1 files changed, 303 insertions, 208 deletions

diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 6f46a00a1e8a..f045cc50832d 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -24,32 +24,32 @@
 /* Structure holding internal timekeeping values. */
 struct timekeeper {
         /* Current clocksource used for timekeeping. */
-        struct clocksource *clock;
+        struct clocksource      *clock;
         /* NTP adjusted clock multiplier */
-        u32     mult;
+        u32                     mult;
         /* The shift value of the current clocksource. */
-        int     shift;
-
+        u32                     shift;
         /* Number of clock cycles in one NTP interval. */
-        cycle_t cycle_interval;
+        cycle_t                 cycle_interval;
         /* Number of clock shifted nano seconds in one NTP interval. */
-        u64     xtime_interval;
+        u64                     xtime_interval;
         /* shifted nano seconds left over when rounding cycle_interval */
-        s64     xtime_remainder;
+        s64                     xtime_remainder;
         /* Raw nano seconds accumulated per NTP interval. */
-        u32     raw_interval;
+        u32                     raw_interval;
+
+        /* Current CLOCK_REALTIME time in seconds */
+        u64                     xtime_sec;
+        /* Clock shifted nano seconds */
+        u64                     xtime_nsec;
 
-        /* Clock shifted nano seconds remainder not stored in xtime.tv_nsec. */
-        u64     xtime_nsec;
         /* Difference between accumulated time and NTP time in ntp
          * shifted nano seconds. */
-        s64     ntp_error;
+        s64                     ntp_error;
         /* Shift conversion between clock shifted nano seconds and
          * ntp shifted nano seconds. */
-        int     ntp_error_shift;
+        u32                     ntp_error_shift;
 
-        /* The current time */
-        struct timespec xtime;
         /*
          * wall_to_monotonic is what we need to add to xtime (or xtime corrected
          * for sub jiffie times) to get to monotonic time.  Monotonic is pegged
@@ -64,14 +64,17 @@ struct timekeeper {
          * - wall_to_monotonic is no longer the boot time, getboottime must be
          * used instead.
          */
-        struct timespec wall_to_monotonic;
+        struct timespec         wall_to_monotonic;
         /* time spent in suspend */
-        struct timespec total_sleep_time;
+        struct timespec         total_sleep_time;
         /* The raw monotonic time for the CLOCK_MONOTONIC_RAW posix clock. */
-        struct timespec raw_time;
-
+        struct timespec         raw_time;
+        /* Offset clock monotonic -> clock realtime */
+        ktime_t                 offs_real;
+        /* Offset clock monotonic -> clock boottime */
+        ktime_t                 offs_boot;
         /* Seqlock for all timekeeper values */
-        seqlock_t lock;
+        seqlock_t               lock;
 };
 
 static struct timekeeper timekeeper;
@@ -82,11 +85,37 @@ static struct timekeeper timekeeper;
  */
 __cacheline_aligned_in_smp DEFINE_SEQLOCK(xtime_lock);
 
-
 /* flag for if timekeeping is suspended */
 int __read_mostly timekeeping_suspended;
 
+static inline void tk_normalize_xtime(struct timekeeper *tk)
+{
+        while (tk->xtime_nsec >= ((u64)NSEC_PER_SEC << tk->shift)) {
+                tk->xtime_nsec -= (u64)NSEC_PER_SEC << tk->shift;
+                tk->xtime_sec++;
+        }
+}
+
+static struct timespec tk_xtime(struct timekeeper *tk)
+{
+        struct timespec ts;
+
+        ts.tv_sec = tk->xtime_sec;
+        ts.tv_nsec = (long)(tk->xtime_nsec >> tk->shift);
+        return ts;
+}
+
+static void tk_set_xtime(struct timekeeper *tk, const struct timespec *ts)
+{
+        tk->xtime_sec = ts->tv_sec;
+        tk->xtime_nsec = ts->tv_nsec << tk->shift;
+}
 
+static void tk_xtime_add(struct timekeeper *tk, const struct timespec *ts)
+{
+        tk->xtime_sec += ts->tv_sec;
+        tk->xtime_nsec += ts->tv_nsec << tk->shift;
+}
 
 /**
  * timekeeper_setup_internals - Set up internals to use clocksource clock.
@@ -98,12 +127,14 @@ int __read_mostly timekeeping_suspended;
  *
  * Unless you're the timekeeping code, you should not be using this!
  */
-static void timekeeper_setup_internals(struct clocksource *clock)
+static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock)
 {
         cycle_t interval;
         u64 tmp, ntpinterval;
+        struct clocksource *old_clock;
 
-        timekeeper.clock = clock;
+        old_clock = tk->clock;
+        tk->clock = clock;
         clock->cycle_last = clock->read(clock);
 
