1 # SPDX-License-Identifier: GPL-2.0-only
3 # Traffic control configuration.
7 bool "QoS and/or fair queueing"
10 When the kernel has several packets to send out over a network
11 device, it has to decide which ones to send first, which ones to
12 delay, and which ones to drop. This is the job of the queueing
13 disciplines, several different algorithms for how to do this
14 "fairly" have been proposed.
16 If you say N here, you will get the standard packet scheduler, which
17 is a FIFO (first come, first served). If you say Y here, you will be
18 able to choose from among several alternative algorithms which can
19 then be attached to different network devices. This is useful for
20 example if some of your network devices are real time devices that
21 need a certain minimum data flow rate, or if you need to limit the
22 maximum data flow rate for traffic which matches specified criteria.
23 This code is considered to be experimental.
25 To administer these schedulers, you'll need the user-level utilities
26 from the package iproute2+tc at
27 <https://www.kernel.org/pub/linux/utils/net/iproute2/>. That package
28 also contains some documentation; for more, check out
29 <http://www.linuxfoundation.org/collaborate/workgroups/networking/iproute2>.
31 This Quality of Service (QoS) support will enable you to use
32 Differentiated Services (diffserv) and Resource Reservation Protocol
33 (RSVP) on your Linux router if you also say Y to the corresponding
34 classifiers below. Documentation and software is at
35 <http://diffserv.sourceforge.net/>.
37 If you say Y here and to "/proc file system" below, you will be able
38 to read status information about packet schedulers from the file
41 The available schedulers are listed in the following questions; you
42 can say Y to as many as you like. If unsure, say N now.
46 comment "Queueing/Scheduling"
49 tristate "Hierarchical Token Bucket (HTB)"
51 Say Y here if you want to use the Hierarchical Token Buckets (HTB)
52 packet scheduling algorithm. See
53 <http://luxik.cdi.cz/~devik/qos/htb/> for complete manual and
56 HTB is very similar to CBQ regarding its goals however is has
57 different properties and different algorithm.
59 To compile this code as a module, choose M here: the
60 module will be called sch_htb.
63 tristate "Hierarchical Fair Service Curve (HFSC)"
65 Say Y here if you want to use the Hierarchical Fair Service Curve
66 (HFSC) packet scheduling algorithm.
68 To compile this code as a module, choose M here: the
69 module will be called sch_hfsc.
72 tristate "Multi Band Priority Queueing (PRIO)"
74 Say Y here if you want to use an n-band priority queue packet
77 To compile this code as a module, choose M here: the
78 module will be called sch_prio.
81 tristate "Hardware Multiqueue-aware Multi Band Queuing (MULTIQ)"
83 Say Y here if you want to use an n-band queue packet scheduler
84 to support devices that have multiple hardware transmit queues.
86 To compile this code as a module, choose M here: the
87 module will be called sch_multiq.
90 tristate "Random Early Detection (RED)"
92 Say Y here if you want to use the Random Early Detection (RED)
93 packet scheduling algorithm.
95 See the top of <file:net/sched/sch_red.c> for more details.
97 To compile this code as a module, choose M here: the
98 module will be called sch_red.
101 tristate "Stochastic Fair Blue (SFB)"
103 Say Y here if you want to use the Stochastic Fair Blue (SFB)
104 packet scheduling algorithm.
106 See the top of <file:net/sched/sch_sfb.c> for more details.
108 To compile this code as a module, choose M here: the
109 module will be called sch_sfb.
112 tristate "Stochastic Fairness Queueing (SFQ)"
114 Say Y here if you want to use the Stochastic Fairness Queueing (SFQ)
115 packet scheduling algorithm.
117 See the top of <file:net/sched/sch_sfq.c> for more details.
119 To compile this code as a module, choose M here: the
120 module will be called sch_sfq.
123 tristate "True Link Equalizer (TEQL)"
125 Say Y here if you want to use the True Link Equalizer (TLE) packet
126 scheduling algorithm. This queueing discipline allows the combination
127 of several physical devices into one virtual device.
129 See the top of <file:net/sched/sch_teql.c> for more details.
131 To compile this code as a module, choose M here: the
132 module will be called sch_teql.
135 tristate "Token Bucket Filter (TBF)"
137 Say Y here if you want to use the Token Bucket Filter (TBF) packet
138 scheduling algorithm.
