Debian Codegraph @

upstream/debian-trixie/man8/tc-cake.8

raw
1 .TH CAKE 8 "19 July 2018" "iproute2" "Linux"
2 .SH NAME
3 CAKE \- Common Applications Kept Enhanced (CAKE)
4 .SH SYNOPSIS
5 .B tc qdisc ... cake
6 .br
7 [
8 .BR bandwidth
9 RATE |
10 .BR unlimited*
11 |
12 .BR autorate-ingress
13 ]
14 .br
15 [
16 .BR rtt
17 TIME |
18 .BR datacentre
19 |
20 .BR lan
21 |
22 .BR metro
23 |
24 .BR regional
25 |
26 .BR internet*
27 |
28 .BR oceanic
29 |
30 .BR satellite
31 |
32 .BR interplanetary
33 ]
34 .br
35 [
36 .BR besteffort
37 |
38 .BR diffserv8
39 |
40 .BR diffserv4
41 |
42 .BR diffserv3*
43 ]
44 .br
45 [
46 .BR flowblind
47 |
48 .BR srchost
49 |
50 .BR dsthost
51 |
52 .BR hosts
53 |
54 .BR flows
55 |
56 .BR dual-srchost
57 |
58 .BR dual-dsthost
59 |
60 .BR triple-isolate*
61 ]
62 .br
63 [
64 .BR nat
65 |
66 .BR nonat*
67 ]
68 .br
69 [
70 .BR wash
71 |
72 .BR nowash*
73 ]
74 .br
75 [
76 .BR split-gso*
77 |
78 .BR no-split-gso
79 ]
80 .br
81 [
82 .BR ack-filter
83 |
84 .BR ack-filter-aggressive
85 |
86 .BR no-ack-filter*
87 ]
88 .br
89 [
90 .BR memlimit
91 LIMIT ]
92 .br
93 [
94 .BR fwmark
95 MASK ]
96 .br
97 [
98 .BR ptm
99 |
100 .BR atm
101 |
102 .BR noatm*
103 ]
104 .br
105 [
106 .BR overhead
107 N |
108 .BR conservative
109 |
110 .BR raw*
111 ]
112 .br
113 [
114 .BR mpu
115 N ]
116 .br
117 [
118 .BR ingress
119 |
120 .BR egress*
121 ]
122 .br
123 (* marks defaults)
124
125
126 .SH DESCRIPTION
127 CAKE (Common Applications Kept Enhanced) is a shaping-capable queue discipline
128 which uses both AQM and FQ. It combines COBALT, which is an AQM algorithm
129 combining Codel and BLUE, a shaper which operates in deficit mode, and a variant
130 of DRR++ for flow isolation. 8-way set-associative hashing is used to virtually
131 eliminate hash collisions. Priority queuing is available through a simplified
132 diffserv implementation. Overhead compensation for various encapsulation
133 schemes is tightly integrated.
134
135 All settings are optional; the default settings are chosen to be sensible in
136 most common deployments. Most people will only need to set the
137 .B bandwidth
138 parameter to get useful results, but reading the
139 .B Overhead Compensation
140 and
141 .B Round Trip Time
142 sections is strongly encouraged.
143
144 .SH SHAPER PARAMETERS
145 CAKE uses a deficit-mode shaper, which does not exhibit the initial burst
146 typical of token-bucket shapers. It will automatically burst precisely as much
147 as required to maintain the configured throughput. As such, it is very
148 straightforward to configure.
149
150 .TP
151 \fBunlimited\fR (default)
152 .br
153 No limit on the bandwidth.
154
155 .TP
156 \fBbandwidth\fR RATE
157 .br
158 Set the shaper bandwidth. See
159 .BR tc(8)
160 or examples below for details of the RATE value.
161
162 .TP
163 .B autorate-ingress
164 .br
165 Automatic capacity estimation based on traffic arriving at this qdisc.
166 This is most likely to be useful with cellular links, which tend to change
167 quality randomly. A
168 .B bandwidth
169 parameter can be used in conjunction to specify an initial estimate. The shaper
170 will periodically be set to a bandwidth slightly below the estimated rate. This
171 estimator cannot estimate the bandwidth of links downstream of itself.
