qemu-timer.c

/*
 * QEMU System Emulator
 *
 * Copyright (c) 2003-2008 Fabrice Bellard
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 * copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */

#include "sysemu.h"
#include "net.h"
#include "monitor.h"
#include "console.h"

#include "hw/hw.h"

#include <unistd.h>
#include <fcntl.h>
#include <time.h>
#include <errno.h>
#include <sys/time.h>
#include <signal.h>
#ifdef __FreeBSD__
#include <sys/param.h>
#endif

#ifdef _WIN32
#include <windows.h>
#include <mmsystem.h>
#endif

#include "qemu-timer.h"

/* Conversion factor from emulated instructions to virtual clock ticks.  */
int icount_time_shift;
/* Arbitrarily pick 1MIPS as the minimum allowable speed.  */
#define MAX_ICOUNT_SHIFT 10
/* Compensate for varying guest execution speed.  */
int64_t qemu_icount_bias;
static QEMUTimer *icount_rt_timer;
static QEMUTimer *icount_vm_timer;

/***********************************************************/
/* guest cycle counter */

typedef struct TimersState {
    int64_t cpu_ticks_prev;
    int64_t cpu_ticks_offset;
    int64_t cpu_clock_offset;
    int32_t cpu_ticks_enabled;
    int64_t dummy;
} TimersState;

TimersState timers_state;

/* return the host CPU cycle counter and handle stop/restart */
int64_t cpu_get_ticks(void)
{
    if (use_icount) {
        return cpu_get_icount();
    }
    if (!timers_state.cpu_ticks_enabled) {
        return timers_state.cpu_ticks_offset;
    } else {
        int64_t ticks;
        ticks = cpu_get_real_ticks();
        if (timers_state.cpu_ticks_prev > ticks) {
            /* Note: non increasing ticks may happen if the host uses
               software suspend */
            timers_state.cpu_ticks_offset += timers_state.cpu_ticks_prev - ticks;
        }
        timers_state.cpu_ticks_prev = ticks;
        return ticks + timers_state.cpu_ticks_offset;
    }
}

/* return the host CPU monotonic timer and handle stop/restart */
static int64_t cpu_get_clock(void)
{
    int64_t ti;
    if (!timers_state.cpu_ticks_enabled) {
        return timers_state.cpu_clock_offset;
    } else {
        ti = get_clock();
        return ti + timers_state.cpu_clock_offset;
    }
}

/* enable cpu_get_ticks() */
void cpu_enable_ticks(void)
{
    if (!timers_state.cpu_ticks_enabled) {
        timers_state.cpu_ticks_offset -= cpu_get_real_ticks();
        timers_state.cpu_clock_offset -= get_clock();
        timers_state.cpu_ticks_enabled = 1;
    }
}

/* disable cpu_get_ticks() : the clock is stopped. You must not call
   cpu_get_ticks() after that.  */
void cpu_disable_ticks(void)
{
    if (timers_state.cpu_ticks_enabled) {
        timers_state.cpu_ticks_offset = cpu_get_ticks();
        timers_state.cpu_clock_offset = cpu_get_clock();
        timers_state.cpu_ticks_enabled = 0;
    }
}

/***********************************************************/
/* timers */

#define QEMU_CLOCK_REALTIME 0
#define QEMU_CLOCK_VIRTUAL  1
#define QEMU_CLOCK_HOST     2

struct QEMUClock {
    int type;
    int enabled;

    QEMUTimer *warp_timer;
    QEMUTimer *active_timers;

    NotifierList reset_notifiers;
    int64_t last;
};

struct QEMUTimer {
    QEMUClock *clock;
    int64_t expire_time;	/* in nanoseconds */
    int scale;
    QEMUTimerCB *cb;
    void *opaque;
    struct QEMUTimer *next;
};

struct qemu_alarm_timer {
    char const *name;
    int (*start)(struct qemu_alarm_timer *t);
    void (*stop)(struct qemu_alarm_timer *t);
    void (*rearm)(struct qemu_alarm_timer *t, int64_t nearest_delta_ns);
#if defined(__linux__)
    int fd;
    timer_t timer;
#elif defined(_WIN32)
    HANDLE timer;
#endif
    char expired;
    char pending;
};

static struct qemu_alarm_timer *alarm_timer;

