[linux-2.6-block.git] / arch / x86 / kernel / pvclock.c

/*  paravirtual clock -- common code used by kvm/xen

    This program is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation; either version 2 of the License, or
    (at your option) any later version.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License
    along with this program; if not, write to the Free Software
    Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
*/

#include <linux/kernel.h>
#include <linux/percpu.h>
#include <linux/notifier.h>
#include <linux/sched.h>
#include <linux/gfp.h>
#include <linux/bootmem.h>
#include <linux/nmi.h>

#include <asm/fixmap.h>
#include <asm/pvclock.h>
#include <asm/vgtod.h>

static u8 valid_flags __read_mostly = 0;
static struct pvclock_vsyscall_time_info *pvti_cpu0_va __read_mostly;

void pvclock_set_flags(u8 flags)
{
	valid_flags = flags;
}

unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src)
{
	u64 pv_tsc_khz = 1000000ULL << 32;

	do_div(pv_tsc_khz, src->tsc_to_system_mul);
	if (src->tsc_shift < 0)
		pv_tsc_khz <<= -src->tsc_shift;
	else
		pv_tsc_khz >>= src->tsc_shift;
	return pv_tsc_khz;
}

void pvclock_touch_watchdogs(void)
{
	touch_softlockup_watchdog_sync();
	clocksource_touch_watchdog();
	rcu_cpu_stall_reset();
	reset_hung_task_detector();
}

static atomic64_t last_value = ATOMIC64_INIT(0);

void pvclock_resume(void)
{
	atomic64_set(&last_value, 0);
}

u8 pvclock_read_flags(struct pvclock_vcpu_time_info *src)
{
	unsigned version;
	u8 flags;

	do {
		version = pvclock_read_begin(src);
		flags = src->flags;
	} while (pvclock_read_retry(src, version));

	return flags & valid_flags;
}

u64 pvclock_clocksource_read(struct pvclock_vcpu_time_info *src)
{
	unsigned version;
	u64 ret;
	u64 last;
	u8 flags;

	do {
		version = pvclock_read_begin(src);
		ret = __pvclock_read_cycles(src, rdtsc_ordered());
		flags = src->flags;
	} while (pvclock_read_retry(src, version));

	if (unlikely((flags & PVCLOCK_GUEST_STOPPED) != 0)) {
		src->flags &= ~PVCLOCK_GUEST_STOPPED;
		pvclock_touch_watchdogs();
	}

	if ((valid_flags & PVCLOCK_TSC_STABLE_BIT) &&
		(flags & PVCLOCK_TSC_STABLE_BIT))
		return ret;

	/*
	 * Assumption here is that last_value, a global accumulator, always goes
	 * forward. If we are less than that, we should not be much smaller.
	 * We assume there is an error marging we're inside, and then the correction
	 * does not sacrifice accuracy.
	 *
	 * For reads: global may have changed between test and return,
	 * but this means someone else updated poked the clock at a later time.
	 * We just need to make sure we are not seeing a backwards event.
	 *
	 * For updates: last_value = ret is not enough, since two vcpus could be
	 * updating at the same time, and one of them could be slightly behind,
	 * making the assumption that last_value always go forward fail to hold.
	 */
	last = atomic64_read(&last_value);
	do {
		if (ret < last)
			return last;
		last = atomic64_cmpxchg(&last_value, last, ret);
	} while (unlikely(last != ret));

	return ret;
}

void pvclock_read_wallclock(struct pvclock_wall_clock *wall_clock,
			    struct pvclock_vcpu_time_info *vcpu_time,
			    struct timespec *ts)
{
	u32 version;
	u64 delta;
	struct timespec now;

