summaryrefslogtreecommitdiff
path: root/sys/kern/kern_tc.c
blob: ab12daba7e63d9972091213e232310ebd7580f45 (plain)
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
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
/*	$OpenBSD: kern_tc.c,v 1.31 2017/03/07 20:22:37 dhill Exp $ */

/*
 * Copyright (c) 2000 Poul-Henning Kamp <phk@FreeBSD.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.
 */

/*
 * If we meet some day, and you think this stuff is worth it, you
 * can buy me a beer in return. Poul-Henning Kamp
 */

#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/timeout.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <sys/systm.h>
#include <sys/timetc.h>
#include <sys/malloc.h>
#include <dev/rndvar.h>

/*
 * A large step happens on boot.  This constant detects such steps.
 * It is relatively small so that ntp_update_second gets called enough
 * in the typical 'missed a couple of seconds' case, but doesn't loop
 * forever when the time step is large.
 */
#define LARGE_STEP	200

u_int dummy_get_timecount(struct timecounter *);

void ntp_update_second(int64_t *);
int sysctl_tc_hardware(void *, size_t *, void *, size_t);
int sysctl_tc_choice(void *, size_t *, void *, size_t);

/*
 * Implement a dummy timecounter which we can use until we get a real one
 * in the air.  This allows the console and other early stuff to use
 * time services.
 */

u_int
dummy_get_timecount(struct timecounter *tc)
{
	static u_int now;

	return (++now);
}

static struct timecounter dummy_timecounter = {
	dummy_get_timecount, 0, ~0u, 1000000, "dummy", -1000000
};

struct timehands {
	/* These fields must be initialized by the driver. */
	struct timecounter	*th_counter;
	int64_t			th_adjustment;
	u_int64_t		th_scale;
	u_int	 		th_offset_count;
	struct bintime		th_offset;
	struct timeval		th_microtime;
	struct timespec		th_nanotime;
	/* Fields not to be copied in tc_windup start with th_generation. */
	volatile u_int		th_generation;
	struct timehands	*th_next;
};

static struct timehands th0;
static struct timehands th9 = { NULL, 0, 0, 0, {0, 0}, {0, 0}, {0, 0}, 0, &th0};
static struct timehands th8 = { NULL, 0, 0, 0, {0, 0}, {0, 0}, {0, 0}, 0, &th9};
static struct timehands th7 = { NULL, 0, 0, 0, {0, 0}, {0, 0}, {0, 0}, 0, &th8};
static struct timehands th6 = { NULL, 0, 0, 0, {0, 0}, {0, 0}, {0, 0}, 0, &th7};
static struct timehands th5 = { NULL, 0, 0, 0, {0, 0}, {0, 0}, {0, 0}, 0, &th6};
static struct timehands th4 = { NULL, 0, 0, 0, {0, 0}, {0, 0}, {0, 0}, 0, &th5};
static struct timehands th3 = { NULL, 0, 0, 0, {0, 0}, {0, 0}, {0, 0}, 0, &th4};
static struct timehands th2 = { NULL, 0, 0, 0, {0, 0}, {0, 0}, {0, 0}, 0, &th3};
static struct timehands th1 = { NULL, 0, 0, 0, {0, 0}, {0, 0}, {0, 0}, 0, &th2};
static struct timehands th0 = {
	&dummy_timecounter,
	0,
	(uint64_t)-1 / 1000000,
	0,
	{1, 0},
	{0, 0},
	{0, 0},
	1,
	&th1
};

static struct timehands *volatile timehands = &th0;
struct timecounter *timecounter = &dummy_timecounter;
static struct timecounter *timecounters = &dummy_timecounter;

volatile time_t time_second = 1;
volatile time_t time_uptime = 0;

struct bintime naptime;
static struct bintime boottimebin;
static int timestepwarnings;

void tc_windup(void);

