source: rtems/cpukit/score/src/kern_tc.c

/**
 * @file
 *
 * @ingroup RTEMSScoreTimecounter
 *
 * @brief This source file contains the definition of
 *  ::_Timecounter, ::_Timecounter_Time_second, and ::_Timecounter_Time_uptime
 *  and the implementation of _Timecounter_Set_NTP_update_second(),
 *  _Timecounter_Binuptime(), _Timecounter_Nanouptime(),
 *  _Timecounter_Microuptime(), _Timecounter_Bintime(),
 *  _Timecounter_Nanotime(), _Timecounter_Microtime(),
 *  _Timecounter_Getbinuptime(), _Timecounter_Getnanouptime(),
 *  _Timecounter_Getmicrouptime(), _Timecounter_Getbintime(),
 *  _Timecounter_Getnanotime(), _Timecounter_Getmicrotime(),
 *  _Timecounter_Getboottime(), _Timecounter_Getboottimebin(), and
 *  _Timecounter_Install().
 */

/*-
 * SPDX-License-Identifier: Beerware
 *
 * ----------------------------------------------------------------------------
 * "THE BEER-WARE LICENSE" (Revision 42):
 * <phk@FreeBSD.ORG> wrote this file.  As long as you retain this notice you
 * can do whatever you want with this stuff. If we meet some day, and you think
 * this stuff is worth it, you can buy me a beer in return.   Poul-Henning Kamp
 * ----------------------------------------------------------------------------
 *
 * Copyright (c) 2011, 2015, 2016 The FreeBSD Foundation
 *
 * Portions of this software were developed by Julien Ridoux at the University
 * of Melbourne under sponsorship from the FreeBSD Foundation.
 *
 * Portions of this software were developed by Konstantin Belousov
 * under sponsorship from the FreeBSD Foundation.
 */

#ifdef __rtems__
#include <sys/lock.h>
#define _KERNEL
#define binuptime(_bt) _Timecounter_Binuptime(_bt)
#define nanouptime(_tsp) _Timecounter_Nanouptime(_tsp)
#define microuptime(_tvp) _Timecounter_Microuptime(_tvp)
#define bintime(_bt) _Timecounter_Bintime(_bt)
#define nanotime(_tsp) _Timecounter_Nanotime(_tsp)
#define microtime(_tvp) _Timecounter_Microtime(_tvp)
#define getbinuptime(_bt) _Timecounter_Getbinuptime(_bt)
#define getnanouptime(_tsp) _Timecounter_Getnanouptime(_tsp)
#define getmicrouptime(_tvp) _Timecounter_Getmicrouptime(_tvp)
#define getbintime(_bt) _Timecounter_Getbintime(_bt)
#define getnanotime(_tsp) _Timecounter_Getnanotime(_tsp)
#define getmicrotime(_tvp) _Timecounter_Getmicrotime(_tvp)
#define getboottime(_tvp) _Timecounter_Getboottime(_tvp)
#define getboottimebin(_bt) _Timecounter_Getboottimebin(_bt)
#define tc_init _Timecounter_Install
#define timecounter _Timecounter
#define time_second _Timecounter_Time_second
#define time_uptime _Timecounter_Time_uptime

#include <rtems/score/timecounterimpl.h>
#include <rtems/score/assert.h>
#include <rtems/score/atomic.h>
#include <rtems/score/smp.h>
#include <rtems/score/todimpl.h>
#include <rtems/score/watchdogimpl.h>
#include <rtems/rtems/clock.h>

#define ENOIOCTL EINVAL
#define KASSERT(exp, arg) _Assert(exp)
#endif /* __rtems__ */
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");

#include "opt_ntp.h"
#include "opt_ffclock.h"

#include <sys/param.h>
#ifndef __rtems__
#include <sys/kernel.h>
#include <sys/limits.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/sbuf.h>
#include <sys/sleepqueue.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <sys/systm.h>
#endif /* __rtems__ */
#include <sys/timeffc.h>
#include <sys/timepps.h>
#include <sys/timetc.h>
#include <sys/timex.h>
#ifndef __rtems__
#include <sys/vdso.h>
#endif /* __rtems__ */
#ifdef __rtems__
#include <errno.h>
#include <limits.h>
#include <string.h>
#include <rtems.h>
ISR_LOCK_DEFINE(, _Timecounter_Lock, "Timecounter")
#define _Timecounter_Release(lock_context) \
  _ISR_lock_Release_and_ISR_enable(&_Timecounter_Lock, lock_context)
#define hz rtems_clock_get_ticks_per_second()
#define printf(...)
#define log(...)

static inline void
atomic_thread_fence_acq(void)
{

        _Atomic_Fence(ATOMIC_ORDER_ACQUIRE);
}

static inline void
atomic_thread_fence_rel(void)
{

        _Atomic_Fence(ATOMIC_ORDER_RELEASE);
}

static inline u_int
atomic_load_int(Atomic_Uint *i)
{

        return (_Atomic_Load_uint(i, ATOMIC_ORDER_RELAXED));
}

static inline u_int
atomic_load_acq_int(Atomic_Uint *i)
{

        return (_Atomic_Load_uint(i, ATOMIC_ORDER_ACQUIRE));
}

static inline void
atomic_store_rel_int(Atomic_Uint *i, u_int val)
{

        _Atomic_Store_uint(i, val, ATOMIC_ORDER_RELEASE);
}

static inline void *
atomic_load_ptr(void *ptr)
{

        return ((void *)_Atomic_Load_uintptr(ptr, ATOMIC_ORDER_RELAXED));
}

static Timecounter_NTP_update_second _Timecounter_NTP_update_second_handler;

void
_Timecounter_Set_NTP_update_second(Timecounter_NTP_update_second handler)
{

        _Timecounter_NTP_update_second_handler = handler;
}

#define ntp_update_second(a, b) (*ntp_update_second_handler)(a, b)
#endif /* __rtems__ */

/*
 * 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

/*
 * 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.
 */

static uint32_t
dummy_get_timecount(struct timecounter *tc)
{
#ifndef __rtems__
        static uint32_t now;

        return (++now);
#else /* __rtems__ */
        return 0;
#endif /* __rtems__ */
}

static struct timecounter dummy_timecounter = {
#ifndef __rtems__
        dummy_get_timecount, 0, ~0u, 1000000, "dummy", -1000000
#else /* __rtems__ */
        dummy_get_timecount, ~(uint32_t)0, 1000000, "dummy", -1000000
#endif /* __rtems__ */
};

struct timehands {
        /* These fields must be initialized by the driver. */
        struct timecounter      *th_counter;
        int64_t                 th_adjustment;
        uint64_t                th_scale;
        uint32_t                th_large_delta;
        uint32_t                th_offset_count;
        struct bintime          th_offset;
        struct bintime          th_bintime;
        struct timeval          th_microtime;
        struct timespec         th_nanotime;
        struct bintime          th_boottime;
        /* Fields not to be copied in tc_windup start with th_generation. */
#ifndef __rtems__
        u_int                   th_generation;
#else /* __rtems__ */
        Atomic_Uint             th_generation;
#endif /* __rtems__ */
        struct timehands        *th_next;
};

#ifndef __rtems__
static struct timehands ths[16] = {
    [0] =  {
        .th_counter = &dummy_timecounter,
        .th_scale = (uint64_t)-1 / 1000000,
        .th_large_delta = 1000000,
        .th_offset = { .sec = 1 },
        .th_generation = 1,
    },
};

static struct timehands *volatile timehands = &ths[0];
struct timecounter *timecounter = &dummy_timecounter;
static struct timecounter *timecounters = &dummy_timecounter;

/* Mutex to protect the timecounter list. */
static struct mtx tc_lock;

int tc_min_ticktock_freq = 1;
#else /* __rtems__ */
/*
 * In FreeBSD, the timehands count is a tuning option from two to 16.  The
 * tuning option was added since it is inexpensive and some FreeBSD users asked
 * for it to play around.  The default value is two.  One system which did not
 * work with two timehands was a system with one processor and a specific PPS
 * device.
 *
 * For RTEMS, in uniprocessor configurations, just use one timehand since the
 * update is done with interrupt disabled.
 *
 * In SMP configurations, use a fixed set of two timehands until someone
 * reports an issue.
 */
#if defined(RTEMS_SMP)
static struct timehands th0;
static struct timehands th1 = {
        .th_next = &th0
};
#endif
static struct timehands th0 = {
        .th_counter = &dummy_timecounter,
        .th_scale = (uint64_t)-1 / 1000000,
        .th_offset = { .sec = 1 },
        .th_large_delta = 1000000,
        .th_generation = UINT_MAX,
#ifdef __rtems__
        .th_bintime = { .sec = TOD_SECONDS_1970_THROUGH_1988 },
        .th_microtime = { TOD_SECONDS_1970_THROUGH_1988, 0 },
        .th_nanotime = { TOD_SECONDS_1970_THROUGH_1988, 0 },
        .th_boottime = { .sec = TOD_SECONDS_1970_THROUGH_1988 - 1 },
#endif /* __rtems__ */
#if defined(RTEMS_SMP)
        .th_next = &th1
#else
        .th_next = &th0
#endif
};

static struct timehands *volatile timehands = &th0;
struct timecounter *timecounter = &dummy_timecounter;
#endif /* __rtems__ */

#ifndef __rtems__
volatile time_t time_second = 1;
volatile time_t time_uptime = 1;
#else /* __rtems__ */
volatile time_t time_second = TOD_SECONDS_1970_THROUGH_1988;
volatile int32_t time_uptime = 1;
#endif /* __rtems__ */

#ifndef __rtems__
/*
 * The system time is always computed by summing the estimated boot time and the
 * system uptime. The timehands track boot time, but it changes when the system
 * time is set by the user, stepped by ntpd or adjusted when resuming. It
 * is set to new_time - uptime.
 */
static int sysctl_kern_boottime(SYSCTL_HANDLER_ARGS);
SYSCTL_PROC(_kern, KERN_BOOTTIME, boottime,
    CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_MPSAFE, NULL, 0,
    sysctl_kern_boottime, "S,timeval",
    "Estimated system boottime");