         /* Do the ns -> cycle conversion first, using original mult */
@@ -116,71 +147,96 @@ static void timekeeper_setup_internals(struct clocksource *clock)
                 tmp = 1;
 
         interval = (cycle_t) tmp;
-        timekeeper.cycle_interval = interval;
+        tk->cycle_interval = interval;
 
         /* Go back from cycles -> shifted ns */
-        timekeeper.xtime_interval = (u64) interval * clock->mult;
-        timekeeper.xtime_remainder = ntpinterval - timekeeper.xtime_interval;
-        timekeeper.raw_interval =
+        tk->xtime_interval = (u64) interval * clock->mult;
+        tk->xtime_remainder = ntpinterval - tk->xtime_interval;
+        tk->raw_interval =
                 ((u64) interval * clock->mult) >> clock->shift;
 
-        timekeeper.xtime_nsec = 0;
-        timekeeper.shift = clock->shift;
+         /* if changing clocks, convert xtime_nsec shift units */
+        if (old_clock) {
+                int shift_change = clock->shift - old_clock->shift;
+                if (shift_change < 0)
+                        tk->xtime_nsec >>= -shift_change;
+                else
+                        tk->xtime_nsec <<= shift_change;
+        }
+        tk->shift = clock->shift;
 
-        timekeeper.ntp_error = 0;
-        timekeeper.ntp_error_shift = NTP_SCALE_SHIFT - clock->shift;
+        tk->ntp_error = 0;
+        tk->ntp_error_shift = NTP_SCALE_SHIFT - clock->shift;
 
         /*
          * The timekeeper keeps its own mult values for the currently
          * active clocksource. These value will be adjusted via NTP
          * to counteract clock drifting.
          */
-        timekeeper.mult = clock->mult;
+        tk->mult = clock->mult;
 }
 
 /* Timekeeper helper functions. */
-static inline s64 timekeeping_get_ns(void)
+static inline s64 timekeeping_get_ns(struct timekeeper *tk)
 {
         cycle_t cycle_now, cycle_delta;
         struct clocksource *clock;
+        s64 nsec;
 
         /* read clocksource: */
-        clock = timekeeper.clock;
+        clock = tk->clock;
         cycle_now = clock->read(clock);
 
         /* calculate the delta since the last update_wall_time: */
         cycle_delta = (cycle_now - clock->cycle_last) & clock->mask;
 
-        /* return delta convert to nanoseconds using ntp adjusted mult. */
-        return clocksource_cyc2ns(cycle_delta, timekeeper.mult,
-                                  timekeeper.shift);
+        nsec = cycle_delta * tk->mult + tk->xtime_nsec;
+        nsec >>= tk->shift;
+
+        /* If arch requires, add in gettimeoffset() */
+        return nsec + arch_gettimeoffset();
 }
 
-static inline s64 timekeeping_get_ns_raw(void)
+static inline s64 timekeeping_get_ns_raw(struct timekeeper *tk)
 {
         cycle_t cycle_now, cycle_delta;
         struct clocksource *clock;
+        s64 nsec;
 
         /* read clocksource: */
-        clock = timekeeper.clock;
+        clock = tk->clock;
         cycle_now = clock->read(clock);
 
         /* calculate the delta since the last update_wall_time: */
         cycle_delta = (cycle_now - clock->cycle_last) & clock->mask;
 
-        /* return delta convert to nanoseconds. */
-        return clocksource_cyc2ns(cycle_delta, clock->mult, clock->shift);
+        /* convert delta to nanoseconds. */
+        nsec = clocksource_cyc2ns(cycle_delta, clock->mult, clock->shift);
+
+        /* If arch requires, add in gettimeoffset() */
+        return nsec + arch_gettimeoffset();
+}
+
+static void update_rt_offset(struct timekeeper *tk)
+{
+        struct timespec tmp, *wtm = &tk->wall_to_monotonic;
+
+        set_normalized_timespec(&tmp, -wtm->tv_sec, -wtm->tv_nsec);
+        tk->offs_real = timespec_to_ktime(tmp);
 }
 
 /* must hold write on timekeeper.lock */
-static void timekeeping_update(bool clearntp)
+static void timekeeping_update(struct timekeeper *tk, bool clearntp)
 {
+        struct timespec xt;
+
         if (clearntp) {
-                timekeeper.ntp_error = 0;
+                tk->ntp_error = 0;
                 ntp_clear();
         }
-        update_vsyscall(&timekeeper.xtime, &timekeeper.wall_to_monotonic,
-                         timekeeper.clock, timekeeper.mult);
+        update_rt_offset(tk);
+        xt = tk_xtime(tk);
+        update_vsyscall(&xt, &tk->wall_to_monotonic, tk->clock, tk->mult);
 }
 