static bool qemu_timer_expired_ns(QEMUTimer *timer_head, int64_t current_time)
{
    return timer_head && (timer_head->expire_time <= current_time);
}

int qemu_alarm_pending(void)
{
    return alarm_timer->pending;
}

static inline int alarm_has_dynticks(struct qemu_alarm_timer *t)
{
    return !!t->rearm;
}

static int64_t qemu_next_alarm_deadline(void)
{
    int64_t delta;
    int64_t rtdelta;

    if (!use_icount && vm_clock->active_timers) {
        delta = vm_clock->active_timers->expire_time -
                     qemu_get_clock_ns(vm_clock);
    } else {
        delta = INT32_MAX;
    }
    if (host_clock->active_timers) {
        int64_t hdelta = host_clock->active_timers->expire_time -
                 qemu_get_clock_ns(host_clock);
        if (hdelta < delta) {
            delta = hdelta;
        }
    }
    if (rt_clock->active_timers) {
        rtdelta = (rt_clock->active_timers->expire_time -
                 qemu_get_clock_ns(rt_clock));
        if (rtdelta < delta) {
            delta = rtdelta;
        }
    }

    return delta;
}

static void qemu_rearm_alarm_timer(struct qemu_alarm_timer *t)
{
    int64_t nearest_delta_ns;
    assert(alarm_has_dynticks(t));
    if (!rt_clock->active_timers &&
        !vm_clock->active_timers &&
        !host_clock->active_timers) {
        return;
    }
    nearest_delta_ns = qemu_next_alarm_deadline();
    t->rearm(t, nearest_delta_ns);
}

/* TODO: MIN_TIMER_REARM_NS should be optimized */
#define MIN_TIMER_REARM_NS 250000

#ifdef _WIN32

static int mm_start_timer(struct qemu_alarm_timer *t);
static void mm_stop_timer(struct qemu_alarm_timer *t);
static void mm_rearm_timer(struct qemu_alarm_timer *t, int64_t delta);

static int win32_start_timer(struct qemu_alarm_timer *t);
static void win32_stop_timer(struct qemu_alarm_timer *t);
static void win32_rearm_timer(struct qemu_alarm_timer *t, int64_t delta);

#else

static int unix_start_timer(struct qemu_alarm_timer *t);
static void unix_stop_timer(struct qemu_alarm_timer *t);
static void unix_rearm_timer(struct qemu_alarm_timer *t, int64_t delta);

#ifdef __linux__

static int dynticks_start_timer(struct qemu_alarm_timer *t);
static void dynticks_stop_timer(struct qemu_alarm_timer *t);
static void dynticks_rearm_timer(struct qemu_alarm_timer *t, int64_t delta);

#endif /* __linux__ */

#endif /* _WIN32 */

/* Correlation between real and virtual time is always going to be
   fairly approximate, so ignore small variation.
   When the guest is idle real and virtual time will be aligned in
   the IO wait loop.  */
#define ICOUNT_WOBBLE (get_ticks_per_sec() / 10)

static void icount_adjust(void)
{
    int64_t cur_time;
    int64_t cur_icount;
    int64_t delta;
    static int64_t last_delta;
    /* If the VM is not running, then do nothing.  */
    if (!runstate_is_running())
        return;

    cur_time = cpu_get_clock();
    cur_icount = qemu_get_clock_ns(vm_clock);
    delta = cur_icount - cur_time;
    /* FIXME: This is a very crude algorithm, somewhat prone to oscillation.  */
    if (delta > 0
        && last_delta + ICOUNT_WOBBLE < delta * 2
        && icount_time_shift > 0) {
        /* The guest is getting too far ahead.  Slow time down.  */
        icount_time_shift--;
    }
    if (delta < 0
        && last_delta - ICOUNT_WOBBLE > delta * 2
        && icount_time_shift < MAX_ICOUNT_SHIFT) {
        /* The guest is getting too far behind.  Speed time up.  */
        icount_time_shift++;
    }
    last_delta = delta;
    qemu_icount_bias = cur_icount - (qemu_icount << icount_time_shift);
}

static void icount_adjust_rt(void * opaque)
{
    qemu_mod_timer(icount_rt_timer,
                   qemu_get_clock_ms(rt_clock) + 1000);
    icount_adjust();
}

