addresses only.
897 Matches an ICMP header.
899 - ``hdr``: ICMP header definition (``rte_icmp.h``).
900 - Default ``mask`` matches ICMP type and code only.
905 Matches a UDP header.
907 - ``hdr``: UDP header definition (``rte_udp.h``).
908 - Default ``mask`` matches source and destination ports only.
913 Matches a TCP header.
915 - ``hdr``: TCP header definition (``rte_tcp.h``).
916 - Default ``mask`` matches source and destination ports only.
921 Matches a SCTP header.
923 - ``hdr``: SCTP header definition (``rte_sctp.h``).
924 - Default ``mask`` matches source and destination ports only.
929 Matches a VXLAN header (RFC 7348).
931 - ``flags``: normally 0x08 (I flag).
932 - ``rsvd0``: reserved, normally 0x000000.
933 - ``vni``: VXLAN network identifier.
934 - ``rsvd1``: reserved, normally 0x00.
935 - Default ``mask`` matches VNI only.
940 Matches an IEEE 802.1BR E-Tag header.
942 The corresponding standard outer EtherType (TPID) value is
943 ``RTE_ETHER_TYPE_ETAG``. It can be overridden by the preceding pattern item.
945 - ``epcp_edei_in_ecid_b``: E-Tag control information (E-TCI), E-PCP (3b),
946 E-DEI (1b), ingress E-CID base (12b).
947 - ``rsvd_grp_ecid_b``: reserved (2b), GRP (2b), E-CID base (12b).
948 - ``in_ecid_e``: ingress E-CID ext.
949 - ``ecid_e``: E-CID ext.
950 - ``inner_type``: inner EtherType or TPID.
951 - Default ``mask`` simultaneously matches GRP and E-CID base.
956 Matches a NVGRE header (RFC 7637).
958 - ``c_k_s_rsvd0_ver``: checksum (1b), undefined (1b), key bit (1b),
959 sequence number (1b), reserved 0 (9b), version (3b). This field must have
960 value 0x2000 according to RFC 7637.
961 - ``protocol``: protocol type (0x6558).
962 - ``tni``: virtual subnet ID.
963 - ``flow_id``: flow ID.
964 - Default ``mask`` matches TNI only.
969 Matches a MPLS header.
971 - ``label_tc_s_ttl``: label, TC, Bottom of Stack and TTL.
972 - Default ``mask`` matches label only.
977 Matches a GRE header.
979 - ``c_rsvd0_ver``: checksum, reserved 0 and version.
980 - ``protocol``: protocol type.
981 - Default ``mask`` matches protocol only.
986 Matches a GRE key field.
987 This should be preceded by item ``GRE``.
989 - Value to be matched is a big-endian 32 bit integer.
990 - When this item present it implicitly match K bit in default mask as "1"
995 Fuzzy pattern match, expect faster than default.
997 This is for device that support fuzzy match option. Usually a fuzzy match is
998 fast but the cost is accuracy. i.e. Signature Match only match pattern's hash
999 value, but it is possible two different patterns have the same hash value.
1001 Matching accuracy level can be configured by threshold. Driver can divide the
1002 range of threshold and map to different accuracy levels that device support.
1004 Threshold 0 means perfect match (no fuzziness), while threshold 0xffffffff
1005 means fuzziest match.
1007 .. _table_rte_flow_item_fuzzy:
1011 +----------+---------------+--------------------------------------------------+
1012 | Field | Subfield | Value |
1013 +==========+===============+==================================================+
1014 | ``spec`` | ``threshold`` | 0 as perfect match, 0xffffffff as fuzziest match |
1015 +----------+---------------+--------------------------------------------------+
1016 | ``last`` | ``threshold`` | upper range value |
1017 +----------+---------------+--------------------------------------------------+
1018 | ``mask`` | ``threshold`` | bit-mask apply to "spec" and "last" |
1019 +----------+---------------+--------------------------------------------------+
1021 Usage example, fuzzy match a TCPv4 packets:
1023 .. _table_rte_flow_item_fuzzy_example:
1025 .. table:: Fuzzy matching
1027 +-------+----------+
1029 +=======+==========+
1031 +-------+----------+
1033 +-------+----------+
1035 +-------+----------+
1037 +-------+----------+
1039 +-------+----------+
1041 Item: ``GTP``, ``GTPC``, ``GTPU``
1042 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1044 Matches a GTPv1 header.
1046 Note: GTP, GTPC and GTPU use the same structure. GTPC and GTPU item
1047 are defined for a user-friendly API when creating GTP-C and GTP-U
1050 - ``v_pt_rsv_flags``: version (3b), protocol type (1b), reserved (1b),
1051 extension header flag (1b), sequence number flag (1b), N-PDU number
1053 - ``msg_type``: message type.
1054 - ``msg_len``: message length.
1055 - ``teid``: tunnel endpoint identifier.
1056 - Default ``mask`` matches teid only.
1061 Matches an ESP header.
1063 - ``hdr``: ESP header definition (``rte_esp.h``).
1064 - Default ``mask`` matches SPI only.
1069 Matches a GENEVE header.
1071 - ``ver_opt_len_o_c_rsvd0``: version (2b), length of the options fields (6b),
1072 OAM packet (1b), critical options present (1b), reserved 0 (6b).
1073 - ``protocol``: protocol type.
1074 - ``vni``: virtual network identifier.
1075 - ``rsvd1``: reserved, normally 0x00.
1076 - Default ``mask`` matches VNI only.
1081 Matches a VXLAN-GPE header (draft-ietf-nvo3-vxlan-gpe-05).
1083 - ``flags``: normally 0x0C (I and P flags).
1084 - ``rsvd0``: reserved, normally 0x0000.
1085 - ``protocol``: protocol type.
1086 - ``vni``: VXLAN network identifier.
1087 - ``rsvd1``: reserved, normally 0x00.
1088 - Default ``mask`` matches VNI only.
1090 Item: ``ARP_ETH_IPV4``
1091 ^^^^^^^^^^^^^^^^^^^^^^
1093 Matches an ARP header for Ethernet/IPv4.
1095 - ``hdr``: hardware type, normally 1.
1096 - ``pro``: protocol type, normally 0x0800.
1097 - ``hln``: hardware address length, normally 6.
1098 - ``pln``: protocol address length, normally 4.
1099 - ``op``: opcode (1 for request, 2 for reply).
1100 - ``sha``: sender hardware address.
1101 - ``spa``: sender IPv4 address.
1102 - ``tha``: target hardware address.
1103 - ``tpa``: target IPv4 address.
1104 - Default ``mask`` matches SHA, SPA, THA and TPA.
1109 Matches the presence of any IPv6 extension header.
1111 - ``next_hdr``: next header.
1112 - Default ``mask`` matches ``next_hdr``.
1114 Normally preceded by any of:
1122 Matches any ICMPv6 header.
1124 - ``type``: ICMPv6 type.
1125 - ``code``: ICMPv6 code.
1126 - ``checksum``: ICMPv6 checksum.
1127 - Default ``mask`` matches ``type`` and ``code``.
1129 Item: ``ICMP6_ND_NS``
1130 ^^^^^^^^^^^^^^^^^^^^^
1132 Matches an ICMPv6 neighbor discovery solicitation.
1134 - ``type``: ICMPv6 type, normally 135.
1135 - ``code``: ICMPv6 code, normally 0.
1136 - ``checksum``: ICMPv6 checksum.
1137 - ``reserved``: reserved, normally 0.
1138 - ``target_addr``: target address.
1139 - Default ``mask`` matches target address only.
1141 Item: ``ICMP6_ND_NA``
1142 ^^^^^^^^^^^^^^^^^^^^^
1144 Matches an ICMPv6 neighbor discovery advertisement.
1146 - ``type``: ICMPv6 type, normally 136.
1147 - ``code``: ICMPv6 code, normally 0.
1148 - ``checksum``: ICMPv6 checksum.
1149 - ``rso_reserved``: route flag (1b), solicited flag (1b), override flag
1150 (1b), reserved (29b).
1151 - ``target_addr``: target address.
1152 - Default ``mask`` matches target address only.
