time.c 9.7 KB
Newer Older
L
Linus Torvalds 已提交
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
/*
 *  arch/s390/kernel/time.c
 *    Time of day based timer functions.
 *
 *  S390 version
 *    Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation
 *    Author(s): Hartmut Penner (hp@de.ibm.com),
 *               Martin Schwidefsky (schwidefsky@de.ibm.com),
 *               Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com)
 *
 *  Derived from "arch/i386/kernel/time.c"
 *    Copyright (C) 1991, 1992, 1995  Linus Torvalds
 */

#include <linux/config.h>
#include <linux/errno.h>
#include <linux/module.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/param.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <linux/interrupt.h>
#include <linux/time.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/smp.h>
#include <linux/types.h>
#include <linux/profile.h>
#include <linux/timex.h>
#include <linux/notifier.h>

#include <asm/uaccess.h>
#include <asm/delay.h>
#include <asm/s390_ext.h>
#include <asm/div64.h>
#include <asm/irq.h>
#include <asm/timer.h>

/* change this if you have some constant time drift */
#define USECS_PER_JIFFY     ((unsigned long) 1000000/HZ)
#define CLK_TICKS_PER_JIFFY ((unsigned long) USECS_PER_JIFFY << 12)

/*
 * Create a small time difference between the timer interrupts
 * on the different cpus to avoid lock contention.
 */
#define CPU_DEVIATION       (smp_processor_id() << 12)

#define TICK_SIZE tick

static ext_int_info_t ext_int_info_cc;
static u64 init_timer_cc;
static u64 jiffies_timer_cc;
static u64 xtime_cc;

extern unsigned long wall_jiffies;

/*
 * Scheduler clock - returns current time in nanosec units.
 */
unsigned long long sched_clock(void)
{
64
	return ((get_clock() - jiffies_timer_cc) * 125) >> 9;
L
Linus Torvalds 已提交
65 66
}

67 68 69 70 71 72 73 74 75
/*
 * Monotonic_clock - returns # of nanoseconds passed since time_init()
 */
unsigned long long monotonic_clock(void)
{
	return sched_clock();
}
EXPORT_SYMBOL(monotonic_clock);

L
Linus Torvalds 已提交
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
void tod_to_timeval(__u64 todval, struct timespec *xtime)
{
	unsigned long long sec;

	sec = todval >> 12;
	do_div(sec, 1000000);
	xtime->tv_sec = sec;
	todval -= (sec * 1000000) << 12;
	xtime->tv_nsec = ((todval * 1000) >> 12);
}

static inline unsigned long do_gettimeoffset(void) 
{
	__u64 now;

        now = (get_clock() - jiffies_timer_cc) >> 12;
	/* We require the offset from the latest update of xtime */
	now -= (__u64) wall_jiffies*USECS_PER_JIFFY;
	return (unsigned long) now;
}

/*
 * This version of gettimeofday has microsecond resolution.
 */
void do_gettimeofday(struct timeval *tv)
{
	unsigned long flags;
	unsigned long seq;
	unsigned long usec, sec;

	do {
		seq = read_seqbegin_irqsave(&xtime_lock, flags);

		sec = xtime.tv_sec;
		usec = xtime.tv_nsec / 1000 + do_gettimeoffset();
	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));

	while (usec >= 1000000) {
		usec -= 1000000;
		sec++;
	}

	tv->tv_sec = sec;
	tv->tv_usec = usec;
}

EXPORT_SYMBOL(do_gettimeofday);

int do_settimeofday(struct timespec *tv)
{
	time_t wtm_sec, sec = tv->tv_sec;
	long wtm_nsec, nsec = tv->tv_nsec;

	if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
		return -EINVAL;

	write_seqlock_irq(&xtime_lock);
	/* This is revolting. We need to set the xtime.tv_nsec
	 * correctly. However, the value in this location is
	 * is value at the last tick.
	 * Discover what correction gettimeofday
	 * would have done, and then undo it!
	 */
	nsec -= do_gettimeoffset() * 1000;

	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);

	set_normalized_timespec(&xtime, sec, nsec);
	set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec);

J
john stultz 已提交
147
	ntp_clear();
L
Linus Torvalds 已提交
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
	write_sequnlock_irq(&xtime_lock);
	clock_was_set();
	return 0;
}