140 See the top of <file:net/sched/sch_tbf.c> for more details.
142 To compile this code as a module, choose M here: the
143 module will be called sch_tbf.
146 tristate "Credit Based Shaper (CBS)"
148 Say Y here if you want to use the Credit Based Shaper (CBS) packet
149 scheduling algorithm.
151 See the top of <file:net/sched/sch_cbs.c> for more details.
153 To compile this code as a module, choose M here: the
154 module will be called sch_cbs.
157 tristate "Earliest TxTime First (ETF)"
159 Say Y here if you want to use the Earliest TxTime First (ETF) packet
160 scheduling algorithm.
162 See the top of <file:net/sched/sch_etf.c> for more details.
164 To compile this code as a module, choose M here: the
165 module will be called sch_etf.
167 config NET_SCH_TAPRIO
168 tristate "Time Aware Priority (taprio) Scheduler"
170 Say Y here if you want to use the Time Aware Priority (taprio) packet
171 scheduling algorithm.
173 See the top of <file:net/sched/sch_taprio.c> for more details.
175 To compile this code as a module, choose M here: the
176 module will be called sch_taprio.
179 tristate "Generic Random Early Detection (GRED)"
181 Say Y here if you want to use the Generic Random Early Detection
182 (GRED) packet scheduling algorithm for some of your network devices
183 (see the top of <file:net/sched/sch_red.c> for details and
184 references about the algorithm).
186 To compile this code as a module, choose M here: the
187 module will be called sch_gred.
190 tristate "Network emulator (NETEM)"
192 Say Y if you want to emulate network delay, loss, and packet
193 re-ordering. This is often useful to simulate networks when
194 testing applications or protocols.
196 To compile this driver as a module, choose M here: the module
197 will be called sch_netem.
202 tristate "Deficit Round Robin scheduler (DRR)"
204 Say Y here if you want to use the Deficit Round Robin (DRR) packet
205 scheduling algorithm.
207 To compile this driver as a module, choose M here: the module
208 will be called sch_drr.
212 config NET_SCH_MQPRIO
213 tristate "Multi-queue priority scheduler (MQPRIO)"
215 Say Y here if you want to use the Multi-queue Priority scheduler.
216 This scheduler allows QOS to be offloaded on NICs that have support
217 for offloading QOS schedulers.
219 To compile this driver as a module, choose M here: the module will
220 be called sch_mqprio.
224 config NET_SCH_SKBPRIO
225 tristate "SKB priority queue scheduler (SKBPRIO)"
227 Say Y here if you want to use the SKB priority queue
228 scheduler. This schedules packets according to skb->priority,
229 which is useful for request packets in DoS mitigation systems such
232 To compile this driver as a module, choose M here: the module will
233 be called sch_skbprio.
238 tristate "CHOose and Keep responsive flow scheduler (CHOKE)"
240 Say Y here if you want to use the CHOKe packet scheduler (CHOose
241 and Keep for responsive flows, CHOose and Kill for unresponsive
242 flows). This is a variation of RED which tries to penalize flows
243 that monopolize the queue.
245 To compile this code as a module, choose M here: the
246 module will be called sch_choke.
249 tristate "Quick Fair Queueing scheduler (QFQ)"
251 Say Y here if you want to use the Quick Fair Queueing Scheduler (QFQ)
252 packet scheduling algorithm.
254 To compile this driver as a module, choose M here: the module
255 will be called sch_qfq.
260 tristate "Controlled Delay AQM (CODEL)"
262 Say Y here if you want to use the Controlled Delay (CODEL)
263 packet scheduling algorithm.
265 To compile this driver as a module, choose M here: the module
266 will be called sch_codel.
270 config NET_SCH_FQ_CODEL
271 tristate "Fair Queue Controlled Delay AQM (FQ_CODEL)"
273 Say Y here if you want to use the FQ Controlled Delay (FQ_CODEL)
274 packet scheduling algorithm.
276 To compile this driver as a module, choose M here: the module
277 will be called sch_fq_codel.
282 tristate "Common Applications Kept Enhanced (CAKE)"
284 Say Y here if you want to use the Common Applications Kept Enhanced
285 (CAKE) queue management algorithm.
287 To compile this driver as a module, choose M here: the module
288 will be called sch_cake.