static void icount_adjust_vm(void * opaque)
{
    qemu_mod_timer(icount_vm_timer,
                   qemu_get_clock_ns(vm_clock) + get_ticks_per_sec() / 10);
    icount_adjust();
}

int64_t qemu_icount_round(int64_t count)
{
    return (count + (1 << icount_time_shift) - 1) >> icount_time_shift;
}

static struct qemu_alarm_timer alarm_timers[] = {
#ifndef _WIN32
#ifdef __linux__
    {"dynticks", dynticks_start_timer,
     dynticks_stop_timer, dynticks_rearm_timer},
#endif
    {"unix", unix_start_timer, unix_stop_timer, unix_rearm_timer},
#else
    {"mmtimer", mm_start_timer, mm_stop_timer, NULL},
    {"mmtimer2", mm_start_timer, mm_stop_timer, mm_rearm_timer},
    {"dynticks", win32_start_timer, win32_stop_timer, win32_rearm_timer},
    {"win32", win32_start_timer, win32_stop_timer, NULL},
#endif
    {NULL, }
};

static void show_available_alarms(void)
{
    int i;

    printf("Available alarm timers, in order of precedence:\n");
    for (i = 0; alarm_timers[i].name; i++)
        printf("%s\n", alarm_timers[i].name);
}

void configure_alarms(char const *opt)
{
    int i;
    int cur = 0;
    int count = ARRAY_SIZE(alarm_timers) - 1;
    char *arg;
    char *name;
    struct qemu_alarm_timer tmp;

    if (!strcmp(opt, "?")) {
        show_available_alarms();
        exit(0);
    }

    arg = g_strdup(opt);

    /* Reorder the array */
    name = strtok(arg, ",");
    while (name) {
        for (i = 0; i < count && alarm_timers[i].name; i++) {
            if (!strcmp(alarm_timers[i].name, name))
                break;
        }

        if (i == count) {
            fprintf(stderr, "Unknown clock %s\n", name);
            goto next;
        }

        if (i < cur)
            /* Ignore */
            goto next;

	/* Swap */
        tmp = alarm_timers[i];
        alarm_timers[i] = alarm_timers[cur];
        alarm_timers[cur] = tmp;

        cur++;
next:
        name = strtok(NULL, ",");
    }

    g_free(arg);

    if (cur) {
        /* Disable remaining timers */
        for (i = cur; i < count; i++)
            alarm_timers[i].name = NULL;
    } else {
        show_available_alarms();
        exit(1);
    }
}

QEMUClock *rt_clock;
QEMUClock *vm_clock;
QEMUClock *host_clock;

static QEMUClock *qemu_new_clock(int type)
{
    QEMUClock *clock;

    clock = g_malloc0(sizeof(QEMUClock));
    clock->type = type;
    clock->enabled = 1;
    notifier_list_init(&clock->reset_notifiers);
    /* required to detect & report backward jumps */
    if (type == QEMU_CLOCK_HOST) {
        clock->last = get_clock_realtime();
    }
    return clock;
}

void qemu_clock_enable(QEMUClock *clock, int enabled)
{
    clock->enabled = enabled;
}

static int64_t vm_clock_warp_start;

static void icount_warp_rt(void *opaque)
{
    if (vm_clock_warp_start == -1) {
        return;
    }

    if (runstate_is_running()) {
        int64_t clock = qemu_get_clock_ns(rt_clock);
        int64_t warp_delta = clock - vm_clock_warp_start;
        if (use_icount == 1) {
            qemu_icount_bias += warp_delta;
        } else {
            /*
             * In adaptive mode, do not let the vm_clock run too
             * far ahead of real time.
             */
            int64_t cur_time = cpu_get_clock();
            int64_t cur_icount = qemu_get_clock_ns(vm_clock);
            int64_t delta = cur_time - cur_icount;
            qemu_icount_bias += MIN(warp_delta, delta);
        }
        if (qemu_timer_expired(vm_clock->active_timers,
                               qemu_get_clock_ns(vm_clock))) {
            qemu_notify_event();
        }
    }
    vm_clock_warp_start = -1;
}

void qemu_clock_warp(QEMUClock *clock)
{
    int64_t deadline;

    if (!clock->warp_timer) {
        return;
    }

    /*
     * There are too many global variables to make the "warp" behavior
     * applicable to other clocks.  But a clock argument removes the
     * need for if statements all over the place.
     */
    assert(clock == vm_clock);

