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
|
/* $OpenBSD: switchctl.c,v 1.7 2016/11/10 17:32:40 rzalamena Exp $ */
/*
* Copyright (c) 2016 Kazuya GODA <goda@openbsd.org>
* Copyright (c) 2015, 2016 YASUOKA Masahiko <yasuoka@openbsd.org>
* Copyright (c) 2015, 2016 Reyk Floeter <reyk@openbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/ioctl.h>
#include <sys/selinfo.h>
#include <sys/rwlock.h>
#include <sys/proc.h>
#include <net/if.h>
#include <net/rtable.h>
#include <netinet/in.h>
#include <netinet/if_ether.h>
#include <net/if_switch.h>
extern struct rwlock switch_ifs_lk;
/*
* device part of switch(4)
*/
#include <sys/poll.h>
#include <sys/selinfo.h>
#include <sys/vnode.h>
struct switch_softc *vsw_dev2sc(dev_t);
int switchopen(dev_t, int, int, struct proc *);
int switchread(dev_t, struct uio *, int);
int switchwrite(dev_t, struct uio *, int);
int switchioctl(dev_t, u_long, caddr_t, int, struct proc *);
int switchclose(dev_t, int, int, struct proc *);
int switchpoll(dev_t, int, struct proc *);
int switchkqfilter(dev_t, struct knote *);
void filt_switch_rdetach(struct knote *);
int filt_switch_read(struct knote *, long);
void filt_switch_wdetach(struct knote *);
int filt_switch_write(struct knote *, long);
int switch_dev_output(struct switch_softc *, struct mbuf *);
void switch_dev_wakeup(struct switch_softc *);
struct filterops switch_rd_filtops = {
1, NULL, filt_switch_rdetach, filt_switch_read
};
struct filterops switch_wr_filtops = {
1, NULL, filt_switch_wdetach, filt_switch_write
};
struct switch_softc *
switch_dev2sc(dev_t dev)
{
struct switch_softc *sc;
rw_enter_read(&switch_ifs_lk);
sc = switch_lookup(minor(dev));
rw_exit_read(&switch_ifs_lk);
return (sc);
}
int
switchopen(dev_t dev, int flags, int mode, struct proc *p)
{
struct switch_softc *sc;
char name[IFNAMSIZ];
int rv, s, error = 0;
if ((sc = switch_dev2sc(dev)) == NULL) {
snprintf(name, sizeof(name), "switch%d", minor(dev));
if ((rv = if_clone_create(name,
rtable_l2(p->p_p->ps_rtableid))) != 0)
return (rv);
if ((sc = switch_dev2sc(dev)) == NULL)
return (ENXIO);
}
rw_enter_write(&switch_ifs_lk);
if (sc->sc_swdev != NULL) {
error = EBUSY;
goto failed;
}
if ((sc->sc_swdev = malloc(sizeof(struct switch_dev), M_DEVBUF,
M_DONTWAIT|M_ZERO)) == NULL ) {
error = ENOBUFS;
goto failed;
}
s = splnet();
mq_init(&sc->sc_swdev->swdev_outq, 128, IPL_NET);
sc->sc_swdev->swdev_output = switch_dev_output;
if (sc->sc_capabilities & SWITCH_CAP_OFP)
swofp_init(sc);
splx(s);
failed:
rw_exit_write(&switch_ifs_lk);
return (error);
}
int
switchread(dev_t dev, struct uio *uio, int ioflag)
{
struct switch_softc *sc;
struct mbuf *m;
u_int len;
int s, error = 0;
sc = switch_dev2sc(dev);
if (sc == NULL)
return (ENXIO);
if (sc->sc_swdev->swdev_lastm != NULL) {
m = sc->sc_swdev->swdev_lastm;
sc->sc_swdev->swdev_lastm = NULL;
goto skip_dequeue;
}
dequeue_next:
s = splnet();
while ((m = mq_dequeue(&sc->sc_swdev->swdev_outq)) == NULL) {
if (ISSET(ioflag, IO_NDELAY)) {
error = EWOULDBLOCK;
goto failed;
}
sc->sc_swdev->swdev_waiting = 1;
error = tsleep(sc, (PZERO + 1)|PCATCH, "switchread", 0);
if (error != 0)
goto failed;
/* sc might be deleted while sleeping */
sc = switch_dev2sc(dev);
if (sc == NULL) {
error = ENXIO;
goto failed;
}
}
splx(s);
skip_dequeue:
while (uio->uio_resid > 0) {
len = ulmin(uio->uio_resid, m->m_len);
if ((error = uiomove(mtod(m, caddr_t), len, uio)) != 0) {
/* Save it so user can recover from EFAULT. */
sc->sc_swdev->swdev_lastm = m;
return (error);
}
/* Handle partial reads. */
if (uio->uio_resid == 0) {
if (len < m->m_len)
m_adj(m, len);
else
m = m_free(m);
sc->sc_swdev->swdev_lastm = m;
break;
}
/*
* After consuming data from this mbuf test if we
* have to dequeue a new chain.