172
173 .SH OVERHEAD COMPENSATION PARAMETERS
174 The size of each packet on the wire may differ from that seen by Linux. The
175 following parameters allow CAKE to compensate for this difference by internally
176 considering each packet to be bigger than Linux informs it. To assist users who
177 are not expert network engineers, keywords have been provided to represent a
178 number of common link technologies.
179
180 .SS Manual Overhead Specification
181 .TP
182 \fBoverhead\fR BYTES
183 .br
184 Adds BYTES to the size of each packet. BYTES may be negative; values
185 between -64 and 256 (inclusive) are accepted.
186
187 .TP
188 \fBmpu\fR BYTES
189 .br
190 Rounds each packet (including overhead) up to a minimum length
191 BYTES. BYTES may not be negative; values between 0 and 256 (inclusive)
192 are accepted.
193
194 .TP
195 .B atm
196 .br
197 Compensates for ATM cell framing, which is normally found on ADSL links.
198 This is performed after the
199 .B overhead
200 parameter above. ATM uses fixed 53-byte cells, each of which can carry 48 bytes
201 payload.
202
203 .TP
204 .B ptm
205 .br
206 Compensates for PTM encoding, which is normally found on VDSL2 links and
207 uses a 64b/65b encoding scheme. It is even more efficient to simply
208 derate the specified shaper bandwidth by a factor of 64/65 or 0.984. See
209 ITU G.992.3 Annex N and IEEE 802.3 Section 61.3 for details.
210
211 .TP
212 .B noatm
213 .br
214 Disables ATM and PTM compensation.
215
216 .SS Failsafe Overhead Keywords
217 These two keywords are provided for quick-and-dirty setup. Use them if you
218 can't be bothered to read the rest of this section.
219
220 .TP
221 \fBraw\fR (default)
222 .br
223 Turns off all overhead compensation in CAKE. The packet size reported
224 by Linux will be used directly.
225
226 Other overhead keywords may be added after "raw". The effect of this is
227 to make the overhead compensation operate relative to the reported packet size,
228 not the underlying IP packet size.
229
230 .TP
231 .B conservative
232 .br
233 Compensates for more overhead than is likely to occur on any
234 widely-deployed link technology.
235 Equivalent to
236 .B overhead 48 atm.
237
238 .SS ADSL Overhead Keywords
239 Most ADSL modems have a way to check which framing scheme is in use. Often this
240 is also specified in the settings document provided by the ISP. The keywords in
241 this section are intended to correspond with these sources of information. All
242 of them implicitly set the
243 .B atm
244 flag.
245
246 .TP
247 .B pppoa-vcmux
248 .br
249 Equivalent to
250 .B overhead 10 atm
251
252 .TP
253 .B pppoa-llc
254 .br
255 Equivalent to
256 .B overhead 14 atm
257
258 .TP
259 .B pppoe-vcmux
260 .br
261 Equivalent to
262 .B overhead 32 atm
263
264 .TP
265 .B pppoe-llcsnap
266 .br
267 Equivalent to
268 .B overhead 40 atm
269
270 .TP
271 .B bridged-vcmux
272 .br
273 Equivalent to
274 .B overhead 24 atm
275
276 .TP
277 .B bridged-llcsnap
278 .br
279 Equivalent to
280 .B overhead 32 atm
281
282 .TP
283 .B ipoa-vcmux
284 .br
285 Equivalent to
286 .B overhead 8 atm
287
288 .TP
289 .B ipoa-llcsnap
290 .br
291 Equivalent to
292 .B overhead 16 atm
293
294 .P
295 See also the Ethernet Correction Factors section below.
296
297 .SS VDSL2 Overhead Keywords
298 ATM was dropped from VDSL2 in favour of PTM, which is a much more
299 straightforward framing scheme. Some ISPs retained PPPoE for compatibility with
300 their existing back-end systems.
301
302 .TP
303 .B pppoe-ptm
304 .br
305 Equivalent to
306 .B overhead 30 ptm
307
308 PPPoE: 2B PPP + 6B PPPoE +
309 .br
310 ETHERNET: 6B dest MAC + 6B src MAC + 2B ethertype + 4B Frame Check Sequence +
311 .br
312 PTM: 1B Start of Frame (S) + 1B End of Frame (Ck) + 2B TC-CRC (PTM-FCS)
313
314 .TP
315 .B bridged-ptm
316 .br
317 Equivalent to
318 .B overhead 22 ptm
319
320 ETHERNET: 6B dest MAC + 6B src MAC + 2B ethertype + 4B Frame Check Sequence +
321 .br
322 PTM: 1B Start of Frame (S) + 1B End of Frame (Ck) + 2B TC-CRC (PTM-FCS)
323
324 .P
325 See also the Ethernet Correction Factors section below.
326
327 .SS DOCSIS Cable Overhead Keyword
328 DOCSIS is the universal standard for providing Internet service over cable-TV
329 infrastructure.