1154 Item: ``ICMP6_ND_OPT``
1155 ^^^^^^^^^^^^^^^^^^^^^^
1157 Matches the presence of any ICMPv6 neighbor discovery option.
1159 - ``type``: ND option type.
1160 - ``length``: ND option length.
1161 - Default ``mask`` matches type only.
1163 Normally preceded by any of:
1165 - `Item: ICMP6_ND_NA`_
1166 - `Item: ICMP6_ND_NS`_
1167 - `Item: ICMP6_ND_OPT`_
1169 Item: ``ICMP6_ND_OPT_SLA_ETH``
1170 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1172 Matches an ICMPv6 neighbor discovery source Ethernet link-layer address
1175 - ``type``: ND option type, normally 1.
1176 - ``length``: ND option length, normally 1.
1177 - ``sla``: source Ethernet LLA.
1178 - Default ``mask`` matches source link-layer address only.
1180 Normally preceded by any of:
1182 - `Item: ICMP6_ND_NA`_
1183 - `Item: ICMP6_ND_OPT`_
1185 Item: ``ICMP6_ND_OPT_TLA_ETH``
1186 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1188 Matches an ICMPv6 neighbor discovery target Ethernet link-layer address
1191 - ``type``: ND option type, normally 2.
1192 - ``length``: ND option length, normally 1.
1193 - ``tla``: target Ethernet LLA.
1194 - Default ``mask`` matches target link-layer address only.
1196 Normally preceded by any of:
1198 - `Item: ICMP6_ND_NS`_
1199 - `Item: ICMP6_ND_OPT`_
1204 Matches an application specific 32 bit metadata item.
1206 - Default ``mask`` matches the specified metadata value.
1211 Matches a GTP PDU extension header with type 0x85.
1213 - ``pdu_type``: PDU type.
1214 - ``qfi``: QoS flow identifier.
1215 - Default ``mask`` matches QFI only.
1217 Item: ``PPPOES``, ``PPPOED``
1218 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1220 Matches a PPPoE header.
1222 - ``version_type``: version (4b), type (4b).
1223 - ``code``: message type.
1224 - ``session_id``: session identifier.
1225 - ``length``: payload length.
1227 Item: ``PPPOE_PROTO_ID``
1228 ^^^^^^^^^^^^^^^^^^^^^^^^
1230 Matches a PPPoE session protocol identifier.
1232 - ``proto_id``: PPP protocol identifier.
1233 - Default ``mask`` matches proto_id only.
1236 .. _table_rte_flow_item_meta:
1240 +----------+----------+---------------------------------------+
1241 | Field | Subfield | Value |
1242 +==========+==========+=======================================+
1243 | ``spec`` | ``data`` | 32 bit metadata value |
1244 +----------+--------------------------------------------------+
1245 | ``last`` | ``data`` | upper range value |
1246 +----------+----------+---------------------------------------+
1247 | ``mask`` | ``data`` | bit-mask applies to "spec" and "last" |
1248 +----------+----------+---------------------------------------+
1253 Matches a network service header (RFC 8300).
1255 - ``version``: normally 0x0 (2 bits).
1256 - ``oam_pkt``: indicate oam packet (1 bit).
1257 - ``reserved``: reserved bit (1 bit).
1258 - ``ttl``: maximum SFF hopes (6 bits).
1259 - ``length``: total length in 4 bytes words (6 bits).
1260 - ``reserved1``: reserved1 bits (4 bits).
1261 - ``mdtype``: ndicates format of NSH header (4 bits).
1262 - ``next_proto``: indicates protocol type of encap data (8 bits).
1263 - ``spi``: service path identifier (3 bytes).
1264 - ``sindex``: service index (1 byte).
1265 - Default ``mask`` matches mdtype, next_proto, spi, sindex.
1271 Matches a Internet Group Management Protocol (RFC 2236).
1273 - ``type``: IGMP message type (Query/Report).
1274 - ``max_resp_time``: max time allowed before sending report.
1275 - ``checksum``: checksum, 1s complement of whole IGMP message.
1276 - ``group_addr``: group address, for Query value will be 0.
1277 - Default ``mask`` matches group_addr.
1283 Matches a IP Authentication Header (RFC 4302).
1285 - ``next_hdr``: next payload after AH.
1286 - ``payload_len``: total length of AH in 4B words.
1287 - ``reserved``: reserved bits.
1288 - ``spi``: security parameters index.
1289 - ``seq_num``: counter value increased by 1 on each packet sent.
1290 - Default ``mask`` matches spi.
1296 Each possible action is represented by a type.
1297 An action can have an associated configuration object.
1298 Several actions combined in a list can be assigned
1299 to a flow rule and are performed in order.
1301 They fall in three categories:
1303 - Actions that modify the fate of matching traffic, for instance by dropping
1304 or assigning it a specific destination.
1306 - Actions that modify matching traffic contents or its properties. This
1307 includes adding/removing encapsulation, encryption, compression and marks.
1309 - Actions related to the flow rule itself, such as updating counters or
1310 making it non-terminating.
1312 Flow rules being terminating by default, not specifying any action of the
1313 fate kind results in undefined behavior. This applies to both ingress and
1316 PASSTHRU, when supported, makes a flow rule non-terminating.
1318 Like matching patterns, action lists are terminated by END items.
1320 Example of action that redirects packets to queue index 10:
1322 .. _table_rte_flow_action_example:
1324 .. table:: Queue action
1326 +-----------+-------+
1328 +===========+=======+
1330 +-----------+-------+
1332 Actions are performed in list order:
1334 .. _table_rte_flow_count_then_drop:
1336 .. table:: Count then drop
1350 .. _table_rte_flow_mark_count_redirect:
1352 .. table:: Mark, count then redirect
1354 +-------+--------+------------+-------+
1355 | Index | Action | Field | Value |
1356 +=======+========+============+=======+
1357 | 0 | MARK | ``mark`` | 0x2a |
1358 +-------+--------+------------+-------+
1359 | 1 | COUNT | ``shared`` | 0 |
1360 | | +------------+-------+
1362 +-------+--------+------------+-------+
1363 | 2 | QUEUE | ``queue`` | 10 |
1364 +-------+--------+------------+-------+
1366 +-------+-----------------------------+
1370 .. _table_rte_flow_redirect_queue_5:
1372 .. table:: Redirect to queue 5
1374 +-------+--------+-----------+-------+
1375 | Index | Action | Field | Value |
1376 +=======+========+===========+=======+
1378 +-------+--------+-----------+-------+
1379 | 1 | QUEUE | ``queue`` | 5 |
1380 +-------+--------+-----------+-------+
1382 +-------+----------------------------+
1384 In the above example, while DROP and QUEUE must be performed in order, both
1385 have to happen before reaching END. Only QUEUE has a visible effect.
1387 Note that such a list may be thought as ambiguous and rejected on that
1390 .. _table_rte_flow_redirect_queue_5_3:
1392 .. table:: Redirect to queues 5 and 3
1394 +-------+--------+-----------+-------+
1395 | Index | Action | Field | Value |
1396 +=======+========+===========+=======+
1397 | 0 | QUEUE | ``queue`` | 5 |
1398 +-------+--------+-----------+-------+
1400 +-------+--------+-----------+-------+
1401 | 2 | QUEUE | ``queue`` | 3 |
1402 +-------+--------+-----------+-------+
1404 +-------+----------------------------+
1406 As previously described, all actions must be taken into account. This
1407 effectively duplicates traffic to both queues. The above example also shows
1408 that VOID is ignored.
1413 Common action types are described in this section. Like pattern item types,
1414 this list is not exhaustive as new actions will be added in the future.
1419 End marker for action lists. Prevents further processing of actions, thereby
1422 - Its numeric value is 0 for convenience.
1423 - PMD support is mandatory.
1424 - No configurable properties.
1426 .. _table_rte_flow_action_end:
1439 Used as a placeholder for convenience. It is ignored and simply discarded by
1442 - PMD support is mandatory.
1443 - No configurable properties.
1445 .. _table_rte_flow_action_void:
1455 Action: ``PASSTHRU``
1456 ^^^^^^^^^^^^^^^^^^^^
1458 Leaves traffic up for additional processing by subsequent flow rules; makes
1459 a flow rule non-terminating.