293 tristate "Fair Queue"
295 Say Y here if you want to use the FQ packet scheduling algorithm.
297 FQ does flow separation, and is able to respect pacing requirements
298 set by TCP stack into sk->sk_pacing_rate (for localy generated
301 To compile this driver as a module, choose M here: the module
302 will be called sch_fq.
307 tristate "Heavy-Hitter Filter (HHF)"
309 Say Y here if you want to use the Heavy-Hitter Filter (HHF)
310 packet scheduling algorithm.
312 To compile this driver as a module, choose M here: the module
313 will be called sch_hhf.
316 tristate "Proportional Integral controller Enhanced (PIE) scheduler"
318 Say Y here if you want to use the Proportional Integral controller
319 Enhanced scheduler packet scheduling algorithm.
320 For more information, please see https://tools.ietf.org/html/rfc8033
322 To compile this driver as a module, choose M here: the module
323 will be called sch_pie.
327 config NET_SCH_FQ_PIE
328 depends on NET_SCH_PIE
329 tristate "Flow Queue Proportional Integral controller Enhanced (FQ-PIE)"
331 Say Y here if you want to use the Flow Queue Proportional Integral
332 controller Enhanced (FQ-PIE) packet scheduling algorithm.
333 For more information, please see https://tools.ietf.org/html/rfc8033
335 To compile this driver as a module, choose M here: the module
336 will be called sch_fq_pie.
340 config NET_SCH_INGRESS
341 tristate "Ingress/classifier-action Qdisc"
342 depends on NET_CLS_ACT
346 Say Y here if you want to use classifiers for incoming and/or outgoing
347 packets. This qdisc doesn't do anything else besides running classifiers,
348 which can also have actions attached to them. In case of outgoing packets,
349 classifiers that this qdisc holds are executed in the transmit path
350 before real enqueuing to an egress qdisc happens.
354 To compile this code as a module, choose M here: the module will be
355 called sch_ingress with alias of sch_clsact.
358 tristate "Plug network traffic until release (PLUG)"
361 This queuing discipline allows userspace to plug/unplug a network
362 output queue, using the netlink interface. When it receives an
363 enqueue command it inserts a plug into the outbound queue that
364 causes following packets to enqueue until a dequeue command arrives
365 over netlink, causing the plug to be removed and resuming the normal
368 This module also provides a generic "network output buffering"
369 functionality (aka output commit), wherein upon arrival of a dequeue
370 command, only packets up to the first plug are released for delivery.
371 The Remus HA project uses this module to enable speculative execution
372 of virtual machines by allowing the generated network output to be rolled
375 For more information, please refer to <http://wiki.xenproject.org/wiki/Remus>
377 Say Y here if you are using this kernel for Xen dom0 and
378 want to protect Xen guests with Remus.
380 To compile this code as a module, choose M here: the
381 module will be called sch_plug.
384 tristate "Enhanced transmission selection scheduler (ETS)"
386 The Enhanced Transmission Selection scheduler is a classful
387 queuing discipline that merges functionality of PRIO and DRR
388 qdiscs in one scheduler. ETS makes it easy to configure a set of
389 strict and bandwidth-sharing bands to implement the transmission
390 selection described in 802.1Qaz.
392 Say Y here if you want to use the ETS packet scheduling
395 To compile this driver as a module, choose M here: the module
396 will be called sch_ets.
400 menuconfig NET_SCH_DEFAULT
401 bool "Allow override default queue discipline"
403 Support for selection of default queuing discipline.
405 Nearly all users can safely say no here, and the default
406 of pfifo_fast will be used. Many distributions already set
407 the default value via /proc/sys/net/core/default_qdisc.
414 prompt "Default queuing discipline"
415 default DEFAULT_PFIFO_FAST
417 Select the queueing discipline that will be used by default
418 for all network devices.
421 bool "Fair Queue" if NET_SCH_FQ
424 bool "Controlled Delay" if NET_SCH_CODEL
426 config DEFAULT_FQ_CODEL
427 bool "Fair Queue Controlled Delay" if NET_SCH_FQ_CODEL
429 config DEFAULT_FQ_PIE
430 bool "Flow Queue Proportional Integral controller Enhanced" if NET_SCH_FQ_PIE
433 bool "Stochastic Fair Queue" if NET_SCH_SFQ
435 config DEFAULT_PFIFO_FAST
436 bool "Priority FIFO Fast"
439 config DEFAULT_NET_SCH
441 default "pfifo_fast" if DEFAULT_PFIFO_FAST
442 default "fq" if DEFAULT_FQ
443 default "fq_codel" if DEFAULT_FQ_CODEL
444 default "fq_pie" if DEFAULT_FQ_PIE
445 default "sfq" if DEFAULT_SFQ
449 comment "Classification"
455 tristate "Elementary classification (BASIC)"
458 Say Y here if you want to be able to classify packets using
459 only extended matches and actions.