330
331 In this case, the actual on-wire overhead is less important than the packet size
332 the head-end equipment uses for shaping and metering. This is specified to be
333 an Ethernet frame including the CRC (aka FCS).
334
335 .TP
336 .B docsis
337 .br
338 Equivalent to
339 .B overhead 18 mpu 64 noatm
340
341 .SS Ethernet Overhead Keywords
342
343 .TP
344 .B ethernet
345 .br
346 Accounts for Ethernet's preamble, inter-frame gap, and Frame Check
347 Sequence. Use this keyword when the bottleneck being shaped for is an
348 actual Ethernet cable.
349 Equivalent to
350 .B overhead 38 mpu 84 noatm
351
352 .TP
353 .B ether-vlan
354 .br
355 Adds 4 bytes to the overhead compensation, accounting for an IEEE 802.1Q
356 VLAN header appended to the Ethernet frame header. NB: Some ISPs use one or
357 even two of these within PPPoE; this keyword may be repeated as necessary to
358 express this.
359
360 .SH ROUND TRIP TIME PARAMETERS
361 Active Queue Management (AQM) consists of embedding congestion signals in the
362 packet flow, which receivers use to instruct senders to slow down when the queue
363 is persistently occupied. CAKE uses ECN signalling when available, and packet
364 drops otherwise, according to a combination of the Codel and BLUE AQM algorithms
365 called COBALT.
366
367 Very short latencies require a very rapid AQM response to adequately control
368 latency. However, such a rapid response tends to impair throughput when the
369 actual RTT is relatively long. CAKE allows specifying the RTT it assumes for
370 tuning various parameters. Actual RTTs within an order of magnitude of this
371 will generally work well for both throughput and latency management.
372
373 At the 'lan' setting and below, the time constants are similar in magnitude to
374 the jitter in the Linux kernel itself, so congestion might be signalled
375 prematurely. The flows will then become sparse and total throughput reduced,
376 leaving little or no back-pressure for the fairness logic to work against. Use
377 the "metro" setting for local lans unless you have a custom kernel.
378
379 .TP
380 \fBrtt\fR TIME
381 .br
382 Manually specify an RTT.
383
384 .TP
385 .B datacentre
386 .br
387 For extremely high-performance 10GigE+ networks only.
388 .br
389 Equivalent to
390 .B rtt 100us.
391
392 .TP
393 .B lan
394 .br
395 For pure Ethernet (not Wi-Fi) networks, at home or in the office. Don't
396 use this when shaping for an Internet access link.
397 .br
398 Equivalent to
399 .B rtt 1ms.
400
401 .TP
402 .B metro
403 .br
404 For traffic mostly within a single city.
405 .br
406 Equivalent to
407 .B rtt 10ms.
408
409 .TP
410 .B regional
411 .br
412 For traffic mostly within a European-sized country.
413 .br
414 Equivalent to
415 .B rtt 30ms.
416
417 .TP
418 \fBinternet\fR (default)
419 .br
420 This is suitable for most Internet traffic.
421 .br
422 Equivalent to
423 .B rtt 100ms.
424
425 .TP
426 .B oceanic
427 .br
428 For Internet traffic with generally above-average latency, such as that
429 suffered by Australasian residents.
430 .br
431 Equivalent to
432 .B rtt 300ms.
433
434 .TP
435 .B satellite
436 .br
437 For traffic via geostationary satellites.
438 .br
439 Equivalent to
440 .B rtt 1000ms.
441
442 .TP
443 .B interplanetary
444 .br
445 So named because Jupiter is about 1 light-hour from Earth. Use this to
446 (almost) completely disable AQM actions.
447 .br
448 Equivalent to
449 .B rtt 3600s.
450
451 .SH FLOW ISOLATION PARAMETERS
452 With flow isolation enabled, CAKE places packets from different flows into
453 different queues, each of which carries its own AQM state. Packets from each
454 queue are then delivered fairly, according to a DRR++ algorithm which minimizes
455 latency for "sparse" flows. CAKE uses a set-associative hashing algorithm to
456 minimize flow collisions.
