summaryrefslogtreecommitdiffstats
path: root/arch/ia64/kernel/time.c
blob: fbaac1afb8441926d3f2fdc9eec4ed85c6122da1 (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
/*
 * linux/arch/ia64/kernel/time.c
 *
 * Copyright (C) 1998-2003 Hewlett-Packard Co
 *	Stephane Eranian <eranian@hpl.hp.com>
 *	David Mosberger <davidm@hpl.hp.com>
 * Copyright (C) 1999 Don Dugger <don.dugger@intel.com>
 * Copyright (C) 1999-2000 VA Linux Systems
 * Copyright (C) 1999-2000 Walt Drummond <drummond@valinux.com>
 */

#include <linux/cpu.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/profile.h>
#include <linux/sched.h>
#include <linux/time.h>
#include <linux/interrupt.h>
#include <linux/efi.h>
#include <linux/timex.h>
#include <linux/timekeeper_internal.h>
#include <linux/platform_device.h>

#include <asm/machvec.h>
#include <asm/delay.h>
#include <asm/hw_irq.h>
#include <asm/paravirt.h>
#include <asm/ptrace.h>
#include <asm/sal.h>
#include <asm/sections.h>

#include "fsyscall_gtod_data.h"

static cycle_t itc_get_cycles(struct clocksource *cs);

struct fsyscall_gtod_data_t fsyscall_gtod_data;

struct itc_jitter_data_t itc_jitter_data;

volatile int time_keeper_id = 0; /* smp_processor_id() of time-keeper */

#ifdef CONFIG_IA64_DEBUG_IRQ

unsigned long last_cli_ip;
EXPORT_SYMBOL(last_cli_ip);

#endif

#ifdef CONFIG_PARAVIRT
/* We need to define a real function for sched_clock, to override the
   weak default version */
unsigned long long sched_clock(void)
{
        return paravirt_sched_clock();
}
#endif

#ifdef CONFIG_PARAVIRT
static void
paravirt_clocksource_resume(struct clocksource *cs)
{
	if (pv_time_ops.clocksource_resume)
		pv_time_ops.clocksource_resume();
}
#endif

static struct clocksource clocksource_itc = {
	.name           = "itc",
	.rating         = 350,
	.read           = itc_get_cycles,
	.mask           = CLOCKSOURCE_MASK(64),
	.flags          = CLOCK_SOURCE_IS_CONTINUOUS,
#ifdef CONFIG_PARAVIRT
	.resume		= paravirt_clocksource_resume,
#endif
};
static struct clocksource *itc_clocksource;

#ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE

#include <linux/kernel_stat.h>

extern cputime_t cycle_to_cputime(u64 cyc);

void vtime_account_user(struct task_struct *tsk)
{
	cputime_t delta_utime;
	struct thread_info *ti = task_thread_info(tsk);

	if (ti->ac_utime) {
		delta_utime = cycle_to_cputime(ti->ac_utime);
		account_user_time(tsk, delta_utime, delta_utime);
		ti->ac_utime = 0;
	}
}

/*
 * Called from the context switch with interrupts disabled, to charge all
 * accumulated times to the current process, and to prepare accounting on
 * the next process.
 */
void arch_vtime_task_switch(struct task_struct *prev)
{
	struct thread_info *pi = task_thread_info(prev);
	struct thread_info *ni = task_thread_info(current);

	pi->ac_stamp = ni->ac_stamp;
	ni->ac_stime = ni->ac_utime = 0;
}

/*
 * Account time for a transition between system, hard irq or soft irq state.
 * Note that this function is called with interrupts enabled.
 */
static cputime_t vtime_delta(struct task_struct *tsk)
{
	struct thread_info *ti = task_thread_info(tsk);
	cputime_t delta_stime;
	__u64 now;