461 To compile this code as a module, choose M here: the
462 module will be called cls_basic.
464 config NET_CLS_ROUTE4
465 tristate "Routing decision (ROUTE)"
467 select IP_ROUTE_CLASSID
470 If you say Y here, you will be able to classify packets
471 according to the route table entry they matched.
473 To compile this code as a module, choose M here: the
474 module will be called cls_route.
477 tristate "Netfilter mark (FW)"
480 If you say Y here, you will be able to classify packets
481 according to netfilter/firewall marks.
483 To compile this code as a module, choose M here: the
484 module will be called cls_fw.
487 tristate "Universal 32bit comparisons w/ hashing (U32)"
490 Say Y here to be able to classify packets using a universal
491 32bit pieces based comparison scheme.
493 To compile this code as a module, choose M here: the
494 module will be called cls_u32.
497 bool "Performance counters support"
498 depends on NET_CLS_U32
500 Say Y here to make u32 gather additional statistics useful for
501 fine tuning u32 classifiers.
504 bool "Netfilter marks support"
505 depends on NET_CLS_U32
507 Say Y here to be able to use netfilter marks as u32 key.
510 tristate "Flow classifier"
513 If you say Y here, you will be able to classify packets based on
514 a configurable combination of packet keys. This is mostly useful
515 in combination with SFQ.
517 To compile this code as a module, choose M here: the
518 module will be called cls_flow.
520 config NET_CLS_CGROUP
521 tristate "Control Group Classifier"
523 select CGROUP_NET_CLASSID
526 Say Y here if you want to classify packets based on the control
527 cgroup of their process.
529 To compile this code as a module, choose M here: the
530 module will be called cls_cgroup.
533 tristate "BPF-based classifier"
536 If you say Y here, you will be able to classify packets based on
537 programmable BPF (JIT'ed) filters as an alternative to ematches.
539 To compile this code as a module, choose M here: the module will
542 config NET_CLS_FLOWER
543 tristate "Flower classifier"
546 If you say Y here, you will be able to classify packets based on
547 a configurable combination of packet keys and masks.
549 To compile this code as a module, choose M here: the module will
550 be called cls_flower.
552 config NET_CLS_MATCHALL
553 tristate "Match-all classifier"
556 If you say Y here, you will be able to classify packets based on
557 nothing. Every packet will match.
559 To compile this code as a module, choose M here: the module will
560 be called cls_matchall.
563 bool "Extended Matches"
566 Say Y here if you want to use extended matches on top of classifiers
567 and select the extended matches below.
569 Extended matches are small classification helpers not worth writing
570 a separate classifier for.
572 A recent version of the iproute2 package is required to use
575 config NET_EMATCH_STACK
577 depends on NET_EMATCH
580 Size of the local stack variable used while evaluating the tree of
581 ematches. Limits the depth of the tree, i.e. the number of
582 encapsulated precedences. Every level requires 4 bytes of additional
585 config NET_EMATCH_CMP
586 tristate "Simple packet data comparison"
587 depends on NET_EMATCH
589 Say Y here if you want to be able to classify packets based on
590 simple packet data comparisons for 8, 16, and 32bit values.
592 To compile this code as a module, choose M here: the
593 module will be called em_cmp.
595 config NET_EMATCH_NBYTE
596 tristate "Multi byte comparison"
597 depends on NET_EMATCH
599 Say Y here if you want to be able to classify packets based on
600 multiple byte comparisons mainly useful for IPv6 address comparisons.
602 To compile this code as a module, choose M here: the
603 module will be called em_nbyte.
605 config NET_EMATCH_U32
607 depends on NET_EMATCH
609 Say Y here if you want to be able to classify packets using
610 the famous u32 key in combination with logic relations.
612 To compile this code as a module, choose M here: the
613 module will be called em_u32.