1461 - No configurable properties.
1463 .. _table_rte_flow_action_passthru:
1473 Example to copy a packet to a queue and continue processing by subsequent
1476 .. _table_rte_flow_action_passthru_example:
1478 .. table:: Copy to queue 8
1480 +-------+--------+-----------+-------+
1481 | Index | Action | Field | Value |
1482 +=======+========+===========+=======+
1484 +-------+--------+-----------+-------+
1485 | 1 | QUEUE | ``queue`` | 8 |
1486 +-------+--------+-----------+-------+
1488 +-------+----------------------------+
1493 Redirects packets to a group on the current device.
1495 In a hierarchy of groups, which can be used to represent physical or logical
1496 flow group/tables on the device, this action redirects the matched flow to
1497 the specified group on that device.
1499 If a matched flow is redirected to a table which doesn't contain a matching
1500 rule for that flow then the behavior is undefined and the resulting behavior
1501 is up to the specific device. Best practice when using groups would be define
1502 a default flow rule for each group which a defines the default actions in that
1503 group so a consistent behavior is defined.
1505 Defining an action for matched flow in a group to jump to a group which is
1506 higher in the group hierarchy may not be supported by physical devices,
1507 depending on how groups are mapped to the physical devices. In the
1508 definitions of jump actions, applications should be aware that it may be
1509 possible to define flow rules which trigger an undefined behavior causing
1510 flows to loop between groups.
1512 .. _table_rte_flow_action_jump:
1516 +-----------+------------------------------+
1518 +===========+==============================+
1519 | ``group`` | Group to redirect packets to |
1520 +-----------+------------------------------+
1525 Attaches an integer value to packets and sets ``PKT_RX_FDIR`` and
1526 ``PKT_RX_FDIR_ID`` mbuf flags.
1528 This value is arbitrary and application-defined. Maximum allowed value
1529 depends on the underlying implementation. It is returned in the
1530 ``hash.fdir.hi`` mbuf field.
1532 .. _table_rte_flow_action_mark:
1536 +--------+--------------------------------------+
1538 +========+======================================+
1539 | ``id`` | integer value to return with packets |
1540 +--------+--------------------------------------+
1545 Flags packets. Similar to `Action: MARK`_ without a specific value; only
1546 sets the ``PKT_RX_FDIR`` mbuf flag.
1548 - No configurable properties.
1550 .. _table_rte_flow_action_flag:
1563 Assigns packets to a given queue index.
1565 .. _table_rte_flow_action_queue:
1569 +-----------+--------------------+
1571 +===========+====================+
1572 | ``index`` | queue index to use |
1573 +-----------+--------------------+
1580 - No configurable properties.
1582 .. _table_rte_flow_action_drop:
1595 Adds a counter action to a matched flow.
1597 If more than one count action is specified in a single flow rule, then each
1598 action must specify a unique id.
1600 Counters can be retrieved and reset through ``rte_flow_query()``, see
1601 ``struct rte_flow_query_count``.
1603 The shared flag indicates whether the counter is unique to the flow rule the
1604 action is specified with, or whether it is a shared counter.
1606 For a count action with the shared flag set, then then a global device
1607 namespace is assumed for the counter id, so that any matched flow rules using
1608 a count action with the same counter id on the same port will contribute to
1611 For ports within the same switch domain then the counter id namespace extends
1612 to all ports within that switch domain.
1614 .. _table_rte_flow_action_count:
1618 +------------+---------------------+
1620 +============+=====================+
1621 | ``shared`` | shared counter flag |
1622 +------------+---------------------+
1623 | ``id`` | counter id |
1624 +------------+---------------------+
1626 Query structure to retrieve and reset flow rule counters:
1628 .. _table_rte_flow_query_count:
1630 .. table:: COUNT query
1632 +---------------+-----+-----------------------------------+
1633 | Field | I/O | Value |
1634 +===============+=====+===================================+
1635 | ``reset`` | in | reset counter after query |
1636 +---------------+-----+-----------------------------------+
1637 | ``hits_set`` | out | ``hits`` field is set |
1638 +---------------+-----+-----------------------------------+
1639 | ``bytes_set`` | out | ``bytes`` field is set |
1640 +---------------+-----+-----------------------------------+
1641 | ``hits`` | out | number of hits for this rule |
1642 +---------------+-----+-----------------------------------+
1643 | ``bytes`` | out | number of bytes through this rule |
1644 +---------------+-----+-----------------------------------+
1649 Similar to QUEUE, except RSS is additionally performed on packets to spread
1650 them among several queues according to the provided parameters.
1652 Unlike global RSS settings used by other DPDK APIs, unsetting the ``types``
1653 field does not disable RSS in a flow rule. Doing so instead requests safe
1654 unspecified "best-effort" settings from the underlying PMD, which depending
1655 on the flow rule, may result in anything ranging from empty (single queue)
1656 to all-inclusive RSS.
1658 Note: RSS hash result is stored in the ``hash.rss`` mbuf field which
1659 overlaps ``hash.fdir.lo``. Since `Action: MARK`_ sets the ``hash.fdir.hi``
1660 field only, both can be requested simultaneously.
1662 Also, regarding packet encapsulation ``level``:
1664 - ``0`` requests the default behavior. Depending on the packet type, it can
1665 mean outermost, innermost, anything in between or even no RSS.
1667 It basically stands for the innermost encapsulation level RSS can be
1668 performed on according to PMD and device capabilities.
1670 - ``1`` requests RSS to be performed on the outermost packet encapsulation
1673 - ``2`` and subsequent values request RSS to be performed on the specified
1674 inner packet encapsulation level, from outermost to innermost (lower to
1677 Values other than ``0`` are not necessarily supported.
1679 Requesting a specific RSS level on unrecognized traffic results in undefined
1680 behavior. For predictable results, it is recommended to make the flow rule
1681 pattern match packet headers up to the requested encapsulation level so that
1682 only matching traffic goes through.
1684 .. _table_rte_flow_action_rss:
1688 +---------------+---------------------------------------------+
1690 +===============+=============================================+
1691 | ``func`` | RSS hash function to apply |
1692 +---------------+---------------------------------------------+
1693 | ``level`` | encapsulation level for ``types`` |
1694 +---------------+---------------------------------------------+
1695 | ``types`` | specific RSS hash types (see ``ETH_RSS_*``) |
1696 +---------------+---------------------------------------------+
1697 | ``key_len`` | hash key length in bytes |
1698 +---------------+---------------------------------------------+
1699 | ``queue_num`` | number of entries in ``queue`` |
1700 +---------------+---------------------------------------------+
1701 | ``key`` | hash key |
1702 +---------------+---------------------------------------------+
1703 | ``queue`` | queue indices to use |
1704 +---------------+---------------------------------------------+
1709 Directs matching traffic to the physical function (PF) of the current
1714 - No configurable properties.
1716 .. _table_rte_flow_action_pf:
1729 Directs matching traffic to a given virtual function of the current device.
1731 Packets matched by a VF pattern item can be redirected to their original VF
1732 ID instead of the specified one. This parameter may not be available and is
1733 not guaranteed to work properly if the VF part is matched by a prior flow
1734 rule or if packets are not addressed to a VF in the first place.
1738 .. _table_rte_flow_action_vf:
1742 +--------------+--------------------------------+
1744 +==============+================================+
1745 | ``original`` | use original VF ID if possible |
1746 +--------------+--------------------------------+
1748 +--------------+--------------------------------+
1750 Action: ``PHY_PORT``
1751 ^^^^^^^^^^^^^^^^^^^^
1753 Directs matching traffic to a given physical port index of the underlying
1756 See `Item: PHY_PORT`_.
1758 .. _table_rte_flow_action_phy_port:
1762 +--------------+-------------------------------------+
1764 +==============+=====================================+
1765 | ``original`` | use original port index if possible |
1766 +--------------+-------------------------------------+
1767 | ``index`` | physical port index |
1768 +--------------+-------------------------------------+
1772 Directs matching traffic to a given DPDK port ID.
1774 See `Item: PORT_ID`_.
1776 .. _table_rte_flow_action_port_id:
1780 +--------------+---------------------------------------+
1782 +==============+=======================================+
1783 | ``original`` | use original DPDK port ID if possible |
1784 +--------------+---------------------------------------+
1785 | ``id`` | DPDK port ID |
1786 +--------------+---------------------------------------+
1791 Applies a stage of metering and policing.
1793 The metering and policing (MTR) object has to be first created using the
1794 rte_mtr_create() API function. The ID of the MTR object is specified as
1795 action parameter. More than one flow can use the same MTR object through
1796 the meter action. The MTR object can be further updated or queried using
1799 .. _table_rte_flow_action_meter:
1803 +--------------+---------------+
1805 +==============+===============+
1806 | ``mtr_id`` | MTR object ID |
1807 +--------------+---------------+
1809 Action: ``SECURITY``
1810 ^^^^^^^^^^^^^^^^^^^^
1812 Perform the security action on flows matched by the pattern items
1813 according to the configuration of the security session.