SYSCTL_NODE(_kern, OID_AUTO, timecounter, CTLFLAG_RW | CTLFLAG_MPSAFE, 0,
    "");
static SYSCTL_NODE(_kern_timecounter, OID_AUTO, tc,
    CTLFLAG_RW | CTLFLAG_MPSAFE, 0,
    "");

static int timestepwarnings;
SYSCTL_INT(_kern_timecounter, OID_AUTO, stepwarnings, CTLFLAG_RWTUN,
    &timestepwarnings, 0, "Log time steps");

static int timehands_count = 2;
SYSCTL_INT(_kern_timecounter, OID_AUTO, timehands_count,
    CTLFLAG_RDTUN | CTLFLAG_NOFETCH,
    &timehands_count, 0, "Count of timehands in rotation");

struct bintime bt_timethreshold;
struct bintime bt_tickthreshold;
sbintime_t sbt_timethreshold;
sbintime_t sbt_tickthreshold;
struct bintime tc_tick_bt;
sbintime_t tc_tick_sbt;
int tc_precexp;
int tc_timepercentage = TC_DEFAULTPERC;
static int sysctl_kern_timecounter_adjprecision(SYSCTL_HANDLER_ARGS);
SYSCTL_PROC(_kern_timecounter, OID_AUTO, alloweddeviation,
    CTLTYPE_INT | CTLFLAG_RWTUN | CTLFLAG_MPSAFE, 0, 0,
    sysctl_kern_timecounter_adjprecision, "I",
    "Allowed time interval deviation in percents");

volatile int rtc_generation = 1;

static int tc_chosen;   /* Non-zero if a specific tc was chosen via sysctl. */
static char tc_from_tunable[16];
#endif /* __rtems__ */

static void tc_windup(struct bintime *new_boottimebin);
#ifndef __rtems__
static void cpu_tick_calibrate(int);
#else /* __rtems__ */
static void _Timecounter_Windup(struct bintime *new_boottimebin,
    ISR_lock_Context *lock_context);
#endif /* __rtems__ */

void dtrace_getnanotime(struct timespec *tsp);
void dtrace_getnanouptime(struct timespec *tsp);

#ifndef __rtems__
static int
sysctl_kern_boottime(SYSCTL_HANDLER_ARGS)
{
        struct timeval boottime;

        getboottime(&boottime);

/* i386 is the only arch which uses a 32bits time_t */
#ifdef __amd64__
#ifdef SCTL_MASK32
        int tv[2];

        if (req->flags & SCTL_MASK32) {
                tv[0] = boottime.tv_sec;
                tv[1] = boottime.tv_usec;
                return (SYSCTL_OUT(req, tv, sizeof(tv)));
        }
#endif
#endif
        return (SYSCTL_OUT(req, &boottime, sizeof(boottime)));
}

static int
sysctl_kern_timecounter_get(SYSCTL_HANDLER_ARGS)
{
        uint32_t ncount;
        struct timecounter *tc = arg1;

        ncount = tc->tc_get_timecount(tc);
        return (sysctl_handle_int(oidp, &ncount, 0, req));
}

static int
sysctl_kern_timecounter_freq(SYSCTL_HANDLER_ARGS)
{
        uint64_t freq;
        struct timecounter *tc = arg1;

        freq = tc->tc_frequency;
        return (sysctl_handle_64(oidp, &freq, 0, req));
}
#endif /* __rtems__ */

/*
 * Return the difference between the timehands' counter value now and what
 * was when we copied it to the timehands' offset_count.
 */
static __inline uint32_t
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);
}

static __inline void
bintime_add_tc_delta(struct bintime *bt, uint64_t scale,
    uint64_t large_delta, uint64_t delta)
{
        uint64_t x;

        if (__predict_false(delta >= large_delta)) {
                /* Avoid overflow for scale * delta. */
                x = (scale >> 32) * delta;
                bt->sec += x >> 32;
                bintime_addx(bt, x << 32);
                bintime_addx(bt, (scale & 0xffffffff) * delta);
        } else {
                bintime_addx(bt, scale * delta);
        }
}

/*
 * 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.
 */

static __inline void
bintime_off(struct bintime *bt, u_int off)
{
        struct timehands *th;
        struct bintime *btp;
        uint64_t scale;
#ifndef __rtems__
        u_int delta, gen, large_delta;
#else /* __rtems__ */
        uint32_t delta, large_delta;
        u_int gen;
#endif /* __rtems__ */

        do {
                th = timehands;
                gen = atomic_load_acq_int(&th->th_generation);
                btp = (struct bintime *)((vm_offset_t)th + off);
                *bt = *btp;
                scale = th->th_scale;
                delta = tc_delta(th);
                large_delta = th->th_large_delta;
                atomic_thread_fence_acq();
#if defined(RTEMS_SMP)
        } while (gen == 0 || gen != th->th_generation);
#else
        } while (gen != th->th_generation);
#endif

        bintime_add_tc_delta(bt, scale, large_delta, delta);
}
#define GETTHBINTIME(dst, member)                                       \
do {                                                                    \
        _Static_assert(_Generic(((struct timehands *)NULL)->member,     \
            struct bintime: 1, default: 0) == 1,                        \
            "struct timehands member is not of struct bintime type");   \
        bintime_off(dst, __offsetof(struct timehands, member));         \
} while (0)

static __inline void
getthmember(void *out, size_t out_size, u_int off)
{
        struct timehands *th;
        u_int gen;

        do {
                th = timehands;
                gen = atomic_load_acq_int(&th->th_generation);
                memcpy(out, (char *)th + off, out_size);
                atomic_thread_fence_acq();
#if defined(RTEMS_SMP)
        } while (gen == 0 || gen != th->th_generation);
#else
        } while (gen != th->th_generation);
#endif
}
#define GETTHMEMBER(dst, member)                                        \
do {                                                                    \
        _Static_assert(_Generic(*dst,                                   \
            __typeof(((struct timehands *)NULL)->member): 1,            \
            default: 0) == 1,                                           \
            "*dst and struct timehands member have different types");   \
        getthmember(dst, sizeof(*dst), __offsetof(struct timehands,     \
            member));                                                   \
} while (0)

#ifdef FFCLOCK
void
fbclock_binuptime(struct bintime *bt)
{

        GETTHBINTIME(bt, th_offset);
}

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

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

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

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

void
fbclock_bintime(struct bintime *bt)
{

        GETTHBINTIME(bt, th_bintime);
}

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

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

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

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

void
fbclock_getbinuptime(struct bintime *bt)
{

        GETTHMEMBER(bt, th_offset);
}

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

        GETTHMEMBER(&bt, th_offset);
        bintime2timespec(&bt, tsp);
}

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

        GETTHMEMBER(&bt, th_offset);
        bintime2timeval(&bt, tvp);
}

void
fbclock_getbintime(struct bintime *bt)
{

        GETTHMEMBER(bt, th_bintime);
}

void
fbclock_getnanotime(struct timespec *tsp)
{

        GETTHMEMBER(tsp, th_nanotime);
}

void
fbclock_getmicrotime(struct timeval *tvp)
{

        GETTHMEMBER(tvp, th_microtime);
}
#else /* !FFCLOCK */

void
binuptime(struct bintime *bt)
{

        GETTHBINTIME(bt, th_offset);
}
#ifdef __rtems__
sbintime_t
_Timecounter_Sbinuptime(void)
{
        struct timehands *th;
        sbintime_t sbt;
        uint64_t scale;
        uint32_t delta;
        uint32_t large_delta;
        u_int gen;

        do {
                th = timehands;
                gen = atomic_load_acq_int(&th->th_generation);
                sbt = bttosbt(th->th_offset);
                scale = th->th_scale;
                delta = tc_delta(th);
                large_delta = th->th_large_delta;
                atomic_thread_fence_acq();
#if defined(RTEMS_SMP)
        } while (gen == 0 || gen != th->th_generation);
#else
        } while (gen != th->th_generation);
#endif

        if (__predict_false(delta >= large_delta)) {
                /* Avoid overflow for scale * delta. */
                sbt += (scale >> 32) * delta;
                sbt += ((scale & 0xffffffff) * delta) >> 32;
        } else {
                sbt += (scale * delta) >> 32;
        }

        return (sbt);
}
#endif /* __rtems__ */

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)
{

        GETTHBINTIME(bt, th_bintime);
}

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
getbinuptime(struct bintime *bt)
{

        GETTHMEMBER(bt, th_offset);
}

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

        GETTHMEMBER(&bt, th_offset);
        bintime2timespec(&bt, tsp);
}

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

        GETTHMEMBER(&bt, th_offset);
        bintime2timeval(&bt, tvp);
}

void
getbintime(struct bintime *bt)
{

        GETTHMEMBER(bt, th_bintime);
}

void
getnanotime(struct timespec *tsp)
{

        GETTHMEMBER(tsp, th_nanotime);
}

void
getmicrotime(struct timeval *tvp)
{

        GETTHMEMBER(tvp, th_microtime);
}
#endif /* FFCLOCK */

#ifdef __rtems__
void
rtems_clock_get_boot_time(struct timespec *boottime)
{
        struct bintime boottimebin;

        getboottimebin(&boottimebin);
        bintime2timespec(&boottimebin, boottime);
}
#endif /* __rtems__ */
void
getboottime(struct timeval *boottime)
{
        struct bintime boottimebin;

        getboottimebin(&boottimebin);
        bintime2timeval(&boottimebin, boottime);
}

void
getboottimebin(struct bintime *boottimebin)
{

        GETTHMEMBER(boottimebin, th_boottime);
}

#ifdef FFCLOCK
/*
 * Support for feed-forward synchronization algorithms. This is heavily inspired
 * by the timehands mechanism but kept independent from it. *_windup() functions
 * have some connection to avoid accessing the timecounter hardware more than
 * necessary.
 */

/* Feed-forward clock estimates kept updated by the synchronization daemon. */
struct ffclock_estimate ffclock_estimate;
struct bintime ffclock_boottime;        /* Feed-forward boot time estimate. */
uint32_t ffclock_status;                /* Feed-forward clock status. */
int8_t ffclock_updated;                 /* New estimates are available. */
struct mtx ffclock_mtx;                 /* Mutex on ffclock_estimate. */

struct fftimehands {
        struct ffclock_estimate cest;
        struct bintime          tick_time;
        struct bintime          tick_time_lerp;
        ffcounter               tick_ffcount;
        uint64_t                period_lerp;
        volatile uint8_t        gen;
        struct fftimehands      *next;
};