615 config NET_EMATCH_META
617 depends on NET_EMATCH
619 Say Y here if you want to be able to classify packets based on
620 metadata such as load average, netfilter attributes, socket
621 attributes and routing decisions.
623 To compile this code as a module, choose M here: the
624 module will be called em_meta.
626 config NET_EMATCH_TEXT
627 tristate "Textsearch"
628 depends on NET_EMATCH
630 select TEXTSEARCH_KMP
632 select TEXTSEARCH_FSM
634 Say Y here if you want to be able to classify packets based on
635 textsearch comparisons.
637 To compile this code as a module, choose M here: the
638 module will be called em_text.
640 config NET_EMATCH_CANID
641 tristate "CAN Identifier"
642 depends on NET_EMATCH && (CAN=y || CAN=m)
644 Say Y here if you want to be able to classify CAN frames based
647 To compile this code as a module, choose M here: the
648 module will be called em_canid.
650 config NET_EMATCH_IPSET
652 depends on NET_EMATCH && IP_SET
654 Say Y here if you want to be able to classify packets based on
657 To compile this code as a module, choose M here: the
658 module will be called em_ipset.
660 config NET_EMATCH_IPT
661 tristate "IPtables Matches"
662 depends on NET_EMATCH && NETFILTER && NETFILTER_XTABLES
664 Say Y here to be able to classify packets based on iptables
666 Current supported match is "policy" which allows packet classification
667 based on IPsec policy that was used during decapsulation
669 To compile this code as a module, choose M here: the
670 module will be called em_ipt.
676 Say Y here if you want to use traffic control actions. Actions
677 get attached to classifiers and are invoked after a successful
678 classification. They are used to overwrite the classification
679 result, instantly drop or redirect packets, etc.
681 A recent version of the iproute2 package is required to use
684 config NET_ACT_POLICE
685 tristate "Traffic Policing"
686 depends on NET_CLS_ACT
688 Say Y here if you want to do traffic policing, i.e. strict
689 bandwidth limiting. This action replaces the existing policing
692 To compile this code as a module, choose M here: the
693 module will be called act_police.
696 tristate "Generic actions"
697 depends on NET_CLS_ACT
699 Say Y here to take generic actions such as dropping and
702 To compile this code as a module, choose M here: the
703 module will be called act_gact.
706 bool "Probability support"
707 depends on NET_ACT_GACT
709 Say Y here to use the generic action randomly or deterministically.
711 config NET_ACT_MIRRED
712 tristate "Redirecting and Mirroring"
713 depends on NET_CLS_ACT
715 Say Y here to allow packets to be mirrored or redirected to
718 To compile this code as a module, choose M here: the
719 module will be called act_mirred.
721 config NET_ACT_SAMPLE
722 tristate "Traffic Sampling"
723 depends on NET_CLS_ACT
726 Say Y here to allow packet sampling tc action. The packet sample
727 action consists of statistically choosing packets and sampling
728 them using the psample module.
730 To compile this code as a module, choose M here: the
731 module will be called act_sample.
734 tristate "IPtables targets"
735 depends on NET_CLS_ACT && NETFILTER && NETFILTER_XTABLES
737 Say Y here to be able to invoke iptables targets after successful
740 To compile this code as a module, choose M here: the
741 module will be called act_ipt.
744 tristate "Stateless NAT"
745 depends on NET_CLS_ACT
747 Say Y here to do stateless NAT on IPv4 packets. You should use
748 netfilter for NAT unless you know what you are doing.
750 To compile this code as a module, choose M here: the
751 module will be called act_nat.
754 tristate "Packet Editing"
755 depends on NET_CLS_ACT
757 Say Y here if you want to mangle the content of packets.
759 To compile this code as a module, choose M here: the
760 module will be called act_pedit.
763 tristate "Simple Example (Debug)"
764 depends on NET_CLS_ACT
766 Say Y here to add a simple action for demonstration purposes.
767 It is meant as an example and for debugging purposes. It will
768 print a configured policy string followed by the packet count
769 to the console for every packet that passes by.
773 To compile this code as a module, choose M here: the
774 module will be called act_simple.
776 config NET_ACT_SKBEDIT
777 tristate "SKB Editing"
778 depends on NET_CLS_ACT
780 Say Y here to change skb priority or queue_mapping settings.
784 To compile this code as a module, choose M here: the
785 module will be called act_skbedit.