1815 This action modifies the payload of matched flows. For INLINE_CRYPTO, the
1816 security protocol headers and IV are fully provided by the application as
1817 specified in the flow pattern. The payload of matching packets is
1818 encrypted on egress, and decrypted and authenticated on ingress.
1819 For INLINE_PROTOCOL, the security protocol is fully offloaded to HW,
1820 providing full encapsulation and decapsulation of packets in security
1821 protocols. The flow pattern specifies both the outer security header fields
1822 and the inner packet fields. The security session specified in the action
1823 must match the pattern parameters.
1825 The security session specified in the action must be created on the same
1826 port as the flow action that is being specified.
1828 The ingress/egress flow attribute should match that specified in the
1829 security session if the security session supports the definition of the
1832 Multiple flows can be configured to use the same security session.
1834 .. _table_rte_flow_action_security:
1838 +----------------------+--------------------------------------+
1840 +======================+======================================+
1841 | ``security_session`` | security session to apply |
1842 +----------------------+--------------------------------------+
1844 The following is an example of configuring IPsec inline using the
1845 INLINE_CRYPTO security session:
1847 The encryption algorithm, keys and salt are part of the opaque
1848 ``rte_security_session``. The SA is identified according to the IP and ESP
1849 fields in the pattern items.
1851 .. _table_rte_flow_item_esp_inline_example:
1853 .. table:: IPsec inline crypto flow pattern items.
1855 +-------+----------+
1857 +=======+==========+
1859 +-------+----------+
1861 +-------+----------+
1863 +-------+----------+
1865 +-------+----------+
1867 .. _table_rte_flow_action_esp_inline_example:
1869 .. table:: IPsec inline flow actions.
1871 +-------+----------+
1873 +=======+==========+
1875 +-------+----------+
1877 +-------+----------+
1879 Action: ``OF_SET_MPLS_TTL``
1880 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
1882 Implements ``OFPAT_SET_MPLS_TTL`` ("MPLS TTL") as defined by the `OpenFlow
1883 Switch Specification`_.
1885 .. _table_rte_flow_action_of_set_mpls_ttl:
1887 .. table:: OF_SET_MPLS_TTL
1889 +--------------+----------+
1891 +==============+==========+
1892 | ``mpls_ttl`` | MPLS TTL |
1893 +--------------+----------+
1895 Action: ``OF_DEC_MPLS_TTL``
1896 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
1898 Implements ``OFPAT_DEC_MPLS_TTL`` ("decrement MPLS TTL") as defined by the
1899 `OpenFlow Switch Specification`_.
1901 .. _table_rte_flow_action_of_dec_mpls_ttl:
1903 .. table:: OF_DEC_MPLS_TTL
1911 Action: ``OF_SET_NW_TTL``
1912 ^^^^^^^^^^^^^^^^^^^^^^^^^
1914 Implements ``OFPAT_SET_NW_TTL`` ("IP TTL") as defined by the `OpenFlow
1915 Switch Specification`_.
1917 .. _table_rte_flow_action_of_set_nw_ttl:
1919 .. table:: OF_SET_NW_TTL
1921 +------------+--------+
1923 +============+========+
1924 | ``nw_ttl`` | IP TTL |
1925 +------------+--------+
1927 Action: ``OF_DEC_NW_TTL``
1928 ^^^^^^^^^^^^^^^^^^^^^^^^^
1930 Implements ``OFPAT_DEC_NW_TTL`` ("decrement IP TTL") as defined by the
1931 `OpenFlow Switch Specification`_.
1933 .. _table_rte_flow_action_of_dec_nw_ttl:
1935 .. table:: OF_DEC_NW_TTL
1943 Action: ``OF_COPY_TTL_OUT``
1944 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
1946 Implements ``OFPAT_COPY_TTL_OUT`` ("copy TTL "outwards" -- from
1947 next-to-outermost to outermost") as defined by the `OpenFlow Switch
1950 .. _table_rte_flow_action_of_copy_ttl_out:
1952 .. table:: OF_COPY_TTL_OUT
1960 Action: ``OF_COPY_TTL_IN``
1961 ^^^^^^^^^^^^^^^^^^^^^^^^^^
1963 Implements ``OFPAT_COPY_TTL_IN`` ("copy TTL "inwards" -- from outermost to
1964 next-to-outermost") as defined by the `OpenFlow Switch Specification`_.
1966 .. _table_rte_flow_action_of_copy_ttl_in:
1968 .. table:: OF_COPY_TTL_IN
1976 Action: ``OF_POP_VLAN``
1977 ^^^^^^^^^^^^^^^^^^^^^^^
1979 Implements ``OFPAT_POP_VLAN`` ("pop the outer VLAN tag") as defined
1980 by the `OpenFlow Switch Specification`_.
1982 .. _table_rte_flow_action_of_pop_vlan:
1984 .. table:: OF_POP_VLAN
1992 Action: ``OF_PUSH_VLAN``
1993 ^^^^^^^^^^^^^^^^^^^^^^^^
1995 Implements ``OFPAT_PUSH_VLAN`` ("push a new VLAN tag") as defined by the
1996 `OpenFlow Switch Specification`_.
1998 .. _table_rte_flow_action_of_push_vlan:
2000 .. table:: OF_PUSH_VLAN
2002 +---------------+-----------+
2004 +===============+===========+
2005 | ``ethertype`` | EtherType |
2006 +---------------+-----------+
2008 Action: ``OF_SET_VLAN_VID``
2009 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
2011 Implements ``OFPAT_SET_VLAN_VID`` ("set the 802.1q VLAN id") as defined by
2012 the `OpenFlow Switch Specification`_.
2014 .. _table_rte_flow_action_of_set_vlan_vid:
2016 .. table:: OF_SET_VLAN_VID
2018 +--------------+---------+
2020 +==============+=========+
2021 | ``vlan_vid`` | VLAN id |
2022 +--------------+---------+
2024 Action: ``OF_SET_VLAN_PCP``
2025 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
2027 Implements ``OFPAT_SET_LAN_PCP`` ("set the 802.1q priority") as defined by
2028 the `OpenFlow Switch Specification`_.
2030 .. _table_rte_flow_action_of_set_vlan_pcp:
2032 .. table:: OF_SET_VLAN_PCP
2034 +--------------+---------------+
2036 +==============+===============+
2037 | ``vlan_pcp`` | VLAN priority |
2038 +--------------+---------------+
2040 Action: ``OF_POP_MPLS``
2041 ^^^^^^^^^^^^^^^^^^^^^^^
2043 Implements ``OFPAT_POP_MPLS`` ("pop the outer MPLS tag") as defined by the
2044 `OpenFlow Switch Specification`_.
2046 .. _table_rte_flow_action_of_pop_mpls:
2048 .. table:: OF_POP_MPLS
2050 +---------------+-----------+
2052 +===============+===========+
2053 | ``ethertype`` | EtherType |
2054 +---------------+-----------+
2056 Action: ``OF_PUSH_MPLS``
2057 ^^^^^^^^^^^^^^^^^^^^^^^^
2059 Implements ``OFPAT_PUSH_MPLS`` ("push a new MPLS tag") as defined by the
2060 `OpenFlow Switch Specification`_.