EXPORT_SYMBOL(do_settimeofday);


#ifdef CONFIG_PROFILING
#define s390_do_profile(regs)	profile_tick(CPU_PROFILING, regs)
#else
#define s390_do_profile(regs)  do { ; } while(0)
#endif /* CONFIG_PROFILING */


/*
 * timer_interrupt() needs to keep up the real-time clock,
 * as well as call the "do_timer()" routine every clocktick
 */
void account_ticks(struct pt_regs *regs)
{
	__u64 tmp;
	__u32 ticks, xticks;

	/* Calculate how many ticks have passed. */
	if (S390_lowcore.int_clock < S390_lowcore.jiffy_timer) {
		/*
		 * We have to program the clock comparator even if
		 * no tick has passed. That happens if e.g. an i/o
		 * interrupt wakes up an idle processor that has
		 * switched off its hz timer.
		 */
		tmp = S390_lowcore.jiffy_timer + CPU_DEVIATION;
		asm volatile ("SCKC %0" : : "m" (tmp));
		return;
	}
	tmp = S390_lowcore.int_clock - S390_lowcore.jiffy_timer;
	if (tmp >= 2*CLK_TICKS_PER_JIFFY) {  /* more than two ticks ? */
		ticks = __div(tmp, CLK_TICKS_PER_JIFFY) + 1;
		S390_lowcore.jiffy_timer +=
			CLK_TICKS_PER_JIFFY * (__u64) ticks;
	} else if (tmp >= CLK_TICKS_PER_JIFFY) {
		ticks = 2;
		S390_lowcore.jiffy_timer += 2*CLK_TICKS_PER_JIFFY;
	} else {
		ticks = 1;
		S390_lowcore.jiffy_timer += CLK_TICKS_PER_JIFFY;
	}

	/* set clock comparator for next tick */
	tmp = S390_lowcore.jiffy_timer + CPU_DEVIATION;
        asm volatile ("SCKC %0" : : "m" (tmp));

#ifdef CONFIG_SMP
	/*
	 * Do not rely on the boot cpu to do the calls to do_timer.
	 * Spread it over all cpus instead.
	 */
	write_seqlock(&xtime_lock);
	if (S390_lowcore.jiffy_timer > xtime_cc) {
		tmp = S390_lowcore.jiffy_timer - xtime_cc;
		if (tmp >= 2*CLK_TICKS_PER_JIFFY) {
			xticks = __div(tmp, CLK_TICKS_PER_JIFFY);
			xtime_cc += (__u64) xticks * CLK_TICKS_PER_JIFFY;
		} else {
			xticks = 1;
			xtime_cc += CLK_TICKS_PER_JIFFY;
		}
		while (xticks--)
			do_timer(regs);
	}
	write_sequnlock(&xtime_lock);
#else
	for (xticks = ticks; xticks > 0; xticks--)
		do_timer(regs);
#endif

#ifdef CONFIG_VIRT_CPU_ACCOUNTING
226
	account_tick_vtime(current);
L
Linus Torvalds 已提交
227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
#else
	while (ticks--)
		update_process_times(user_mode(regs));
#endif

	s390_do_profile(regs);
}

#ifdef CONFIG_NO_IDLE_HZ

#ifdef CONFIG_NO_IDLE_HZ_INIT
int sysctl_hz_timer = 0;
#else
int sysctl_hz_timer = 1;
#endif

/*
 * Stop the HZ tick on the current CPU.
 * Only cpu_idle may call this function.
 */
static inline void stop_hz_timer(void)
{
249 250
	unsigned long flags;
	unsigned long seq, next;
251
	__u64 timer, todval;
L
Linus Torvalds 已提交
252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270

	if (sysctl_hz_timer != 0)
		return;

	cpu_set(smp_processor_id(), nohz_cpu_mask);

	/*
	 * Leave the clock comparator set up for the next timer
	 * tick if either rcu or a softirq is pending.
	 */
	if (rcu_pending(smp_processor_id()) || local_softirq_pending()) {
		cpu_clear(smp_processor_id(), nohz_cpu_mask);
		return;
	}