457
458 These keywords specify whether fairness based on source address, destination
459 address, individual flows, or any combination of those is desired.
460
461 .TP
462 .B flowblind
463 .br
464 Disables flow isolation; all traffic passes through a single queue for
465 each tin.
466
467 .TP
468 .B srchost
469 .br
470 Flows are defined only by source address. Could be useful on the egress
471 path of an ISP backhaul.
472
473 .TP
474 .B dsthost
475 .br
476 Flows are defined only by destination address. Could be useful on the
477 ingress path of an ISP backhaul.
478
479 .TP
480 .B hosts
481 .br
482 Flows are defined by source-destination host pairs. This is host
483 isolation, rather than flow isolation.
484
485 .TP
486 .B flows
487 .br
488 Flows are defined by the entire 5-tuple of source address, destination
489 address, transport protocol, source port and destination port. This is the type
490 of flow isolation performed by SFQ and fq_codel.
491
492 .TP
493 .B dual-srchost
494 .br
495 Flows are defined by the 5-tuple, and fairness is applied first over
496 source addresses, then over individual flows. Good for use on egress traffic
497 from a LAN to the internet, where it'll prevent any one LAN host from
498 monopolising the uplink, regardless of the number of flows they use.
499
500 .TP
501 .B dual-dsthost
502 .br
503 Flows are defined by the 5-tuple, and fairness is applied first over
504 destination addresses, then over individual flows. Good for use on ingress
505 traffic to a LAN from the internet, where it'll prevent any one LAN host from
506 monopolising the downlink, regardless of the number of flows they use.
507
508 .TP
509 \fBtriple-isolate\fR (default)
510 .br
511 Flows are defined by the 5-tuple, and fairness is applied over source
512 *and* destination addresses intelligently (ie. not merely by host-pairs), and
513 also over individual flows. Use this if you're not certain whether to use
514 dual-srchost or dual-dsthost; it'll do both jobs at once, preventing any one
515 host on *either* side of the link from monopolising it with a large number of
516 flows.
517
518 .TP
519 .B nat
520 .br
521 Instructs Cake to perform a NAT lookup before applying flow-isolation
522 rules, to determine the true addresses and port numbers of the packet, to
523 improve fairness between hosts "inside" the NAT. This has no practical effect
524 in "flowblind" or "flows" modes, or if NAT is performed on a different host.
525
526 .TP
527 \fBnonat\fR (default)
528 .br
529 Cake will not perform a NAT lookup. Flow isolation will be performed
530 using the addresses and port numbers directly visible to the interface Cake is
531 attached to.
532
533 .SH PRIORITY QUEUE PARAMETERS
534 CAKE can divide traffic into "tins" based on the Diffserv field. Each tin has
535 its own independent set of flow-isolation queues, and is serviced based on a WRR
536 algorithm. To avoid perverse Diffserv marking incentives, tin weights have a
537 "priority sharing" value when bandwidth used by that tin is below a threshold,
538 and a lower "bandwidth sharing" value when above. Bandwidth is compared against
539 the threshold using the same algorithm as the deficit-mode shaper.
540
541 Detailed customisation of tin parameters is not provided. The following presets
542 perform all necessary tuning, relative to the current shaper bandwidth and RTT
543 settings.
544
545 .TP
546 .B besteffort
547 .br
548 Disables priority queuing by placing all traffic in one tin.
549
550 .TP
551 .B precedence
552 .br
553 Enables legacy interpretation of TOS "Precedence" field. Use of this
554 preset on the modern Internet is firmly discouraged.
555
556 .TP
557 .B diffserv4
558 .br
559 Provides a general-purpose Diffserv implementation with four tins:
560
561 \(bu Bulk (CS1, LE in kernel v5.9+), 6.25% threshold, generally low priority.
562 .br
563 \(bu Best Effort (general), 100% threshold.
564 .br
565 \(bu Video (AF4x, AF3x, CS3, AF2x, CS2, TOS4, TOS1), 50% threshold.
566 .br
567 \(bu Voice (CS7, CS6, EF, VA, CS5, CS4), 25% threshold.
568
569 .TP
570 \fBdiffserv3\fR (default)
571 .br
572 Provides a simple, general-purpose Diffserv implementation with three tins:
573
574 \(bu Bulk (CS1, LE in kernel v5.9+), 6.25% threshold, generally low priority.