	/* get wallclock at system boot */
	do {
		version = wall_clock->version;
		rmb();		/* fetch version before time */
		now.tv_sec  = wall_clock->sec;
		now.tv_nsec = wall_clock->nsec;
		rmb();		/* fetch time before checking version */
	} while ((wall_clock->version & 1) || (version != wall_clock->version));

	delta = pvclock_clocksource_read(vcpu_time);	/* time since system boot */
	delta += now.tv_sec * (u64)NSEC_PER_SEC + now.tv_nsec;

	now.tv_nsec = do_div(delta, NSEC_PER_SEC);
	now.tv_sec = delta;

	set_normalized_timespec(ts, now.tv_sec, now.tv_nsec);
}

void pvclock_set_pvti_cpu0_va(struct pvclock_vsyscall_time_info *pvti)
{
	WARN_ON(vclock_was_used(VCLOCK_PVCLOCK));
	pvti_cpu0_va = pvti;
}

struct pvclock_vsyscall_time_info *pvclock_get_pvti_cpu0_va(void)
{
	return pvti_cpu0_va;
}
EXPORT_SYMBOL_GPL(pvclock_get_pvti_cpu0_va);
Commit	Line	Data
7af192c9 GH	1	/* paravirtual clock -- common code used by kvm/xen
	2
	3	This program is free software; you can redistribute it and/or modify
	4	it under the terms of the GNU General Public License as published by
	5	the Free Software Foundation; either version 2 of the License, or
	6	(at your option) any later version.
	7
	8	This program is distributed in the hope that it will be useful,
	9	but WITHOUT ANY WARRANTY; without even the implied warranty of
	10	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	11	GNU General Public License for more details.
	12
	13	You should have received a copy of the GNU General Public License
	14	along with this program; if not, write to the Free Software
	15	Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
	16	*/
	17
	18	#include <linux/kernel.h>
	19	#include <linux/percpu.h>
71056ae2 MT	20	#include <linux/notifier.h>
	21	#include <linux/sched.h>
	22	#include <linux/gfp.h>
	23	#include <linux/bootmem.h>
38b8d208 IM	24	#include <linux/nmi.h>
38b8d208 IM	25
71056ae2	26	#include <asm/fixmap.h>
7af192c9	27	#include <asm/pvclock.h>
9f08890a	28	#include <asm/vgtod.h>
7af192c9	29
424c32f1	30	static u8 valid_flags __read_mostly = 0;
9f08890a	31	static struct pvclock_vsyscall_time_info *pvti_cpu0_va __read_mostly;
424c32f1 GC	32
	33	void pvclock_set_flags(u8 flags)
	34	{
	35	valid_flags = flags;
	36	}
	37
3807f345 GC	38	unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src)
3807f345 GC	39	{
a0854600	40	u64 pv_tsc_khz = 1000000ULL << 32;
3807f345	41
a0854600	42	do_div(pv_tsc_khz, src->tsc_to_system_mul);
3807f345	43	if (src->tsc_shift < 0)
a0854600	44	pv_tsc_khz <<= -src->tsc_shift;
3807f345	45	else
a0854600 HH	46	pv_tsc_khz >>= src->tsc_shift;
a0854600 HH	47	return pv_tsc_khz;
3807f345 GC	48	}
3807f345 GC	49
d63285e9 MT	50	void pvclock_touch_watchdogs(void)
	51	{
	52	touch_softlockup_watchdog_sync();
	53	clocksource_touch_watchdog();
	54	rcu_cpu_stall_reset();
8b414521	55	reset_hung_task_detector();
d63285e9 MT	56	}
d63285e9 MT	57
489fb490 GC	58	static atomic64_t last_value = ATOMIC64_INIT(0);
489fb490 GC	59
e7a3481c JF	60	void pvclock_resume(void)
	61	{
	62	atomic64_set(&last_value, 0);
	63	}
	64
2697902b MT	65	u8 pvclock_read_flags(struct pvclock_vcpu_time_info *src)
	66	{
	67	unsigned version;