/*
 * Return the difference between the timehands' counter value now and what
 * was when we copied it to the timehands' offset_count.
 */
static __inline u_int
tc_delta(struct timehands *th)
{
	struct timecounter *tc;

	tc = th->th_counter;
	return ((tc->tc_get_timecount(tc) - th->th_offset_count) &
	    tc->tc_counter_mask);
}

/*
 * Functions for reading the time.  We have to loop until we are sure that
 * the timehands that we operated on was not updated under our feet.  See
 * the comment in <sys/time.h> for a description of these 12 functions.
 */

void
binuptime(struct bintime *bt)
{
	struct timehands *th;
	u_int gen;

	do {
		th = timehands;
		gen = th->th_generation;
		*bt = th->th_offset;
		bintime_addx(bt, th->th_scale * tc_delta(th));
	} while (gen == 0 || gen != th->th_generation);
}

void
nanouptime(struct timespec *tsp)
{
	struct bintime bt;

	binuptime(&bt);
	bintime2timespec(&bt, tsp);
}

void
microuptime(struct timeval *tvp)
{
	struct bintime bt;

	binuptime(&bt);
	bintime2timeval(&bt, tvp);
}

void
bintime(struct bintime *bt)
{

	binuptime(bt);
	bintime_add(bt, &boottimebin);
}

void
nanotime(struct timespec *tsp)
{
	struct bintime bt;

	bintime(&bt);
	bintime2timespec(&bt, tsp);
}

void
microtime(struct timeval *tvp)
{
	struct bintime bt;

	bintime(&bt);
	bintime2timeval(&bt, tvp);
}

void
getnanouptime(struct timespec *tsp)
{
	struct timehands *th;
	u_int gen;

	do {
		th = timehands;
		gen = th->th_generation;
		bintime2timespec(&th->th_offset, tsp);
	} while (gen == 0 || gen != th->th_generation);
}

void
getmicrouptime(struct timeval *tvp)
{
	struct timehands *th;
	u_int gen;

	do {
		th = timehands;
		gen = th->th_generation;
		bintime2timeval(&th->th_offset, tvp);
	} while (gen == 0 || gen != th->th_generation);
}

void
getnanotime(struct timespec *tsp)
{
	struct timehands *th;
	u_int gen;

	do {
		th = timehands;
		gen = th->th_generation;
		*tsp = th->th_nanotime;
	} while (gen == 0 || gen != th->th_generation);
}

void
getmicrotime(struct timeval *tvp)
{
	struct timehands *th;
	u_int gen;

	do {
		th = timehands;
		gen = th->th_generation;
		*tvp = th->th_microtime;
	} while (gen == 0 || gen != th->th_generation);
}

/*
 * Initialize a new timecounter and possibly use it.
 */
void
tc_init(struct timecounter *tc)
{
	u_int u;

	u = tc->tc_frequency / tc->tc_counter_mask;
	/* XXX: We need some margin here, 10% is a guess */
	u *= 11;
	u /= 10;
	if (tc->tc_quality >= 0) {
		if (u > hz) {
			tc->tc_quality = -2000;
			printf("Timecounter \"%s\" frequency %lu Hz",
			    tc->tc_name, (unsigned long)tc->tc_frequency);
			printf(" -- Insufficient hz, needs at least %u\n", u);
		}
	}

	tc->tc_next = timecounters;
	timecounters = tc;
	/*
	 * Never automatically use a timecounter with negative quality.
	 * Even though we run on the dummy counter, switching here may be
	 * worse since this timecounter may not be monotonic.
	 */
	if (tc->tc_quality < 0)
		return;
	if (tc->tc_quality < timecounter->tc_quality)
		return;
	if (tc->tc_quality == timecounter->tc_quality &&
	    tc->tc_frequency < timecounter->tc_frequency)
		return;
	(void)tc->tc_get_timecount(tc);
	add_timer_randomness(tc->tc_get_timecount(tc));