 
@@ -191,27 +247,26 @@ static void timekeeping_update(bool clearntp)
  * update_wall_time(). This is useful before significant clock changes,
  * as it avoids having to deal with this time offset explicitly.
  */
-static void timekeeping_forward_now(void)
+static void timekeeping_forward_now(struct timekeeper *tk)
 {
         cycle_t cycle_now, cycle_delta;
         struct clocksource *clock;
         s64 nsec;
 
-        clock = timekeeper.clock;
+        clock = tk->clock;
         cycle_now = clock->read(clock);
         cycle_delta = (cycle_now - clock->cycle_last) & clock->mask;
         clock->cycle_last = cycle_now;
 
-        nsec = clocksource_cyc2ns(cycle_delta, timekeeper.mult,
-                                  timekeeper.shift);
+        tk->xtime_nsec += cycle_delta * tk->mult;
 
         /* If arch requires, add in gettimeoffset() */
-        nsec += arch_gettimeoffset();
+        tk->xtime_nsec += arch_gettimeoffset() << tk->shift;
 
-        timespec_add_ns(&timekeeper.xtime, nsec);
+        tk_normalize_xtime(tk);
 
         nsec = clocksource_cyc2ns(cycle_delta, clock->mult, clock->shift);
-        timespec_add_ns(&timekeeper.raw_time, nsec);
+        timespec_add_ns(&tk->raw_time, nsec);
 }
 
 /**
@@ -223,18 +278,15 @@ static void timekeeping_forward_now(void)
 void getnstimeofday(struct timespec *ts)
 {
         unsigned long seq;
-        s64 nsecs;
+        s64 nsecs = 0;
 
         WARN_ON(timekeeping_suspended);
 
         do {
                 seq = read_seqbegin(&timekeeper.lock);
 
-                *ts = timekeeper.xtime;
-                nsecs = timekeeping_get_ns();
-
-                /* If arch requires, add in gettimeoffset() */
-                nsecs += arch_gettimeoffset();
+                ts->tv_sec = timekeeper.xtime_sec;
+                ts->tv_nsec = timekeeping_get_ns(&timekeeper);
 
         } while (read_seqretry(&timekeeper.lock, seq));
 
@@ -251,13 +303,10 @@ ktime_t ktime_get(void)
 
         do {
                 seq = read_seqbegin(&timekeeper.lock);
-                secs = timekeeper.xtime.tv_sec +
+                secs = timekeeper.xtime_sec +
                                 timekeeper.wall_to_monotonic.tv_sec;
-                nsecs = timekeeper.xtime.tv_nsec +
+                nsecs = timekeeping_get_ns(&timekeeper) +
                                 timekeeper.wall_to_monotonic.tv_nsec;
-                nsecs += timekeeping_get_ns();
-                /* If arch requires, add in gettimeoffset() */
-                nsecs += arch_gettimeoffset();
 
         } while (read_seqretry(&timekeeper.lock, seq));
         /*
@@ -280,22 +329,19 @@ void ktime_get_ts(struct timespec *ts)
 {
         struct timespec tomono;
         unsigned int seq;
-        s64 nsecs;
 
         WARN_ON(timekeeping_suspended);
 
         do {
                 seq = read_seqbegin(&timekeeper.lock);
-                *ts = timekeeper.xtime;
+                ts->tv_sec = timekeeper.xtime_sec;
+                ts->tv_nsec = timekeeping_get_ns(&timekeeper);
                 tomono = timekeeper.wall_to_monotonic;
-                nsecs = timekeeping_get_ns();
-                /* If arch requires, add in gettimeoffset() */
-                nsecs += arch_gettimeoffset();
 
         } while (read_seqretry(&timekeeper.lock, seq));
 
         set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec,
-                                ts->tv_nsec + tomono.tv_nsec + nsecs);
+                                ts->tv_nsec + tomono.tv_nsec);
 }
 EXPORT_SYMBOL_GPL(ktime_get_ts);
 
@@ -318,20 +364,14 @@ void getnstime_raw_and_real(struct timespec *ts_raw, struct timespec *ts_real)
         WARN_ON_ONCE(timekeeping_suspended);
 
         do {
-                u32 arch_offset;
-
                 seq = read_seqbegin(&timekeeper.lock);
 
                 *ts_raw = timekeeper.raw_time;
-                *ts_real = timekeeper.xtime;
-
-                nsecs_raw = timekeeping_get_ns_raw();
-                nsecs_real = timekeeping_get_ns();
+                ts_real->tv_sec = timekeeper.xtime_sec;
+                ts_real->tv_nsec = 0;
 
-                /* If arch requires, add in gettimeoffset() */
-                arch_offset = arch_gettimeoffset();
-                nsecs_raw += arch_offset;
-                nsecs_real += arch_offset;
+                nsecs_raw = timekeeping_get_ns_raw(&timekeeper);
+                nsecs_real = timekeeping_get_ns(&timekeeper);
 
         } while (read_seqretry(&timekeeper.lock, seq));
 