	WARN_ON_ONCE(!irqs_disabled());

	now = ia64_get_itc();

	delta_stime = cycle_to_cputime(ti->ac_stime + (now - ti->ac_stamp));
	ti->ac_stime = 0;
	ti->ac_stamp = now;

	return delta_stime;
}

void vtime_account_system(struct task_struct *tsk)
{
	cputime_t delta = vtime_delta(tsk);

	account_system_time(tsk, 0, delta, delta);
}
EXPORT_SYMBOL_GPL(vtime_account_system);

void vtime_account_idle(struct task_struct *tsk)
{
	account_idle_time(vtime_delta(tsk));
}

#endif /* CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */

static irqreturn_t
timer_interrupt (int irq, void *dev_id)
{
	unsigned long new_itm;

	if (cpu_is_offline(smp_processor_id())) {
		return IRQ_HANDLED;
	}

	platform_timer_interrupt(irq, dev_id);

	new_itm = local_cpu_data->itm_next;

	if (!time_after(ia64_get_itc(), new_itm))
		printk(KERN_ERR "Oops: timer tick before it's due (itc=%lx,itm=%lx)\n",
		       ia64_get_itc(), new_itm);

	profile_tick(CPU_PROFILING);

	if (paravirt_do_steal_accounting(&new_itm))
		goto skip_process_time_accounting;

	while (1) {
		update_process_times(user_mode(get_irq_regs()));

		new_itm += local_cpu_data->itm_delta;

		if (smp_processor_id() == time_keeper_id)
			xtime_update(1);

		local_cpu_data->itm_next = new_itm;

		if (time_after(new_itm, ia64_get_itc()))
			break;

		/*
		 * Allow IPIs to interrupt the timer loop.
		 */
		local_irq_enable();
		local_irq_disable();
	}

skip_process_time_accounting:

	do {
		/*
		 * If we're too close to the next clock tick for
		 * comfort, we increase the safety margin by
		 * intentionally dropping the next tick(s).  We do NOT
		 * update itm.next because that would force us to call
		 * xtime_update() which in turn would let our clock run
		 * too fast (with the potentially devastating effect
		 * of losing monotony of time).
		 */
		while (!time_after(new_itm, ia64_get_itc() + local_cpu_data->itm_delta/2))
			new_itm += local_cpu_data->itm_delta;
		ia64_set_itm(new_itm);
		/* double check, in case we got hit by a (slow) PMI: */
	} while (time_after_eq(ia64_get_itc(), new_itm));
	return IRQ_HANDLED;
}

/*
 * Encapsulate access to the itm structure for SMP.
 */
void
ia64_cpu_local_tick (void)
{
	int cpu = smp_processor_id();
	unsigned long shift = 0, delta;

	/* arrange for the cycle counter to generate a timer interrupt: */
	ia64_set_itv(IA64_TIMER_VECTOR);

	delta = local_cpu_data->itm_delta;
	/*
	 * Stagger the timer tick for each CPU so they don't occur all at (almost) the
	 * same time:
	 */
	if (cpu) {
		unsigned long hi = 1UL << ia64_fls(cpu);
		shift = (2*(cpu - hi) + 1) * delta/hi/2;
	}
	local_cpu_data->itm_next = ia64_get_itc() + delta + shift;
	ia64_set_itm(local_cpu_data->itm_next);
}

static int nojitter;

static int __init nojitter_setup(char *str)
{
	nojitter = 1;
	printk("Jitter checking for ITC timers disabled\n");
	return 1;
}

__setup("nojitter", nojitter_setup);


void ia64_init_itm(void)
{
	unsigned long platform_base_freq, itc_freq;
	struct pal_freq_ratio itc_ratio, proc_ratio;
	long status, platform_base_drift, itc_drift;