    /*
     * If the CPUs have been sleeping, advance the vm_clock timer now.  This
     * ensures that the deadline for the timer is computed correctly below.
     * This also makes sure that the insn counter is synchronized before the
     * CPU starts running, in case the CPU is woken by an event other than
     * the earliest vm_clock timer.
     */
    icount_warp_rt(NULL);
    if (!all_cpu_threads_idle() || !clock->active_timers) {
        qemu_del_timer(clock->warp_timer);
        return;
    }

    vm_clock_warp_start = qemu_get_clock_ns(rt_clock);
    deadline = qemu_next_icount_deadline();
    if (deadline > 0) {
        /*
         * Ensure the vm_clock proceeds even when the virtual CPU goes to
         * sleep.  Otherwise, the CPU might be waiting for a future timer
         * interrupt to wake it up, but the interrupt never comes because
         * the vCPU isn't running any insns and thus doesn't advance the
         * vm_clock.
         *
         * An extreme solution for this problem would be to never let VCPUs
         * sleep in icount mode if there is a pending vm_clock timer; rather
         * time could just advance to the next vm_clock event.  Instead, we
         * do stop VCPUs and only advance vm_clock after some "real" time,
         * (related to the time left until the next event) has passed.  This
         * rt_clock timer will do this.  This avoids that the warps are too
         * visible externally---for example, you will not be sending network
         * packets continously instead of every 100ms.
         */
        qemu_mod_timer(clock->warp_timer, vm_clock_warp_start + deadline);
    } else {
        qemu_notify_event();
    }
}

int64_t qemu_clock_has_timers(QEMUClock *clock)
{
    return !!clock->active_timers;
}

int64_t qemu_clock_expired(QEMUClock *clock)
{
    return (clock->active_timers &&
            clock->active_timers->expire_time < qemu_get_clock_ns(clock));
}

int64_t qemu_clock_deadline(QEMUClock *clock)
{
    /* To avoid problems with overflow limit this to 2^32.  */
    int64_t delta = INT32_MAX;

    if (clock->active_timers) {
        delta = clock->active_timers->expire_time - qemu_get_clock_ns(clock);
    }
    if (delta < 0) {
        delta = 0;
    }
    return delta;
}

QEMUTimer *qemu_new_timer(QEMUClock *clock, int scale,
                          QEMUTimerCB *cb, void *opaque)
{
    QEMUTimer *ts;

    ts = g_malloc0(sizeof(QEMUTimer));
    ts->clock = clock;
    ts->cb = cb;
    ts->opaque = opaque;
    ts->scale = scale;
    return ts;
}

void qemu_free_timer(QEMUTimer *ts)
{
    g_free(ts);
}

/* stop a timer, but do not dealloc it */
void qemu_del_timer(QEMUTimer *ts)
{
    QEMUTimer **pt, *t;

    /* NOTE: this code must be signal safe because
       qemu_timer_expired() can be called from a signal. */
    pt = &ts->clock->active_timers;
    for(;;) {
        t = *pt;
        if (!t)
            break;
        if (t == ts) {
            *pt = t->next;
            break;
        }
        pt = &t->next;
    }
}

/* modify the current timer so that it will be fired when current_time
   >= expire_time. The corresponding callback will be called. */
static void qemu_mod_timer_ns(QEMUTimer *ts, int64_t expire_time)
{
    QEMUTimer **pt, *t;

    qemu_del_timer(ts);

    /* add the timer in the sorted list */
    /* NOTE: this code must be signal safe because
       qemu_timer_expired() can be called from a signal. */
    pt = &ts->clock->active_timers;
    for(;;) {
        t = *pt;
        if (!qemu_timer_expired_ns(t, expire_time)) {
            break;
        }
        pt = &t->next;
    }
    ts->expire_time = expire_time;
    ts->next = *pt;
    *pt = ts;

    /* Rearm if necessary  */
    if (pt == &ts->clock->active_timers) {
        if (!alarm_timer->pending) {
            qemu_rearm_alarm_timer(alarm_timer);
        }
        /* Interrupt execution to force deadline recalculation.  */
        qemu_clock_warp(ts->clock);
        if (use_icount) {
            qemu_notify_event();
        }
    }
}

