Root/net/sched/Kconfig

1#
2# Traffic control configuration.
3#
4
5menuconfig NET_SCHED
6    bool "QoS and/or fair queueing"
7    select NET_SCH_FIFO
8    ---help---
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.
14
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.
23
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>.
28
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/>.
34
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
37      /proc/net/psched.
38
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.
41
42if NET_SCHED
43
44comment "Queueing/Scheduling"
45
46config NET_SCH_CBQ
47    tristate "Class Based Queueing (CBQ)"
48    ---help---
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.
53
54      See the top of <file:net/sched/sch_cbq.c> for more details.
55
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.
59
60      To compile this code as a module, choose M here: the
61      module will be called sch_cbq.
62
63config NET_SCH_HTB
64    tristate "Hierarchical Token Bucket (HTB)"
65    ---help---
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
69      in-depth articles.
70
71      HTB is very similar to CBQ regarding its goals however is has
72      different properties and different algorithm.
73
74      To compile this code as a module, choose M here: the
75      module will be called sch_htb.
76
77config NET_SCH_HFSC
78    tristate "Hierarchical Fair Service Curve (HFSC)"
79    ---help---
80      Say Y here if you want to use the Hierarchical Fair Service Curve
81      (HFSC) packet scheduling algorithm.
82
83      To compile this code as a module, choose M here: the
84      module will be called sch_hfsc.
85
86config NET_SCH_ATM
87    tristate "ATM Virtual Circuits (ATM)"
88    depends on ATM
89    ---help---
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.
94
95      See the top of <file:net/sched/sch_atm.c> for more details.
96
97      To compile this code as a module, choose M here: the
98      module will be called sch_atm.
99
100config NET_SCH_PRIO
101    tristate "Multi Band Priority Queueing (PRIO)"
102    ---help---
103      Say Y here if you want to use an n-band priority queue packet
104      scheduler.
105
106      To compile this code as a module, choose M here: the
107      module will be called sch_prio.
108
109config NET_SCH_MULTIQ
110    tristate "Hardware Multiqueue-aware Multi Band Queuing (MULTIQ)"
111    ---help---
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.
114
115      To compile this code as a module, choose M here: the
116      module will be called sch_multiq.
117
118config NET_SCH_RED
119    tristate "Random Early Detection (RED)"
120    ---help---
121      Say Y here if you want to use the Random Early Detection (RED)
122      packet scheduling algorithm.
123
124      See the top of <file:net/sched/sch_red.c> for more details.
125
126      To compile this code as a module, choose M here: the
127      module will be called sch_red.
128
129config NET_SCH_SFB
130    tristate "Stochastic Fair Blue (SFB)"
131    ---help---
132      Say Y here if you want to use the Stochastic Fair Blue (SFB)
133      packet scheduling algorithm.
134
135      See the top of <file:net/sched/sch_sfb.c> for more details.
136
137      To compile this code as a module, choose M here: the
138      module will be called sch_sfb.
139
140config NET_SCH_SFQ
141    tristate "Stochastic Fairness Queueing (SFQ)"
142    ---help---
143      Say Y here if you want to use the Stochastic Fairness Queueing (SFQ)
144      packet scheduling algorithm.
145
146      See the top of <file:net/sched/sch_sfq.c> for more details.
147
148      To compile this code as a module, choose M here: the
149      module will be called sch_sfq.
150
151config NET_SCH_TEQL
152    tristate "True Link Equalizer (TEQL)"
153    ---help---
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.
157
158      See the top of <file:net/sched/sch_teql.c> for more details.
159
160      To compile this code as a module, choose M here: the
161      module will be called sch_teql.
162
163config NET_SCH_TBF
164    tristate "Token Bucket Filter (TBF)"
165    ---help---
166      Say Y here if you want to use the Token Bucket Filter (TBF) packet
167      scheduling algorithm.
168
169      See the top of <file:net/sched/sch_tbf.c> for more details.
170
171      To compile this code as a module, choose M here: the
172      module will be called sch_tbf.
