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
|
/*-
* Copyright (c) 2000 Jake Burkholder <jake@freebsd.org>.
* 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 THE AUTHOR 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 THE AUTHOR 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.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_ktrace.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/kernel.h>
#include <sys/ktr.h>
#include <sys/condvar.h>
#include <sys/sched.h>
#include <sys/signalvar.h>
#include <sys/sleepqueue.h>
#include <sys/resourcevar.h>
#ifdef KTRACE
#include <sys/uio.h>
#include <sys/ktrace.h>
#endif
/*
* Common sanity checks for cv_wait* functions.
*/
#define CV_ASSERT(cvp, lock, td) do { \
KASSERT((td) != NULL, ("%s: curthread NULL", __func__)); \
KASSERT(TD_IS_RUNNING(td), ("%s: not TDS_RUNNING", __func__)); \
KASSERT((cvp) != NULL, ("%s: cvp NULL", __func__)); \
KASSERT((lock) != NULL, ("%s: lock NULL", __func__)); \
} while (0)
/*
* Initialize a condition variable. Must be called before use.
*/
void
cv_init(struct cv *cvp, const char *desc)
{
cvp->cv_description = desc;
cvp->cv_waiters = 0;
}
/*
* Destroy a condition variable. The condition variable must be re-initialized
* in order to be re-used.
*/
void
cv_destroy(struct cv *cvp)
{
#ifdef INVARIANTS
struct sleepqueue *sq;
sleepq_lock(cvp);
sq = sleepq_lookup(cvp);
sleepq_release(cvp);
KASSERT(sq == NULL, ("%s: associated sleep queue non-empty", __func__));
#endif
}
/*
* Wait on a condition variable. The current thread is placed on the condition
* variable's wait queue and suspended. A cv_signal or cv_broadcast on the same
* condition variable will resume the thread. The mutex is released before
* sleeping and will be held on return. It is recommended that the mutex be
* held when cv_signal or cv_broadcast are called.
*/
void
_cv_wait(struct cv *cvp, struct lock_object *lock)
{
WITNESS_SAVE_DECL(lock_witness);
struct lock_class *class;
struct thread *td;
int lock_state;
td = curthread;
lock_state = 0;
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(1, 0, cv_wmesg(cvp));
#endif
CV_ASSERT(cvp, lock, td);
WITNESS_WARN(WARN_GIANTOK | WARN_SLEEPOK, lock,
"Waiting on \"%s\"", cvp->cv_description);
class = LOCK_CLASS(lock);
if (cold || panicstr) {
/*
* During autoconfiguration, just give interrupts
* a chance, then just return. Don't run any other
* thread or panic below, in case this is the idle
* process and already asleep.
*/
return;
}
sleepq_lock(cvp);
cvp->cv_waiters++;
if (lock == &Giant.lock_object)
mtx_assert(&Giant, MA_OWNED);
DROP_GIANT();
sleepq_add(cvp, lock, cvp->cv_description, SLEEPQ_CONDVAR, 0);
if (lock != &Giant.lock_object) {
if (class->lc_flags & LC_SLEEPABLE)
sleepq_release(cvp);
WITNESS_SAVE(lock, lock_witness);
lock_state = class->lc_unlock(lock);
if (class->lc_flags & LC_SLEEPABLE)
sleepq_lock(cvp);
}
sleepq_wait(cvp, 0);
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(0, 0, cv_wmesg(cvp));
#endif
PICKUP_GIANT();
if (lock != &Giant.lock_object) {
class->lc_lock(lock, lock_state);
WITNESS_RESTORE(lock, lock_witness);
}
}
/*
* Wait on a condition variable. This function differs from cv_wait by
* not aquiring the mutex after condition variable was signaled.
*/
void
_cv_wait_unlock(struct cv *cvp, struct lock_object *lock)
{
struct lock_class *class;
struct thread *td;
td = curthread;
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(1, 0, cv_wmesg(cvp));
#endif
CV_ASSERT(cvp, lock, td);
WITNESS_WARN(WARN_GIANTOK | WARN_SLEEPOK, lock,
"Waiting on \"%s\"", cvp->cv_description);
KASSERT(lock != &Giant.lock_object,
("cv_wait_unlock cannot be used with Giant"));
class = LOCK_CLASS(lock);
if (cold || panicstr) {
/*
* During autoconfiguration, just give interrupts
* a chance, then just return. Don't run any other
* thread or panic below, in case this is the idle
* process and already asleep.