/* modify the current timer so that it will be fired when current_time
   >= expire_time. The corresponding callback will be called. */
void qemu_mod_timer(QEMUTimer *ts, int64_t expire_time)
{
    qemu_mod_timer_ns(ts, expire_time * ts->scale);
}

int qemu_timer_pending(QEMUTimer *ts)
{
    QEMUTimer *t;
    for (t = ts->clock->active_timers; t != NULL; t = t->next) {
        if (t == ts)
            return 1;
    }
    return 0;
}

int qemu_timer_expired(QEMUTimer *timer_head, int64_t current_time)
{
    return qemu_timer_expired_ns(timer_head, current_time * timer_head->scale);
}

static void qemu_run_timers(QEMUClock *clock)
{
    QEMUTimer **ptimer_head, *ts;
    int64_t current_time;
   
    if (!clock->enabled)
        return;

    current_time = qemu_get_clock_ns(clock);
    ptimer_head = &clock->active_timers;
    for(;;) {
        ts = *ptimer_head;
        if (!qemu_timer_expired_ns(ts, current_time)) {
            break;
        }
        /* remove timer from the list before calling the callback */
        *ptimer_head = ts->next;
        ts->next = NULL;

        /* run the callback (the timer list can be modified) */
        ts->cb(ts->opaque);
    }
}

int64_t qemu_get_clock_ns(QEMUClock *clock)
{
    int64_t now, last;

    switch(clock->type) {
    case QEMU_CLOCK_REALTIME:
        return get_clock();
    default:
    case QEMU_CLOCK_VIRTUAL:
        if (use_icount) {
            return cpu_get_icount();
        } else {
            return cpu_get_clock();
        }
    case QEMU_CLOCK_HOST:
        now = get_clock_realtime();
        last = clock->last;
        clock->last = now;
        if (now < last) {
            notifier_list_notify(&clock->reset_notifiers, &now);
        }
        return now;
    }
}

void qemu_register_clock_reset_notifier(QEMUClock *clock, Notifier *notifier)
{
    notifier_list_add(&clock->reset_notifiers, notifier);
}

void qemu_unregister_clock_reset_notifier(QEMUClock *clock, Notifier *notifier)
{
    notifier_list_remove(&clock->reset_notifiers, notifier);
}

void init_clocks(void)
{
    rt_clock = qemu_new_clock(QEMU_CLOCK_REALTIME);
    vm_clock = qemu_new_clock(QEMU_CLOCK_VIRTUAL);
    host_clock = qemu_new_clock(QEMU_CLOCK_HOST);

    rtc_clock = host_clock;
}

/* save a timer */
void qemu_put_timer(QEMUFile *f, QEMUTimer *ts)
{
    uint64_t expire_time;

    if (qemu_timer_pending(ts)) {
        expire_time = ts->expire_time;
    } else {
        expire_time = -1;
    }
    qemu_put_be64(f, expire_time);
}

void qemu_get_timer(QEMUFile *f, QEMUTimer *ts)
{
    uint64_t expire_time;

    expire_time = qemu_get_be64(f);
    if (expire_time != -1) {
        qemu_mod_timer_ns(ts, expire_time);
    } else {
        qemu_del_timer(ts);
    }
}

static const VMStateDescription vmstate_timers = {
    .name = "timer",
    .version_id = 2,
    .minimum_version_id = 1,
    .minimum_version_id_old = 1,
    .fields      = (VMStateField []) {
        VMSTATE_INT64(cpu_ticks_offset, TimersState),
        VMSTATE_INT64(dummy, TimersState),
        VMSTATE_INT64_V(cpu_clock_offset, TimersState, 2),
        VMSTATE_END_OF_LIST()
    }
};

void configure_icount(const char *option)
{
    vmstate_register(NULL, 0, &vmstate_timers, &timers_state);
    if (!option)
        return;

    vm_clock->warp_timer = qemu_new_timer_ns(rt_clock, icount_warp_rt, NULL);

    if (strcmp(option, "auto") != 0) {
        icount_time_shift = strtol(option, NULL, 0);
        use_icount = 1;
        return;
    }

    use_icount = 2;