173
174config NET_SCH_GRED
175    tristate "Generic Random Early Detection (GRED)"
176    ---help---
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).
181
182      To compile this code as a module, choose M here: the
183      module will be called sch_gred.
184
185config NET_SCH_DSMARK
186    tristate "Differentiated Services marker (DSMARK)"
187    ---help---
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/>.
192
193      To compile this code as a module, choose M here: the
194      module will be called sch_dsmark.
195
196config NET_SCH_NETEM
197    tristate "Network emulator (NETEM)"
198    ---help---
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.
202
203      To compile this driver as a module, choose M here: the module
204      will be called sch_netem.
205
206      If unsure, say N.
207
208config NET_SCH_DRR
209    tristate "Deficit Round Robin scheduler (DRR)"
210    help
211      Say Y here if you want to use the Deficit Round Robin (DRR) packet
212      scheduling algorithm.
213
214      To compile this driver as a module, choose M here: the module
215      will be called sch_drr.
216
217      If unsure, say N.
218
219config NET_SCH_MQPRIO
220    tristate "Multi-queue priority scheduler (MQPRIO)"
221    help
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.
225
226      To compile this driver as a module, choose M here: the module will
227      be called sch_mqprio.
228
229      If unsure, say N.
230
231config NET_SCH_CHOKE
232    tristate "CHOose and Keep responsive flow scheduler (CHOKE)"
233    help
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.
238
239      To compile this code as a module, choose M here: the
240      module will be called sch_choke.
241
242config NET_SCH_QFQ
243    tristate "Quick Fair Queueing scheduler (QFQ)"
244    help
245      Say Y here if you want to use the Quick Fair Queueing Scheduler (QFQ)
246      packet scheduling algorithm.
247
248      To compile this driver as a module, choose M here: the module
249      will be called sch_qfq.
250
251      If unsure, say N.
252
253config NET_SCH_CODEL
254    tristate "Controlled Delay AQM (CODEL)"
255    help
256      Say Y here if you want to use the Controlled Delay (CODEL)
257      packet scheduling algorithm.
258
259      To compile this driver as a module, choose M here: the module
260      will be called sch_codel.
261
262      If unsure, say N.
263
264config NET_SCH_FQ_CODEL
265    tristate "Fair Queue Controlled Delay AQM (FQ_CODEL)"
266    help
267      Say Y here if you want to use the FQ Controlled Delay (FQ_CODEL)
268      packet scheduling algorithm.
269
270      To compile this driver as a module, choose M here: the module
271      will be called sch_fq_codel.
272
273      If unsure, say N.
274
275config NET_SCH_FQ
276    tristate "Fair Queue"
277    help
278      Say Y here if you want to use the FQ packet scheduling algorithm.
279
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
282      traffic)
283
284      To compile this driver as a module, choose M here: the module
285      will be called sch_fq.
286
287      If unsure, say N.
288
289config NET_SCH_HHF
290    tristate "Heavy-Hitter Filter (HHF)"
291    help
292      Say Y here if you want to use the Heavy-Hitter Filter (HHF)
293      packet scheduling algorithm.
294
295      To compile this driver as a module, choose M here: the module
296      will be called sch_hhf.
297
298config NET_SCH_PIE
299    tristate "Proportional Integral controller Enhanced (PIE) scheduler"
300    help
301      Say Y here if you want to use the Proportional Integral controller
302      Enhanced scheduler packet scheduling algorithm.
303      For more information, please see
304      http://tools.ietf.org/html/draft-pan-tsvwg-pie-00
305
306      To compile this driver as a module, choose M here: the module
307      will be called sch_pie.
308
309      If unsure, say N.
310
311config NET_SCH_INGRESS
312    tristate "Ingress Qdisc"
313    depends on NET_CLS_ACT
314    ---help---
315      Say Y here if you want to use classifiers for incoming packets.
316      If unsure, say Y.
317
318      To compile this code as a module, choose M here: the
319      module will be called sch_ingress.