	timecounter = tc;
}

/* Report the frequency of the current timecounter. */
u_int64_t
tc_getfrequency(void)
{

	return (timehands->th_counter->tc_frequency);
}

/*
 * Step our concept of UTC, aka the realtime clock.
 * This is done by modifying our estimate of when we booted.
 * XXX: not locked.
 */
void
tc_setrealtimeclock(struct timespec *ts)
{
	struct timespec ts2;
	struct bintime bt, bt2;

	binuptime(&bt2);
	timespec2bintime(ts, &bt);
	bintime_sub(&bt, &bt2);
	bintime_add(&bt2, &boottimebin);
	boottimebin = bt;
	bintime2timespec(&bt, &boottime);
	add_timer_randomness(ts->tv_sec);

	/* XXX fiddle all the little crinkly bits around the fiords... */
	tc_windup();
	if (timestepwarnings) {
		bintime2timespec(&bt2, &ts2);
		log(LOG_INFO, "Time stepped from %lld.%09ld to %lld.%09ld\n",
		    (long long)ts2.tv_sec, ts2.tv_nsec,
		    (long long)ts->tv_sec, ts->tv_nsec);
	}
}

/*
 * Step the monotonic and realtime clocks, triggering any timeouts that
 * should have occurred across the interval.
 * XXX: not locked.
 */
void
tc_setclock(struct timespec *ts)
{
	struct bintime bt, bt2;
#ifndef SMALL_KERNEL
	long long adj_ticks;
#endif

	/*
	 * When we're called for the first time, during boot when
	 * the root partition is mounted, boottime is still zero:
	 * we just need to set it.
	 */
	if (boottimebin.sec == 0) {
		tc_setrealtimeclock(ts);
		return;
	}

	add_timer_randomness(ts->tv_sec);

	timespec2bintime(ts, &bt);
	bintime_sub(&bt, &boottimebin);
	bt2 = timehands->th_offset;
	timehands->th_offset = bt;

	/* XXX fiddle all the little crinkly bits around the fiords... */
	tc_windup();

#ifndef SMALL_KERNEL
	/* convert the bintime to ticks */
	bintime_sub(&bt, &bt2);
	bintime_add(&naptime, &bt);
	adj_ticks = (uint64_t)hz * bt.sec +
	    (((uint64_t)1000000 * (uint32_t)(bt.frac >> 32)) >> 32) / tick;
	if (adj_ticks > 0) {
		if (adj_ticks > INT_MAX)
			adj_ticks = INT_MAX;
		timeout_adjust_ticks(adj_ticks);
	}
#endif
}

/*
 * Initialize the next struct timehands in the ring and make
 * it the active timehands.  Along the way we might switch to a different
 * timecounter and/or do seconds processing in NTP.  Slightly magic.
 */
void
tc_windup(void)
{
	struct bintime bt;
	struct timehands *th, *tho;
	u_int64_t scale;
	u_int delta, ncount, ogen;
	int i;

	/*
	 * Make the next timehands a copy of the current one, but do not
	 * overwrite the generation or next pointer.  While we update
	 * the contents, the generation must be zero.
	 */
	tho = timehands;
	th = tho->th_next;
	ogen = th->th_generation;
	th->th_generation = 0;
	memcpy(th, tho, offsetof(struct timehands, th_generation));

	/*
	 * Capture a timecounter delta on the current timecounter and if
	 * changing timecounters, a counter value from the new timecounter.
	 * Update the offset fields accordingly.
	 */
	delta = tc_delta(th);
	if (th->th_counter != timecounter)
		ncount = timecounter->tc_get_timecount(timecounter);
	else
		ncount = 0;
	th->th_offset_count += delta;
	th->th_offset_count &= th->th_counter->tc_counter_mask;
	bintime_addx(&th->th_offset, th->th_scale * delta);