#define NUM_ELEMENTS(x) (sizeof(x) / sizeof(*x))

static struct fftimehands ffth[10];
static struct fftimehands *volatile fftimehands = ffth;

static void
ffclock_init(void)
{
        struct fftimehands *cur;
        struct fftimehands *last;

        memset(ffth, 0, sizeof(ffth));

        last = ffth + NUM_ELEMENTS(ffth) - 1;
        for (cur = ffth; cur < last; cur++)
                cur->next = cur + 1;
        last->next = ffth;

        ffclock_updated = 0;
        ffclock_status = FFCLOCK_STA_UNSYNC;
        mtx_init(&ffclock_mtx, "ffclock lock", NULL, MTX_DEF);
}

/*
 * Reset the feed-forward clock estimates. Called from inittodr() to get things
 * kick started and uses the timecounter nominal frequency as a first period
 * estimate. Note: this function may be called several time just after boot.
 * Note: this is the only function that sets the value of boot time for the
 * monotonic (i.e. uptime) version of the feed-forward clock.
 */
void
ffclock_reset_clock(struct timespec *ts)
{
        struct timecounter *tc;
        struct ffclock_estimate cest;

        tc = timehands->th_counter;
        memset(&cest, 0, sizeof(struct ffclock_estimate));

        timespec2bintime(ts, &ffclock_boottime);
        timespec2bintime(ts, &(cest.update_time));
        ffclock_read_counter(&cest.update_ffcount);
        cest.leapsec_next = 0;
        cest.period = ((1ULL << 63) / tc->tc_frequency) << 1;
        cest.errb_abs = 0;
        cest.errb_rate = 0;
        cest.status = FFCLOCK_STA_UNSYNC;
        cest.leapsec_total = 0;
        cest.leapsec = 0;

        mtx_lock(&ffclock_mtx);
        bcopy(&cest, &ffclock_estimate, sizeof(struct ffclock_estimate));
        ffclock_updated = INT8_MAX;
        mtx_unlock(&ffclock_mtx);

        printf("ffclock reset: %s (%llu Hz), time = %ld.%09lu\n", tc->tc_name,
            (unsigned long long)tc->tc_frequency, (long)ts->tv_sec,
            (unsigned long)ts->tv_nsec);
}

/*
 * Sub-routine to convert a time interval measured in RAW counter units to time
 * in seconds stored in bintime format.
 * NOTE: bintime_mul requires u_int, but the value of the ffcounter may be
 * larger than the max value of u_int (on 32 bit architecture). Loop to consume
 * extra cycles.
 */
static void
ffclock_convert_delta(ffcounter ffdelta, uint64_t period, struct bintime *bt)
{
        struct bintime bt2;
        ffcounter delta, delta_max;

        delta_max = (1ULL << (8 * sizeof(unsigned int))) - 1;
        bintime_clear(bt);
        do {
                if (ffdelta > delta_max)
                        delta = delta_max;
                else
                        delta = ffdelta;
                bt2.sec = 0;
                bt2.frac = period;
                bintime_mul(&bt2, (unsigned int)delta);
                bintime_add(bt, &bt2);
                ffdelta -= delta;
        } while (ffdelta > 0);
}

/*
 * Update the fftimehands.
 * Push the tick ffcount and time(s) forward based on current clock estimate.
 * The conversion from ffcounter to bintime relies on the difference clock
 * principle, whose accuracy relies on computing small time intervals. If a new
 * clock estimate has been passed by the synchronisation daemon, make it
 * current, and compute the linear interpolation for monotonic time if needed.
 */
static void
ffclock_windup(unsigned int delta)
{
        struct ffclock_estimate *cest;
        struct fftimehands *ffth;
        struct bintime bt, gap_lerp;
        ffcounter ffdelta;
        uint64_t frac;
        unsigned int polling;
        uint8_t forward_jump, ogen;

        /*
         * Pick the next timehand, copy current ffclock estimates and move tick
         * times and counter forward.
         */
        forward_jump = 0;
        ffth = fftimehands->next;
        ogen = ffth->gen;
        ffth->gen = 0;
        cest = &ffth->cest;
        bcopy(&fftimehands->cest, cest, sizeof(struct ffclock_estimate));
        ffdelta = (ffcounter)delta;
        ffth->period_lerp = fftimehands->period_lerp;

        ffth->tick_time = fftimehands->tick_time;
        ffclock_convert_delta(ffdelta, cest->period, &bt);
        bintime_add(&ffth->tick_time, &bt);

        ffth->tick_time_lerp = fftimehands->tick_time_lerp;
        ffclock_convert_delta(ffdelta, ffth->period_lerp, &bt);
        bintime_add(&ffth->tick_time_lerp, &bt);

        ffth->tick_ffcount = fftimehands->tick_ffcount + ffdelta;

        /*
         * Assess the status of the clock, if the last update is too old, it is
         * likely the synchronisation daemon is dead and the clock is free
         * running.
         */
        if (ffclock_updated == 0) {
                ffdelta = ffth->tick_ffcount - cest->update_ffcount;
                ffclock_convert_delta(ffdelta, cest->period, &bt);
                if (bt.sec > 2 * FFCLOCK_SKM_SCALE)
                        ffclock_status |= FFCLOCK_STA_UNSYNC;
        }

        /*
         * If available, grab updated clock estimates and make them current.
         * Recompute time at this tick using the updated estimates. The clock
         * estimates passed the feed-forward synchronisation daemon may result
         * in time conversion that is not monotonically increasing (just after
         * the update). time_lerp is a particular linear interpolation over the
         * synchronisation algo polling period that ensures monotonicity for the
         * clock ids requesting it.
         */
        if (ffclock_updated > 0) {
                bcopy(&ffclock_estimate, cest, sizeof(struct ffclock_estimate));
                ffdelta = ffth->tick_ffcount - cest->update_ffcount;
                ffth->tick_time = cest->update_time;
                ffclock_convert_delta(ffdelta, cest->period, &bt);
                bintime_add(&ffth->tick_time, &bt);

                /* ffclock_reset sets ffclock_updated to INT8_MAX */
                if (ffclock_updated == INT8_MAX)
                        ffth->tick_time_lerp = ffth->tick_time;

                if (bintime_cmp(&ffth->tick_time, &ffth->tick_time_lerp, >))
                        forward_jump = 1;
                else
                        forward_jump = 0;

                bintime_clear(&gap_lerp);
                if (forward_jump) {
                        gap_lerp = ffth->tick_time;
                        bintime_sub(&gap_lerp, &ffth->tick_time_lerp);
                } else {
                        gap_lerp = ffth->tick_time_lerp;
                        bintime_sub(&gap_lerp, &ffth->tick_time);
                }

                /*
                 * The reset from the RTC clock may be far from accurate, and
                 * reducing the gap between real time and interpolated time
                 * could take a very long time if the interpolated clock insists
                 * on strict monotonicity. The clock is reset under very strict
                 * conditions (kernel time is known to be wrong and
                 * synchronization daemon has been restarted recently.
                 * ffclock_boottime absorbs the jump to ensure boot time is
                 * correct and uptime functions stay consistent.
                 */
                if (((ffclock_status & FFCLOCK_STA_UNSYNC) == FFCLOCK_STA_UNSYNC) &&
                    ((cest->status & FFCLOCK_STA_UNSYNC) == 0) &&
                    ((cest->status & FFCLOCK_STA_WARMUP) == FFCLOCK_STA_WARMUP)) {
                        if (forward_jump)
                                bintime_add(&ffclock_boottime, &gap_lerp);
                        else
                                bintime_sub(&ffclock_boottime, &gap_lerp);
                        ffth->tick_time_lerp = ffth->tick_time;
                        bintime_clear(&gap_lerp);
                }

                ffclock_status = cest->status;
                ffth->period_lerp = cest->period;

                /*
                 * Compute corrected period used for the linear interpolation of
                 * time. The rate of linear interpolation is capped to 5000PPM
                 * (5ms/s).
                 */
                if (bintime_isset(&gap_lerp)) {
                        ffdelta = cest->update_ffcount;
                        ffdelta -= fftimehands->cest.update_ffcount;
                        ffclock_convert_delta(ffdelta, cest->period, &bt);
                        polling = bt.sec;
                        bt.sec = 0;
                        bt.frac = 5000000 * (uint64_t)18446744073LL;
                        bintime_mul(&bt, polling);
                        if (bintime_cmp(&gap_lerp, &bt, >))
                                gap_lerp = bt;

                        /* Approximate 1 sec by 1-(1/2^64) to ease arithmetic */
                        frac = 0;
                        if (gap_lerp.sec > 0) {
                                frac -= 1;
                                frac /= ffdelta / gap_lerp.sec;
                        }
                        frac += gap_lerp.frac / ffdelta;

                        if (forward_jump)
                                ffth->period_lerp += frac;
                        else
                                ffth->period_lerp -= frac;
                }

                ffclock_updated = 0;
        }
        if (++ogen == 0)
                ogen = 1;
        ffth->gen = ogen;
        fftimehands = ffth;
}

/*
 * Adjust the fftimehands when the timecounter is changed. Stating the obvious,
 * the old and new hardware counter cannot be read simultaneously. tc_windup()
 * does read the two counters 'back to back', but a few cycles are effectively
 * lost, and not accumulated in tick_ffcount. This is a fairly radical
 * operation for a feed-forward synchronization daemon, and it is its job to not
 * pushing irrelevant data to the kernel. Because there is no locking here,
 * simply force to ignore pending or next update to give daemon a chance to
 * realize the counter has changed.
 */
static void
ffclock_change_tc(struct timehands *th)
{
        struct fftimehands *ffth;
        struct ffclock_estimate *cest;
        struct timecounter *tc;
        uint8_t ogen;

        tc = th->th_counter;
        ffth = fftimehands->next;
        ogen = ffth->gen;
        ffth->gen = 0;

        cest = &ffth->cest;
        bcopy(&(fftimehands->cest), cest, sizeof(struct ffclock_estimate));
        cest->period = ((1ULL << 63) / tc->tc_frequency ) << 1;
        cest->errb_abs = 0;
        cest->errb_rate = 0;
        cest->status |= FFCLOCK_STA_UNSYNC;

        ffth->tick_ffcount = fftimehands->tick_ffcount;
        ffth->tick_time_lerp = fftimehands->tick_time_lerp;
        ffth->tick_time = fftimehands->tick_time;
        ffth->period_lerp = cest->period;

        /* Do not lock but ignore next update from synchronization daemon. */
        ffclock_updated--;

        if (++ogen == 0)
                ogen = 1;
        ffth->gen = ogen;
        fftimehands = ffth;
}

