1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
|
.\" $OpenBSD: altq.9,v 1.14 2007/05/31 19:20:00 jmc Exp $
.\"
.\" Copyright (C) 2001
.\" Sony Computer Science Laboratories Inc. All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\" notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\" notice, this list of conditions and the following disclaimer in the
.\" documentation and/or other materials provided with the distribution.
.\"
.\" THIS SOFTWARE IS PROVIDED BY SONY CSL AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED. IN NO EVENT SHALL SONY CSL OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.Dd $Mdocdate: May 31 2007 $
.Dt ALTQ 9
.Os
.\"
.Sh NAME
.Nm altq
.Nd kernel interfaces for manipulating output queues on network interfaces
.Sh SYNOPSIS
.Fd #include <sys/types.h>
.Fd #include <sys/socket.h>
.Fd #include <net/if.h>
.Ft void \"macro
.Fn IFQ_ENQUEUE "struct ifaltq *ifq" "struct mbuf *m" "struct altq_pktattr *pa" "int err"
.Ft void \"macro
.Fn IFQ_DEQUEUE "struct ifaltq *ifq" "struct mbuf *m"
.Ft void \"macro
.Fn IFQ_POLL "struct ifaltq *ifq" "struct mbuf *m"
.Ft void \"macro
.Fn IFQ_PURGE "struct ifaltq *ifq"
.Ft void \"macro
.Fn IFQ_CLASSIFY "struct ifaltq *ifq" "struct mbuf *m" "int af" "struct altq_pktattr *pktattr"
.Ft void \"macro
.Fn IFQ_IS_EMPTY "struct ifaltq *ifq"
.Ft void \"macro
.Fn IFQ_SET_MAXLEN "struct ifaltq *ifq" "int len"
.Ft void \"macro
.Fn IFQ_INC_LEN "struct ifaltq *ifq"
.Ft void \"macro
.Fn IFQ_DEC_LEN "struct ifaltq *ifq"
.Ft void \"macro
.Fn IFQ_INC_DROPS "struct ifaltq *ifq"
.Ft void \"macro
.Fn IFQ_SET_READY "struct ifaltq *ifq"
.Sh DESCRIPTION
The
.Nm
system is a framework to manage queuing disciplines on network
interfaces.
.Nm
introduces new macros to manipulate output queues.
The output queue macros are used to abstract queue operations and not to
touch the internal fields of the output queue structure.
The macros are independent from the
.Nm
implementation, and compatible with the traditional
.Dv ifqueue
macros for ease of transition.
.Pp
.Fn IFQ_ENQUEUE
enqueues a packet
.Fa m
to the queue
.Fa ifq .
The underlying queuing discipline may discard the packet.
.Fa err
is set to 0 on success, or
.Dv ENOBUFS
if the packet is discarded.
.Fa m
will be freed by the device driver on success or by the queuing discipline on
failure, so the caller should not touch
.Fa m
after calling
.Fn IFQ_ENQUEUE .
.Pp
.Fn IFQ_DEQUEUE
dequeues a packet from the queue.
The dequeued packet is returned in
.Fa m ,
or
.Fa m
is set to
.Dv NULL
if no packet is dequeued.
The caller must always check
.Fa m
since a non-empty queue could return
.Dv NULL
under rate-limiting.
.Pp
.Fn IFQ_POLL
returns the next packet without removing it from the queue.
It is guaranteed by the underlying queuing discipline that
.Fn IFQ_DEQUEUE
immediately after
.Fn IFQ_POLL
returns the same packet.
.Pp
.Fn IFQ_PURGE
discards all the packets in the queue.
The purge operation is needed since a non-work conserving queue cannot be
emptied by a dequeue loop.
.Pp
.Fn IFQ_CLASSIFY
classifies a packet to a scheduling class, and returns the result in
.Fa pktattr .
.Pp
.Fn IFQ_IS_EMPTY
can be used to check if the queue is empty.
Note that
.Fn IFQ_DEQUEUE
could still return
.Dv NULL
if the queuing discipline is non-work conserving.
.Pp
.Fn IFQ_SET_MAXLEN
sets the queue length limit to the default FIFO queue.
.Pp
.Fn IFQ_INC_LEN
and
.Fn IFQ_DEC_LEN
increment or decrement the current queue length in packets.
.Pp
.Fn IFQ_INC_DROPS
increments the drop counter and is equal to
.Fn IF_DROP .
It is defined for naming consistency.
.Pp
.Fn IFQ_SET_READY
sets a flag to indicate this driver is converted to use the new macros.
.Nm
can be enabled only on interfaces with this flag.
