2 # Traffic control configuration.
6 bool "QoS and/or fair queueing"
9 When the kernel has several packets to send out over a network
10 device, it has to decide which ones to send first, which ones to
11 delay, and which ones to drop. This is the job of the queueing
12 disciplines, several different algorithms for how to do this
13 "fairly" have been proposed.
15 If you say N here, you will get the standard packet scheduler, which
16 is a FIFO (first come, first served). If you say Y here, you will be
17 able to choose from among several alternative algorithms which can
18 then be attached to different network devices. This is useful for
19 example if some of your network devices are real time devices that
20 need a certain minimum data flow rate, or if you need to limit the
21 maximum data flow rate for traffic which matches specified criteria.
22 This code is considered to be experimental.
24 To administer these schedulers, you'll need the user-level utilities
25 from the package iproute2+tc at <ftp://ftp.tux.org/pub/net/ip-routing/>.
26 That package also contains some documentation; for more, check out
27 <http://www.linuxfoundation.org/collaborate/workgroups/networking/iproute2>.
29 This Quality of Service (QoS) support will enable you to use
30 Differentiated Services (diffserv) and Resource Reservation Protocol
31 (RSVP) on your Linux router if you also say Y to the corresponding
32 classifiers below. Documentation and software is at
33 <http://diffserv.sourceforge.net/>.
35 If you say Y here and to "/proc file system" below, you will be able
36 to read status information about packet schedulers from the file
39 The available schedulers are listed in the following questions; you
40 can say Y to as many as you like. If unsure, say N now.
44 comment "Queueing/Scheduling"
47 tristate "Class Based Queueing (CBQ)"
49 Say Y here if you want to use the Class-Based Queueing (CBQ) packet
50 scheduling algorithm. This algorithm classifies the waiting packets
51 into a tree-like hierarchy of classes; the leaves of this tree are
52 in turn scheduled by separate algorithms.
54 See the top of <file:net/sched/sch_cbq.c> for more details.
56 CBQ is a commonly used scheduler, so if you're unsure, you should
57 say Y here. Then say Y to all the queueing algorithms below that you
58 want to use as leaf disciplines.
60 To compile this code as a module, choose M here: the
61 module will be called sch_cbq.
64 tristate "Hierarchical Token Bucket (HTB)"
66 Say Y here if you want to use the Hierarchical Token Buckets (HTB)
67 packet scheduling algorithm. See
68 <http://luxik.cdi.cz/~devik/qos/htb/> for complete manual and
71 HTB is very similar to CBQ regarding its goals however is has
72 different properties and different algorithm.
74 To compile this code as a module, choose M here: the
75 module will be called sch_htb.
78 tristate "Hierarchical Fair Service Curve (HFSC)"
80 Say Y here if you want to use the Hierarchical Fair Service Curve
81 (HFSC) packet scheduling algorithm.
83 To compile this code as a module, choose M here: the
84 module will be called sch_hfsc.
87 tristate "ATM Virtual Circuits (ATM)"
90 Say Y here if you want to use the ATM pseudo-scheduler. This
91 provides a framework for invoking classifiers, which in turn
92 select classes of this queuing discipline. Each class maps
93 the flow(s) it is handling to a given virtual circuit.
95 See the top of <file:net/sched/sch_atm.c> for more details.
97 To compile this code as a module, choose M here: the
98 module will be called sch_atm.
101 tristate "Multi Band Priority Queueing (PRIO)"
103 Say Y here if you want to use an n-band priority queue packet
106 To compile this code as a module, choose M here: the
107 module will be called sch_prio.
109 config NET_SCH_MULTIQ
110 tristate "Hardware Multiqueue-aware Multi Band Queuing (MULTIQ)"
112 Say Y here if you want to use an n-band queue packet scheduler
113 to support devices that have multiple hardware transmit queues.
115 To compile this code as a module, choose M here: the
116 module will be called sch_multiq.
119 tristate "Random Early Detection (RED)"
121 Say Y here if you want to use the Random Early Detection (RED)
122 packet scheduling algorithm.
124 See the top of <file:net/sched/sch_red.c> for more details.
126 To compile this code as a module, choose M here: the
127 module will be called sch_red.
130 tristate "Stochastic Fair Blue (SFB)"
132 Say Y here if you want to use the Stochastic Fair Blue (SFB)
133 packet scheduling algorithm.
135 See the top of <file:net/sched/sch_sfb.c> for more details.
137 To compile this code as a module, choose M here: the
138 module will be called sch_sfb.
141 tristate "Stochastic Fairness Queueing (SFQ)"
143 Say Y here if you want to use the Stochastic Fairness Queueing (SFQ)
144 packet scheduling algorithm.