/*
 * Retrieve feed-forward counter and time of last kernel tick.
 */
void
ffclock_last_tick(ffcounter *ffcount, struct bintime *bt, uint32_t flags)
{
        struct fftimehands *ffth;
        uint8_t gen;

        /*
         * No locking but check generation has not changed. Also need to make
         * sure ffdelta is positive, i.e. ffcount > tick_ffcount.
         */
        do {
                ffth = fftimehands;
                gen = ffth->gen;
                if ((flags & FFCLOCK_LERP) == FFCLOCK_LERP)
                        *bt = ffth->tick_time_lerp;
                else
                        *bt = ffth->tick_time;
                *ffcount = ffth->tick_ffcount;
        } while (gen == 0 || gen != ffth->gen);
}

/*
 * Absolute clock conversion. Low level function to convert ffcounter to
 * bintime. The ffcounter is converted using the current ffclock period estimate
 * or the "interpolated period" to ensure monotonicity.
 * NOTE: this conversion may have been deferred, and the clock updated since the
 * hardware counter has been read.
 */
void
ffclock_convert_abs(ffcounter ffcount, struct bintime *bt, uint32_t flags)
{
        struct fftimehands *ffth;
        struct bintime bt2;
        ffcounter ffdelta;
        uint8_t gen;

        /*
         * No locking but check generation has not changed. Also need to make
         * sure ffdelta is positive, i.e. ffcount > tick_ffcount.
         */
        do {
                ffth = fftimehands;
                gen = ffth->gen;
                if (ffcount > ffth->tick_ffcount)
                        ffdelta = ffcount - ffth->tick_ffcount;
                else
                        ffdelta = ffth->tick_ffcount - ffcount;

                if ((flags & FFCLOCK_LERP) == FFCLOCK_LERP) {
                        *bt = ffth->tick_time_lerp;
                        ffclock_convert_delta(ffdelta, ffth->period_lerp, &bt2);
                } else {
                        *bt = ffth->tick_time;
                        ffclock_convert_delta(ffdelta, ffth->cest.period, &bt2);
                }

                if (ffcount > ffth->tick_ffcount)
                        bintime_add(bt, &bt2);
                else
                        bintime_sub(bt, &bt2);
        } while (gen == 0 || gen != ffth->gen);
}

/*
 * Difference clock conversion.
 * Low level function to Convert a time interval measured in RAW counter units
 * into bintime. The difference clock allows measuring small intervals much more
 * reliably than the absolute clock.
 */
void
ffclock_convert_diff(ffcounter ffdelta, struct bintime *bt)
{
        struct fftimehands *ffth;
        uint8_t gen;

        /* No locking but check generation has not changed. */
        do {
                ffth = fftimehands;
                gen = ffth->gen;
                ffclock_convert_delta(ffdelta, ffth->cest.period, bt);
        } while (gen == 0 || gen != ffth->gen);
}

/*
 * Access to current ffcounter value.
 */
void
ffclock_read_counter(ffcounter *ffcount)
{
        struct timehands *th;
        struct fftimehands *ffth;
        unsigned int gen, delta;

        /*
         * ffclock_windup() called from tc_windup(), safe to rely on
         * th->th_generation only, for correct delta and ffcounter.
         */
        do {
                th = timehands;
                gen = atomic_load_acq_int(&th->th_generation);
                ffth = fftimehands;
                delta = tc_delta(th);
                *ffcount = ffth->tick_ffcount;
                atomic_thread_fence_acq();
        } while (gen == 0 || gen != th->th_generation);

        *ffcount += delta;
}

void
binuptime(struct bintime *bt)
{

        binuptime_fromclock(bt, sysclock_active);
}

void
nanouptime(struct timespec *tsp)
{

        nanouptime_fromclock(tsp, sysclock_active);
}

void
microuptime(struct timeval *tvp)
{

        microuptime_fromclock(tvp, sysclock_active);
}

void
bintime(struct bintime *bt)
{

        bintime_fromclock(bt, sysclock_active);
}

void
nanotime(struct timespec *tsp)
{

        nanotime_fromclock(tsp, sysclock_active);
}

void
microtime(struct timeval *tvp)
{

        microtime_fromclock(tvp, sysclock_active);
}

void
getbinuptime(struct bintime *bt)
{

        getbinuptime_fromclock(bt, sysclock_active);
}

void
getnanouptime(struct timespec *tsp)
{

        getnanouptime_fromclock(tsp, sysclock_active);
}

void
getmicrouptime(struct timeval *tvp)
{

        getmicrouptime_fromclock(tvp, sysclock_active);
}

void
getbintime(struct bintime *bt)
{

        getbintime_fromclock(bt, sysclock_active);
}

void
getnanotime(struct timespec *tsp)
{

        getnanotime_fromclock(tsp, sysclock_active);
}

void
getmicrotime(struct timeval *tvp)
{

        getmicrouptime_fromclock(tvp, sysclock_active);
}

#endif /* FFCLOCK */

#ifndef __rtems__
/*
 * This is a clone of getnanotime and used for walltimestamps.
 * The dtrace_ prefix prevents fbt from creating probes for
 * it so walltimestamp can be safely used in all fbt probes.
 */
void
dtrace_getnanotime(struct timespec *tsp)
{

        GETTHMEMBER(tsp, th_nanotime);
}

/*
 * This is a clone of getnanouptime used for time since boot.
 * The dtrace_ prefix prevents fbt from creating probes for
 * it so an uptime that can be safely used in all fbt probes.
 */
void
dtrace_getnanouptime(struct timespec *tsp)
{
        struct bintime bt;

        GETTHMEMBER(&bt, th_offset);
        bintime2timespec(&bt, tsp);
}
#endif /* __rtems__ */

#ifdef FFCLOCK
/*
 * System clock currently providing time to the system. Modifiable via sysctl
 * when the FFCLOCK option is defined.
 */
int sysclock_active = SYSCLOCK_FBCK;
#endif

/* Internal NTP status and error estimates. */
extern int time_status;
extern long time_esterror;

#ifndef __rtems__
/*
 * Take a snapshot of sysclock data which can be used to compare system clocks
 * and generate timestamps after the fact.
 */
void
sysclock_getsnapshot(struct sysclock_snap *clock_snap, int fast)
{
        struct fbclock_info *fbi;
        struct timehands *th;
        struct bintime bt;
        unsigned int delta, gen;
#ifdef FFCLOCK
        ffcounter ffcount;
        struct fftimehands *ffth;
        struct ffclock_info *ffi;
        struct ffclock_estimate cest;

        ffi = &clock_snap->ff_info;
#endif

        fbi = &clock_snap->fb_info;
        delta = 0;

        do {
                th = timehands;
                gen = atomic_load_acq_int(&th->th_generation);
                fbi->th_scale = th->th_scale;
                fbi->tick_time = th->th_offset;
#ifdef FFCLOCK
                ffth = fftimehands;
                ffi->tick_time = ffth->tick_time_lerp;
                ffi->tick_time_lerp = ffth->tick_time_lerp;
                ffi->period = ffth->cest.period;
                ffi->period_lerp = ffth->period_lerp;
                clock_snap->ffcount = ffth->tick_ffcount;
                cest = ffth->cest;
#endif
                if (!fast)
                        delta = tc_delta(th);
                atomic_thread_fence_acq();
        } while (gen == 0 || gen != th->th_generation);

        clock_snap->delta = delta;
#ifdef FFCLOCK
        clock_snap->sysclock_active = sysclock_active;
#endif

        /* Record feedback clock status and error. */
        clock_snap->fb_info.status = time_status;
        /* XXX: Very crude estimate of feedback clock error. */
        bt.sec = time_esterror / 1000000;
        bt.frac = ((time_esterror - bt.sec) * 1000000) *
            (uint64_t)18446744073709ULL;
        clock_snap->fb_info.error = bt;

#ifdef FFCLOCK
        if (!fast)
                clock_snap->ffcount += delta;

        /* Record feed-forward clock leap second adjustment. */
        ffi->leapsec_adjustment = cest.leapsec_total;
        if (clock_snap->ffcount > cest.leapsec_next)
                ffi->leapsec_adjustment -= cest.leapsec;

        /* Record feed-forward clock status and error. */
        clock_snap->ff_info.status = cest.status;
        ffcount = clock_snap->ffcount - cest.update_ffcount;
        ffclock_convert_delta(ffcount, cest.period, &bt);
        /* 18446744073709 = int(2^64/1e12), err_bound_rate in [ps/s]. */
        bintime_mul(&bt, cest.errb_rate * (uint64_t)18446744073709ULL);
        /* 18446744073 = int(2^64 / 1e9), since err_abs in [ns]. */
        bintime_addx(&bt, cest.errb_abs * (uint64_t)18446744073ULL);
        clock_snap->ff_info.error = bt;
#endif
}

/*
 * Convert a sysclock snapshot into a struct bintime based on the specified
 * clock source and flags.
 */
int
sysclock_snap2bintime(struct sysclock_snap *cs, struct bintime *bt,
    int whichclock, uint32_t flags)
{
        struct bintime boottimebin;
#ifdef FFCLOCK
        struct bintime bt2;
        uint64_t period;
#endif

        switch (whichclock) {
        case SYSCLOCK_FBCK:
                *bt = cs->fb_info.tick_time;

                /* If snapshot was created with !fast, delta will be >0. */
                if (cs->delta > 0)
                        bintime_addx(bt, cs->fb_info.th_scale * cs->delta);

                if ((flags & FBCLOCK_UPTIME) == 0) {
                        getboottimebin(&boottimebin);
                        bintime_add(bt, &boottimebin);
                }
                break;
#ifdef FFCLOCK
        case SYSCLOCK_FFWD:
                if (flags & FFCLOCK_LERP) {
                        *bt = cs->ff_info.tick_time_lerp;
                        period = cs->ff_info.period_lerp;
                } else {
                        *bt = cs->ff_info.tick_time;
                        period = cs->ff_info.period;
                }

                /* If snapshot was created with !fast, delta will be >0. */
                if (cs->delta > 0) {
                        ffclock_convert_delta(cs->delta, period, &bt2);
                        bintime_add(bt, &bt2);
                }

                /* Leap second adjustment. */
                if (flags & FFCLOCK_LEAPSEC)
                        bt->sec -= cs->ff_info.leapsec_adjustment;