.Sh COMPATIBILITY
.Ss ifaltq structure
In order to keep compatibility with the existing code, the new
output queue structure
.Dv ifaltq
has the same fields.
The traditional
.Fn IF_XXX
macros and the code directly referencing the fields within
.Dv if_snd
still work with
.Dv ifaltq .
(Once we finish conversions of all the drivers, we no longer need
these fields.)
.Bd -literal
##old-style## ##new-style##
|
struct ifqueue { | struct ifaltq {
struct mbuf *ifq_head; | struct mbuf *ifq_head;
struct mbuf *ifq_tail; | struct mbuf *ifq_tail;
int ifq_len; | int ifq_len;
int ifq_maxlen; | int ifq_maxlen;
int ifq_drops; | int ifq_drops;
}; | /* altq related fields */
| ......
| };
|
.Ed
The new structure replaces
.Dv struct ifqueue
in
.Dv struct ifnet .
.Bd -literal
##old-style## ##new-style##
|
struct ifnet { | struct ifnet {
.... | ....
|
struct ifqueue if_snd; | struct ifaltq if_snd;
|
.... | ....
}; | };
|
.Ed
The (simplified) new
.Fn IFQ_XXX
macros looks like:
.Bd -literal
#ifdef ALTQ
#define IFQ_DEQUEUE(ifq, m) \e
if (ALTQ_IS_ENABLED((ifq)) \e
ALTQ_DEQUEUE((ifq), (m)); \e
else \e
IF_DEQUEUE((ifq), (m));
#else
#define IFQ_DEQUEUE(ifq, m) IF_DEQUEUE((ifq), (m));
#endif
.Ed
.Ss Enqueue operation
The semantics of the enqueue operation are changed.
In the new style,
enqueue and packet drop are combined since they cannot be easily
separated in many queuing disciplines.
The new enqueue operation corresponds to the following macro that is
written with the old macros.
.Bd -literal
#define IFQ_ENQUEUE(ifq, m, pattr, err) \e
do { \e
if (ALTQ_IS_ENABLED((ifq))) \e
ALTQ_ENQUEUE((ifq), (m), (pattr), (err)); \e
else { \e
if (IF_QFULL((ifq))) { \e
m_freem((m)); \e
(err) = ENOBUFS; \e
} else { \e
IF_ENQUEUE((ifq), (m)); \e
(err) = 0; \e
} \e
} \e
if ((err)) \e
(ifq)->ifq_drops++; \e
} while (0)
.Ed
.Pp
.Fn IFQ_ENQUEUE
does the following:
.Bl -hyphen -compact
.It
queue a packet
.It
drop (and free) a packet if the enqueue operation fails
.El
If the enqueue operation fails,
.Fa err
is set to
.Dv ENOBUFS .
.Fa m
is freed by the queuing discipline.
The caller should not touch
.Fa m
after calling
.Fn IFQ_ENQUEUE ,
so the caller may need to copy the
.Fa m_pkthdr.len
or
.Fa m_flags
fields beforehand for statistics.
The caller should not use
.Fn senderr
since
.Fa m
was already freed.
.Pp
The new style
.Fn if_output
looks as follows:
.Bd -literal
##old-style## ##new-style##
|
int | int
ether_output(ifp, m0, dst, rt0) | ether_output(ifp, m0, dst, rt0)
{ | {
...... | ......
|
| mflags = m->m_flags;
| len = m->m_pkthdr.len;
s = splimp(); | s = splimp();
if (IF_QFULL(&ifp->if_snd)) { | IFQ_ENQUEUE(&ifp->if_snd, m,
| NULL, error);
IF_DROP(&ifp->if_snd); | if (error != 0) {
splx(s); | splx(s);
senderr(ENOBUFS); | return (error);
} | }
IF_ENQUEUE(&ifp->if_snd, m); |
ifp->if_obytes += | ifp->if_obytes += len;
m->m_pkthdr.len; |
if (m->m_flags & M_MCAST) | if (mflags & M_MCAST)
ifp->if_omcasts++; | ifp->if_omcasts++;
|
if ((ifp->if_flags & IFF_OACTIVE) | if ((ifp->if_flags & IFF_OACTIVE)
== 0) | == 0)
(*ifp->if_start)(ifp); | (*ifp->if_start)(ifp);
splx(s); | splx(s);
return (error); | return (error);
|
bad: | bad:
if (m) | if (m)
m_freem(m); | m_freem(m);
return (error); | return (error);
} | }
|
.Ed
.Ss Classifier
The classifier mechanism is currently implemented in
.Fn if_output .
.Dv struct altq_pktattr
is used to store the classifier result, and it is passed to the enqueue
function.