146 See the top of <file:net/sched/sch_sfq.c> for more details.
148 To compile this code as a module, choose M here: the
149 module will be called sch_sfq.
152 tristate "True Link Equalizer (TEQL)"
154 Say Y here if you want to use the True Link Equalizer (TLE) packet
155 scheduling algorithm. This queueing discipline allows the combination
156 of several physical devices into one virtual device.
158 See the top of <file:net/sched/sch_teql.c> for more details.
160 To compile this code as a module, choose M here: the
161 module will be called sch_teql.
164 tristate "Token Bucket Filter (TBF)"
166 Say Y here if you want to use the Token Bucket Filter (TBF) packet
167 scheduling algorithm.
169 See the top of <file:net/sched/sch_tbf.c> for more details.
171 To compile this code as a module, choose M here: the
172 module will be called sch_tbf.
175 tristate "Generic Random Early Detection (GRED)"
177 Say Y here if you want to use the Generic Random Early Detection
178 (GRED) packet scheduling algorithm for some of your network devices
179 (see the top of <file:net/sched/sch_red.c> for details and
180 references about the algorithm).
182 To compile this code as a module, choose M here: the
183 module will be called sch_gred.
185 config NET_SCH_DSMARK
186 tristate "Differentiated Services marker (DSMARK)"
188 Say Y if you want to schedule packets according to the
189 Differentiated Services architecture proposed in RFC 2475.
190 Technical information on this method, with pointers to associated
191 RFCs, is available at <http://www.gta.ufrj.br/diffserv/>.
193 To compile this code as a module, choose M here: the
194 module will be called sch_dsmark.
197 tristate "Network emulator (NETEM)"
199 Say Y if you want to emulate network delay, loss, and packet
200 re-ordering. This is often useful to simulate networks when
201 testing applications or protocols.
203 To compile this driver as a module, choose M here: the module
204 will be called sch_netem.
209 tristate "Deficit Round Robin scheduler (DRR)"
211 Say Y here if you want to use the Deficit Round Robin (DRR) packet
212 scheduling algorithm.
214 To compile this driver as a module, choose M here: the module
215 will be called sch_drr.
219 config NET_SCH_MQPRIO
220 tristate "Multi-queue priority scheduler (MQPRIO)"
222 Say Y here if you want to use the Multi-queue Priority scheduler.
223 This scheduler allows QOS to be offloaded on NICs that have support
224 for offloading QOS schedulers.
226 To compile this driver as a module, choose M here: the module will
227 be called sch_mqprio.
232 tristate "CHOose and Keep responsive flow scheduler (CHOKE)"
234 Say Y here if you want to use the CHOKe packet scheduler (CHOose
235 and Keep for responsive flows, CHOose and Kill for unresponsive
236 flows). This is a variation of RED which trys to penalize flows
237 that monopolize the queue.
239 To compile this code as a module, choose M here: the
240 module will be called sch_choke.
243 tristate "Quick Fair Queueing scheduler (QFQ)"
245 Say Y here if you want to use the Quick Fair Queueing Scheduler (QFQ)
246 packet scheduling algorithm.
248 To compile this driver as a module, choose M here: the module
249 will be called sch_qfq.
254 tristate "Controlled Delay AQM (CODEL)"
256 Say Y here if you want to use the Controlled Delay (CODEL)
257 packet scheduling algorithm.
259 To compile this driver as a module, choose M here: the module
260 will be called sch_codel.
264 config NET_SCH_FQ_CODEL
265 tristate "Fair Queue Controlled Delay AQM (FQ_CODEL)"
267 Say Y here if you want to use the FQ Controlled Delay (FQ_CODEL)
268 packet scheduling algorithm.
270 To compile this driver as a module, choose M here: the module
271 will be called sch_fq_codel.
276 tristate "Fair Queue"
278 Say Y here if you want to use the FQ packet scheduling algorithm.
280 FQ does flow separation, and is able to respect pacing requirements
281 set by TCP stack into sk->sk_pacing_rate (for localy generated
284 To compile this driver as a module, choose M here: the module
285 will be called sch_fq.
290 tristate "Heavy-Hitter Filter (HHF)"
292 Say Y here if you want to use the Heavy-Hitter Filter (HHF)
293 packet scheduling algorithm.
295 To compile this driver as a module, choose M here: the module
296 will be called sch_hhf.
298 config NET_SCH_INGRESS
299 tristate "Ingress Qdisc"
300 depends on NET_CLS_ACT
302 Say Y here if you want to use classifiers for incoming packets.
305 To compile this code as a module, choose M here: the
306 module will be called sch_ingress.