575 .br
576 \(bu Best Effort (general), 100% threshold.
577 .br
578 \(bu Voice (CS7, CS6, EF, VA, TOS4), 25% threshold, reduced Codel interval.
579
580 .TP
581 \fBfwmark\fR MASK
582 .br
583 This options turns on fwmark-based overriding of CAKE's tin selection.
584 If set, the option specifies a bitmask that will be applied to the fwmark
585 associated with each packet. If the result of this masking is non-zero, the
586 result will be right-shifted by the number of least-significant unset bits in
587 the mask value, and the result will be used as a the tin number for that packet.
588 This can be used to set policies in a firewall script that will override CAKE's
589 built-in tin selection.
590
591 .SH OTHER PARAMETERS
592
593 .TP
594 .B ingress
595 .br
596 Indicates that CAKE is running in ingress mode (i.e. running on the downlink of
597 a connection). This changes the shaper to also count dropped packets as data
598 transferred, as these will have already traversed the link before CAKE can
599 choose what to do with them.
600
601 In addition, the AQM will be tuned to always keep at least two packets
602 queued per flow. The reason for this is that retransmits are more expensive in
603 ingress mode, since dropped packets have to traverse the link again; thus,
604 keeping a minimum number of packets queued will improve throughput in cases
605 where the number of active flows are so large that they saturate the link even
606 at their minimum window size.
607
608 .TP
609 \fBmemlimit\fR LIMIT
610 .br
611 Limit the memory consumed by Cake to LIMIT bytes. Note that this does
612 not translate directly to queue size (so do not size this based on bandwidth
613 delay product considerations, but rather on worst case acceptable memory
614 consumption), as there is some overhead in the data structures containing the
615 packets, especially for small packets.
616
617 By default, the limit is calculated based on the bandwidth and RTT
618 settings.
619
620 .TP
621 .B wash
622 .br
623 Traffic entering your diffserv domain is frequently mis-marked in
624 transit from the perspective of your network, and traffic exiting yours may be
625 mis-marked from the perspective of the transiting provider.
626
627 Apply the wash option to clear all extra diffserv (but not ECN bits), after
628 priority queuing has taken place.
629
630 If you are shaping inbound, and cannot trust the diffserv markings (as is the
631 case for Comcast Cable, among others), it is best to use a single queue
632 "besteffort" mode with wash.
633
634 .TP
635 .B split-gso
636 .br
637 This option controls whether CAKE will split General Segmentation
638 Offload (GSO) super-packets into their on-the-wire components and
639 dequeue them individually.
640
641 Super-packets are created by the networking stack to improve efficiency.
642 However, because they are larger they take longer to dequeue, which
643 translates to higher latency for competing flows, especially at lower
644 bandwidths. CAKE defaults to splitting GSO packets to achieve the lowest
645 possible latency. At link speeds higher than 10 Gbps, setting the
646 no-split-gso parameter can increase the maximum achievable throughput by
647 retaining the full GSO packets.
648
649 .SH OVERRIDING CLASSIFICATION WITH TC FILTERS
650
651 CAKE supports overriding of its internal classification of packets through the
652 tc filter mechanism. Packets can be assigned to different priority tins by
653 setting the
654 .B priority
655 field on the skb, and the flow hashing can be overridden by setting the
656 .B classid
657 parameter.
658
659 .SS Tin override
660 To assign a priority tin, the major number of the priority field needs
661 to match the qdisc handle of the cake instance; if it does, the minor number
662 will be interpreted as the tin index. For example, to classify all ICMP packets
663 as 'bulk', the following filter can be used:
664
665 .RS
666 .EX
667 # tc qdisc replace dev eth0 handle 1: root cake diffserv3
668 # tc filter add dev eth0 parent 1: protocol ip prio 1 \\
669 u32 match icmp type 0 0 action skbedit priority 1:1
670 .EE
671 .RE
672
673 .SS Flow hash override
674 To override flow hashing, the classid can be set. CAKE will interpret
675 the major number of the classid as the host hash used in host isolation mode,
676 and the minor number as the flow hash used for flow-based queueing. One or both
677 of those can be set, and will be used if the relevant flow isolation parameter
678 is set (i.e., the major number will be ignored if CAKE is not configured in
679 hosts mode, and the minor number will be ignored if CAKE is not configured in
680 flows mode).