@@ -366,7 +406,7 @@ EXPORT_SYMBOL(do_gettimeofday);
  */
 int do_settimeofday(const struct timespec *tv)
 {
-        struct timespec ts_delta;
+        struct timespec ts_delta, xt;
         unsigned long flags;
 
         if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
@@ -374,15 +414,18 @@ int do_settimeofday(const struct timespec *tv)
 
         write_seqlock_irqsave(&timekeeper.lock, flags);
 
-        timekeeping_forward_now();
+        timekeeping_forward_now(&timekeeper);
+
+        xt = tk_xtime(&timekeeper);
+        ts_delta.tv_sec = tv->tv_sec - xt.tv_sec;
+        ts_delta.tv_nsec = tv->tv_nsec - xt.tv_nsec;
 
-        ts_delta.tv_sec = tv->tv_sec - timekeeper.xtime.tv_sec;
-        ts_delta.tv_nsec = tv->tv_nsec - timekeeper.xtime.tv_nsec;
         timekeeper.wall_to_monotonic =
                         timespec_sub(timekeeper.wall_to_monotonic, ts_delta);
 
-        timekeeper.xtime = *tv;
-        timekeeping_update(true);
+        tk_set_xtime(&timekeeper, tv);
+
+        timekeeping_update(&timekeeper, true);
 
         write_sequnlock_irqrestore(&timekeeper.lock, flags);
 
@@ -409,13 +452,14 @@ int timekeeping_inject_offset(struct timespec *ts)
 
         write_seqlock_irqsave(&timekeeper.lock, flags);
 
-        timekeeping_forward_now();
+        timekeeping_forward_now(&timekeeper);
 
-        timekeeper.xtime = timespec_add(timekeeper.xtime, *ts);
+
+        tk_xtime_add(&timekeeper, ts);
         timekeeper.wall_to_monotonic =
                                 timespec_sub(timekeeper.wall_to_monotonic, *ts);
 
-        timekeeping_update(true);
+        timekeeping_update(&timekeeper, true);
 
         write_sequnlock_irqrestore(&timekeeper.lock, flags);
 
@@ -440,14 +484,14 @@ static int change_clocksource(void *data)
 
         write_seqlock_irqsave(&timekeeper.lock, flags);
 
-        timekeeping_forward_now();
+        timekeeping_forward_now(&timekeeper);
         if (!new->enable || new->enable(new) == 0) {
                 old = timekeeper.clock;
-                timekeeper_setup_internals(new);
+                tk_setup_internals(&timekeeper, new);
                 if (old->disable)
                         old->disable(old);
         }
-        timekeeping_update(true);
+        timekeeping_update(&timekeeper, true);
 
         write_sequnlock_irqrestore(&timekeeper.lock, flags);
 
@@ -497,7 +541,7 @@ void getrawmonotonic(struct timespec *ts)
 
         do {
                 seq = read_seqbegin(&timekeeper.lock);
-                nsecs = timekeeping_get_ns_raw();
+                nsecs = timekeeping_get_ns_raw(&timekeeper);
                 *ts = timekeeper.raw_time;
 
         } while (read_seqretry(&timekeeper.lock, seq));
@@ -532,6 +576,7 @@ u64 timekeeping_max_deferment(void)
 {
         unsigned long seq;
         u64 ret;
+
         do {
                 seq = read_seqbegin(&timekeeper.lock);
 
@@ -592,18 +637,17 @@ void __init timekeeping_init(void)
         clock = clocksource_default_clock();
         if (clock->enable)
                 clock->enable(clock);
-        timekeeper_setup_internals(clock);
+        tk_setup_internals(&timekeeper, clock);
 
-        timekeeper.xtime.tv_sec = now.tv_sec;
-        timekeeper.xtime.tv_nsec = now.tv_nsec;
+        tk_set_xtime(&timekeeper, &now);
         timekeeper.raw_time.tv_sec = 0;
         timekeeper.raw_time.tv_nsec = 0;
-        if (boot.tv_sec == 0 && boot.tv_nsec == 0) {
-                boot.tv_sec = timekeeper.xtime.tv_sec;
-                boot.tv_nsec = timekeeper.xtime.tv_nsec;
-        }
+        if (boot.tv_sec == 0 && boot.tv_nsec == 0)
+                boot = tk_xtime(&timekeeper);
+
         set_normalized_timespec(&timekeeper.wall_to_monotonic,
                                 -boot.tv_sec, -boot.tv_nsec);
+        update_rt_offset(&timekeeper);
         timekeeper.total_sleep_time.tv_sec = 0;
         timekeeper.total_sleep_time.tv_nsec = 0;
         write_sequnlock_irqrestore(&timekeeper.lock, flags);
@@ -612,6 +656,12 @@ void __init timekeeping_init(void)
 /* time in seconds when suspend began */
 static struct timespec timekeeping_suspend_time;
 