    /* 125MIPS seems a reasonable initial guess at the guest speed.
       It will be corrected fairly quickly anyway.  */
    icount_time_shift = 3;

    /* Have both realtime and virtual time triggers for speed adjustment.
       The realtime trigger catches emulated time passing too slowly,
       the virtual time trigger catches emulated time passing too fast.
       Realtime triggers occur even when idle, so use them less frequently
       than VM triggers.  */
    icount_rt_timer = qemu_new_timer_ms(rt_clock, icount_adjust_rt, NULL);
    qemu_mod_timer(icount_rt_timer,
                   qemu_get_clock_ms(rt_clock) + 1000);
    icount_vm_timer = qemu_new_timer_ns(vm_clock, icount_adjust_vm, NULL);
    qemu_mod_timer(icount_vm_timer,
                   qemu_get_clock_ns(vm_clock) + get_ticks_per_sec() / 10);
}

void qemu_run_all_timers(void)
{
    alarm_timer->pending = 0;

    /* rearm timer, if not periodic */
    if (alarm_timer->expired) {
        alarm_timer->expired = 0;
        qemu_rearm_alarm_timer(alarm_timer);
    }

    /* vm time timers */
    if (runstate_is_running()) {
        qemu_run_timers(vm_clock);
    }

    qemu_run_timers(rt_clock);
    qemu_run_timers(host_clock);
}

#ifdef _WIN32
static void CALLBACK host_alarm_handler(PVOID lpParam, BOOLEAN unused)
#else
static void host_alarm_handler(int host_signum)
#endif
{
    struct qemu_alarm_timer *t = alarm_timer;
    if (!t)
	return;

#if 0
#define DISP_FREQ 1000
    {
        static int64_t delta_min = INT64_MAX;
        static int64_t delta_max, delta_cum, last_clock, delta, ti;
        static int count;
        ti = qemu_get_clock_ns(vm_clock);
        if (last_clock != 0) {
            delta = ti - last_clock;
            if (delta < delta_min)
                delta_min = delta;
            if (delta > delta_max)
                delta_max = delta;
            delta_cum += delta;
            if (++count == DISP_FREQ) {
                printf("timer: min=%" PRId64 " us max=%" PRId64 " us avg=%" PRId64 " us avg_freq=%0.3f Hz\n",
                       muldiv64(delta_min, 1000000, get_ticks_per_sec()),
                       muldiv64(delta_max, 1000000, get_ticks_per_sec()),
                       muldiv64(delta_cum, 1000000 / DISP_FREQ, get_ticks_per_sec()),
                       (double)get_ticks_per_sec() / ((double)delta_cum / DISP_FREQ));
                count = 0;
                delta_min = INT64_MAX;
                delta_max = 0;
                delta_cum = 0;
            }
        }
        last_clock = ti;
    }
#endif
    if (alarm_has_dynticks(t) ||
        qemu_next_alarm_deadline () <= 0) {
        t->expired = alarm_has_dynticks(t);
        t->pending = 1;
        qemu_notify_event();
    }
}

int64_t qemu_next_icount_deadline(void)
{
    /* To avoid problems with overflow limit this to 2^32.  */
    int64_t delta = INT32_MAX;

    assert(use_icount);
    if (vm_clock->active_timers) {
        delta = vm_clock->active_timers->expire_time -
                     qemu_get_clock_ns(vm_clock);
    }

    if (delta < 0)
        delta = 0;

    return delta;
}

#if defined(__linux__)

#include "compatfd.h"

static int dynticks_start_timer(struct qemu_alarm_timer *t)
{
    struct sigevent ev;
    timer_t host_timer;
    struct sigaction act;

    sigfillset(&act.sa_mask);
    act.sa_flags = 0;
    act.sa_handler = host_alarm_handler;

    sigaction(SIGALRM, &act, NULL);

    /* 
     * Initialize ev struct to 0 to avoid valgrind complaining
     * about uninitialized data in timer_create call
     */
    memset(&ev, 0, sizeof(ev));
    ev.sigev_value.sival_int = 0;
    ev.sigev_notify = SIGEV_SIGNAL;
#ifdef SIGEV_THREAD_ID
    if (qemu_signalfd_available()) {
        ev.sigev_notify = SIGEV_THREAD_ID;
        ev._sigev_un._tid = qemu_get_thread_id();
    }
#endif /* SIGEV_THREAD_ID */
    ev.sigev_signo = SIGALRM;

    if (timer_create(CLOCK_REALTIME, &ev, &host_timer)) {
        perror("timer_create");