                /* Boot time adjustment, for uptime/monotonic clocks. */
                if (flags & FFCLOCK_UPTIME)
                        bintime_sub(bt, &ffclock_boottime);
                break;
#endif
        default:
                return (EINVAL);
                break;
        }

        return (0);
}
#endif /* __rtems__ */

/*
 * Initialize a new timecounter and possibly use it.
 */
void
tc_init(struct timecounter *tc)
{
#ifndef __rtems__
        uint32_t u;
        struct sysctl_oid *tc_root;

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

        /*
         * Set up sysctl tree for this counter.
         */
        tc_root = SYSCTL_ADD_NODE_WITH_LABEL(NULL,
            SYSCTL_STATIC_CHILDREN(_kern_timecounter_tc), OID_AUTO, tc->tc_name,
            CTLFLAG_RW | CTLFLAG_MPSAFE, 0,
            "timecounter description", "timecounter");
        SYSCTL_ADD_UINT(NULL, SYSCTL_CHILDREN(tc_root), OID_AUTO,
            "mask", CTLFLAG_RD, &(tc->tc_counter_mask), 0,
            "mask for implemented bits");
        SYSCTL_ADD_PROC(NULL, SYSCTL_CHILDREN(tc_root), OID_AUTO,
            "counter", CTLTYPE_UINT | CTLFLAG_RD | CTLFLAG_MPSAFE, tc,
            sizeof(*tc), sysctl_kern_timecounter_get, "IU",
            "current timecounter value");
        SYSCTL_ADD_PROC(NULL, SYSCTL_CHILDREN(tc_root), OID_AUTO,
            "frequency", CTLTYPE_U64 | CTLFLAG_RD | CTLFLAG_MPSAFE, tc,
            sizeof(*tc), sysctl_kern_timecounter_freq, "QU",
            "timecounter frequency");
        SYSCTL_ADD_INT(NULL, SYSCTL_CHILDREN(tc_root), OID_AUTO,
            "quality", CTLFLAG_RD, &(tc->tc_quality), 0,
            "goodness of time counter");

        mtx_lock(&tc_lock);
        tc->tc_next = timecounters;
        timecounters = tc;

        /*
         * Do not automatically switch if the current tc was specifically
         * chosen.  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_chosen)
                goto unlock;
        if (tc->tc_quality < 0)
                goto unlock;
        if (tc_from_tunable[0] != '\0' &&
            strcmp(tc->tc_name, tc_from_tunable) == 0) {
                tc_chosen = 1;
                tc_from_tunable[0] = '\0';
        } else {
                if (tc->tc_quality < timecounter->tc_quality)
                        goto unlock;
                if (tc->tc_quality == timecounter->tc_quality &&
                    tc->tc_frequency < timecounter->tc_frequency)
                        goto unlock;
        }
        (void)tc->tc_get_timecount(tc);
        timecounter = tc;
unlock:
        mtx_unlock(&tc_lock);
#else /* __rtems__ */
        if (tc->tc_quality < timecounter->tc_quality)
                return;
        if (tc->tc_quality == timecounter->tc_quality &&
            tc->tc_frequency < timecounter->tc_frequency)
                return;
        timecounter = tc;
        tc_windup(NULL);
#endif /* __rtems__ */
}

/* Report the frequency of the current timecounter. */
uint64_t
tc_getfrequency(void)
{

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

#ifndef __rtems__
static bool
sleeping_on_old_rtc(struct thread *td)
{

        /*
         * td_rtcgen is modified by curthread when it is running,
         * and by other threads in this function.  By finding the thread
         * on a sleepqueue and holding the lock on the sleepqueue
         * chain, we guarantee that the thread is not running and that
         * modifying td_rtcgen is safe.  Setting td_rtcgen to zero informs
         * the thread that it was woken due to a real-time clock adjustment.
         * (The declaration of td_rtcgen refers to this comment.)
         */
        if (td->td_rtcgen != 0 && td->td_rtcgen != rtc_generation) {
                td->td_rtcgen = 0;
                return (true);
        }
        return (false);
}

static struct mtx tc_setclock_mtx;
MTX_SYSINIT(tc_setclock_init, &tc_setclock_mtx, "tcsetc", MTX_SPIN);
#endif /* __rtems__ */

/*
 * Step our concept of UTC.  This is done by modifying our estimate of
 * when we booted.
 */
void
#ifndef __rtems__
tc_setclock(struct timespec *ts)
#else /* __rtems__ */
_Timecounter_Set_clock(const struct bintime *_bt,
    ISR_lock_Context *lock_context)
#endif /* __rtems__ */
{
#ifndef __rtems__
        struct timespec tbef, taft;
#endif /* __rtems__ */
        struct bintime bt, bt2;

#ifndef __rtems__
        timespec2bintime(ts, &bt);
        nanotime(&tbef);
        mtx_lock_spin(&tc_setclock_mtx);
        cpu_tick_calibrate(1);
#else /* __rtems__ */
        bt = *_bt;
#endif /* __rtems__ */
        binuptime(&bt2);
        bintime_sub(&bt, &bt2);

        /* XXX fiddle all the little crinkly bits around the fiords... */
#ifndef __rtems__
        tc_windup(&bt);
        mtx_unlock_spin(&tc_setclock_mtx);

        /* Avoid rtc_generation == 0, since td_rtcgen == 0 is special. */
        atomic_add_rel_int(&rtc_generation, 2);
        sleepq_chains_remove_matching(sleeping_on_old_rtc);
        if (timestepwarnings) {
                nanotime(&taft);
                log(LOG_INFO,
                    "Time stepped from %jd.%09ld to %jd.%09ld (%jd.%09ld)\n",
                    (intmax_t)tbef.tv_sec, tbef.tv_nsec,
                    (intmax_t)taft.tv_sec, taft.tv_nsec,
                    (intmax_t)ts->tv_sec, ts->tv_nsec);
        }
#else /* __rtems__ */
        _Timecounter_Windup(&bt, lock_context);
#endif /* __rtems__ */
}

/*
 * 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, that
 * leaves no 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.
 */
static void
recalculate_scaling_factor_and_large_delta(struct timehands *th)
{
        uint64_t scale;

        scale = (uint64_t)1 << 63;
        scale += (th->th_adjustment / 1024) * 2199;
        scale /= th->th_counter->tc_frequency;
        th->th_scale = scale * 2;
        th->th_large_delta = MIN(((uint64_t)1 << 63) / scale, UINT_MAX);
}

/*
 * 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.
 */
static void
tc_windup(struct bintime *new_boottimebin)
#ifdef __rtems__
{
        ISR_lock_Context lock_context;

        _Timecounter_Acquire(&lock_context);
        _Timecounter_Windup(new_boottimebin, &lock_context);
}

static void
_Timecounter_Windup(struct bintime *new_boottimebin,
    ISR_lock_Context *lock_context)
#endif /* __rtems__ */
{
        struct bintime bt;
        struct timecounter *tc;
        struct timehands *th, *tho;
        uint32_t delta, ncount;
#if defined(RTEMS_SMP)
        u_int ogen;
#endif
        int i;
        time_t t;
#ifdef __rtems__
        Timecounter_NTP_update_second ntp_update_second_handler;
#endif

        /*
         * 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.  We need
         * to ensure that the zero generation is visible before the
         * data updates become visible, which requires release fence.
         * For similar reasons, re-reading of the generation after the
         * data is read should use acquire fence.
         */
        tho = timehands;
#if defined(RTEMS_SMP)
        th = tho->th_next;
        ogen = th->th_generation;
        th->th_generation = 0;
        atomic_thread_fence_rel();
        memcpy(th, tho, offsetof(struct timehands, th_generation));
#else
        th = tho;
#endif
        if (new_boottimebin != NULL)
                th->th_boottime = *new_boottimebin;

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

#ifndef __rtems__
        /*
         * 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 /* __rtems__ */

        /*
         * Deal with NTP second processing.  The 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, &th->th_boottime);
#ifdef __rtems__
        ntp_update_second_handler = _Timecounter_NTP_update_second_handler;
        if (ntp_update_second_handler != NULL) {
#endif /* __rtems__ */
        i = bt.sec - tho->th_microtime.tv_sec;
        if (i > 0) {
                if (i > LARGE_STEP)
                        i = 2;

                do {
                        t = bt.sec;
                        ntp_update_second(&th->th_adjustment, &bt.sec);
                        if (bt.sec != t)
                                th->th_boottime.sec += bt.sec - t;
                        --i;
                } while (i > 0);

                recalculate_scaling_factor_and_large_delta(th);
        }
#ifdef __rtems__
        }
#endif /* __rtems__ */

        /* Update the UTC timestamps used by the get*() functions. */
        th->th_bintime = bt;
        bintime2timeval(&bt, &th->th_microtime);
        bintime2timespec(&bt, &th->th_nanotime);

        /* Now is a good time to change timecounters. */
        if (th->th_counter != tc) {
#ifndef __rtems__
#ifndef __arm__
                if ((tc->tc_flags & TC_FLAGS_C2STOP) != 0)
                        cpu_disable_c2_sleep++;
                if ((th->th_counter->tc_flags & TC_FLAGS_C2STOP) != 0)
                        cpu_disable_c2_sleep--;
#endif
#endif /* __rtems__ */
                th->th_counter = tc;
                th->th_offset_count = ncount;
#ifndef __rtems__
                tc_min_ticktock_freq = max(1, tc->tc_frequency /
                    (((uint64_t)tc->tc_counter_mask + 1) / 3));
#endif /* __rtems__ */
                recalculate_scaling_factor_and_large_delta(th);
#ifdef FFCLOCK
                ffclock_change_tc(th);
#endif
        }

#if defined(RTEMS_SMP)
        /*
         * 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;
        atomic_store_rel_int(&th->th_generation, ogen);
#else
        atomic_store_rel_int(&th->th_generation, th->th_generation + 1);
#endif

        /* Go live with the new struct timehands. */
#ifdef FFCLOCK
        switch (sysclock_active) {
        case SYSCLOCK_FBCK:
#endif
                time_second = th->th_microtime.tv_sec;
                time_uptime = th->th_offset.sec;
#ifdef FFCLOCK
                break;
        case SYSCLOCK_FFWD:
                time_second = fftimehands->tick_time_lerp.sec;
                time_uptime = fftimehands->tick_time_lerp.sec - ffclock_boottime.sec;
                break;
        }
#endif

#if defined(RTEMS_SMP)
        timehands = th;
#endif
#ifndef __rtems__
        timekeep_push_vdso();
#endif /* __rtems__ */
#ifdef __rtems__
        _Timecounter_Release(lock_context);
#endif /* __rtems__ */
}