2062 .. _table_rte_flow_action_of_push_mpls:
2064 .. table:: OF_PUSH_MPLS
2066 +---------------+-----------+
2068 +===============+===========+
2069 | ``ethertype`` | EtherType |
2070 +---------------+-----------+
2072 Action: ``VXLAN_ENCAP``
2073 ^^^^^^^^^^^^^^^^^^^^^^^
2075 Performs a VXLAN encapsulation action by encapsulating the matched flow in the
2076 VXLAN tunnel as defined in the``rte_flow_action_vxlan_encap`` flow items
2079 This action modifies the payload of matched flows. The flow definition specified
2080 in the ``rte_flow_action_tunnel_encap`` action structure must define a valid
2081 VLXAN network overlay which conforms with RFC 7348 (Virtual eXtensible Local
2082 Area Network (VXLAN): A Framework for Overlaying Virtualized Layer 2 Networks
2083 over Layer 3 Networks). The pattern must be terminated with the
2084 RTE_FLOW_ITEM_TYPE_END item type.
2086 .. _table_rte_flow_action_vxlan_encap:
2088 .. table:: VXLAN_ENCAP
2090 +----------------+-------------------------------------+
2092 +================+=====================================+
2093 | ``definition`` | Tunnel end-point overlay definition |
2094 +----------------+-------------------------------------+
2096 .. _table_rte_flow_action_vxlan_encap_example:
2098 .. table:: IPv4 VxLAN flow pattern example.
2100 +-------+----------+
2102 +=======+==========+
2104 +-------+----------+
2106 +-------+----------+
2108 +-------+----------+
2110 +-------+----------+
2112 +-------+----------+
2114 Action: ``VXLAN_DECAP``
2115 ^^^^^^^^^^^^^^^^^^^^^^^
2117 Performs a decapsulation action by stripping all headers of the VXLAN tunnel
2118 network overlay from the matched flow.
2120 The flow items pattern defined for the flow rule with which a ``VXLAN_DECAP``
2121 action is specified, must define a valid VXLAN tunnel as per RFC7348. If the
2122 flow pattern does not specify a valid VXLAN tunnel then a
2123 RTE_FLOW_ERROR_TYPE_ACTION error should be returned.
2125 This action modifies the payload of matched flows.
2127 Action: ``NVGRE_ENCAP``
2128 ^^^^^^^^^^^^^^^^^^^^^^^
2130 Performs a NVGRE encapsulation action by encapsulating the matched flow in the
2131 NVGRE tunnel as defined in the``rte_flow_action_tunnel_encap`` flow item
2134 This action modifies the payload of matched flows. The flow definition specified
2135 in the ``rte_flow_action_tunnel_encap`` action structure must defined a valid
2136 NVGRE network overlay which conforms with RFC 7637 (NVGRE: Network
2137 Virtualization Using Generic Routing Encapsulation). The pattern must be
2138 terminated with the RTE_FLOW_ITEM_TYPE_END item type.
2140 .. _table_rte_flow_action_nvgre_encap:
2142 .. table:: NVGRE_ENCAP
2144 +----------------+-------------------------------------+
2146 +================+=====================================+
2147 | ``definition`` | NVGRE end-point overlay definition |
2148 +----------------+-------------------------------------+
2150 .. _table_rte_flow_action_nvgre_encap_example:
2152 .. table:: IPv4 NVGRE flow pattern example.
2154 +-------+----------+
2156 +=======+==========+
2158 +-------+----------+
2160 +-------+----------+
2162 +-------+----------+
2164 +-------+----------+
2166 Action: ``NVGRE_DECAP``
2167 ^^^^^^^^^^^^^^^^^^^^^^^
2169 Performs a decapsulation action by stripping all headers of the NVGRE tunnel
2170 network overlay from the matched flow.
2172 The flow items pattern defined for the flow rule with which a ``NVGRE_DECAP``
2173 action is specified, must define a valid NVGRE tunnel as per RFC7637. If the
2174 flow pattern does not specify a valid NVGRE tunnel then a
2175 RTE_FLOW_ERROR_TYPE_ACTION error should be returned.
2177 This action modifies the payload of matched flows.
2179 Action: ``RAW_ENCAP``
2180 ^^^^^^^^^^^^^^^^^^^^^
2182 Adds outer header whose template is provided in its data buffer,
2183 as defined in the ``rte_flow_action_raw_encap`` definition.
2185 This action modifies the payload of matched flows. The data supplied must
2186 be a valid header, either holding layer 2 data in case of adding layer 2 after
2187 decap layer 3 tunnel (for example MPLSoGRE) or complete tunnel definition
2188 starting from layer 2 and moving to the tunnel item itself. When applied to
2189 the original packet the resulting packet must be a valid packet.
2191 .. _table_rte_flow_action_raw_encap:
2193 .. table:: RAW_ENCAP
2195 +----------------+----------------------------------------+
2197 +================+========================================+
2198 | ``data`` | Encapsulation data |
2199 +----------------+----------------------------------------+
2200 | ``preserve`` | Bit-mask of data to preserve on output |
2201 +----------------+----------------------------------------+
2202 | ``size`` | Size of data and preserve |
2203 +----------------+----------------------------------------+
2205 Action: ``RAW_DECAP``
2206 ^^^^^^^^^^^^^^^^^^^^^^^
2208 Remove outer header whose template is provided in its data buffer,
2209 as defined in the ``rte_flow_action_raw_decap``
2211 This action modifies the payload of matched flows. The data supplied must
2212 be a valid header, either holding layer 2 data in case of removing layer 2
2213 before encapsulation of layer 3 tunnel (for example MPLSoGRE) or complete
2214 tunnel definition starting from layer 2 and moving to the tunnel item itself.
2215 When applied to the original packet the resulting packet must be a
2218 .. _table_rte_flow_action_raw_decap:
2220 .. table:: RAW_DECAP
2222 +----------------+----------------------------------------+
2224 +================+========================================+
2225 | ``data`` | Decapsulation data |
2226 +----------------+----------------------------------------+
2227 | ``size`` | Size of data |
2228 +----------------+----------------------------------------+
2230 Action: ``SET_IPV4_SRC``
2231 ^^^^^^^^^^^^^^^^^^^^^^^^
2233 Set a new IPv4 source address in the outermost IPv4 header.
2235 It must be used with a valid RTE_FLOW_ITEM_TYPE_IPV4 flow pattern item.
2236 Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned.
2238 .. _table_rte_flow_action_set_ipv4_src:
2240 .. table:: SET_IPV4_SRC
2242 +-----------------------------------------+
2244 +===============+=========================+
2245 | ``ipv4_addr`` | new IPv4 source address |
2246 +---------------+-------------------------+
2248 Action: ``SET_IPV4_DST``
2249 ^^^^^^^^^^^^^^^^^^^^^^^^
2251 Set a new IPv4 destination address in the outermost IPv4 header.
2253 It must be used with a valid RTE_FLOW_ITEM_TYPE_IPV4 flow pattern item.
2254 Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned.
2256 .. _table_rte_flow_action_set_ipv4_dst:
2258 .. table:: SET_IPV4_DST
2260 +---------------+------------------------------+
2262 +===============+==============================+
2263 | ``ipv4_addr`` | new IPv4 destination address |
2264 +---------------+------------------------------+
2266 Action: ``SET_IPV6_SRC``
2267 ^^^^^^^^^^^^^^^^^^^^^^^^
2269 Set a new IPv6 source address in the outermost IPv6 header.
2271 It must be used with a valid RTE_FLOW_ITEM_TYPE_IPV6 flow pattern item.
2272 Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned.
2274 .. _table_rte_flow_action_set_ipv6_src:
2276 .. table:: SET_IPV6_SRC
2278 +---------------+-------------------------+
2280 +===============+=========================+
2281 | ``ipv6_addr`` | new IPv6 source address |
2282 +---------------+-------------------------+
2284 Action: ``SET_IPV6_DST``
2285 ^^^^^^^^^^^^^^^^^^^^^^^^
2287 Set a new IPv6 destination address in the outermost IPv6 header.
2289 It must be used with a valid RTE_FLOW_ITEM_TYPE_IPV6 flow pattern item.
2290 Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned.