681
682 This example will assign all ICMP packets to the first queue:
683
684 .RS
685 .EX
686 # tc qdisc replace dev eth0 handle 1: root cake
687 # tc filter add dev eth0 parent 1: protocol ip prio 1 \\
688 u32 match icmp type 0 0 classid 0:1
689 .EE
690 .RE
691
692 If only one of the host and flow overrides is set, CAKE will compute the other
693 hash from the packet as normal. Note, however, that the host isolation mode
694 works by assigning a host ID to the flow queue; so if overriding both host and
695 flow, the same flow cannot have more than one host assigned. In addition, it is
696 not possible to assign different source and destination host IDs through the
697 override mechanism; if a host ID is assigned, it will be used as both source and
698 destination host.
699
700
701
702 .SH EXAMPLES
703 .EX
704 # tc qdisc delete root dev eth0
705 # tc qdisc add root dev eth0 cake bandwidth 100Mbit ethernet
706 # tc -s qdisc show dev eth0
707
708 qdisc cake 1: root refcnt 2 bandwidth 100Mbit diffserv3 triple-isolate rtt 100.0ms noatm overhead 38 mpu 84
709 Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
710 backlog 0b 0p requeues 0
711 memory used: 0b of 5000000b
712 capacity estimate: 100Mbit
713 min/max network layer size: 65535 / 0
714 min/max overhead-adjusted size: 65535 / 0
715 average network hdr offset: 0
716
717 Bulk Best Effort Voice
718 thresh 6250Kbit 100Mbit 25Mbit
719 target 5.0ms 5.0ms 5.0ms
720 interval 100.0ms 100.0ms 100.0ms
721 pk_delay 0us 0us 0us
722 av_delay 0us 0us 0us
723 sp_delay 0us 0us 0us
724 pkts 0 0 0
725 bytes 0 0 0
726 way_inds 0 0 0
727 way_miss 0 0 0
728 way_cols 0 0 0
729 drops 0 0 0
730 marks 0 0 0
731 ack_drop 0 0 0
732 sp_flows 0 0 0
733 bk_flows 0 0 0
734 un_flows 0 0 0
735 max_len 0 0 0
736 quantum 300 1514 762
737 .EE
738
739 .SS After some use:
740 .EX
741 # tc -s qdisc show dev eth0
742
743 qdisc cake 1: root refcnt 2 bandwidth 100Mbit diffserv3 triple-isolate rtt 100.0ms noatm overhead 38 mpu 84
744 Sent 44709231 bytes 31931 pkt (dropped 45, overlimits 93782 requeues 0)
745 backlog 33308b 22p requeues 0
746 memory used: 292352b of 5000000b
747 capacity estimate: 100Mbit
748 min/max network layer size: 28 / 1500
749 min/max overhead-adjusted size: 84 / 1538
750 average network hdr offset: 14
751
752 Bulk Best Effort Voice
753 thresh 6250Kbit 100Mbit 25Mbit
754 target 5.0ms 5.0ms 5.0ms
755 interval 100.0ms 100.0ms 100.0ms
756 pk_delay 8.7ms 6.9ms 5.0ms
757 av_delay 4.9ms 5.3ms 3.8ms
758 sp_delay 727us 1.4ms 511us
759 pkts 2590 21271 8137
760 bytes 3081804 30302659 11426206
761 way_inds 0 46 0
762 way_miss 3 17 4
763 way_cols 0 0 0
764 drops 20 15 10
765 marks 0 0 0
766 ack_drop 0 0 0
767 sp_flows 2 4 1
768 bk_flows 1 2 1
769 un_flows 0 0 0
770 max_len 1514 1514 1514
771 quantum 300 1514 762
772 .EE
773
774 .SH SEE ALSO
775 .BR tc (8),
776 .BR tc-codel (8),
777 .BR tc-fq_codel (8),
778 .BR tc-htb (8)
779
780 .SH AUTHORS
781 Cake's principal author is Jonathan Morton, with contributions from
782 Tony Ambardar, Kevin Darbyshire-Bryant, Toke Høiland-Jørgensen,
783 Sebastian Moeller, Ryan Mounce, Dean Scarff, Nils Andreas Svee, and Dave Täht.
784
785 This manual page was written by Loganaden Velvindron. Please report corrections
786 to the Linux Networking mailing list <[email protected]>.