#ifndef __rtems__
/* Report or change the active timecounter hardware. */
static int
sysctl_kern_timecounter_hardware(SYSCTL_HANDLER_ARGS)
{
        char newname[32];
        struct timecounter *newtc, *tc;
        int error;

        mtx_lock(&tc_lock);
        tc = timecounter;
        strlcpy(newname, tc->tc_name, sizeof(newname));
        mtx_unlock(&tc_lock);

        error = sysctl_handle_string(oidp, &newname[0], sizeof(newname), req);
        if (error != 0 || req->newptr == NULL)
                return (error);

        mtx_lock(&tc_lock);
        /* Record that the tc in use now was specifically chosen. */
        tc_chosen = 1;
        if (strcmp(newname, tc->tc_name) == 0) {
                mtx_unlock(&tc_lock);
                return (0);
        }
        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);

                timecounter = newtc;

                /*
                 * The vdso timehands update is deferred until the next
                 * 'tc_windup()'.
                 *
                 * This is prudent given that 'timekeep_push_vdso()' does not
                 * use any locking and that it can be called in hard interrupt
                 * context via 'tc_windup()'.
                 */
                break;
        }
        mtx_unlock(&tc_lock);
        return (newtc != NULL ? 0 : EINVAL);
}
SYSCTL_PROC(_kern_timecounter, OID_AUTO, hardware,
    CTLTYPE_STRING | CTLFLAG_RWTUN | CTLFLAG_NOFETCH | CTLFLAG_MPSAFE, 0, 0,
    sysctl_kern_timecounter_hardware, "A",
    "Timecounter hardware selected");

/* Report the available timecounter hardware. */
static int
sysctl_kern_timecounter_choice(SYSCTL_HANDLER_ARGS)
{
        struct sbuf sb;
        struct timecounter *tc;
        int error;

        error = sysctl_wire_old_buffer(req, 0);
        if (error != 0)
                return (error);
        sbuf_new_for_sysctl(&sb, NULL, 0, req);
        mtx_lock(&tc_lock);
        for (tc = timecounters; tc != NULL; tc = tc->tc_next) {
                if (tc != timecounters)
                        sbuf_putc(&sb, ' ');
                sbuf_printf(&sb, "%s(%d)", tc->tc_name, tc->tc_quality);
        }
        mtx_unlock(&tc_lock);
        error = sbuf_finish(&sb);
        sbuf_delete(&sb);
        return (error);
}

SYSCTL_PROC(_kern_timecounter, OID_AUTO, choice,
    CTLTYPE_STRING | CTLFLAG_RD | CTLFLAG_MPSAFE, 0, 0,
    sysctl_kern_timecounter_choice, "A",
    "Timecounter hardware detected");
#endif /* __rtems__ */

/*
 * RFC 2783 PPS-API implementation.
 */

/*
 *  Return true if the driver is aware of the abi version extensions in the
 *  pps_state structure, and it supports at least the given abi version number.
 */
static inline int
abi_aware(struct pps_state *pps, int vers)
{

        return ((pps->kcmode & KCMODE_ABIFLAG) && pps->driver_abi >= vers);
}

static int
pps_fetch(struct pps_fetch_args *fapi, struct pps_state *pps)
{
#ifndef __rtems__
        int err, timo;
#else /* __rtems__ */
        int err;
#endif /* __rtems__ */
        pps_seq_t aseq, cseq;
#ifndef __rtems__
        struct timeval tv;
#endif /* __rtems__ */

        if (fapi->tsformat && fapi->tsformat != PPS_TSFMT_TSPEC)
                return (EINVAL);

        /*
         * If no timeout is requested, immediately return whatever values were
         * most recently captured.  If timeout seconds is -1, that's a request
         * to block without a timeout.  WITNESS won't let us sleep forever
         * without a lock (we really don't need a lock), so just repeatedly
         * sleep a long time.
         */
        if (fapi->timeout.tv_sec || fapi->timeout.tv_nsec) {
#ifndef __rtems__
                if (fapi->timeout.tv_sec == -1)
                        timo = 0x7fffffff;
                else {
                        tv.tv_sec = fapi->timeout.tv_sec;
                        tv.tv_usec = fapi->timeout.tv_nsec / 1000;
                        timo = tvtohz(&tv);
                }
#endif /* __rtems__ */
                aseq = atomic_load_int(&pps->ppsinfo.assert_sequence);
                cseq = atomic_load_int(&pps->ppsinfo.clear_sequence);
                while (aseq == atomic_load_int(&pps->ppsinfo.assert_sequence) &&
                    cseq == atomic_load_int(&pps->ppsinfo.clear_sequence)) {
#ifndef __rtems__
                        if (abi_aware(pps, 1) && pps->driver_mtx != NULL) {
                                if (pps->flags & PPSFLAG_MTX_SPIN) {
                                        err = msleep_spin(pps, pps->driver_mtx,
                                            "ppsfch", timo);
                                } else {
                                        err = msleep(pps, pps->driver_mtx, PCATCH,
                                            "ppsfch", timo);
                                }
                        } else {
                                err = tsleep(pps, PCATCH, "ppsfch", timo);
                        }
                        if (err == EWOULDBLOCK) {
                                if (fapi->timeout.tv_sec == -1) {
                                        continue;
                                } else {
                                        return (ETIMEDOUT);
                                }
                        } else if (err != 0) {
                                return (err);
                        }
#else /* __rtems__ */
                        _Assert(pps->wait != NULL);
                        err = (*pps->wait)(pps, fapi->timeout);
                        if (err != 0)
                                return (err);
#endif /* __rtems__ */
                }
        }

        pps->ppsinfo.current_mode = pps->ppsparam.mode;
        fapi->pps_info_buf = pps->ppsinfo;

        return (0);
}

int
pps_ioctl(u_long cmd, caddr_t data, struct pps_state *pps)
{
        pps_params_t *app;
        struct pps_fetch_args *fapi;
#ifdef FFCLOCK
        struct pps_fetch_ffc_args *fapi_ffc;
#endif
#ifdef PPS_SYNC
        struct pps_kcbind_args *kapi;
#endif

        KASSERT(pps != NULL, ("NULL pps pointer in pps_ioctl"));
        switch (cmd) {
        case PPS_IOC_CREATE:
                return (0);
        case PPS_IOC_DESTROY:
                return (0);
        case PPS_IOC_SETPARAMS:
                app = (pps_params_t *)data;
                if (app->mode & ~pps->ppscap)
                        return (EINVAL);
#ifdef FFCLOCK
                /* Ensure only a single clock is selected for ffc timestamp. */
                if ((app->mode & PPS_TSCLK_MASK) == PPS_TSCLK_MASK)
                        return (EINVAL);
#endif
                pps->ppsparam = *app;
                return (0);
        case PPS_IOC_GETPARAMS:
                app = (pps_params_t *)data;
                *app = pps->ppsparam;
                app->api_version = PPS_API_VERS_1;
                return (0);
        case PPS_IOC_GETCAP:
                *(int*)data = pps->ppscap;
                return (0);
        case PPS_IOC_FETCH:
                fapi = (struct pps_fetch_args *)data;
                return (pps_fetch(fapi, pps));
#ifdef FFCLOCK
        case PPS_IOC_FETCH_FFCOUNTER:
                fapi_ffc = (struct pps_fetch_ffc_args *)data;
                if (fapi_ffc->tsformat && fapi_ffc->tsformat !=
                    PPS_TSFMT_TSPEC)
                        return (EINVAL);
                if (fapi_ffc->timeout.tv_sec || fapi_ffc->timeout.tv_nsec)
                        return (EOPNOTSUPP);
                pps->ppsinfo_ffc.current_mode = pps->ppsparam.mode;
                fapi_ffc->pps_info_buf_ffc = pps->ppsinfo_ffc;
                /* Overwrite timestamps if feedback clock selected. */
                switch (pps->ppsparam.mode & PPS_TSCLK_MASK) {
                case PPS_TSCLK_FBCK:
                        fapi_ffc->pps_info_buf_ffc.assert_timestamp =
                            pps->ppsinfo.assert_timestamp;
                        fapi_ffc->pps_info_buf_ffc.clear_timestamp =
                            pps->ppsinfo.clear_timestamp;
                        break;
                case PPS_TSCLK_FFWD:
                        break;
                default:
                        break;
                }
                return (0);
#endif /* FFCLOCK */
        case PPS_IOC_KCBIND:
#ifdef PPS_SYNC
                kapi = (struct pps_kcbind_args *)data;
                /* XXX Only root should be able to do this */
                if (kapi->tsformat && kapi->tsformat != PPS_TSFMT_TSPEC)
                        return (EINVAL);
                if (kapi->kernel_consumer != PPS_KC_HARDPPS)
                        return (EINVAL);
                if (kapi->edge & ~pps->ppscap)
                        return (EINVAL);
                pps->kcmode = (kapi->edge & KCMODE_EDGEMASK) |
                    (pps->kcmode & KCMODE_ABIFLAG);
                return (0);
#else
                return (EOPNOTSUPP);
#endif
        default:
                return (ENOIOCTL);
        }
}

#ifdef __rtems__
/*
 * The real implementation of hardpps() is defined in kern_ntptime.c.  Use it
 * only if the NTP support is needed by the application.
 */
RTEMS_WEAK void
hardpps(struct timespec *tsp, long nsec)
{

        (void)tsp;
        (void)nsec;
}

static int
default_wait(struct pps_state *pps, struct timespec timeout)
{

        (void)pps;
        (void)timeout;

        return (ETIMEDOUT);
}

static void
default_wakeup(struct pps_state *pps)
{

        (void)pps;
}
#endif /* __rtems__ */
void
pps_init(struct pps_state *pps)
{
#ifdef __rtems__
        pps->wait = default_wait;
        pps->wakeup = default_wakeup;
#endif /* __rtems__ */
        pps->ppscap |= PPS_TSFMT_TSPEC | PPS_CANWAIT;
        if (pps->ppscap & PPS_CAPTUREASSERT)
                pps->ppscap |= PPS_OFFSETASSERT;
        if (pps->ppscap & PPS_CAPTURECLEAR)
                pps->ppscap |= PPS_OFFSETCLEAR;
#ifdef FFCLOCK
        pps->ppscap |= PPS_TSCLK_MASK;
#endif
        pps->kcmode &= ~KCMODE_ABIFLAG;
}

void
pps_init_abi(struct pps_state *pps)
{

        pps_init(pps);
        if (pps->driver_abi > 0) {
                pps->kcmode |= KCMODE_ABIFLAG;
                pps->kernel_abi = PPS_ABI_VERSION;
        }
}

void
pps_capture(struct pps_state *pps)
{
        struct timehands *th;
        struct timecounter *tc;