        /* disable dynticks */
        fprintf(stderr, "Dynamic Ticks disabled\n");

        return -1;
    }

    t->timer = host_timer;

    return 0;
}

static void dynticks_stop_timer(struct qemu_alarm_timer *t)
{
    timer_t host_timer = t->timer;

    timer_delete(host_timer);
}

static void dynticks_rearm_timer(struct qemu_alarm_timer *t,
                                 int64_t nearest_delta_ns)
{
    timer_t host_timer = t->timer;
    struct itimerspec timeout;
    int64_t current_ns;

    if (nearest_delta_ns < MIN_TIMER_REARM_NS)
        nearest_delta_ns = MIN_TIMER_REARM_NS;

    /* check whether a timer is already running */
    if (timer_gettime(host_timer, &timeout)) {
        perror("gettime");
        fprintf(stderr, "Internal timer error: aborting\n");
        exit(1);
    }
    current_ns = timeout.it_value.tv_sec * 1000000000LL + timeout.it_value.tv_nsec;
    if (current_ns && current_ns <= nearest_delta_ns)
        return;

    timeout.it_interval.tv_sec = 0;
    timeout.it_interval.tv_nsec = 0; /* 0 for one-shot timer */
    timeout.it_value.tv_sec =  nearest_delta_ns / 1000000000;
    timeout.it_value.tv_nsec = nearest_delta_ns % 1000000000;
    if (timer_settime(host_timer, 0 /* RELATIVE */, &timeout, NULL)) {
        perror("settime");
        fprintf(stderr, "Internal timer error: aborting\n");
        exit(1);
    }
}

#endif /* defined(__linux__) */

#if !defined(_WIN32)

static int unix_start_timer(struct qemu_alarm_timer *t)
{
    struct sigaction act;

    /* timer signal */
    sigfillset(&act.sa_mask);
    act.sa_flags = 0;
    act.sa_handler = host_alarm_handler;

    sigaction(SIGALRM, &act, NULL);
    return 0;
}

static void unix_rearm_timer(struct qemu_alarm_timer *t,
                             int64_t nearest_delta_ns)
{
    struct itimerval itv;
    int err;

    if (nearest_delta_ns < MIN_TIMER_REARM_NS)
        nearest_delta_ns = MIN_TIMER_REARM_NS;

    itv.it_interval.tv_sec = 0;
    itv.it_interval.tv_usec = 0; /* 0 for one-shot timer */
    itv.it_value.tv_sec =  nearest_delta_ns / 1000000000;
    itv.it_value.tv_usec = (nearest_delta_ns % 1000000000) / 1000;
    err = setitimer(ITIMER_REAL, &itv, NULL);
    if (err) {
        perror("setitimer");
        fprintf(stderr, "Internal timer error: aborting\n");
        exit(1);
    }
}

static void unix_stop_timer(struct qemu_alarm_timer *t)
{
    struct itimerval itv;

    memset(&itv, 0, sizeof(itv));
    setitimer(ITIMER_REAL, &itv, NULL);
}

#endif /* !defined(_WIN32) */


#ifdef _WIN32

static MMRESULT mm_timer;
static unsigned mm_period;

static void CALLBACK mm_alarm_handler(UINT uTimerID, UINT uMsg,
                                      DWORD_PTR dwUser, DWORD_PTR dw1,
                                      DWORD_PTR dw2)
{
    struct qemu_alarm_timer *t = alarm_timer;
    if (!t) {
        return;
    }
    if (alarm_has_dynticks(t) || qemu_next_alarm_deadline() <= 0) {
        t->expired = alarm_has_dynticks(t);
        t->pending = 1;
        qemu_notify_event();
    }
}

static int mm_start_timer(struct qemu_alarm_timer *t)
{
    TIMECAPS tc;
    UINT flags;

    memset(&tc, 0, sizeof(tc));
    timeGetDevCaps(&tc, sizeof(tc));

    mm_period = tc.wPeriodMin;
    timeBeginPeriod(mm_period);