320
321config NET_SCH_PLUG
322    tristate "Plug network traffic until release (PLUG)"
323    ---help---
324
325      This queuing discipline allows userspace to plug/unplug a network
326      output queue, using the netlink interface. When it receives an
327      enqueue command it inserts a plug into the outbound queue that
328      causes following packets to enqueue until a dequeue command arrives
329      over netlink, causing the plug to be removed and resuming the normal
330      packet flow.
331
332      This module also provides a generic "network output buffering"
333      functionality (aka output commit), wherein upon arrival of a dequeue
334      command, only packets up to the first plug are released for delivery.
335      The Remus HA project uses this module to enable speculative execution
336      of virtual machines by allowing the generated network output to be rolled
337      back if needed.
338
339      For more information, please refer to http://wiki.xensource.com/xenwiki/Remus
340
341      Say Y here if you are using this kernel for Xen dom0 and
342      want to protect Xen guests with Remus.
343
344      To compile this code as a module, choose M here: the
345      module will be called sch_plug.
346
347comment "Classification"
348
349config NET_CLS
350    boolean
351
352config NET_CLS_BASIC
353    tristate "Elementary classification (BASIC)"
354    select NET_CLS
355    ---help---
356      Say Y here if you want to be able to classify packets using
357      only extended matches and actions.
358
359      To compile this code as a module, choose M here: the
360      module will be called cls_basic.
361
362config NET_CLS_TCINDEX
363    tristate "Traffic-Control Index (TCINDEX)"
364    select NET_CLS
365    ---help---
366      Say Y here if you want to be able to classify packets based on
367      traffic control indices. You will want this feature if you want
368      to implement Differentiated Services together with DSMARK.
369
370      To compile this code as a module, choose M here: the
371      module will be called cls_tcindex.
372
373config NET_CLS_ROUTE4
374    tristate "Routing decision (ROUTE)"
375    depends on INET
376    select IP_ROUTE_CLASSID
377    select NET_CLS
378    ---help---
379      If you say Y here, you will be able to classify packets
380      according to the route table entry they matched.
381
382      To compile this code as a module, choose M here: the
383      module will be called cls_route.
384
385config NET_CLS_FW
386    tristate "Netfilter mark (FW)"
387    select NET_CLS
388    ---help---
389      If you say Y here, you will be able to classify packets
390      according to netfilter/firewall marks.
391
392      To compile this code as a module, choose M here: the
393      module will be called cls_fw.
394
395config NET_CLS_U32
396    tristate "Universal 32bit comparisons w/ hashing (U32)"
397    select NET_CLS
398    ---help---
399      Say Y here to be able to classify packets using a universal
400      32bit pieces based comparison scheme.
401
402      To compile this code as a module, choose M here: the
403      module will be called cls_u32.
404
405config CLS_U32_PERF
406    bool "Performance counters support"
407    depends on NET_CLS_U32
408    ---help---
409      Say Y here to make u32 gather additional statistics useful for
410      fine tuning u32 classifiers.
411
412config CLS_U32_MARK
413    bool "Netfilter marks support"
414    depends on NET_CLS_U32
415    ---help---
416      Say Y here to be able to use netfilter marks as u32 key.
417
418config NET_CLS_RSVP
419    tristate "IPv4 Resource Reservation Protocol (RSVP)"
420    select NET_CLS
421    ---help---
422      The Resource Reservation Protocol (RSVP) permits end systems to
423      request a minimum and maximum data flow rate for a connection; this
424      is important for real time data such as streaming sound or video.
425
426      Say Y here if you want to be able to classify outgoing packets based
427      on their RSVP requests.
428
429      To compile this code as a module, choose M here: the
430      module will be called cls_rsvp.
431
432config NET_CLS_RSVP6
433    tristate "IPv6 Resource Reservation Protocol (RSVP6)"
434    select NET_CLS
435    ---help---
436      The Resource Reservation Protocol (RSVP) permits end systems to
437      request a minimum and maximum data flow rate for a connection; this
438      is important for real time data such as streaming sound or video.
439
440      Say Y here if you want to be able to classify outgoing packets based
441      on their RSVP requests and you are using the IPv6 protocol.