2292 .. _table_rte_flow_action_set_ipv6_dst:
2294 .. table:: SET_IPV6_DST
2296 +---------------+------------------------------+
2298 +===============+==============================+
2299 | ``ipv6_addr`` | new IPv6 destination address |
2300 +---------------+------------------------------+
2302 Action: ``SET_TP_SRC``
2303 ^^^^^^^^^^^^^^^^^^^^^^^^^
2305 Set a new source port number in the outermost TCP/UDP header.
2307 It must be used with a valid RTE_FLOW_ITEM_TYPE_TCP or RTE_FLOW_ITEM_TYPE_UDP
2308 flow pattern item. Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned.
2310 .. _table_rte_flow_action_set_tp_src:
2312 .. table:: SET_TP_SRC
2314 +----------+-------------------------+
2316 +==========+=========================+
2317 | ``port`` | new TCP/UDP source port |
2318 +---------------+--------------------+
2320 Action: ``SET_TP_DST``
2321 ^^^^^^^^^^^^^^^^^^^^^^^^^
2323 Set a new destination port number in the outermost TCP/UDP header.
2325 It must be used with a valid RTE_FLOW_ITEM_TYPE_TCP or RTE_FLOW_ITEM_TYPE_UDP
2326 flow pattern item. Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned.
2328 .. _table_rte_flow_action_set_tp_dst:
2330 .. table:: SET_TP_DST
2332 +----------+------------------------------+
2334 +==========+==============================+
2335 | ``port`` | new TCP/UDP destination port |
2336 +---------------+-------------------------+
2338 Action: ``MAC_SWAP``
2339 ^^^^^^^^^^^^^^^^^^^^^^^^^
2341 Swap the source and destination MAC addresses in the outermost Ethernet
2344 It must be used with a valid RTE_FLOW_ITEM_TYPE_ETH flow pattern item.
2345 Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned.
2347 .. _table_rte_flow_action_mac_swap:
2362 If there is no valid RTE_FLOW_ITEM_TYPE_IPV4 or RTE_FLOW_ITEM_TYPE_IPV6
2363 in pattern, Some PMDs will reject rule because behavior will be undefined.
2365 .. _table_rte_flow_action_dec_ttl:
2378 Assigns a new TTL value.
2380 If there is no valid RTE_FLOW_ITEM_TYPE_IPV4 or RTE_FLOW_ITEM_TYPE_IPV6
2381 in pattern, Some PMDs will reject rule because behavior will be undefined.
2383 .. _table_rte_flow_action_set_ttl:
2387 +---------------+--------------------+
2389 +===============+====================+
2390 | ``ttl_value`` | new TTL value |
2391 +---------------+--------------------+
2393 Action: ``SET_MAC_SRC``
2394 ^^^^^^^^^^^^^^^^^^^^^^^
2396 Set source MAC address.
2398 It must be used with a valid RTE_FLOW_ITEM_TYPE_ETH flow pattern item.
2399 Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned.
2401 .. _table_rte_flow_action_set_mac_src:
2403 .. table:: SET_MAC_SRC
2405 +--------------+---------------+
2407 +==============+===============+
2408 | ``mac_addr`` | MAC address |
2409 +--------------+---------------+
2411 Action: ``SET_MAC_DST``
2412 ^^^^^^^^^^^^^^^^^^^^^^^
2414 Set destination MAC address.
2416 It must be used with a valid RTE_FLOW_ITEM_TYPE_ETH flow pattern item.
2417 Otherwise, RTE_FLOW_ERROR_TYPE_ACTION error will be returned.
2419 .. _table_rte_flow_action_set_mac_dst:
2421 .. table:: SET_MAC_DST
2423 +--------------+---------------+
2425 +==============+===============+
2426 | ``mac_addr`` | MAC address |
2427 +--------------+---------------+
2429 Action: ``INC_TCP_SEQ``
2430 ^^^^^^^^^^^^^^^^^^^^^^^
2432 Increase sequence number in the outermost TCP header.
2433 Value to increase TCP sequence number by is a big-endian 32 bit integer.
2435 Using this action on non-matching traffic will result in undefined behavior.
2437 Action: ``DEC_TCP_SEQ``
2438 ^^^^^^^^^^^^^^^^^^^^^^^
2440 Decrease sequence number in the outermost TCP header.
2441 Value to decrease TCP sequence number by is a big-endian 32 bit integer.
2443 Using this action on non-matching traffic will result in undefined behavior.
2445 Action: ``INC_TCP_ACK``
2446 ^^^^^^^^^^^^^^^^^^^^^^^
2448 Increase acknowledgment number in the outermost TCP header.
2449 Value to increase TCP acknowledgment number by is a big-endian 32 bit integer.
2451 Using this action on non-matching traffic will result in undefined behavior.
2453 Action: ``DEC_TCP_ACK``
2454 ^^^^^^^^^^^^^^^^^^^^^^^
2456 Decrease acknowledgment number in the outermost TCP header.
2457 Value to decrease TCP acknowledgment number by is a big-endian 32 bit integer.
2459 Using this action on non-matching traffic will result in undefined behavior.
2464 All specified pattern items (``enum rte_flow_item_type``) and actions
2465 (``enum rte_flow_action_type``) use positive identifiers.
2467 The negative space is reserved for dynamic types generated by PMDs during
2468 run-time. PMDs may encounter them as a result but must not accept negative
2469 identifiers they are not aware of.
2471 A method to generate them remains to be defined.
2476 Pattern item types will be added as new protocols are implemented.
2478 Variable headers support through dedicated pattern items, for example in
2479 order to match specific IPv4 options and IPv6 extension headers would be
2480 stacked after IPv4/IPv6 items.
2482 Other action types are planned but are not defined yet. These include the
2483 ability to alter packet data in several ways, such as performing
2484 encapsulation/decapsulation of tunnel headers.
2489 A rather simple API with few functions is provided to fully manage flow
2492 Each created flow rule is associated with an opaque, PMD-specific handle
2493 pointer. The application is responsible for keeping it until the rule is
2496 Flows rules are represented by ``struct rte_flow`` objects.
2501 Given that expressing a definite set of device capabilities is not
2502 practical, a dedicated function is provided to check if a flow rule is
2503 supported and can be created.
2508 rte_flow_validate(uint16_t port_id,
2509 const struct rte_flow_attr *attr,
2510 const struct rte_flow_item pattern[],
2511 const struct rte_flow_action actions[],
2512 struct rte_flow_error *error);
2514 The flow rule is validated for correctness and whether it could be accepted
2515 by the device given sufficient resources. The rule is checked against the
2516 current device mode and queue configuration. The flow rule may also
2517 optionally be validated against existing flow rules and device resources.
2518 This function has no effect on the target device.
2520 The returned value is guaranteed to remain valid only as long as no
2521 successful calls to ``rte_flow_create()`` or ``rte_flow_destroy()`` are made
2522 in the meantime and no device parameter affecting flow rules in any way are
2523 modified, due to possible collisions or resource limitations (although in
2524 such cases ``EINVAL`` should not be returned).
2528 - ``port_id``: port identifier of Ethernet device.
2529 - ``attr``: flow rule attributes.
2530 - ``pattern``: pattern specification (list terminated by the END pattern
2532 - ``actions``: associated actions (list terminated by the END action).
2533 - ``error``: perform verbose error reporting if not NULL. PMDs initialize
2534 this structure in case of error only.
2538 - 0 if flow rule is valid and can be created. A negative errno value
2539 otherwise (``rte_errno`` is also set), the following errors are defined.
2540 - ``-ENOSYS``: underlying device does not support this functionality.
2541 - ``-EINVAL``: unknown or invalid rule specification.
2542 - ``-ENOTSUP``: valid but unsupported rule specification (e.g. partial
2543 bit-masks are unsupported).
2544 - ``EEXIST``: collision with an existing rule. Only returned if device
2545 supports flow rule collision checking and there was a flow rule
2546 collision. Not receiving this return code is no guarantee that creating
2547 the rule will not fail due to a collision.
2548 - ``ENOMEM``: not enough memory to execute the function, or if the device
2549 supports resource validation, resource limitation on the device.
2550 - ``-EBUSY``: action cannot be performed due to busy device resources, may
2551 succeed if the affected queues or even the entire port are in a stopped
2552 state (see ``rte_eth_dev_rx_queue_stop()`` and ``rte_eth_dev_stop()``).