(We will change the method to tag the classifier result to mbuf in the future.)
.Bd -literal
int
ether_output(ifp, m0, dst, rt0)
{
......
struct altq_pktattr pktattr;
......
/* classify the packet before prepending link-headers */
IFQ_CLASSIFY(&ifp->if_snd, m, dst->sa_family, &pktattr);
/* prepend link-level headers */
......
IFQ_ENQUEUE(&ifp->if_snd, m, &pktattr, error);
......
}
.Ed
.Sh HOW TO CONVERT THE EXISTING DRIVERS
First, make sure the corresponding
.Fn if_output
is already converted to the new style.
.Pp
Look for
.Fa if_snd
in the driver.
You will probably need to make changes to the lines that include
.Fa if_snd .
.Ss Empty check operation
If the code checks
.Fa ifq_head
to see whether the queue is empty or not, use
.Fn IFQ_IS_EMPTY .
.Bd -literal
##old-style## ##new-style##
|
if (ifp->if_snd.ifq_head != NULL) | if (IFQ_IS_EMPTY(&ifp->if_snd) == 0)
|
.Ed
Note that
.Fn IFQ_POLL
can be used for the same purpose, but
.Fn IFQ_POLL
could be costly for a complex scheduling algorithm since
.Fn IFQ_POLL
needs to run the scheduling algorithm to select the next packet.
On the other hand,
.Fn IFQ_IS_EMPTY
checks only if there is any packet stored in the queue.
Another difference is that even when
.Fn IFQ_IS_EMPTY
is
.Dv FALSE ,
.Fn IFQ_DEQUEUE
could still return
.Dv NULL
if the queue is under rate-limiting.
.Ss Dequeue operation
Replace
.Fn IF_DEQUEUE
by
.Fn IFQ_DEQUEUE .
Always check whether the dequeued mbuf is
.Dv NULL
or not.
Note that even when
.Fn IFQ_IS_EMPTY
is
.Dv FALSE ,
.Fn IFQ_DEQUEUE
could return
.Dv NULL
due to rate-limiting.
.Bd -literal
##old-style## ##new-style##
|
IF_DEQUEUE(&ifp->if_snd, m); | IFQ_DEQUEUE(&ifp->if_snd, m);
| if (m == NULL)
| return;
|
.Ed
A driver is supposed to call
.Fn if_start
from transmission complete interrupts in order to trigger the next dequeue.
.Ss Poll-and-dequeue operation
If the code polls the packet at the head of the queue and actually uses
the packet before dequeuing it, use
.Fn IFQ_POLL
and
.Fn IFQ_DEQUEUE .
.Bd -literal
##old-style## ##new-style##
|
m = ifp->if_snd.ifq_head; | IFQ_POLL(&ifp->if_snd, m);
if (m != NULL) { | if (m != NULL) {
|
/* use m to get resources */ | /* use m to get resources */
if (something goes wrong) | if (something goes wrong)
return; | return;
|
IF_DEQUEUE(&ifp->if_snd, m); | IFQ_DEQUEUE(&ifp->if_snd, m);
|
/* kick the hardware */ | /* kick the hardware */
} | }
|
.Ed
It is guaranteed that
.Fn IFQ_DEQUEUE
immediately after
.Fn IFQ_POLL
returns the same packet.
Note that they need to be guarded by
.Fn splimp
if called from outside of
.Fn if_start .
.Ss Eliminating IF_PREPEND
If the code uses
.Fn IF_PREPEND ,
you have to eliminate it since the prepend operation is not possible for many
queuing disciplines.
A common use of
.Fn IF_PREPEND
is to cancel the previous dequeue operation.
You have to convert the logic into poll-and-dequeue.
.Bd -literal
##old-style## ##new-style##
|
IF_DEQUEUE(&ifp->if_snd, m); | IFQ_POLL(&ifp->if_snd, m);
if (m != NULL) { | if (m != NULL) {
|
if (something_goes_wrong) { | if (something_goes_wrong) {
IF_PREPEND(&ifp->if_snd, m); |
return; | return;
} | }
|
| /* at this point, the driver
| * is committed to send this
| * packet.
| */
| IFQ_DEQUEUE(&ifp->if_snd, m);
|
/* kick the hardware */ | /* kick the hardware */
} | }
|
.Ed
.Ss Purge operation
Use
.Fn IFQ_PURGE
to empty the queue.
Note that a non-work conserving queue cannot be emptied by a dequeue loop.