309 tristate "Plug network traffic until release (PLUG)"
312 This queuing discipline allows userspace to plug/unplug a network
313 output queue, using the netlink interface. When it receives an
314 enqueue command it inserts a plug into the outbound queue that
315 causes following packets to enqueue until a dequeue command arrives
316 over netlink, causing the plug to be removed and resuming the normal
319 This module also provides a generic "network output buffering"
320 functionality (aka output commit), wherein upon arrival of a dequeue
321 command, only packets up to the first plug are released for delivery.
322 The Remus HA project uses this module to enable speculative execution
323 of virtual machines by allowing the generated network output to be rolled
326 For more information, please refer to http://wiki.xensource.com/xenwiki/Remus
328 Say Y here if you are using this kernel for Xen dom0 and
329 want to protect Xen guests with Remus.
331 To compile this code as a module, choose M here: the
332 module will be called sch_plug.
334 comment "Classification"
340 tristate "Elementary classification (BASIC)"
343 Say Y here if you want to be able to classify packets using
344 only extended matches and actions.
346 To compile this code as a module, choose M here: the
347 module will be called cls_basic.
349 config NET_CLS_TCINDEX
350 tristate "Traffic-Control Index (TCINDEX)"
353 Say Y here if you want to be able to classify packets based on
354 traffic control indices. You will want this feature if you want
355 to implement Differentiated Services together with DSMARK.
357 To compile this code as a module, choose M here: the
358 module will be called cls_tcindex.
360 config NET_CLS_ROUTE4
361 tristate "Routing decision (ROUTE)"
363 select IP_ROUTE_CLASSID
366 If you say Y here, you will be able to classify packets
367 according to the route table entry they matched.
369 To compile this code as a module, choose M here: the
370 module will be called cls_route.
373 tristate "Netfilter mark (FW)"
376 If you say Y here, you will be able to classify packets
377 according to netfilter/firewall marks.
379 To compile this code as a module, choose M here: the
380 module will be called cls_fw.
383 tristate "Universal 32bit comparisons w/ hashing (U32)"
386 Say Y here to be able to classify packets using a universal
387 32bit pieces based comparison scheme.
389 To compile this code as a module, choose M here: the
390 module will be called cls_u32.
393 bool "Performance counters support"
394 depends on NET_CLS_U32
396 Say Y here to make u32 gather additional statistics useful for
397 fine tuning u32 classifiers.
400 bool "Netfilter marks support"
401 depends on NET_CLS_U32
403 Say Y here to be able to use netfilter marks as u32 key.
406 tristate "IPv4 Resource Reservation Protocol (RSVP)"
409 The Resource Reservation Protocol (RSVP) permits end systems to
410 request a minimum and maximum data flow rate for a connection; this
411 is important for real time data such as streaming sound or video.
413 Say Y here if you want to be able to classify outgoing packets based
414 on their RSVP requests.
416 To compile this code as a module, choose M here: the
417 module will be called cls_rsvp.
420 tristate "IPv6 Resource Reservation Protocol (RSVP6)"
423 The Resource Reservation Protocol (RSVP) permits end systems to
424 request a minimum and maximum data flow rate for a connection; this
425 is important for real time data such as streaming sound or video.
427 Say Y here if you want to be able to classify outgoing packets based
428 on their RSVP requests and you are using the IPv6 protocol.
430 To compile this code as a module, choose M here: the
431 module will be called cls_rsvp6.
434 tristate "Flow classifier"
437 If you say Y here, you will be able to classify packets based on
438 a configurable combination of packet keys. This is mostly useful
439 in combination with SFQ.
441 To compile this code as a module, choose M here: the
442 module will be called cls_flow.
444 config NET_CLS_CGROUP
445 tristate "Control Group Classifier"
449 Say Y here if you want to classify packets based on the control
450 cgroup of their process.
452 To compile this code as a module, choose M here: the
453 module will be called cls_cgroup.
456 tristate "BPF-based classifier"
459 If you say Y here, you will be able to classify packets based on
460 programmable BPF (JIT'ed) filters as an alternative to ematches.
462 To compile this code as a module, choose M here: the module will
466 bool "Extended Matches"
469 Say Y here if you want to use extended matches on top of classifiers
470 and select the extended matches below.
472 Extended matches are small classification helpers not worth writing
473 a separate classifier for.
475 A recent version of the iproute2 package is required to use
478 config NET_EMATCH_STACK
480 depends on NET_EMATCH
483 Size of the local stack variable used while evaluating the tree of
484 ematches. Limits the depth of the tree, i.e. the number of
485 encapsulated precedences. Every level requires 4 bytes of additional
488 config NET_EMATCH_CMP
489 tristate "Simple packet data comparison"
490 depends on NET_EMATCH
492 Say Y here if you want to be able to classify packets based on
493 simple packet data comparisons for 8, 16, and 32bit values.