#ifdef notyet
	/*
	 * Hardware latching timecounters may not generate interrupts on
	 * PPS events, so instead we poll them.  There is a finite risk that
	 * the hardware might capture a count which is later than the one we
	 * got above, and therefore possibly in the next NTP second which might
	 * have a different rate than the current NTP second.  It doesn't
	 * matter in practice.
	 */
	if (tho->th_counter->tc_poll_pps)
		tho->th_counter->tc_poll_pps(tho->th_counter);
#endif

	/*
	 * Deal with NTP second processing.  The for loop normally
	 * iterates at most once, but in extreme situations it might
	 * keep NTP sane if timeouts are not run for several seconds.
	 * At boot, the time step can be large when the TOD hardware
	 * has been read, so on really large steps, we call
	 * ntp_update_second only twice.  We need to call it twice in
	 * case we missed a leap second.
	 */
	bt = th->th_offset;
	bintime_add(&bt, &boottimebin);
	i = bt.sec - tho->th_microtime.tv_sec;
	if (i > LARGE_STEP)
		i = 2;
	for (; i > 0; i--)
		ntp_update_second(&th->th_adjustment);

	/* Update the UTC timestamps used by the get*() functions. */
	/* XXX shouldn't do this here.  Should force non-`get' versions. */
	bintime2timeval(&bt, &th->th_microtime);
	bintime2timespec(&bt, &th->th_nanotime);

	/* Now is a good time to change timecounters. */
	if (th->th_counter != timecounter) {
		th->th_counter = timecounter;
		th->th_offset_count = ncount;
	}

	/*-
	 * Recalculate the scaling factor.  We want the number of 1/2^64
	 * fractions of a second per period of the hardware counter, taking
	 * into account the th_adjustment factor which the NTP PLL/adjtime(2)
	 * processing provides us with.
	 *
	 * The th_adjustment is nanoseconds per second with 32 bit binary
	 * fraction and we want 64 bit binary fraction of second:
	 *
	 *	 x = a * 2^32 / 10^9 = a * 4.294967296
	 *
	 * The range of th_adjustment is +/- 5000PPM so inside a 64bit int
	 * we can only multiply by about 850 without overflowing, but that
	 * leaves suitably precise fractions for multiply before divide.
	 *
	 * Divide before multiply with a fraction of 2199/512 results in a
	 * systematic undercompensation of 10PPM of th_adjustment.  On a
	 * 5000PPM adjustment this is a 0.05PPM error.  This is acceptable.
 	 *
	 * We happily sacrifice the lowest of the 64 bits of our result
	 * to the goddess of code clarity.
	 *
	 */
	scale = (u_int64_t)1 << 63;
	scale += (th->th_adjustment / 1024) * 2199;
	scale /= th->th_counter->tc_frequency;
	th->th_scale = scale * 2;

	/*
	 * Now that the struct timehands is again consistent, set the new
	 * generation number, making sure to not make it zero.
	 */
	if (++ogen == 0)
		ogen = 1;
	th->th_generation = ogen;

	/* Go live with the new struct timehands. */
	time_second = th->th_microtime.tv_sec;
	time_uptime = th->th_offset.sec;
	timehands = th;
}

/* Report or change the active timecounter hardware. */
int
sysctl_tc_hardware(void *oldp, size_t *oldlenp, void *newp, size_t newlen)
{
	char newname[32];
	struct timecounter *newtc, *tc;
	int error;

	tc = timecounter;
	strlcpy(newname, tc->tc_name, sizeof(newname));

	error = sysctl_string(oldp, oldlenp, newp, newlen, newname, sizeof(newname));
	if (error != 0 || strcmp(newname, tc->tc_name) == 0)
		return (error);
	for (newtc = timecounters; newtc != NULL; newtc = newtc->tc_next) {
		if (strcmp(newname, newtc->tc_name) != 0)
			continue;