2557 Creating a flow rule is similar to validating one, except the rule is
2558 actually created and a handle returned.
2563 rte_flow_create(uint16_t port_id,
2564 const struct rte_flow_attr *attr,
2565 const struct rte_flow_item pattern[],
2566 const struct rte_flow_action *actions[],
2567 struct rte_flow_error *error);
2571 - ``port_id``: port identifier of Ethernet device.
2572 - ``attr``: flow rule attributes.
2573 - ``pattern``: pattern specification (list terminated by the END pattern
2575 - ``actions``: associated actions (list terminated by the END action).
2576 - ``error``: perform verbose error reporting if not NULL. PMDs initialize
2577 this structure in case of error only.
2581 A valid handle in case of success, NULL otherwise and ``rte_errno`` is set
2582 to the positive version of one of the error codes defined for
2583 ``rte_flow_validate()``.
2588 Flow rules destruction is not automatic, and a queue or a port should not be
2589 released if any are still attached to them. Applications must take care of
2590 performing this step before releasing resources.
2595 rte_flow_destroy(uint16_t port_id,
2596 struct rte_flow *flow,
2597 struct rte_flow_error *error);
2600 Failure to destroy a flow rule handle may occur when other flow rules depend
2601 on it, and destroying it would result in an inconsistent state.
2603 This function is only guaranteed to succeed if handles are destroyed in
2604 reverse order of their creation.
2608 - ``port_id``: port identifier of Ethernet device.
2609 - ``flow``: flow rule handle to destroy.
2610 - ``error``: perform verbose error reporting if not NULL. PMDs initialize
2611 this structure in case of error only.
2615 - 0 on success, a negative errno value otherwise and ``rte_errno`` is set.
2620 Convenience function to destroy all flow rule handles associated with a
2621 port. They are released as with successive calls to ``rte_flow_destroy()``.
2626 rte_flow_flush(uint16_t port_id,
2627 struct rte_flow_error *error);
2629 In the unlikely event of failure, handles are still considered destroyed and
2630 no longer valid but the port must be assumed to be in an inconsistent state.
2634 - ``port_id``: port identifier of Ethernet device.
2635 - ``error``: perform verbose error reporting if not NULL. PMDs initialize
2636 this structure in case of error only.
2640 - 0 on success, a negative errno value otherwise and ``rte_errno`` is set.
2645 Query an existing flow rule.
2647 This function allows retrieving flow-specific data such as counters. Data
2648 is gathered by special actions which must be present in the flow rule
2654 rte_flow_query(uint16_t port_id,
2655 struct rte_flow *flow,
2656 const struct rte_flow_action *action,
2658 struct rte_flow_error *error);
2662 - ``port_id``: port identifier of Ethernet device.
2663 - ``flow``: flow rule handle to query.
2664 - ``action``: action to query, this must match prototype from flow rule.
2665 - ``data``: pointer to storage for the associated query data type.
2666 - ``error``: perform verbose error reporting if not NULL. PMDs initialize
2667 this structure in case of error only.
2671 - 0 on success, a negative errno value otherwise and ``rte_errno`` is set.
2673 .. _flow_isolated_mode:
2678 The general expectation for ingress traffic is that flow rules process it
2679 first; the remaining unmatched or pass-through traffic usually ends up in a
2680 queue (with or without RSS, locally or in some sub-device instance)
2681 depending on the global configuration settings of a port.
2683 While fine from a compatibility standpoint, this approach makes drivers more
2684 complex as they have to check for possible side effects outside of this API
2685 when creating or destroying flow rules. It results in a more limited set of
2686 available rule types due to the way device resources are assigned (e.g. no
2687 support for the RSS action even on capable hardware).
2689 Given that nonspecific traffic can be handled by flow rules as well,
2690 isolated mode is a means for applications to tell a driver that ingress on
2691 the underlying port must be injected from the defined flow rules only; that
2692 no default traffic is expected outside those rules.
2694 This has the following benefits:
2696 - Applications get finer-grained control over the kind of traffic they want
2697 to receive (no traffic by default).
2699 - More importantly they control at what point nonspecific traffic is handled
2700 relative to other flow rules, by adjusting priority levels.
2702 - Drivers can assign more hardware resources to flow rules and expand the
2703 set of supported rule types.
2705 Because toggling isolated mode may cause profound changes to the ingress
2706 processing path of a driver, it may not be possible to leave it once
2707 entered. Likewise, existing flow rules or global configuration settings may
2708 prevent a driver from entering isolated mode.
2710 Applications relying on this mode are therefore encouraged to toggle it as
2711 soon as possible after device initialization, ideally before the first call
2712 to ``rte_eth_dev_configure()`` to avoid possible failures due to conflicting
2715 Once effective, the following functionality has no effect on the underlying
2716 port and may return errors such as ``ENOTSUP`` ("not supported"):
2718 - Toggling promiscuous mode.
2719 - Toggling allmulticast mode.
2720 - Configuring MAC addresses.
2721 - Configuring multicast addresses.
2722 - Configuring VLAN filters.
2723 - Configuring Rx filters through the legacy API (e.g. FDIR).
2724 - Configuring global RSS settings.
2729 rte_flow_isolate(uint16_t port_id, int set, struct rte_flow_error *error);
2733 - ``port_id``: port identifier of Ethernet device.
2734 - ``set``: nonzero to enter isolated mode, attempt to leave it otherwise.
2735 - ``error``: perform verbose error reporting if not NULL. PMDs initialize
2736 this structure in case of error only.
2740 - 0 on success, a negative errno value otherwise and ``rte_errno`` is set.
2742 Verbose error reporting
2743 -----------------------
2745 The defined *errno* values may not be accurate enough for users or
2746 application developers who want to investigate issues related to flow rules
2747 management. A dedicated error object is defined for this purpose:
2751 enum rte_flow_error_type {
2752 RTE_FLOW_ERROR_TYPE_NONE, /**< No error. */
2753 RTE_FLOW_ERROR_TYPE_UNSPECIFIED, /**< Cause unspecified. */
2754 RTE_FLOW_ERROR_TYPE_HANDLE, /**< Flow rule (handle). */
2755 RTE_FLOW_ERROR_TYPE_ATTR_GROUP, /**< Group field. */
2756 RTE_FLOW_ERROR_TYPE_ATTR_PRIORITY, /**< Priority field. */
2757 RTE_FLOW_ERROR_TYPE_ATTR_INGRESS, /**< Ingress field. */
2758 RTE_FLOW_ERROR_TYPE_ATTR_EGRESS, /**< Egress field. */
2759 RTE_FLOW_ERROR_TYPE_ATTR, /**< Attributes structure. */
2760 RTE_FLOW_ERROR_TYPE_ITEM_NUM, /**< Pattern length. */
2761 RTE_FLOW_ERROR_TYPE_ITEM, /**< Specific pattern item. */
2762 RTE_FLOW_ERROR_TYPE_ACTION_NUM, /**< Number of actions. */
2763 RTE_FLOW_ERROR_TYPE_ACTION, /**< Specific action. */
2766 struct rte_flow_error {
2767 enum rte_flow_error_type type; /**< Cause field and error types. */
2768 const void *cause; /**< Object responsible for the error. */
2769 const char *message; /**< Human-readable error message. */
2772 Error type ``RTE_FLOW_ERROR_TYPE_NONE`` stands for no error, in which case
2773 remaining fields can be ignored. Other error types describe the type of the
2774 object pointed by ``cause``.
2776 If non-NULL, ``cause`` points to the object responsible for the error. For a
2777 flow rule, this may be a pattern item or an individual action.
2779 If non-NULL, ``message`` provides a human-readable error message.
2781 This object is normally allocated by applications and set by PMDs in case of
2782 error, the message points to a constant string which does not need to be
2783 freed by the application, however its pointer can be considered valid only
2784 as long as its associated DPDK port remains configured. Closing the
2785 underlying device or unloading the PMD invalidates it.