	/*
	 * According to SAL v2.6, we need to use a SAL call to determine the platform base
	 * frequency and then a PAL call to determine the frequency ratio between the ITC
	 * and the base frequency.
	 */
	status = ia64_sal_freq_base(SAL_FREQ_BASE_PLATFORM,
				    &platform_base_freq, &platform_base_drift);
	if (status != 0) {
		printk(KERN_ERR "SAL_FREQ_BASE_PLATFORM failed: %s\n", ia64_sal_strerror(status));
	} else {
		status = ia64_pal_freq_ratios(&proc_ratio, NULL, &itc_ratio);
		if (status != 0)
			printk(KERN_ERR "PAL_FREQ_RATIOS failed with status=%ld\n", status);
	}
	if (status != 0) {
		/* invent "random" values */
		printk(KERN_ERR
		       "SAL/PAL failed to obtain frequency info---inventing reasonable values\n");
		platform_base_freq = 100000000;
		platform_base_drift = -1;	/* no drift info */
		itc_ratio.num = 3;
		itc_ratio.den = 1;
	}
	if (platform_base_freq < 40000000) {
		printk(KERN_ERR "Platform base frequency %lu bogus---resetting to 75MHz!\n",
		       platform_base_freq);
		platform_base_freq = 75000000;
		platform_base_drift = -1;
	}
	if (!proc_ratio.den)
		proc_ratio.den = 1;	/* avoid division by zero */
	if (!itc_ratio.den)
		itc_ratio.den = 1;	/* avoid division by zero */

	itc_freq = (platform_base_freq*itc_ratio.num)/itc_ratio.den;

	local_cpu_data->itm_delta = (itc_freq + HZ/2) / HZ;
	printk(KERN_DEBUG "CPU %d: base freq=%lu.%03luMHz, ITC ratio=%u/%u, "
	       "ITC freq=%lu.%03luMHz", smp_processor_id(),
	       platform_base_freq / 1000000, (platform_base_freq / 1000) % 1000,
	       itc_ratio.num, itc_ratio.den, itc_freq / 1000000, (itc_freq / 1000) % 1000);

	if (platform_base_drift != -1) {
		itc_drift = platform_base_drift*itc_ratio.num/itc_ratio.den;
		printk("+/-%ldppm\n", itc_drift);
	} else {
		itc_drift = -1;
		printk("\n");
	}

	local_cpu_data->proc_freq = (platform_base_freq*proc_ratio.num)/proc_ratio.den;
	local_cpu_data->itc_freq = itc_freq;
	local_cpu_data->cyc_per_usec = (itc_freq + USEC_PER_SEC/2) / USEC_PER_SEC;
	local_cpu_data->nsec_per_cyc = ((NSEC_PER_SEC<<IA64_NSEC_PER_CYC_SHIFT)
					+ itc_freq/2)/itc_freq;

	if (!(sal_platform_features & IA64_SAL_PLATFORM_FEATURE_ITC_DRIFT)) {
#ifdef CONFIG_SMP
		/* On IA64 in an SMP configuration ITCs are never accurately synchronized.
		 * Jitter compensation requires a cmpxchg which may limit
		 * the scalability of the syscalls for retrieving time.
		 * The ITC synchronization is usually successful to within a few
		 * ITC ticks but this is not a sure thing. If you need to improve
		 * timer performance in SMP situations then boot the kernel with the
		 * "nojitter" option. However, doing so may result in time fluctuating (maybe
		 * even going backward) if the ITC offsets between the individual CPUs
		 * are too large.
		 */
		if (!nojitter)
			itc_jitter_data.itc_jitter = 1;
#endif
	} else
		/*
		 * ITC is drifty and we have not synchronized the ITCs in smpboot.c.
		 * ITC values may fluctuate significantly between processors.
		 * Clock should not be used for hrtimers. Mark itc as only
		 * useful for boot and testing.
		 *
		 * Note that jitter compensation is off! There is no point of
		 * synchronizing ITCs since they may be large differentials
		 * that change over time.
		 *
		 * The only way to fix this would be to repeatedly sync the
		 * ITCs. Until that time we have to avoid ITC.
		 */
		clocksource_itc.rating = 50;

	paravirt_init_missing_ticks_accounting(smp_processor_id());

	/* avoid softlock up message when cpu is unplug and plugged again. */
	touch_softlockup_watchdog();