442
443      To compile this code as a module, choose M here: the
444      module will be called cls_rsvp6.
445
446config NET_CLS_FLOW
447    tristate "Flow classifier"
448    select NET_CLS
449    ---help---
450      If you say Y here, you will be able to classify packets based on
451      a configurable combination of packet keys. This is mostly useful
452      in combination with SFQ.
453
454      To compile this code as a module, choose M here: the
455      module will be called cls_flow.
456
457config NET_CLS_CGROUP
458    tristate "Control Group Classifier"
459    select NET_CLS
460    select CGROUP_NET_CLASSID
461    depends on CGROUPS
462    ---help---
463      Say Y here if you want to classify packets based on the control
464      cgroup of their process.
465
466      To compile this code as a module, choose M here: the
467      module will be called cls_cgroup.
468
469config NET_CLS_BPF
470    tristate "BPF-based classifier"
471    select NET_CLS
472    ---help---
473      If you say Y here, you will be able to classify packets based on
474      programmable BPF (JIT'ed) filters as an alternative to ematches.
475
476      To compile this code as a module, choose M here: the module will
477      be called cls_bpf.
478
479config NET_EMATCH
480    bool "Extended Matches"
481    select NET_CLS
482    ---help---
483      Say Y here if you want to use extended matches on top of classifiers
484      and select the extended matches below.
485
486      Extended matches are small classification helpers not worth writing
487      a separate classifier for.
488
489      A recent version of the iproute2 package is required to use
490      extended matches.
491
492config NET_EMATCH_STACK
493    int "Stack size"
494    depends on NET_EMATCH
495    default "32"
496    ---help---
497      Size of the local stack variable used while evaluating the tree of
498      ematches. Limits the depth of the tree, i.e. the number of
499      encapsulated precedences. Every level requires 4 bytes of additional
500      stack space.
501
502config NET_EMATCH_CMP
503    tristate "Simple packet data comparison"
504    depends on NET_EMATCH
505    ---help---
506      Say Y here if you want to be able to classify packets based on
507      simple packet data comparisons for 8, 16, and 32bit values.
508
509      To compile this code as a module, choose M here: the
510      module will be called em_cmp.
511
512config NET_EMATCH_NBYTE
513    tristate "Multi byte comparison"
514    depends on NET_EMATCH
515    ---help---
516      Say Y here if you want to be able to classify packets based on
517      multiple byte comparisons mainly useful for IPv6 address comparisons.
518
519      To compile this code as a module, choose M here: the
520      module will be called em_nbyte.
521
522config NET_EMATCH_U32
523    tristate "U32 key"
524    depends on NET_EMATCH
525    ---help---
526      Say Y here if you want to be able to classify packets using
527      the famous u32 key in combination with logic relations.
528
529      To compile this code as a module, choose M here: the
530      module will be called em_u32.
531
532config NET_EMATCH_META
533    tristate "Metadata"
534    depends on NET_EMATCH
535    ---help---
536      Say Y here if you want to be able to classify packets based on
537      metadata such as load average, netfilter attributes, socket
538      attributes and routing decisions.
539
540      To compile this code as a module, choose M here: the
541      module will be called em_meta.
542
543config NET_EMATCH_TEXT
544    tristate "Textsearch"
545    depends on NET_EMATCH
546    select TEXTSEARCH
547    select TEXTSEARCH_KMP
548    select TEXTSEARCH_BM
549    select TEXTSEARCH_FSM
550    ---help---
551      Say Y here if you want to be able to classify packets based on
552      textsearch comparisons.
553
554      To compile this code as a module, choose M here: the
555      module will be called em_text.
556
557config NET_EMATCH_CANID
558    tristate "CAN Identifier"
559    depends on NET_EMATCH && (CAN=y || CAN=m)
560    ---help---
561      Say Y here if you want to be able to classify CAN frames based
562      on CAN Identifier.
563
564      To compile this code as a module, choose M here: the
565      module will be called em_canid.
566
567config NET_EMATCH_IPSET
568    tristate "IPset"
569    depends on NET_EMATCH && IP_SET
570    ---help---
571      Say Y here if you want to be able to classify packets based on
572      ipset membership.