*/
class->lc_unlock(lock);
return;
}
sleepq_lock(cvp);
cvp->cv_waiters++;
DROP_GIANT();
sleepq_add(cvp, lock, cvp->cv_description, SLEEPQ_CONDVAR, 0);
if (class->lc_flags & LC_SLEEPABLE)
sleepq_release(cvp);
class->lc_unlock(lock);
if (class->lc_flags & LC_SLEEPABLE)
sleepq_lock(cvp);
sleepq_wait(cvp, 0);
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(0, 0, cv_wmesg(cvp));
#endif
PICKUP_GIANT();
}
/*
* Wait on a condition variable, allowing interruption by signals. Return 0 if
* the thread was resumed with cv_signal or cv_broadcast, EINTR or ERESTART if
* a signal was caught. If ERESTART is returned the system call should be
* restarted if possible.
*/
int
_cv_wait_sig(struct cv *cvp, struct lock_object *lock)
{
WITNESS_SAVE_DECL(lock_witness);
struct lock_class *class;
struct thread *td;
int lock_state, rval;
td = curthread;
lock_state = 0;
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(1, 0, cv_wmesg(cvp));
#endif
CV_ASSERT(cvp, lock, td);
WITNESS_WARN(WARN_GIANTOK | WARN_SLEEPOK, lock,
"Waiting on \"%s\"", cvp->cv_description);
class = LOCK_CLASS(lock);
if (cold || panicstr) {
/*
* After a panic, or during autoconfiguration, just give
* interrupts a chance, then just return; don't run any other
* procs or panic below, in case this is the idle process and
* already asleep.
*/
return (0);
}
sleepq_lock(cvp);
cvp->cv_waiters++;
if (lock == &Giant.lock_object)
mtx_assert(&Giant, MA_OWNED);
DROP_GIANT();
sleepq_add(cvp, lock, cvp->cv_description, SLEEPQ_CONDVAR |
SLEEPQ_INTERRUPTIBLE, 0);
if (lock != &Giant.lock_object) {
if (class->lc_flags & LC_SLEEPABLE)
sleepq_release(cvp);
WITNESS_SAVE(lock, lock_witness);
lock_state = class->lc_unlock(lock);
if (class->lc_flags & LC_SLEEPABLE)
sleepq_lock(cvp);
}
rval = sleepq_wait_sig(cvp, 0);
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(0, 0, cv_wmesg(cvp));
#endif
PICKUP_GIANT();
if (lock != &Giant.lock_object) {
class->lc_lock(lock, lock_state);
WITNESS_RESTORE(lock, lock_witness);
}
return (rval);
}
/*
* Wait on a condition variable for at most timo/hz seconds. Returns 0 if the
* process was resumed by cv_signal or cv_broadcast, EWOULDBLOCK if the timeout
* expires.
*/
int
_cv_timedwait(struct cv *cvp, struct lock_object *lock, int timo)
{
WITNESS_SAVE_DECL(lock_witness);
struct lock_class *class;
struct thread *td;
int lock_state, rval;
td = curthread;
lock_state = 0;
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(1, 0, cv_wmesg(cvp));
#endif
CV_ASSERT(cvp, lock, td);
WITNESS_WARN(WARN_GIANTOK | WARN_SLEEPOK, lock,
"Waiting on \"%s\"", cvp->cv_description);
class = LOCK_CLASS(lock);
if (cold || panicstr) {
/*
* After a panic, or during autoconfiguration, just give
* interrupts a chance, then just return; don't run any other
* thread or panic below, in case this is the idle process and
* already asleep.