	/* Setup the CPU local timer tick */
	ia64_cpu_local_tick();

	if (!itc_clocksource) {
		clocksource_register_hz(&clocksource_itc,
						local_cpu_data->itc_freq);
		itc_clocksource = &clocksource_itc;
	}
}

static cycle_t itc_get_cycles(struct clocksource *cs)
{
	unsigned long lcycle, now, ret;

	if (!itc_jitter_data.itc_jitter)
		return get_cycles();

	lcycle = itc_jitter_data.itc_lastcycle;
	now = get_cycles();
	if (lcycle && time_after(lcycle, now))
		return lcycle;

	/*
	 * Keep track of the last timer value returned.
	 * In an SMP environment, you could lose out in contention of
	 * cmpxchg. If so, your cmpxchg returns new value which the
	 * winner of contention updated to. Use the new value instead.
	 */
	ret = cmpxchg(&itc_jitter_data.itc_lastcycle, lcycle, now);
	if (unlikely(ret != lcycle))
		return ret;

	return now;
}


static struct irqaction timer_irqaction = {
	.handler =	timer_interrupt,
	.flags =	IRQF_DISABLED | IRQF_IRQPOLL,
	.name =		"timer"
};

static struct platform_device rtc_efi_dev = {
	.name = "rtc-efi",
	.id = -1,
};

static int __init rtc_init(void)
{
	if (platform_device_register(&rtc_efi_dev) < 0)
		printk(KERN_ERR "unable to register rtc device...\n");

	/* not necessarily an error */
	return 0;
}
module_init(rtc_init);

void read_persistent_clock(struct timespec *ts)
{
	efi_gettimeofday(ts);
}

void __init
time_init (void)
{
	register_percpu_irq(IA64_TIMER_VECTOR, &timer_irqaction);
	ia64_init_itm();
}

/*
 * Generic udelay assumes that if preemption is allowed and the thread
 * migrates to another CPU, that the ITC values are synchronized across
 * all CPUs.
 */
static void
ia64_itc_udelay (unsigned long usecs)
{
	unsigned long start = ia64_get_itc();
	unsigned long end = start + usecs*local_cpu_data->cyc_per_usec;

	while (time_before(ia64_get_itc(), end))
		cpu_relax();
}

void (*ia64_udelay)(unsigned long usecs) = &ia64_itc_udelay;

void
udelay (unsigned long usecs)
{
	(*ia64_udelay)(usecs);
}
EXPORT_SYMBOL(udelay);

/* IA64 doesn't cache the timezone */
void update_vsyscall_tz(void)
{
}

void update_vsyscall_old(struct timespec *wall, struct timespec *wtm,
			struct clocksource *c, u32 mult)
{
	write_seqcount_begin(&fsyscall_gtod_data.seq);

        /* copy fsyscall clock data */
        fsyscall_gtod_data.clk_mask = c->mask;
        fsyscall_gtod_data.clk_mult = mult;
        fsyscall_gtod_data.clk_shift = c->shift;
        fsyscall_gtod_data.clk_fsys_mmio = c->archdata.fsys_mmio;
        fsyscall_gtod_data.clk_cycle_last = c->cycle_last;

	/* copy kernel time structures */
        fsyscall_gtod_data.wall_time.tv_sec = wall->tv_sec;
        fsyscall_gtod_data.wall_time.tv_nsec = wall->tv_nsec;
	fsyscall_gtod_data.monotonic_time.tv_sec = wtm->tv_sec
							+ wall->tv_sec;
	fsyscall_gtod_data.monotonic_time.tv_nsec = wtm->tv_nsec
							+ wall->tv_nsec;

	/* normalize */
	while (fsyscall_gtod_data.monotonic_time.tv_nsec >= NSEC_PER_SEC) {
		fsyscall_gtod_data.monotonic_time.tv_nsec -= NSEC_PER_SEC;
		fsyscall_gtod_data.monotonic_time.tv_sec++;
	}

	write_seqcount_end(&fsyscall_gtod_data.seq);
}