495 To compile this code as a module, choose M here: the
496 module will be called em_cmp.
498 config NET_EMATCH_NBYTE
499 tristate "Multi byte comparison"
500 depends on NET_EMATCH
502 Say Y here if you want to be able to classify packets based on
503 multiple byte comparisons mainly useful for IPv6 address comparisons.
505 To compile this code as a module, choose M here: the
506 module will be called em_nbyte.
508 config NET_EMATCH_U32
510 depends on NET_EMATCH
512 Say Y here if you want to be able to classify packets using
513 the famous u32 key in combination with logic relations.
515 To compile this code as a module, choose M here: the
516 module will be called em_u32.
518 config NET_EMATCH_META
520 depends on NET_EMATCH
522 Say Y here if you want to be able to classify packets based on
523 metadata such as load average, netfilter attributes, socket
524 attributes and routing decisions.
526 To compile this code as a module, choose M here: the
527 module will be called em_meta.
529 config NET_EMATCH_TEXT
530 tristate "Textsearch"
531 depends on NET_EMATCH
533 select TEXTSEARCH_KMP
535 select TEXTSEARCH_FSM
537 Say Y here if you want to be able to classify packets based on
538 textsearch comparisons.
540 To compile this code as a module, choose M here: the
541 module will be called em_text.
543 config NET_EMATCH_CANID
544 tristate "CAN Identifier"
545 depends on NET_EMATCH && (CAN=y || CAN=m)
547 Say Y here if you want to be able to classify CAN frames based
550 To compile this code as a module, choose M here: the
551 module will be called em_canid.
553 config NET_EMATCH_IPSET
555 depends on NET_EMATCH && IP_SET
557 Say Y here if you want to be able to classify packets based on
560 To compile this code as a module, choose M here: the
561 module will be called em_ipset.
566 Say Y here if you want to use traffic control actions. Actions
567 get attached to classifiers and are invoked after a successful
568 classification. They are used to overwrite the classification
569 result, instantly drop or redirect packets, etc.
571 A recent version of the iproute2 package is required to use
574 config NET_ACT_POLICE
575 tristate "Traffic Policing"
576 depends on NET_CLS_ACT
578 Say Y here if you want to do traffic policing, i.e. strict
579 bandwidth limiting. This action replaces the existing policing
582 To compile this code as a module, choose M here: the
583 module will be called act_police.
586 tristate "Generic actions"
587 depends on NET_CLS_ACT
589 Say Y here to take generic actions such as dropping and
592 To compile this code as a module, choose M here: the
593 module will be called act_gact.
596 bool "Probability support"
597 depends on NET_ACT_GACT
599 Say Y here to use the generic action randomly or deterministically.
601 config NET_ACT_MIRRED
602 tristate "Redirecting and Mirroring"
603 depends on NET_CLS_ACT
605 Say Y here to allow packets to be mirrored or redirected to
608 To compile this code as a module, choose M here: the
609 module will be called act_mirred.
612 tristate "IPtables targets"
613 depends on NET_CLS_ACT && NETFILTER && IP_NF_IPTABLES
615 Say Y here to be able to invoke iptables targets after successful
618 To compile this code as a module, choose M here: the
619 module will be called act_ipt.
622 tristate "Stateless NAT"
623 depends on NET_CLS_ACT
625 Say Y here to do stateless NAT on IPv4 packets. You should use
626 netfilter for NAT unless you know what you are doing.
628 To compile this code as a module, choose M here: the
629 module will be called act_nat.
632 tristate "Packet Editing"
633 depends on NET_CLS_ACT
635 Say Y here if you want to mangle the content of packets.
637 To compile this code as a module, choose M here: the
638 module will be called act_pedit.
641 tristate "Simple Example (Debug)"
642 depends on NET_CLS_ACT
644 Say Y here to add a simple action for demonstration purposes.
645 It is meant as an example and for debugging purposes. It will
646 print a configured policy string followed by the packet count
647 to the console for every packet that passes by.
651 To compile this code as a module, choose M here: the
652 module will be called act_simple.
654 config NET_ACT_SKBEDIT
655 tristate "SKB Editing"
656 depends on NET_CLS_ACT
658 Say Y here to change skb priority or queue_mapping settings.
662 To compile this code as a module, choose M here: the
663 module will be called act_skbedit.
666 tristate "Checksum Updating"
667 depends on NET_CLS_ACT && INET
669 Say Y here to update some common checksum after some direct
672 To compile this code as a module, choose M here: the
673 module will be called act_csum.
676 bool "Incoming device classification"
677 depends on NET_CLS_U32 || NET_CLS_FW
679 Say Y here to extend the u32 and fw classifier to support
680 classification based on the incoming device. This option is
681 likely to disappear in favour of the metadata ematch.