+static void update_sleep_time(struct timespec t)
+{
+        timekeeper.total_sleep_time = t;
+        timekeeper.offs_boot = timespec_to_ktime(t);
+}
+
 /**
  * __timekeeping_inject_sleeptime - Internal function to add sleep interval
  * @delta: pointer to a timespec delta value
@@ -619,7 +669,8 @@ static struct timespec timekeeping_suspend_time;
  * Takes a timespec offset measuring a suspend interval and properly
  * adds the sleep offset to the timekeeping variables.
  */
-static void __timekeeping_inject_sleeptime(struct timespec *delta)
+static void __timekeeping_inject_sleeptime(struct timekeeper *tk,
+                                                        struct timespec *delta)
 {
         if (!timespec_valid(delta)) {
                 printk(KERN_WARNING "__timekeeping_inject_sleeptime: Invalid "
@@ -627,11 +678,9 @@ static void __timekeeping_inject_sleeptime(struct timespec *delta)
                 return;
         }
 
-        timekeeper.xtime = timespec_add(timekeeper.xtime, *delta);
-        timekeeper.wall_to_monotonic =
-                        timespec_sub(timekeeper.wall_to_monotonic, *delta);
-        timekeeper.total_sleep_time = timespec_add(
-                                        timekeeper.total_sleep_time, *delta);
+        tk_xtime_add(tk, delta);
+        tk->wall_to_monotonic = timespec_sub(tk->wall_to_monotonic, *delta);
+        update_sleep_time(timespec_add(tk->total_sleep_time, *delta));
 }
 
 
@@ -657,11 +706,11 @@ void timekeeping_inject_sleeptime(struct timespec *delta)
 
         write_seqlock_irqsave(&timekeeper.lock, flags);
 
-        timekeeping_forward_now();
+        timekeeping_forward_now(&timekeeper);
 
-        __timekeeping_inject_sleeptime(delta);
+        __timekeeping_inject_sleeptime(&timekeeper, delta);
 
-        timekeeping_update(true);
+        timekeeping_update(&timekeeper, true);
 
         write_sequnlock_irqrestore(&timekeeper.lock, flags);
 
@@ -690,12 +739,13 @@ static void timekeeping_resume(void)
 
         if (timespec_compare(&ts, &timekeeping_suspend_time) > 0) {
                 ts = timespec_sub(ts, timekeeping_suspend_time);
-                __timekeeping_inject_sleeptime(&ts);
+                __timekeeping_inject_sleeptime(&timekeeper, &ts);
         }
         /* re-base the last cycle value */
         timekeeper.clock->cycle_last = timekeeper.clock->read(timekeeper.clock);
         timekeeper.ntp_error = 0;
         timekeeping_suspended = 0;
+        timekeeping_update(&timekeeper, false);
         write_sequnlock_irqrestore(&timekeeper.lock, flags);
 
         touch_softlockup_watchdog();
@@ -715,7 +765,7 @@ static int timekeeping_suspend(void)
         read_persistent_clock(&timekeeping_suspend_time);
 
         write_seqlock_irqsave(&timekeeper.lock, flags);
-        timekeeping_forward_now();
+        timekeeping_forward_now(&timekeeper);
         timekeeping_suspended = 1;
 
         /*
@@ -724,7 +774,7 @@ static int timekeeping_suspend(void)
          * try to compensate so the difference in system time
          * and persistent_clock time stays close to constant.
          */
-        delta = timespec_sub(timekeeper.xtime, timekeeping_suspend_time);
+        delta = timespec_sub(tk_xtime(&timekeeper), timekeeping_suspend_time);
         delta_delta = timespec_sub(delta, old_delta);
         if (abs(delta_delta.tv_sec)  >= 2) {
                 /*
@@ -763,7 +813,8 @@ device_initcall(timekeeping_init_ops);
  * If the error is already larger, we look ahead even further
  * to compensate for late or lost adjustments.
  */
-static __always_inline int timekeeping_bigadjust(s64 error, s64 *interval,
+static __always_inline int timekeeping_bigadjust(struct timekeeper *tk,
+                                                 s64 error, s64 *interval,
                                                  s64 *offset)
 {
         s64 tick_error, i;
@@ -779,7 +830,7 @@ static __always_inline int timekeeping_bigadjust(s64 error, s64 *interval,
          * here.  This is tuned so that an error of about 1 msec is adjusted
          * within about 1 sec (or 2^20 nsec in 2^SHIFT_HZ ticks).
          */
-        error2 = timekeeper.ntp_error >> (NTP_SCALE_SHIFT + 22 - 2 * SHIFT_HZ);
+        error2 = tk->ntp_error >> (NTP_SCALE_SHIFT + 22 - 2 * SHIFT_HZ);
         error2 = abs(error2);
         for (look_ahead = 0; error2 > 0; look_ahead++)
                 error2 >>= 2;
@@ -788,8 +839,8 @@ static __always_inline int timekeeping_bigadjust(s64 error, s64 *interval,
          * Now calculate the error in (1 << look_ahead) ticks, but first
          * remove the single look ahead already included in the error.
          */
-        tick_error = ntp_tick_length() >> (timekeeper.ntp_error_shift + 1);
-        tick_error -= timekeeper.xtime_interval >> 1;
+        tick_error = ntp_tick_length() >> (tk->ntp_error_shift + 1);
+        tick_error -= tk->xtime_interval >> 1;
         error = ((error - tick_error) >> look_ahead) + tick_error;
 