.Bd -literal
##old-style## ##new-style##
|
while (ifp->if_snd.ifq_head != NULL) {| IFQ_PURGE(&ifp->if_snd);
IF_DEQUEUE(&ifp->if_snd, m); |
m_freem(m); |
} |
|
.Ed
.Ss Attach routine
Use
.Fn IFQ_SET_MAXLEN
to set
.Fa ifq_maxlen
to
.Fa len .
Add
.Fn IFQ_SET_READY
to show this driver is converted to the new style.
(This is used to distinguish new-style drivers.)
.Bd -literal
##old-style## ##new-style##
|
ifp->if_snd.ifq_maxlen = qsize; | IFQ_SET_MAXLEN(&ifp->if_snd, qsize);
| IFQ_SET_READY(&ifp->if_snd);
if_attach(ifp); | if_attach(ifp);
|
.Ed
.Ss Other issues
The new macros for statistics:
.Bd -literal
##old-style## ##new-style##
|
IF_DROP(&ifp->if_snd); | IFQ_INC_DROPS(&ifp->if_snd);
|
ifp->if_snd.ifq_len++; | IFQ_INC_LEN(&ifp->if_snd);
|
ifp->if_snd.ifq_len--; | IFQ_DEC_LEN(&ifp->if_snd);
|
.Ed
Some drivers instruct the hardware to invoke transmission complete
interrupts only when it thinks necessary.
Rate-limiting breaks its assumption.
.Ss How to convert drivers using multiple ifqueues
Some (pseudo) devices (such as slip) have another
.Dv ifqueue
to prioritize packets.
It is possible to eliminate the second queue
since
.Nm
provides more flexible mechanisms but the following shows
how to keep the original behavior.
.Bd -literal
struct sl_softc {
struct ifnet sc_if; /* network-visible interface */
...
struct ifqueue sc_fastq; /* interactive output queue */
...
};
.Ed
The driver doesn't compile in the new model since it has the following
line
.Po
.Fa if_snd
is no longer a type of
.Dv struct ifqueue
.Pc .
.Bd -literal
struct ifqueue *ifq = &ifp->if_snd;
.Ed
A simple way is to use the original
.Fn IF_XXX
macros for
.Fa sc_fastq
and use the new
.Fn IFQ_XXX
macros for
.Fa if_snd .
The enqueue operation looks like:
.Bd -literal
##old-style## ##new-style##
|
struct ifqueue *ifq = &ifp->if_snd; | struct ifqueue *ifq = NULL;
|
if (ip->ip_tos & IPTOS_LOWDELAY) | if ((ip->ip_tos & IPTOS_LOWDELAY) &&
ifq = &sc->sc_fastq; | !ALTQ_IS_ENABLED(&sc->sc_if.if_snd)) {
| ifq = &sc->sc_fastq;
if (IF_QFULL(ifq)) { | if (IF_QFULL(ifq)) {
IF_DROP(ifq); | IF_DROP(ifq);
m_freem(m); | m_freem(m);
splx(s); | error = ENOBUFS;
sc->sc_if.if_oerrors++; | } else {
return (ENOBUFS); | IF_ENQUEUE(ifq, m);
} | error = 0;
IF_ENQUEUE(ifq, m); | }
| } else
| IFQ_ENQUEUE(&sc->sc_if.if_snd,
| NULL, m, error);
|
| if (error) {
| splx(s);
| sc->sc_if.if_oerrors++;
| return (error);
| }
if ((sc->sc_oqlen = | if ((sc->sc_oqlen =
sc->sc_ttyp->t_outq.c_cc) == 0) | sc->sc_ttyp->t_outq.c_cc) == 0)
slstart(sc->sc_ttyp); | slstart(sc->sc_ttyp);
splx(s); | splx(s);
|
.Ed
The dequeue operations looks like:
.Bd -literal
##old-style## ##new-style##
|
s = splimp(); | s = splimp();
IF_DEQUEUE(&sc->sc_fastq, m); | IF_DEQUEUE(&sc->sc_fastq, m);
if (m == NULL) | if (m == NULL)
IF_DEQUEUE(&sc->sc_if.if_snd, m); | IFQ_DEQUEUE(&sc->sc_if.if_snd, m);
splx(s); | splx(s);
|
.Ed
.Sh QUEUEING DISCIPLINES
Queuing disciplines need to maintain
.Fa ifq_len
.Po
used by
.Fn IFQ_IS_EMPTY
.Pc .
Queuing disciplines also need to guarantee the same mbuf is returned if
.Fn IFQ_DEQUEUE
is called immediately after
.Fn IFQ_POLL .
.Sh SEE ALSO
.Xr pf.conf 5 ,
.Xr pfctl 8
.Sh HISTORY
The
.Nm
system first appeared in March 1997.
|