        KASSERT(pps != NULL, ("NULL pps pointer in pps_capture"));
        th = timehands;
        pps->capgen = atomic_load_acq_int(&th->th_generation);
        pps->capth = th;
#ifdef FFCLOCK
        pps->capffth = fftimehands;
#endif
        tc = th->th_counter;
        pps->capcount = tc->tc_get_timecount(tc);
}

void
pps_event(struct pps_state *pps, int event)
{
        struct timehands *capth;
        struct timecounter *captc;
        uint64_t capth_scale;
        struct bintime bt;
        struct timespec *tsp, *osp;
        uint32_t tcount, *pcount;
        int foff;
        pps_seq_t *pseq;
#ifdef FFCLOCK
        struct timespec *tsp_ffc;
        pps_seq_t *pseq_ffc;
        ffcounter *ffcount;
#endif
#ifdef PPS_SYNC
        int fhard;
#endif

        KASSERT(pps != NULL, ("NULL pps pointer in pps_event"));
        /* Nothing to do if not currently set to capture this event type. */
        if ((event & pps->ppsparam.mode) == 0)
                return;

        /* Make a snapshot of the captured timehand */
        capth = pps->capth;
        captc = capth->th_counter;
        capth_scale = capth->th_scale;
        tcount = capth->th_offset_count;
        bt = capth->th_bintime;

        /* If the timecounter was wound up underneath us, bail out. */
        atomic_thread_fence_acq();
#if defined(RTEMS_SMP)
        if (pps->capgen == 0 || pps->capgen != capth->th_generation)
#else
        if (pps->capgen != capth->th_generation)
#endif
                return;

        /* Things would be easier with arrays. */
        if (event == PPS_CAPTUREASSERT) {
                tsp = &pps->ppsinfo.assert_timestamp;
                osp = &pps->ppsparam.assert_offset;
                foff = pps->ppsparam.mode & PPS_OFFSETASSERT;
#ifdef PPS_SYNC
                fhard = pps->kcmode & PPS_CAPTUREASSERT;
#endif
                pcount = &pps->ppscount[0];
                pseq = &pps->ppsinfo.assert_sequence;
#ifdef FFCLOCK
                ffcount = &pps->ppsinfo_ffc.assert_ffcount;
                tsp_ffc = &pps->ppsinfo_ffc.assert_timestamp;
                pseq_ffc = &pps->ppsinfo_ffc.assert_sequence;
#endif
        } else {
                tsp = &pps->ppsinfo.clear_timestamp;
                osp = &pps->ppsparam.clear_offset;
                foff = pps->ppsparam.mode & PPS_OFFSETCLEAR;
#ifdef PPS_SYNC
                fhard = pps->kcmode & PPS_CAPTURECLEAR;
#endif
                pcount = &pps->ppscount[1];
                pseq = &pps->ppsinfo.clear_sequence;
#ifdef FFCLOCK
                ffcount = &pps->ppsinfo_ffc.clear_ffcount;
                tsp_ffc = &pps->ppsinfo_ffc.clear_timestamp;
                pseq_ffc = &pps->ppsinfo_ffc.clear_sequence;
#endif
        }

        *pcount = pps->capcount;

        /*
         * If the timecounter changed, we cannot compare the count values, so
         * we have to drop the rest of the PPS-stuff until the next event.
         */
        if (__predict_false(pps->ppstc != captc)) {
                pps->ppstc = captc;
                pps->ppscount[2] = pps->capcount;
                return;
        }

        (*pseq)++;

        /* Convert the count to a timespec. */
        tcount = pps->capcount - tcount;
        tcount &= captc->tc_counter_mask;
        bintime_addx(&bt, capth_scale * tcount);
        bintime2timespec(&bt, tsp);

        if (foff) {
                timespecadd(tsp, osp, tsp);
                if (tsp->tv_nsec < 0) {
                        tsp->tv_nsec += 1000000000;
                        tsp->tv_sec -= 1;
                }
        }

#ifdef FFCLOCK
        *ffcount = pps->capffth->tick_ffcount + tcount;
        bt = pps->capffth->tick_time;
        ffclock_convert_delta(tcount, pps->capffth->cest.period, &bt);
        bintime_add(&bt, &pps->capffth->tick_time);
        (*pseq_ffc)++;
        bintime2timespec(&bt, tsp_ffc);
#endif

#ifdef PPS_SYNC
        if (fhard) {
                uint64_t delta_nsec;
                uint64_t freq;

                /*
                 * Feed the NTP PLL/FLL.
                 * The FLL wants to know how many (hardware) nanoseconds
                 * elapsed since the previous event.
                 */
                tcount = pps->capcount - pps->ppscount[2];
                pps->ppscount[2] = pps->capcount;
                tcount &= captc->tc_counter_mask;
                delta_nsec = 1000000000;
                delta_nsec *= tcount;
                freq = captc->tc_frequency;
                delta_nsec = (delta_nsec + freq / 2) / freq;
                hardpps(tsp, (long)delta_nsec);
        }
#endif

        /* Wakeup anyone sleeping in pps_fetch().  */
#ifndef __rtems__
        wakeup(pps);
#else /* __rtems__ */
        _Assert(pps->wakeup != NULL);
        (*pps->wakeup)(pps);
#endif /* __rtems__ */
}

/*
 * 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.
 */

#ifndef __rtems__
static int tc_tick;
SYSCTL_INT(_kern_timecounter, OID_AUTO, tick, CTLFLAG_RD, &tc_tick, 0,
    "Approximate number of hardclock ticks in a millisecond");
#endif /* __rtems__ */

#ifndef __rtems__
void
tc_ticktock(int cnt)
{
        static int count;

        if (mtx_trylock_spin(&tc_setclock_mtx)) {
                count += cnt;
                if (count >= tc_tick) {
                        count = 0;
                        tc_windup(NULL);
                }
                mtx_unlock_spin(&tc_setclock_mtx);
        }
}
#else /* __rtems__ */
void
_Timecounter_Tick(void)
{
        Per_CPU_Control *cpu_self = _Per_CPU_Get();

#if defined(RTEMS_SMP)
        if (_Per_CPU_Is_boot_processor(cpu_self)) {
#endif
                tc_windup(NULL);
#if defined(RTEMS_SMP)
        }
#endif

        _Watchdog_Tick(cpu_self);
}

void
_Timecounter_Tick_simple(uint32_t delta, uint32_t offset,
    ISR_lock_Context *lock_context)
{
        struct bintime bt;
        struct timehands *th;
#if defined(RTEMS_SMP)
        u_int ogen;
#endif

        th = timehands;
#if defined(RTEMS_SMP)
        ogen = th->th_generation;
        th->th_generation = 0;
        atomic_thread_fence_rel();
#endif

        th->th_offset_count = offset;
        bintime_addx(&th->th_offset, th->th_scale * delta);
        bt = th->th_offset;
        bintime_add(&bt, &th->th_boottime);

        /* Update the UTC timestamps used by the get*() functions. */
        th->th_bintime = bt;
        bintime2timeval(&bt, &th->th_microtime);
        bintime2timespec(&bt, &th->th_nanotime);

#if defined(RTEMS_SMP)
        /*
         * 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;
        atomic_store_rel_int(&th->th_generation, ogen);
#else
        atomic_store_rel_int(&th->th_generation, th->th_generation + 1);
#endif

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

        _Timecounter_Release(lock_context);

        _Watchdog_Tick(_Per_CPU_Get_snapshot());
}
#endif /* __rtems__ */

#ifndef __rtems__
static void __inline
tc_adjprecision(void)
{
        int t;

        if (tc_timepercentage > 0) {
                t = (99 + tc_timepercentage) / tc_timepercentage;
                tc_precexp = fls(t + (t >> 1)) - 1;
                FREQ2BT(hz / tc_tick, &bt_timethreshold);
                FREQ2BT(hz, &bt_tickthreshold);
                bintime_shift(&bt_timethreshold, tc_precexp);
                bintime_shift(&bt_tickthreshold, tc_precexp);
        } else {
                tc_precexp = 31;
                bt_timethreshold.sec = INT_MAX;
                bt_timethreshold.frac = ~(uint64_t)0;
                bt_tickthreshold = bt_timethreshold;
        }
        sbt_timethreshold = bttosbt(bt_timethreshold);
        sbt_tickthreshold = bttosbt(bt_tickthreshold);
}
#endif /* __rtems__ */

#ifndef __rtems__
static int
sysctl_kern_timecounter_adjprecision(SYSCTL_HANDLER_ARGS)
{
        int error, val;

        val = tc_timepercentage;
        error = sysctl_handle_int(oidp, &val, 0, req);
        if (error != 0 || req->newptr == NULL)
                return (error);
        tc_timepercentage = val;
        if (cold)
                goto done;
        tc_adjprecision();
done:
        return (0);
}

/* Set up the requested number of timehands. */
static void
inittimehands(void *dummy)
{
        struct timehands *thp;
        int i;

        TUNABLE_INT_FETCH("kern.timecounter.timehands_count",
            &timehands_count);
        if (timehands_count < 1)
                timehands_count = 1;
        if (timehands_count > nitems(ths))
                timehands_count = nitems(ths);
        for (i = 1, thp = &ths[0]; i < timehands_count;  thp = &ths[i++])
                thp->th_next = &ths[i];
        thp->th_next = &ths[0];

        TUNABLE_STR_FETCH("kern.timecounter.hardware", tc_from_tunable,
            sizeof(tc_from_tunable));

        mtx_init(&tc_lock, "tc", NULL, MTX_DEF);
}
SYSINIT(timehands, SI_SUB_TUNABLES, SI_ORDER_ANY, inittimehands, NULL);

static void
inittimecounter(void *dummy)
{
        u_int p;
        int tick_rate;

        /*
         * 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;
        tc_adjprecision();
        FREQ2BT(hz, &tick_bt);
        tick_sbt = bttosbt(tick_bt);
        tick_rate = hz / tc_tick;
        FREQ2BT(tick_rate, &tc_tick_bt);
        tc_tick_sbt = bttosbt(tc_tick_bt);
        p = (tc_tick * 1000000) / hz;
        printf("Timecounters tick every %d.%03u msec\n", p / 1000, p % 1000);