         /* Finally calculate the adjustment shift value.  */
@@ -814,9 +865,9 @@ static __always_inline int timekeeping_bigadjust(s64 error, s64 *interval,
  * this is optimized for the most common adjustments of -1,0,1,
  * for other values we can do a bit more work.
  */
-static void timekeeping_adjust(s64 offset)
+static void timekeeping_adjust(struct timekeeper *tk, s64 offset)
 {
-        s64 error, interval = timekeeper.cycle_interval;
+        s64 error, interval = tk->cycle_interval;
         int adj;
 
         /*
@@ -832,7 +883,7 @@ static void timekeeping_adjust(s64 offset)
          *
          * Note: It does not "save" on aggravation when reading the code.
          */
-        error = timekeeper.ntp_error >> (timekeeper.ntp_error_shift - 1);
+        error = tk->ntp_error >> (tk->ntp_error_shift - 1);
         if (error > interval) {
                 /*
                  * We now divide error by 4(via shift), which checks if
@@ -854,7 +905,8 @@ static void timekeeping_adjust(s64 offset)
                 if (likely(error <= interval))
                         adj = 1;
                 else
-                        adj = timekeeping_bigadjust(error, &interval, &offset);
+                        adj = timekeeping_bigadjust(tk, error, &interval,
+                                                        &offset);
         } else if (error < -interval) {
                 /* See comment above, this is just switched for the negative */
                 error >>= 2;
@@ -863,18 +915,17 @@ static void timekeeping_adjust(s64 offset)
                         interval = -interval;
                         offset = -offset;
                 } else
-                        adj = timekeeping_bigadjust(error, &interval, &offset);
-        } else /* No adjustment needed */
+                        adj = timekeeping_bigadjust(tk, error, &interval,
+                                                        &offset);
+        } else
                 return;
 
-        if (unlikely(timekeeper.clock->maxadj &&
-                        (timekeeper.mult + adj >
-                        timekeeper.clock->mult + timekeeper.clock->maxadj))) {
+        if (unlikely(tk->clock->maxadj &&
+                (tk->mult + adj > tk->clock->mult + tk->clock->maxadj))) {
                 printk_once(KERN_WARNING
                         "Adjusting %s more than 11%% (%ld vs %ld)\n",
-                        timekeeper.clock->name, (long)timekeeper.mult + adj,
-                        (long)timekeeper.clock->mult +
-                                timekeeper.clock->maxadj);
+                        tk->clock->name, (long)tk->mult + adj,
+                        (long)tk->clock->mult + tk->clock->maxadj);
         }
         /*
          * So the following can be confusing.
@@ -925,11 +976,60 @@ static void timekeeping_adjust(s64 offset)
          *
          * XXX - TODO: Doc ntp_error calculation.
          */
-        timekeeper.mult += adj;
-        timekeeper.xtime_interval += interval;
-        timekeeper.xtime_nsec -= offset;
-        timekeeper.ntp_error -= (interval - offset) <<
-                                timekeeper.ntp_error_shift;
+        tk->mult += adj;
+        tk->xtime_interval += interval;
+        tk->xtime_nsec -= offset;
+        tk->ntp_error -= (interval - offset) << tk->ntp_error_shift;
+
+        /*
+         * It may be possible that when we entered this function, xtime_nsec
+         * was very small.  Further, if we're slightly speeding the clocksource
+         * in the code above, its possible the required corrective factor to
+         * xtime_nsec could cause it to underflow.
+         *
+         * Now, since we already accumulated the second, cannot simply roll
+         * the accumulated second back, since the NTP subsystem has been
+         * notified via second_overflow. So instead we push xtime_nsec forward
+         * by the amount we underflowed, and add that amount into the error.
+         *
+         * We'll correct this error next time through this function, when
+         * xtime_nsec is not as small.
+         */
+        if (unlikely((s64)tk->xtime_nsec < 0)) {
+                s64 neg = -(s64)tk->xtime_nsec;
+                tk->xtime_nsec = 0;
+                tk->ntp_error += neg << tk->ntp_error_shift;
+        }
+
+}
+
+
+/**
+ * accumulate_nsecs_to_secs - Accumulates nsecs into secs
+ *
+ * Helper function that accumulates a the nsecs greater then a second
+ * from the xtime_nsec field to the xtime_secs field.
+ * It also calls into the NTP code to handle leapsecond processing.
+ *
+ */
+static inline void accumulate_nsecs_to_secs(struct timekeeper *tk)
+{
+        u64 nsecps = (u64)NSEC_PER_SEC << tk->shift;
+
+        while (tk->xtime_nsec >= nsecps) {
+                int leap;
+
+                tk->xtime_nsec -= nsecps;
+                tk->xtime_sec++;
+
+                /* Figure out if its a leap sec and apply if needed */
+                leap = second_overflow(tk->xtime_sec);
+                tk->xtime_sec += leap;
+                tk->wall_to_monotonic.tv_sec -= leap;
+                if (leap)
+                        clock_was_set_delayed();
+
+        }
 }
 