788 tristate "Checksum Updating"
789 depends on NET_CLS_ACT && INET
792 Say Y here to update some common checksum after some direct
795 To compile this code as a module, choose M here: the
796 module will be called act_csum.
799 tristate "MPLS manipulation"
800 depends on NET_CLS_ACT
802 Say Y here to push or pop MPLS headers.
806 To compile this code as a module, choose M here: the
807 module will be called act_mpls.
810 tristate "Vlan manipulation"
811 depends on NET_CLS_ACT
813 Say Y here to push or pop vlan headers.
817 To compile this code as a module, choose M here: the
818 module will be called act_vlan.
821 tristate "BPF based action"
822 depends on NET_CLS_ACT
824 Say Y here to execute BPF code on packets. The BPF code will decide
825 if the packet should be dropped or not.
829 To compile this code as a module, choose M here: the
830 module will be called act_bpf.
832 config NET_ACT_CONNMARK
833 tristate "Netfilter Connection Mark Retriever"
834 depends on NET_CLS_ACT && NETFILTER
835 depends on NF_CONNTRACK && NF_CONNTRACK_MARK
837 Say Y here to allow retrieving of conn mark
841 To compile this code as a module, choose M here: the
842 module will be called act_connmark.
844 config NET_ACT_CTINFO
845 tristate "Netfilter Connection Mark Actions"
846 depends on NET_CLS_ACT && NETFILTER
847 depends on NF_CONNTRACK && NF_CONNTRACK_MARK
849 Say Y here to allow transfer of a connmark stored information.
850 Current actions transfer connmark stored DSCP into
851 ipv4/v6 diffserv and/or to transfer connmark to packet
852 mark. Both are useful for restoring egress based marks
853 back onto ingress connections for qdisc priority mapping
858 To compile this code as a module, choose M here: the
859 module will be called act_ctinfo.
861 config NET_ACT_SKBMOD
862 tristate "skb data modification action"
863 depends on NET_CLS_ACT
865 Say Y here to allow modification of skb data
869 To compile this code as a module, choose M here: the
870 module will be called act_skbmod.
873 tristate "Inter-FE action based on IETF ForCES InterFE LFB"
874 depends on NET_CLS_ACT
877 Say Y here to allow for sourcing and terminating metadata
878 For details refer to netdev01 paper:
879 "Distributing Linux Traffic Control Classifier-Action Subsystem"
880 Authors: Jamal Hadi Salim and Damascene M. Joachimpillai
882 To compile this code as a module, choose M here: the
883 module will be called act_ife.
885 config NET_ACT_TUNNEL_KEY
886 tristate "IP tunnel metadata manipulation"
887 depends on NET_CLS_ACT
889 Say Y here to set/release ip tunnel metadata.
893 To compile this code as a module, choose M here: the
894 module will be called act_tunnel_key.
897 tristate "connection tracking tc action"
898 depends on NET_CLS_ACT && NF_CONNTRACK && (!NF_NAT || NF_NAT) && NF_FLOW_TABLE
900 Say Y here to allow sending the packets to conntrack module.
904 To compile this code as a module, choose M here: the
905 module will be called act_ct.
908 tristate "Frame gate entry list control tc action"
909 depends on NET_CLS_ACT
911 Say Y here to allow to control the ingress flow to be passed at
912 specific time slot and be dropped at other specific time slot by
916 To compile this code as a module, choose M here: the
917 module will be called act_gate.
919 config NET_IFE_SKBMARK
920 tristate "Support to encoding decoding skb mark on IFE action"
921 depends on NET_ACT_IFE
923 config NET_IFE_SKBPRIO
924 tristate "Support to encoding decoding skb prio on IFE action"
925 depends on NET_ACT_IFE
927 config NET_IFE_SKBTCINDEX
928 tristate "Support to encoding decoding skb tcindex on IFE action"
929 depends on NET_ACT_IFE
931 config NET_TC_SKB_EXT
932 bool "TC recirculation support"
933 depends on NET_CLS_ACT
934 select SKB_EXTENSIONS
937 Say Y here to allow tc chain misses to continue in OvS datapath in
938 the correct recirc_id, and hardware chain misses to continue in
939 the correct chain in tc software datapath.
941 Say N here if you won't be using tc<->ovs offload or tc chains offload.