2697902b MT	68	u8 flags;
	69
	70	do {
3aed64f6	71	version = pvclock_read_begin(src);
ed911b43	72	flags = src->flags;
3aed64f6	73	} while (pvclock_read_retry(src, version));
2697902b MT	74
	75	return flags & valid_flags;
	76	}
	77
a5a1d1c2	78	u64 pvclock_clocksource_read(struct pvclock_vcpu_time_info *src)
7af192c9	79	{
7af192c9	80	unsigned version;
a5a1d1c2	81	u64 ret;
489fb490	82	u64 last;
42b5637d	83	u8 flags;
7af192c9 GH	84
7af192c9 GH	85	do {
3aed64f6	86	version = pvclock_read_begin(src);
108b249c	87	ret = __pvclock_read_cycles(src, rdtsc_ordered());
3aed64f6 PB	88	flags = src->flags;
3aed64f6 PB	89	} while (pvclock_read_retry(src, version));
7af192c9	90
d63285e9 MT	91	if (unlikely((flags & PVCLOCK_GUEST_STOPPED) != 0)) {
	92	src->flags &= ~PVCLOCK_GUEST_STOPPED;
	93	pvclock_touch_watchdogs();
	94	}
	95
3a0d7256	96	if ((valid_flags & PVCLOCK_TSC_STABLE_BIT) &&
42b5637d	97	(flags & PVCLOCK_TSC_STABLE_BIT))
3a0d7256 GC	98	return ret;
3a0d7256 GC	99
489fb490 GC	100	/*
	101	* Assumption here is that last_value, a global accumulator, always goes
	102	* forward. If we are less than that, we should not be much smaller.
	103	* We assume there is an error marging we're inside, and then the correction
	104	* does not sacrifice accuracy.
	105	*
	106	* For reads: global may have changed between test and return,
	107	* but this means someone else updated poked the clock at a later time.
	108	* We just need to make sure we are not seeing a backwards event.
	109	*
	110	* For updates: last_value = ret is not enough, since two vcpus could be
	111	* updating at the same time, and one of them could be slightly behind,
	112	* making the assumption that last_value always go forward fail to hold.
	113	*/
	114	last = atomic64_read(&last_value);
	115	do {
	116	if (ret < last)
	117	return last;
	118	last = atomic64_cmpxchg(&last_value, last, ret);
	119	} while (unlikely(last != ret));
	120
7af192c9 GH	121	return ret;
	122	}
	123
	124	void pvclock_read_wallclock(struct pvclock_wall_clock *wall_clock,
	125	struct pvclock_vcpu_time_info *vcpu_time,
	126	struct timespec *ts)
	127	{
	128	u32 version;
	129	u64 delta;
	130	struct timespec now;
	131
	132	/* get wallclock at system boot */
	133	do {
	134	version = wall_clock->version;
	135	rmb(); /* fetch version before time */
	136	now.tv_sec = wall_clock->sec;
	137	now.tv_nsec = wall_clock->nsec;
	138	rmb(); /* fetch time before checking version */
	139	} while ((wall_clock->version & 1) \|\| (version != wall_clock->version));
	140
	141	delta = pvclock_clocksource_read(vcpu_time); /* time since system boot */
	142	delta += now.tv_sec * (u64)NSEC_PER_SEC + now.tv_nsec;
	143
	144	now.tv_nsec = do_div(delta, NSEC_PER_SEC);
	145	now.tv_sec = delta;
	146
	147	set_normalized_timespec(ts, now.tv_sec, now.tv_nsec);
	148	}
9f08890a JM	149
	150	void pvclock_set_pvti_cpu0_va(struct pvclock_vsyscall_time_info *pvti)
	151	{
	152	WARN_ON(vclock_was_used(VCLOCK_PVCLOCK));
	153	pvti_cpu0_va = pvti;
	154	}
	155
	156	struct pvclock_vsyscall_time_info *pvclock_get_pvti_cpu0_va(void)
	157	{
	158	return pvti_cpu0_va;
	159	}
	160	EXPORT_SYMBOL_GPL(pvclock_get_pvti_cpu0_va);