 
@@ -942,44 +1042,36 @@ static void timekeeping_adjust(s64 offset)
  *
  * Returns the unconsumed cycles.
  */
-static cycle_t logarithmic_accumulation(cycle_t offset, int shift)
+static cycle_t logarithmic_accumulation(struct timekeeper *tk, cycle_t offset,
+                                                u32 shift)
 {
-        u64 nsecps = (u64)NSEC_PER_SEC << timekeeper.shift;
         u64 raw_nsecs;
 
-        /* If the offset is smaller than a shifted interval, do nothing */
-        if (offset < timekeeper.cycle_interval<<shift)
+        /* If the offset is smaller then a shifted interval, do nothing */
+        if (offset < tk->cycle_interval<<shift)
                 return offset;
 
         /* Accumulate one shifted interval */
-        offset -= timekeeper.cycle_interval << shift;
-        timekeeper.clock->cycle_last += timekeeper.cycle_interval << shift;
+        offset -= tk->cycle_interval << shift;
+        tk->clock->cycle_last += tk->cycle_interval << shift;
 
-        timekeeper.xtime_nsec += timekeeper.xtime_interval << shift;
-        while (timekeeper.xtime_nsec >= nsecps) {
-                int leap;
-                timekeeper.xtime_nsec -= nsecps;
-                timekeeper.xtime.tv_sec++;
-                leap = second_overflow(timekeeper.xtime.tv_sec);
-                timekeeper.xtime.tv_sec += leap;
-                timekeeper.wall_to_monotonic.tv_sec -= leap;
-        }
+        tk->xtime_nsec += tk->xtime_interval << shift;
+        accumulate_nsecs_to_secs(tk);
 
         /* Accumulate raw time */
-        raw_nsecs = timekeeper.raw_interval << shift;
-        raw_nsecs += timekeeper.raw_time.tv_nsec;
+        raw_nsecs = tk->raw_interval << shift;
+        raw_nsecs += tk->raw_time.tv_nsec;
         if (raw_nsecs >= NSEC_PER_SEC) {
                 u64 raw_secs = raw_nsecs;
                 raw_nsecs = do_div(raw_secs, NSEC_PER_SEC);
-                timekeeper.raw_time.tv_sec += raw_secs;
+                tk->raw_time.tv_sec += raw_secs;
         }
-        timekeeper.raw_time.tv_nsec = raw_nsecs;
+        tk->raw_time.tv_nsec = raw_nsecs;
 
         /* Accumulate error between NTP and clock interval */
-        timekeeper.ntp_error += ntp_tick_length() << shift;
-        timekeeper.ntp_error -=
-            (timekeeper.xtime_interval + timekeeper.xtime_remainder) <<
-                                (timekeeper.ntp_error_shift + shift);
+        tk->ntp_error += ntp_tick_length() << shift;
+        tk->ntp_error -= (tk->xtime_interval + tk->xtime_remainder) <<
+                                                (tk->ntp_error_shift + shift);
 
         return offset;
 }
@@ -995,6 +1087,7 @@ static void update_wall_time(void)
         cycle_t offset;
         int shift = 0, maxshift;
         unsigned long flags;
+        s64 remainder;
 
         write_seqlock_irqsave(&timekeeper.lock, flags);
 
@@ -1009,8 +1102,6 @@ static void update_wall_time(void)
 #else
         offset = (clock->read(clock) - clock->cycle_last) & clock->mask;
 #endif
-        timekeeper.xtime_nsec = (s64)timekeeper.xtime.tv_nsec <<
-                                                timekeeper.shift;
 
         /*
          * With NO_HZ we may have to accumulate many cycle_intervals
@@ -1026,62 +1117,36 @@ static void update_wall_time(void)
         maxshift = (64 - (ilog2(ntp_tick_length())+1)) - 1;
         shift = min(shift, maxshift);
         while (offset >= timekeeper.cycle_interval) {
-                offset = logarithmic_accumulation(offset, shift);
+                offset = logarithmic_accumulation(&timekeeper, offset, shift);
                 if(offset < timekeeper.cycle_interval<<shift)
                         shift--;
         }
 