#ifdef FFCLOCK
        ffclock_init();
#endif

        /* warm up new timecounter (again) and get rolling. */
        (void)timecounter->tc_get_timecount(timecounter);
        mtx_lock_spin(&tc_setclock_mtx);
        tc_windup(NULL);
        mtx_unlock_spin(&tc_setclock_mtx);
}

SYSINIT(timecounter, SI_SUB_CLOCKS, SI_ORDER_SECOND, inittimecounter, NULL);

/* Cpu tick handling -------------------------------------------------*/

static bool cpu_tick_variable;
static uint64_t cpu_tick_frequency;

DPCPU_DEFINE_STATIC(uint64_t, tc_cpu_ticks_base);
DPCPU_DEFINE_STATIC(unsigned, tc_cpu_ticks_last);

static uint64_t
tc_cpu_ticks(void)
{
        struct timecounter *tc;
        uint64_t res, *base;
        unsigned u, *last;

        critical_enter();
        base = DPCPU_PTR(tc_cpu_ticks_base);
        last = DPCPU_PTR(tc_cpu_ticks_last);
        tc = timehands->th_counter;
        u = tc->tc_get_timecount(tc) & tc->tc_counter_mask;
        if (u < *last)
                *base += (uint64_t)tc->tc_counter_mask + 1;
        *last = u;
        res = u + *base;
        critical_exit();
        return (res);
}

void
cpu_tick_calibration(void)
{
        static time_t last_calib;

        if (time_uptime != last_calib && !(time_uptime & 0xf)) {
                cpu_tick_calibrate(0);
                last_calib = time_uptime;
        }
}

/*
 * This function gets called every 16 seconds on only one designated
 * CPU in the system from hardclock() via cpu_tick_calibration()().
 *
 * Whenever the real time clock is stepped we get called with reset=1
 * to make sure we handle suspend/resume and similar events correctly.
 */

static void
cpu_tick_calibrate(int reset)
{
        static uint64_t c_last;
        uint64_t c_this, c_delta;
        static struct bintime  t_last;
        struct bintime t_this, t_delta;
        uint32_t divi;

        if (reset) {
                /* The clock was stepped, abort & reset */
                t_last.sec = 0;
                return;
        }

        /* we don't calibrate fixed rate cputicks */
        if (!cpu_tick_variable)
                return;

        getbinuptime(&t_this);
        c_this = cpu_ticks();
        if (t_last.sec != 0) {
                c_delta = c_this - c_last;
                t_delta = t_this;
                bintime_sub(&t_delta, &t_last);
                /*
                 * Headroom:
                 *      2^(64-20) / 16[s] =
                 *      2^(44) / 16[s] =
                 *      17.592.186.044.416 / 16 =
                 *      1.099.511.627.776 [Hz]
                 */
                divi = t_delta.sec << 20;
                divi |= t_delta.frac >> (64 - 20);
                c_delta <<= 20;
                c_delta /= divi;
                if (c_delta > cpu_tick_frequency) {
                        if (0 && bootverbose)
                                printf("cpu_tick increased to %ju Hz\n",
                                    c_delta);
                        cpu_tick_frequency = c_delta;
                }
        }
        c_last = c_this;
        t_last = t_this;
}

void
set_cputicker(cpu_tick_f *func, uint64_t freq, bool isvariable)
{

        if (func == NULL) {
                cpu_ticks = tc_cpu_ticks;
        } else {
                cpu_tick_frequency = freq;
                cpu_tick_variable = isvariable;
                cpu_ticks = func;
        }
}

uint64_t
cpu_tickrate(void)
{

        if (cpu_ticks == tc_cpu_ticks) 
                return (tc_getfrequency());
        return (cpu_tick_frequency);
}

/*
 * We need to be slightly careful converting cputicks to microseconds.
 * There is plenty of margin in 64 bits of microseconds (half a million
 * years) and in 64 bits at 4 GHz (146 years), but if we do a multiply
 * before divide conversion (to retain precision) we find that the
 * margin shrinks to 1.5 hours (one millionth of 146y).
 */

uint64_t
cputick2usec(uint64_t tick)
{
        uint64_t tr;
        tr = cpu_tickrate();
        return ((tick / tr) * 1000000ULL) + ((tick % tr) * 1000000ULL) / tr;
}

cpu_tick_f      *cpu_ticks = tc_cpu_ticks;
#endif /* __rtems__ */

#ifndef __rtems__
static int vdso_th_enable = 1;
static int
sysctl_fast_gettime(SYSCTL_HANDLER_ARGS)
{
        int old_vdso_th_enable, error;

        old_vdso_th_enable = vdso_th_enable;
        error = sysctl_handle_int(oidp, &old_vdso_th_enable, 0, req);
        if (error != 0)
                return (error);
        vdso_th_enable = old_vdso_th_enable;
        return (0);
}
SYSCTL_PROC(_kern_timecounter, OID_AUTO, fast_gettime,
    CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_MPSAFE,
    NULL, 0, sysctl_fast_gettime, "I", "Enable fast time of day");

uint32_t
tc_fill_vdso_timehands(struct vdso_timehands *vdso_th)
{
        struct timehands *th;
        uint32_t enabled;

        th = timehands;
        vdso_th->th_scale = th->th_scale;
        vdso_th->th_offset_count = th->th_offset_count;
        vdso_th->th_counter_mask = th->th_counter->tc_counter_mask;
        vdso_th->th_offset = th->th_offset;
        vdso_th->th_boottime = th->th_boottime;
        if (th->th_counter->tc_fill_vdso_timehands != NULL) {
                enabled = th->th_counter->tc_fill_vdso_timehands(vdso_th,
                    th->th_counter);
        } else
                enabled = 0;
        if (!vdso_th_enable)
                enabled = 0;
        return (enabled);
}

#ifdef COMPAT_FREEBSD32
uint32_t
tc_fill_vdso_timehands32(struct vdso_timehands32 *vdso_th32)
{
        struct timehands *th;
        uint32_t enabled;

        th = timehands;
        *(uint64_t *)&vdso_th32->th_scale[0] = th->th_scale;
        vdso_th32->th_offset_count = th->th_offset_count;
        vdso_th32->th_counter_mask = th->th_counter->tc_counter_mask;
        vdso_th32->th_offset.sec = th->th_offset.sec;
        *(uint64_t *)&vdso_th32->th_offset.frac[0] = th->th_offset.frac;
        vdso_th32->th_boottime.sec = th->th_boottime.sec;
        *(uint64_t *)&vdso_th32->th_boottime.frac[0] = th->th_boottime.frac;
        if (th->th_counter->tc_fill_vdso_timehands32 != NULL) {
                enabled = th->th_counter->tc_fill_vdso_timehands32(vdso_th32,
                    th->th_counter);
        } else
                enabled = 0;
        if (!vdso_th_enable)
                enabled = 0;
        return (enabled);
}
#endif

#include "opt_ddb.h"
#ifdef DDB
#include <ddb/ddb.h>

DB_SHOW_COMMAND(timecounter, db_show_timecounter)
{
        struct timehands *th;
        struct timecounter *tc;
        u_int val1, val2;

        th = timehands;
        tc = th->th_counter;
        val1 = tc->tc_get_timecount(tc);
        __compiler_membar();
        val2 = tc->tc_get_timecount(tc);

        db_printf("timecounter %p %s\n", tc, tc->tc_name);
        db_printf("  mask %#x freq %ju qual %d flags %#x priv %p\n",
            tc->tc_counter_mask, (uintmax_t)tc->tc_frequency, tc->tc_quality,
            tc->tc_flags, tc->tc_priv);
        db_printf("  val %#x %#x\n", val1, val2);
        db_printf("timehands adj %#jx scale %#jx ldelta %d off_cnt %d gen %d\n",
            (uintmax_t)th->th_adjustment, (uintmax_t)th->th_scale,
            th->th_large_delta, th->th_offset_count, th->th_generation);
        db_printf("  offset %jd %jd boottime %jd %jd\n",
            (intmax_t)th->th_offset.sec, (uintmax_t)th->th_offset.frac,
            (intmax_t)th->th_boottime.sec, (uintmax_t)th->th_boottime.frac);
}
#endif
#else /* __rtems__ */
RTEMS_ALIAS(_Timecounter_Nanotime)
void rtems_clock_get_realtime(struct timespec *);

RTEMS_ALIAS(_Timecounter_Bintime)
void rtems_clock_get_realtime_bintime(struct bintime *);

RTEMS_ALIAS(_Timecounter_Microtime)
void rtems_clock_get_realtime_timeval(struct timeval *);

RTEMS_ALIAS(_Timecounter_Getnanotime)
void rtems_clock_get_realtime_coarse(struct timespec *);

RTEMS_ALIAS(_Timecounter_Getbintime)
void rtems_clock_get_realtime_coarse_bintime(struct bintime *);

RTEMS_ALIAS(_Timecounter_Getmicrotime)
void rtems_clock_get_realtime_coarse_timeval(struct timeval *);

RTEMS_ALIAS(_Timecounter_Nanouptime)
void rtems_clock_get_monotonic(struct timespec *);

RTEMS_ALIAS(_Timecounter_Binuptime)
void rtems_clock_get_monotonic_bintime(struct bintime *);

RTEMS_ALIAS(_Timecounter_Sbinuptime)
sbintime_t rtems_clock_get_monotonic_sbintime(void);

RTEMS_ALIAS(_Timecounter_Microuptime)
void rtems_clock_get_monotonic_timeval(struct timeval *);

RTEMS_ALIAS(_Timecounter_Getnanouptime)
void rtems_clock_get_monotonic_coarse(struct timespec *);

RTEMS_ALIAS(_Timecounter_Getbinuptime)
void rtems_clock_get_monotonic_coarse_bintime(struct bintime *);

RTEMS_ALIAS(_Timecounter_Getmicrouptime)
void rtems_clock_get_monotonic_coarse_timeval(struct timeval *);

RTEMS_ALIAS(_Timecounter_Getboottimebin)
void rtems_clock_get_boot_time_bintime(struct bintime *);

RTEMS_ALIAS(_Timecounter_Getboottime)
void rtems_clock_get_boot_time_timeval(struct timeval *);
#endif /* __rtems__ */