573
574      To compile this code as a module, choose M here: the
575      module will be called em_ipset.
576
577config NET_CLS_ACT
578    bool "Actions"
579    ---help---
580      Say Y here if you want to use traffic control actions. Actions
581      get attached to classifiers and are invoked after a successful
582      classification. They are used to overwrite the classification
583      result, instantly drop or redirect packets, etc.
584
585      A recent version of the iproute2 package is required to use
586      extended matches.
587
588config NET_ACT_POLICE
589    tristate "Traffic Policing"
590        depends on NET_CLS_ACT
591        ---help---
592      Say Y here if you want to do traffic policing, i.e. strict
593      bandwidth limiting. This action replaces the existing policing
594      module.
595
596      To compile this code as a module, choose M here: the
597      module will be called act_police.
598
599config NET_ACT_GACT
600        tristate "Generic actions"
601        depends on NET_CLS_ACT
602        ---help---
603      Say Y here to take generic actions such as dropping and
604      accepting packets.
605
606      To compile this code as a module, choose M here: the
607      module will be called act_gact.
608
609config GACT_PROB
610        bool "Probability support"
611        depends on NET_ACT_GACT
612        ---help---
613      Say Y here to use the generic action randomly or deterministically.
614
615config NET_ACT_MIRRED
616        tristate "Redirecting and Mirroring"
617        depends on NET_CLS_ACT
618        ---help---
619      Say Y here to allow packets to be mirrored or redirected to
620      other devices.
621
622      To compile this code as a module, choose M here: the
623      module will be called act_mirred.
624
625config NET_ACT_IPT
626        tristate "IPtables targets"
627        depends on NET_CLS_ACT && NETFILTER && IP_NF_IPTABLES
628        ---help---
629      Say Y here to be able to invoke iptables targets after successful
630      classification.
631
632      To compile this code as a module, choose M here: the
633      module will be called act_ipt.
634
635config NET_ACT_NAT
636        tristate "Stateless NAT"
637        depends on NET_CLS_ACT
638        ---help---
639      Say Y here to do stateless NAT on IPv4 packets. You should use
640      netfilter for NAT unless you know what you are doing.
641
642      To compile this code as a module, choose M here: the
643      module will be called act_nat.
644
645config NET_ACT_PEDIT
646        tristate "Packet Editing"
647        depends on NET_CLS_ACT
648        ---help---
649      Say Y here if you want to mangle the content of packets.
650
651      To compile this code as a module, choose M here: the
652      module will be called act_pedit.
653
654config NET_ACT_SIMP
655        tristate "Simple Example (Debug)"
656        depends on NET_CLS_ACT
657        ---help---
658      Say Y here to add a simple action for demonstration purposes.
659      It is meant as an example and for debugging purposes. It will
660      print a configured policy string followed by the packet count
661      to the console for every packet that passes by.
662
663      If unsure, say N.
664
665      To compile this code as a module, choose M here: the
666      module will be called act_simple.
667
668config NET_ACT_SKBEDIT
669        tristate "SKB Editing"
670        depends on NET_CLS_ACT
671        ---help---
672      Say Y here to change skb priority or queue_mapping settings.
673
674      If unsure, say N.
675
676      To compile this code as a module, choose M here: the
677      module will be called act_skbedit.
678
679config NET_ACT_CSUM
680        tristate "Checksum Updating"
681        depends on NET_CLS_ACT && INET
682        ---help---
683      Say Y here to update some common checksum after some direct
684      packet alterations.
685
686      To compile this code as a module, choose M here: the
687      module will be called act_csum.
688
689config NET_CLS_IND
690    bool "Incoming device classification"
691    depends on NET_CLS_U32 || NET_CLS_FW
692    ---help---
693      Say Y here to extend the u32 and fw classifier to support
694      classification based on the incoming device. This option is
695      likely to disappear in favour of the metadata ematch.
696
697endif # NET_SCHED
698
699config NET_SCH_FIFO
700    bool
701

Archive Download this file



interactive