*/
m = m_free(m);
if (m == NULL)
goto dequeue_next;
}
return (0);
failed:
splx(s);
return (error);
}
int
switchwrite(dev_t dev, struct uio *uio, int ioflag)
{
struct switch_softc *sc = NULL;
struct mbuf *m, *n, *mhead, *mtail = NULL;
int s, error, trailing;
size_t len;
if (uio->uio_resid == 0)
return (0);
len = uio->uio_resid;
sc = switch_dev2sc(dev);
if (sc == NULL)
return (ENXIO);
if (sc->sc_swdev->swdev_inputm == NULL) {
MGETHDR(m, M_DONTWAIT, MT_DATA);
if (m == NULL)
return (ENOBUFS);
if (len >= MHLEN) {
MCLGETI(m, M_DONTWAIT, NULL, MIN(MAXMCLBYTES, len));
if ((m->m_flags & M_EXT) == 0) {
m_free(m);
return (ENOBUFS);
}
}
mhead = m;
/* M_TRAILINGSPACE() uses this to calculate space. */
m->m_len = 0;
} else {
/* Recover the mbuf from the last write and get its tail. */
mhead = sc->sc_swdev->swdev_inputm;
for (m = mhead; m->m_next != NULL; m = m->m_next)
/* NOTHING */;
sc->sc_swdev->swdev_inputm = NULL;
}
while (len) {
trailing = ulmin(M_TRAILINGSPACE(m), len);
if ((error = uiomove(mtod(m, caddr_t), trailing, uio)) != 0)
goto save_return;
len -= trailing;
mhead->m_pkthdr.len += trailing;
m->m_len += trailing;
if (len == 0)
break;
MGET(n, M_DONTWAIT, MT_DATA);
if (n == NULL) {
error = ENOBUFS;
goto save_return;
}
if (len >= MLEN) {
MCLGETI(n, M_DONTWAIT, NULL, MIN(MAXMCLBYTES, len));
if ((n->m_flags & M_EXT) == 0) {
m_free(n);
error = ENOBUFS;
goto save_return;
}
}
n->m_len = 0;
m->m_next = n;
m = n;
}
/* Loop until there is no more complete OFP packets. */
while (ofp_split_mbuf(mhead, &mtail) == 0) {
s = splnet();
sc->sc_swdev->swdev_input(sc, mhead);
splx(s);
/* We wrote everything, just quit. */
if (mtail == NULL)
return (0);
mhead = mtail;
}
/* Save the head, because ofp_split_mbuf failed. */
sc->sc_swdev->swdev_inputm = mhead;
return (0);
save_return:
/* Save it so user can recover from errors later. */
sc->sc_swdev->swdev_inputm = mhead;
return (error);
}
int
switchioctl(dev_t dev, u_long cmd, caddr_t addr, int flags, struct proc *p)
{
int error;
switch (cmd) {
case FIONBIO:
case FIOASYNC:
case FIONREAD:
return (0);
default:
error = ENOTTY;
break;
}
return (error);
}
int
switchclose(dev_t dev, int flags, int mode, struct proc *p)
{
struct switch_softc *sc;
rw_enter_write(&switch_ifs_lk);
sc = switch_lookup(minor(dev));
if (sc != NULL && sc->sc_swdev != NULL) {
m_freem(sc->sc_swdev->swdev_lastm);
m_freem(sc->sc_swdev->swdev_inputm);
mq_purge(&sc->sc_swdev->swdev_outq);
free(sc->sc_swdev, M_DEVBUF, sizeof(struct switch_dev));
sc->sc_swdev = NULL;
}
rw_exit_write(&switch_ifs_lk);
return (0);
}
void
switch_dev_destroy(struct switch_softc *sc)
{
int s;
if (sc->sc_swdev == NULL)
return;
rw_enter_write(&switch_ifs_lk);
if (sc->sc_swdev != NULL) {
switch_dev_wakeup(sc);
s = splhigh();
klist_invalidate(&sc->sc_swdev->swdev_rsel.si_note);
klist_invalidate(&sc->sc_swdev->swdev_wsel.si_note);
splx(s);