		/* Warm up new timecounter. */
		(void)newtc->tc_get_timecount(newtc);
		(void)newtc->tc_get_timecount(newtc);

		timecounter = newtc;
		return (0);
	}
	return (EINVAL);
}

/* Report or change the active timecounter hardware. */
int
sysctl_tc_choice(void *oldp, size_t *oldlenp, void *newp, size_t newlen)
{
	char buf[32], *spc, *choices;
	struct timecounter *tc;
	int error, maxlen;

	spc = "";
	maxlen = 0;
	for (tc = timecounters; tc != NULL; tc = tc->tc_next)
		maxlen += sizeof(buf);
	choices = malloc(maxlen, M_TEMP, M_WAITOK);
	*choices = '\0';
	for (tc = timecounters; tc != NULL; tc = tc->tc_next) {
		snprintf(buf, sizeof(buf), "%s%s(%d)",
		    spc, tc->tc_name, tc->tc_quality);
		spc = " ";
		strlcat(choices, buf, maxlen);
	}
	error = sysctl_rdstring(oldp, oldlenp, newp, choices);
	free(choices, M_TEMP, maxlen);
	return (error);
}

/*
 * Timecounters need to be updated every so often to prevent the hardware
 * counter from overflowing.  Updating also recalculates the cached values
 * used by the get*() family of functions, so their precision depends on
 * the update frequency.
 */
static int tc_tick;

void
tc_ticktock(void)
{
	static int count;

	if (++count < tc_tick)
		return;
	count = 0;
	tc_windup();
}

void
inittimecounter(void)
{
#ifdef DEBUG
	u_int p;
#endif

	/*
	 * Set the initial timeout to
	 * max(1, <approx. number of hardclock ticks in a millisecond>).
	 * People should probably not use the sysctl to set the timeout
	 * to smaller than its initial value, since that value is the
	 * smallest reasonable one.  If they want better timestamps they
	 * should use the non-"get"* functions.
	 */
	if (hz > 1000)
		tc_tick = (hz + 500) / 1000;
	else
		tc_tick = 1;
#ifdef DEBUG
	p = (tc_tick * 1000000) / hz;
	printf("Timecounters tick every %d.%03u msec\n", p / 1000, p % 1000);
#endif

	/* warm up new timecounter (again) and get rolling. */
	(void)timecounter->tc_get_timecount(timecounter);
	(void)timecounter->tc_get_timecount(timecounter);
}

/*
 * Return timecounter-related information.
 */
int
sysctl_tc(int *name, u_int namelen, void *oldp, size_t *oldlenp,
    void *newp, size_t newlen)
{
	if (namelen != 1)
		return (ENOTDIR);

	switch (name[0]) {
	case KERN_TIMECOUNTER_TICK:
		return (sysctl_rdint(oldp, oldlenp, newp, tc_tick));
	case KERN_TIMECOUNTER_TIMESTEPWARNINGS:
		return (sysctl_int(oldp, oldlenp, newp, newlen,
		    &timestepwarnings));
	case KERN_TIMECOUNTER_HARDWARE:
		return (sysctl_tc_hardware(oldp, oldlenp, newp, newlen));
	case KERN_TIMECOUNTER_CHOICE:
		return (sysctl_tc_choice(oldp, oldlenp, newp, newlen));
	default:
		return (EOPNOTSUPP);
	}
	/* NOTREACHED */
}

void
ntp_update_second(int64_t *adjust)
{
	int64_t adj;

	/* Skew time according to any adjtime(2) adjustments. */
	if (adjtimedelta > 0)
		adj = MIN(5000, adjtimedelta);
	else
		adj = MAX(-5000, adjtimedelta);
	adjtimedelta -= adj;
	*adjust = (adj * 1000) << 32;
	*adjust += timecounter->tc_freq_adj;
}

int
tc_adjfreq(int64_t *old, int64_t *new)
{
	if (old != NULL) {
		*old = timecounter->tc_freq_adj;
	}
	if (new != NULL) {
		timecounter->tc_freq_adj = *new;
	}
	return 0;
}