    flags = TIME_CALLBACK_FUNCTION;
    if (alarm_has_dynticks(t)) {
        flags |= TIME_ONESHOT;
    } else {
        flags |= TIME_PERIODIC;
    }

    mm_timer = timeSetEvent(1,                  /* interval (ms) */
                            mm_period,          /* resolution */
                            mm_alarm_handler,   /* function */
                            (DWORD_PTR)t,       /* parameter */
                            flags);

    if (!mm_timer) {
        fprintf(stderr, "Failed to initialize win32 alarm timer: %ld\n",
                GetLastError());
        timeEndPeriod(mm_period);
        return -1;
    }

    return 0;
}

static void mm_stop_timer(struct qemu_alarm_timer *t)
{
    timeKillEvent(mm_timer);
    timeEndPeriod(mm_period);
}

static void mm_rearm_timer(struct qemu_alarm_timer *t, int64_t delta)
{
    int nearest_delta_ms = (delta + 999999) / 1000000;
    if (nearest_delta_ms < 1) {
        nearest_delta_ms = 1;
    }

    timeKillEvent(mm_timer);
    mm_timer = timeSetEvent(nearest_delta_ms,
                            mm_period,
                            mm_alarm_handler,
                            (DWORD_PTR)t,
                            TIME_ONESHOT | TIME_CALLBACK_FUNCTION);

    if (!mm_timer) {
        fprintf(stderr, "Failed to re-arm win32 alarm timer %ld\n",
                GetLastError());

        timeEndPeriod(mm_period);
        exit(1);
    }
}

static int win32_start_timer(struct qemu_alarm_timer *t)
{
    HANDLE hTimer;
    BOOLEAN success;

    /* If you call ChangeTimerQueueTimer on a one-shot timer (its period
       is zero) that has already expired, the timer is not updated.  Since
       creating a new timer is relatively expensive, set a bogus one-hour
       interval in the dynticks case.  */
    success = CreateTimerQueueTimer(&hTimer,
                          NULL,
                          host_alarm_handler,
                          t,
                          1,
                          alarm_has_dynticks(t) ? 3600000 : 1,
                          WT_EXECUTEINTIMERTHREAD);

    if (!success) {
        fprintf(stderr, "Failed to initialize win32 alarm timer: %ld\n",
                GetLastError());
        return -1;
    }

    t->timer = hTimer;
    return 0;
}

static void win32_stop_timer(struct qemu_alarm_timer *t)
{
    HANDLE hTimer = t->timer;

    if (hTimer) {
        DeleteTimerQueueTimer(NULL, hTimer, NULL);
    }
}

static void win32_rearm_timer(struct qemu_alarm_timer *t,
                              int64_t nearest_delta_ns)
{
    HANDLE hTimer = t->timer;
    int nearest_delta_ms;
    BOOLEAN success;

    nearest_delta_ms = (nearest_delta_ns + 999999) / 1000000;
    if (nearest_delta_ms < 1) {
        nearest_delta_ms = 1;
    }
    success = ChangeTimerQueueTimer(NULL,
                                    hTimer,
                                    nearest_delta_ms,
                                    3600000);

    if (!success) {
        fprintf(stderr, "Failed to rearm win32 alarm timer: %ld\n",
                GetLastError());
        exit(-1);
    }

}

#endif /* _WIN32 */

static void alarm_timer_on_change_state_rearm(void *opaque, int running,
                                              RunState state)
{
    if (running)
        qemu_rearm_alarm_timer((struct qemu_alarm_timer *) opaque);
}

int init_timer_alarm(void)
{
    struct qemu_alarm_timer *t = NULL;
    int i, err = -1;

    for (i = 0; alarm_timers[i].name; i++) {
        t = &alarm_timers[i];

        err = t->start(t);
        if (!err)
            break;
    }

    if (err) {
        err = -ENOENT;
        goto fail;
    }

    /* first event is at time 0 */
    t->pending = 1;
    alarm_timer = t;
    qemu_add_vm_change_state_handler(alarm_timer_on_change_state_rearm, t);

    return 0;

fail:
    return err;
}

void quit_timers(void)
{
    struct qemu_alarm_timer *t = alarm_timer;
    alarm_timer = NULL;
    t->stop(t);
}

int qemu_calculate_timeout(void)
{
    return 1000;
}