*/
return 0;
}
sleepq_lock(cvp);
cvp->cv_waiters++;
if (lock == &Giant.lock_object)
mtx_assert(&Giant, MA_OWNED);
DROP_GIANT();
sleepq_add(cvp, lock, cvp->cv_description, SLEEPQ_CONDVAR, 0);
sleepq_set_timeout(cvp, timo);
if (lock != &Giant.lock_object) {
if (class->lc_flags & LC_SLEEPABLE)
sleepq_release(cvp);
WITNESS_SAVE(lock, lock_witness);
lock_state = class->lc_unlock(lock);
if (class->lc_flags & LC_SLEEPABLE)
sleepq_lock(cvp);
}
rval = sleepq_timedwait(cvp, 0);
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(0, 0, cv_wmesg(cvp));
#endif
PICKUP_GIANT();
if (lock != &Giant.lock_object) {
class->lc_lock(lock, lock_state);
WITNESS_RESTORE(lock, lock_witness);
}
return (rval);
}
/*
* Wait on a condition variable for at most timo/hz seconds, allowing
* interruption by signals. Returns 0 if the thread was resumed by cv_signal
* or cv_broadcast, EWOULDBLOCK if the timeout expires, and EINTR or ERESTART if
* a signal was caught.
*/
int
_cv_timedwait_sig(struct cv *cvp, struct lock_object *lock, int timo)
{
WITNESS_SAVE_DECL(lock_witness);
struct lock_class *class;
struct thread *td;
int lock_state, rval;
td = curthread;
lock_state = 0;
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(1, 0, cv_wmesg(cvp));
#endif
CV_ASSERT(cvp, lock, td);
WITNESS_WARN(WARN_GIANTOK | WARN_SLEEPOK, lock,
"Waiting on \"%s\"", cvp->cv_description);
class = LOCK_CLASS(lock);
if (cold || panicstr) {
/*
* After a panic, or during autoconfiguration, just give
* interrupts a chance, then just return; don't run any other
* thread or panic below, in case this is the idle process and
* already asleep.
*/
return 0;
}
sleepq_lock(cvp);
cvp->cv_waiters++;
if (lock == &Giant.lock_object)
mtx_assert(&Giant, MA_OWNED);
DROP_GIANT();
sleepq_add(cvp, lock, cvp->cv_description, SLEEPQ_CONDVAR |
SLEEPQ_INTERRUPTIBLE, 0);
sleepq_set_timeout(cvp, timo);
if (lock != &Giant.lock_object) {
if (class->lc_flags & LC_SLEEPABLE)
sleepq_release(cvp);
WITNESS_SAVE(lock, lock_witness);
lock_state = class->lc_unlock(lock);
if (class->lc_flags & LC_SLEEPABLE)
sleepq_lock(cvp);
}
rval = sleepq_timedwait_sig(cvp, 0);
#ifdef KTRACE
if (KTRPOINT(td, KTR_CSW))
ktrcsw(0, 0, cv_wmesg(cvp));
#endif
PICKUP_GIANT();
if (lock != &Giant.lock_object) {
class->lc_lock(lock, lock_state);
WITNESS_RESTORE(lock, lock_witness);
}
return (rval);
}
/*
* Signal a condition variable, wakes up one waiting thread. Will also wakeup
* the swapper if the process is not in memory, so that it can bring the
* sleeping process in. Note that this may also result in additional threads
* being made runnable. Should be called with the same mutex as was passed to
* cv_wait held.
*/
void
cv_signal(struct cv *cvp)
{
int wakeup_swapper;
wakeup_swapper = 0;
sleepq_lock(cvp);
if (cvp->cv_waiters > 0) {
cvp->cv_waiters--;
wakeup_swapper = sleepq_signal(cvp, SLEEPQ_CONDVAR, 0, 0);
}
sleepq_release(cvp);
if (wakeup_swapper)
kick_proc0();
}
/*
* Broadcast a signal to a condition variable. Wakes up all waiting threads.
* Should be called with the same mutex as was passed to cv_wait held.
*/
void
cv_broadcastpri(struct cv *cvp, int pri)
{
int wakeup_swapper;
/*
* XXX sleepq_broadcast pri argument changed from -1 meaning
* no pri to 0 meaning no pri.
*/
wakeup_swapper = 0;
if (pri == -1)
pri = 0;
sleepq_lock(cvp);
if (cvp->cv_waiters > 0) {
cvp->cv_waiters = 0;
wakeup_swapper = sleepq_broadcast(cvp, SLEEPQ_CONDVAR, pri, 0);
}
sleepq_release(cvp);
if (wakeup_swapper)
kick_proc0();
}
|