	/*
	 * This cpu is going really idle. Set up the clock comparator
	 * for the next event.
	 */
271 272 273 274 275
	next = next_timer_interrupt();
	do {
		seq = read_seqbegin_irqsave(&xtime_lock, flags);
		timer = (__u64)(next - jiffies) + jiffies_64;
	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
276 277 278 279 280 281 282 283
	todval = -1ULL;
	/* Be careful about overflows. */
	if (timer < (-1ULL / CLK_TICKS_PER_JIFFY)) {
		timer = jiffies_timer_cc + timer * CLK_TICKS_PER_JIFFY;
		if (timer >= jiffies_timer_cc)
			todval = timer;
	}
	asm volatile ("SCKC %0" : : "m" (todval));
L
Linus Torvalds 已提交
284 285 286 287 288 289 290 291 292 293
}

/*
 * Start the HZ tick on the current CPU.
 * Only cpu_idle may call this function.
 */
static inline void start_hz_timer(void)
{
	if (!cpu_isset(smp_processor_id(), nohz_cpu_mask))
		return;
A
Al Viro 已提交
294
	account_ticks(task_pt_regs(current));
L
Linus Torvalds 已提交
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
	cpu_clear(smp_processor_id(), nohz_cpu_mask);
}

static int nohz_idle_notify(struct notifier_block *self,
			    unsigned long action, void *hcpu)
{
	switch (action) {
	case CPU_IDLE:
		stop_hz_timer();
		break;
	case CPU_NOT_IDLE:
		start_hz_timer();
		break;
	}
	return NOTIFY_OK;
}

static struct notifier_block nohz_idle_nb = {
	.notifier_call = nohz_idle_notify,
};

void __init nohz_init(void)
{
	if (register_idle_notifier(&nohz_idle_nb))
		panic("Couldn't register idle notifier");
}

#endif

/*
 * Start the clock comparator on the current CPU.
 */
void init_cpu_timer(void)
{
	unsigned long cr0;
	__u64 timer;

	timer = jiffies_timer_cc + jiffies_64 * CLK_TICKS_PER_JIFFY;
	S390_lowcore.jiffy_timer = timer + CLK_TICKS_PER_JIFFY;
	timer += CLK_TICKS_PER_JIFFY + CPU_DEVIATION;
	asm volatile ("SCKC %0" : : "m" (timer));
        /* allow clock comparator timer interrupt */
	__ctl_store(cr0, 0, 0);
        cr0 |= 0x800;
	__ctl_load(cr0, 0, 0);
}

extern void vtime_init(void);

/*
 * Initialize the TOD clock and the CPU timer of
 * the boot cpu.
 */
void __init time_init(void)
{
	__u64 set_time_cc;
	int cc;

        /* kick the TOD clock */
        asm volatile ("STCK 0(%1)\n\t"
                      "IPM  %0\n\t"
                      "SRL  %0,28" : "=r" (cc) : "a" (&init_timer_cc) 
				   : "memory", "cc");
        switch (cc) {
        case 0: /* clock in set state: all is fine */
                break;
        case 1: /* clock in non-set state: FIXME */
                printk("time_init: TOD clock in non-set state\n");
                break;
        case 2: /* clock in error state: FIXME */
                printk("time_init: TOD clock in error state\n");
                break;
        case 3: /* clock in stopped or not-operational state: FIXME */
                printk("time_init: TOD clock stopped/non-operational\n");
                break;
        }
	jiffies_timer_cc = init_timer_cc - jiffies_64 * CLK_TICKS_PER_JIFFY;

	/* set xtime */
	xtime_cc = init_timer_cc + CLK_TICKS_PER_JIFFY;
	set_time_cc = init_timer_cc - 0x8126d60e46000000LL +
		(0x3c26700LL*1000000*4096);
        tod_to_timeval(set_time_cc, &xtime);
        set_normalized_timespec(&wall_to_monotonic,
                                -xtime.tv_sec, -xtime.tv_nsec);

	/* request the clock comparator external interrupt */
        if (register_early_external_interrupt(0x1004, 0,
					      &ext_int_info_cc) != 0)
                panic("Couldn't request external interrupt 0x1004");

        init_cpu_timer();

#ifdef CONFIG_NO_IDLE_HZ
	nohz_init();
#endif

#ifdef CONFIG_VIRT_TIMER
	vtime_init();
#endif
}