2796 rte_flow_error_set(struct rte_flow_error *error,
2798 enum rte_flow_error_type type,
2800 const char *message);
2802 This function initializes ``error`` (if non-NULL) with the provided
2803 parameters and sets ``rte_errno`` to ``code``. A negative error ``code`` is
2812 rte_flow_conv(enum rte_flow_conv_op op,
2816 struct rte_flow_error *error);
2818 Convert ``src`` to ``dst`` according to operation ``op``. Possible
2821 - Attributes, pattern item or action duplication.
2822 - Duplication of an entire pattern or list of actions.
2823 - Duplication of a complete flow rule description.
2824 - Pattern item or action name retrieval.
2829 - DPDK does not keep track of flow rules definitions or flow rule objects
2830 automatically. Applications may keep track of the former and must keep
2831 track of the latter. PMDs may also do it for internal needs, however this
2832 must not be relied on by applications.
2834 - Flow rules are not maintained between successive port initializations. An
2835 application exiting without releasing them and restarting must re-create
2838 - API operations are synchronous and blocking (``EAGAIN`` cannot be
2841 - There is no provision for re-entrancy/multi-thread safety, although nothing
2842 should prevent different devices from being configured at the same
2843 time. PMDs may protect their control path functions accordingly.
2845 - Stopping the data path (TX/RX) should not be necessary when managing flow
2846 rules. If this cannot be achieved naturally or with workarounds (such as
2847 temporarily replacing the burst function pointers), an appropriate error
2848 code must be returned (``EBUSY``).
2850 - PMDs, not applications, are responsible for maintaining flow rules
2851 configuration when stopping and restarting a port or performing other
2852 actions which may affect them. They can only be destroyed explicitly by
2855 For devices exposing multiple ports sharing global settings affected by flow
2858 - All ports under DPDK control must behave consistently, PMDs are
2859 responsible for making sure that existing flow rules on a port are not
2860 affected by other ports.
2862 - Ports not under DPDK control (unaffected or handled by other applications)
2863 are user's responsibility. They may affect existing flow rules and cause
2864 undefined behavior. PMDs aware of this may prevent flow rules creation
2865 altogether in such cases.
2870 The PMD interface is defined in ``rte_flow_driver.h``. It is not subject to
2871 API/ABI versioning constraints as it is not exposed to applications and may
2872 evolve independently.
2874 It is currently implemented on top of the legacy filtering framework through
2875 filter type *RTE_ETH_FILTER_GENERIC* that accepts the single operation
2876 *RTE_ETH_FILTER_GET* to return PMD-specific *rte_flow* callbacks wrapped
2877 inside ``struct rte_flow_ops``.
2879 This overhead is temporarily necessary in order to keep compatibility with
2880 the legacy filtering framework, which should eventually disappear.
2882 - PMD callbacks implement exactly the interface described in `Rules
2883 management`_, except for the port ID argument which has already been
2884 converted to a pointer to the underlying ``struct rte_eth_dev``.
2886 - Public API functions do not process flow rules definitions at all before
2887 calling PMD functions (no basic error checking, no validation
2888 whatsoever). They only make sure these callbacks are non-NULL or return
2889 the ``ENOSYS`` (function not supported) error.
2891 This interface additionally defines the following helper function:
2893 - ``rte_flow_ops_get()``: get generic flow operations structure from a
2896 More will be added over time.
2898 Device compatibility
2899 --------------------
2901 No known implementation supports all the described features.
2903 Unsupported features or combinations are not expected to be fully emulated
2904 in software by PMDs for performance reasons. Partially supported features
2905 may be completed in software as long as hardware performs most of the work
2906 (such as queue redirection and packet recognition).
2908 However PMDs are expected to do their best to satisfy application requests
2909 by working around hardware limitations as long as doing so does not affect
2910 the behavior of existing flow rules.
2912 The following sections provide a few examples of such cases and describe how
2913 PMDs should handle them, they are based on limitations built into the
2919 Each flow rule comes with its own, per-layer bit-masks, while hardware may
2920 support only a single, device-wide bit-mask for a given layer type, so that
2921 two IPv4 rules cannot use different bit-masks.
2923 The expected behavior in this case is that PMDs automatically configure
2924 global bit-masks according to the needs of the first flow rule created.
2926 Subsequent rules are allowed only if their bit-masks match those, the
2927 ``EEXIST`` error code should be returned otherwise.
2929 Unsupported layer types
2930 ~~~~~~~~~~~~~~~~~~~~~~~
2932 Many protocols can be simulated by crafting patterns with the `Item: RAW`_
2935 PMDs can rely on this capability to simulate support for protocols with
2936 headers not directly recognized by hardware.
2938 ``ANY`` pattern item
2939 ~~~~~~~~~~~~~~~~~~~~
2941 This pattern item stands for anything, which can be difficult to translate
2942 to something hardware would understand, particularly if followed by more
2945 Consider the following pattern:
2947 .. _table_rte_flow_unsupported_any:
2949 .. table:: Pattern with ANY as L3
2951 +-------+-----------------------+
2953 +=======+=======================+
2955 +-------+-----+---------+-------+
2956 | 1 | ANY | ``num`` | ``1`` |
2957 +-------+-----+---------+-------+
2959 +-------+-----------------------+
2961 +-------+-----------------------+
2963 Knowing that TCP does not make sense with something other than IPv4 and IPv6
2964 as L3, such a pattern may be translated to two flow rules instead:
2966 .. _table_rte_flow_unsupported_any_ipv4:
2968 .. table:: ANY replaced with IPV4
2970 +-------+--------------------+
2972 +=======+====================+
2974 +-------+--------------------+
2975 | 1 | IPV4 (zeroed mask) |
2976 +-------+--------------------+
2978 +-------+--------------------+
2980 +-------+--------------------+
2984 .. _table_rte_flow_unsupported_any_ipv6:
2986 .. table:: ANY replaced with IPV6
2988 +-------+--------------------+
2990 +=======+====================+
2992 +-------+--------------------+
2993 | 1 | IPV6 (zeroed mask) |
2994 +-------+--------------------+
2996 +-------+--------------------+
2998 +-------+--------------------+
3000 Note that as soon as a ANY rule covers several layers, this approach may
3001 yield a large number of hidden flow rules. It is thus suggested to only
3002 support the most common scenarios (anything as L2 and/or L3).
3007 - When combined with `Action: QUEUE`_, packet counting (`Action: COUNT`_)
3008 and tagging (`Action: MARK`_ or `Action: FLAG`_) may be implemented in
3009 software as long as the target queue is used by a single rule.
3011 - When a single target queue is provided, `Action: RSS`_ can also be
3012 implemented through `Action: QUEUE`_.
3017 While it would naturally make sense, flow rules cannot be assumed to be
3018 processed by hardware in the same order as their creation for several
3021 - They may be managed internally as a tree or a hash table instead of a
3023 - Removing a flow rule before adding another one can either put the new rule
3024 at the end of the list or reuse a freed entry.
3025 - Duplication may occur when packets are matched by several rules.
3027 For overlapping rules (particularly in order to use `Action: PASSTHRU`_)
3028 predictable behavior is only guaranteed by using different priority levels.
3030 Priority levels are not necessarily implemented in hardware, or may be
3031 severely limited (e.g. a single priority bit).
3033 For these reasons, priority levels may be implemented purely in software by
3036 - For devices expecting flow rules to be added in the correct order, PMDs
3037 may destroy and re-create existing rules after adding a new one with
3040 - A configurable number of dummy or empty rules can be created at
3041 initialization time to save high priority slots for later.
3043 - In order to save priority levels, PMDs may evaluate whether rules are
3044 likely to collide and adjust their priority accordingly.
3049 - A device profile selection function which could be used to force a
3050 permanent profile instead of relying on its automatic configuration based
3051 on existing flow rules.
3053 - A method to optimize *rte_flow* rules with specific pattern items and
3054 action types generated on the fly by PMDs. DPDK should assign negative
3055 numbers to these in order to not collide with the existing types. See
3058 - Adding specific egress pattern items and actions as described in
3059 `Attribute: Traffic direction`_.
3061 - Optional software fallback when PMDs are unable to handle requested flow
3062 rules so applications do not have to implement their own.
3064 .. _OpenFlow Switch Specification: https://www.opennetworking.org/software-defined-standards/specifications/