m_freem(sc->sc_swdev->swdev_lastm);
m_freem(sc->sc_swdev->swdev_inputm);
mq_purge(&sc->sc_swdev->swdev_outq);
free(sc->sc_swdev, M_DEVBUF, sizeof(struct switch_dev));
sc->sc_swdev = NULL;
}
rw_exit_write(&switch_ifs_lk);
}
int
switchpoll(dev_t dev, int events, struct proc *p)
{
int revents = 0;
struct switch_softc *sc = switch_dev2sc(dev);
if (sc == NULL)
return (ENXIO);
if (events & (POLLIN | POLLRDNORM)) {
if (!mq_empty(&sc->sc_swdev->swdev_outq) ||
sc->sc_swdev->swdev_lastm != NULL)
revents |= events & (POLLIN | POLLRDNORM);
}
if (events & (POLLOUT | POLLWRNORM))
revents |= events & (POLLOUT | POLLWRNORM);
if (revents == 0) {
if (events & (POLLIN | POLLRDNORM))
selrecord(p, &sc->sc_swdev->swdev_rsel);
}
return (revents);
}
int
switchkqfilter(dev_t dev, struct knote *kn)
{
struct switch_softc *sc = switch_dev2sc(dev);
struct mutex *mtx;
struct klist *klist;
if (sc == NULL)
return (ENXIO);
switch (kn->kn_filter) {
case EVFILT_READ:
mtx = &sc->sc_swdev->swdev_rsel_mtx;
klist = &sc->sc_swdev->swdev_rsel.si_note;
kn->kn_fop = &switch_rd_filtops;
break;
case EVFILT_WRITE:
mtx = &sc->sc_swdev->swdev_wsel_mtx;
klist = &sc->sc_swdev->swdev_wsel.si_note;
kn->kn_fop = &switch_wr_filtops;
break;
default:
return (EINVAL);
}
kn->kn_hook = (caddr_t)sc;
mtx_enter(mtx);
SLIST_INSERT_HEAD(klist, kn, kn_selnext);
mtx_leave(mtx);
return (0);
}
void
filt_switch_rdetach(struct knote *kn)
{
struct switch_softc *sc = (struct switch_softc *)kn->kn_hook;
struct klist *klist = &sc->sc_swdev->swdev_rsel.si_note;
if (ISSET(kn->kn_status, KN_DETACHED))
return;
mtx_enter(&sc->sc_swdev->swdev_rsel_mtx);
SLIST_REMOVE(klist, kn, knote, kn_selnext);
mtx_leave(&sc->sc_swdev->swdev_rsel_mtx);
}
int
filt_switch_read(struct knote *kn, long hint)
{
struct switch_softc *sc = (struct switch_softc *)kn->kn_hook;
if (ISSET(kn->kn_status, KN_DETACHED)) {
kn->kn_data = 0;
return (1);
}
if (!mq_empty(&sc->sc_swdev->swdev_outq) ||
sc->sc_swdev->swdev_lastm != NULL) {
kn->kn_data = mq_len(&sc->sc_swdev->swdev_outq) +
(sc->sc_swdev->swdev_lastm != NULL);
return (1);
}
return (0);
}
void
filt_switch_wdetach(struct knote *kn)
{
struct switch_softc *sc = (struct switch_softc *)kn->kn_hook;
struct klist *klist = &sc->sc_swdev->swdev_wsel.si_note;
if (ISSET(kn->kn_status, KN_DETACHED))
return;
mtx_enter(&sc->sc_swdev->swdev_wsel_mtx);
SLIST_REMOVE(klist, kn, knote, kn_selnext);
mtx_leave(&sc->sc_swdev->swdev_wsel_mtx);
}
int
filt_switch_write(struct knote *kn, long hint)
{
/* Always writable */
return (1);
}
int
switch_dev_output(struct switch_softc *sc, struct mbuf *m)
{
if (mq_enqueue(&sc->sc_swdev->swdev_outq, m) != 0)
return (-1);
switch_dev_wakeup(sc);
return (0);
}
void
switch_dev_wakeup(struct switch_softc *sc)
{
if (sc->sc_swdev->swdev_waiting) {
sc->sc_swdev->swdev_waiting = 0;
wakeup((caddr_t)sc);
}
selwakeup(&sc->sc_swdev->swdev_rsel);
}
|