         /* correct the clock when NTP error is too big */
-        timekeeping_adjust(offset);
-
-        /*
-         * Since in the loop above, we accumulate any amount of time
-         * in xtime_nsec over a second into xtime.tv_sec, its possible for
-         * xtime_nsec to be fairly small after the loop. Further, if we're
-         * slightly speeding the clocksource up in timekeeping_adjust(),
-         * its possible the required corrective factor to xtime_nsec could
-         * cause it to underflow.
-         *
-         * Now, we cannot simply roll the accumulated second back, since
-         * the NTP subsystem has been notified via second_overflow. So
-         * instead we push xtime_nsec forward by the amount we underflowed,
-         * and add that amount into the error.
-         *
-         * We'll correct this error next time through this function, when
-         * xtime_nsec is not as small.
-         */
-        if (unlikely((s64)timekeeper.xtime_nsec < 0)) {
-                s64 neg = -(s64)timekeeper.xtime_nsec;
-                timekeeper.xtime_nsec = 0;
-                timekeeper.ntp_error += neg << timekeeper.ntp_error_shift;
-        }
+        timekeeping_adjust(&timekeeper, offset);
 
 
         /*
-         * Store full nanoseconds into xtime after rounding it up and
-         * add the remainder to the error difference.
-         */
-        timekeeper.xtime.tv_nsec = ((s64)timekeeper.xtime_nsec >>
-                                                timekeeper.shift) + 1;
-        timekeeper.xtime_nsec -= (s64)timekeeper.xtime.tv_nsec <<
-                                                timekeeper.shift;
-        timekeeper.ntp_error += timekeeper.xtime_nsec <<
-                                timekeeper.ntp_error_shift;
+        * Store only full nanoseconds into xtime_nsec after rounding
+        * it up and add the remainder to the error difference.
+        * XXX - This is necessary to avoid small 1ns inconsistnecies caused
+        * by truncating the remainder in vsyscalls. However, it causes
+        * additional work to be done in timekeeping_adjust(). Once
+        * the vsyscall implementations are converted to use xtime_nsec
+        * (shifted nanoseconds), this can be killed.
+        */
+        remainder = timekeeper.xtime_nsec & ((1 << timekeeper.shift) - 1);
+        timekeeper.xtime_nsec -= remainder;
+        timekeeper.xtime_nsec += 1 << timekeeper.shift;
+        timekeeper.ntp_error += remainder << timekeeper.ntp_error_shift;
 
         /*
          * Finally, make sure that after the rounding
-         * xtime.tv_nsec isn't larger than NSEC_PER_SEC
+         * xtime_nsec isn't larger than NSEC_PER_SEC
          */
-        if (unlikely(timekeeper.xtime.tv_nsec >= NSEC_PER_SEC)) {
-                int leap;
-                timekeeper.xtime.tv_nsec -= NSEC_PER_SEC;
-                timekeeper.xtime.tv_sec++;
-                leap = second_overflow(timekeeper.xtime.tv_sec);
-                timekeeper.xtime.tv_sec += leap;
-                timekeeper.wall_to_monotonic.tv_sec -= leap;
-        }
+        accumulate_nsecs_to_secs(&timekeeper);
 
-        timekeeping_update(false);
+        timekeeping_update(&timekeeper, false);
 
 out:
         write_sequnlock_irqrestore(&timekeeper.lock, flags);
@@ -1126,21 +1191,20 @@ void get_monotonic_boottime(struct timespec *ts)
 {
         struct timespec tomono, sleep;
         unsigned int seq;
-        s64 nsecs;
 
         WARN_ON(timekeeping_suspended);
 
         do {
                 seq = read_seqbegin(&timekeeper.lock);
-                *ts = timekeeper.xtime;
+                ts->tv_sec = timekeeper.xtime_sec;
+                ts->tv_nsec = timekeeping_get_ns(&timekeeper);
                 tomono = timekeeper.wall_to_monotonic;
                 sleep = timekeeper.total_sleep_time;
-                nsecs = timekeeping_get_ns();
 
         } while (read_seqretry(&timekeeper.lock, seq));
 
         set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec + sleep.tv_sec,
-                        ts->tv_nsec + tomono.tv_nsec + sleep.tv_nsec + nsecs);
+                        ts->tv_nsec + tomono.tv_nsec + sleep.tv_nsec);
 }
 EXPORT_SYMBOL_GPL(get_monotonic_boottime);
 
@@ -1173,13 +1237,13 @@ EXPORT_SYMBOL_GPL(monotonic_to_bootbased);
 
 unsigned long get_seconds(void)
 {
-        return timekeeper.xtime.tv_sec;
+        return timekeeper.xtime_sec;
 }
 EXPORT_SYMBOL(get_seconds);
 
 struct timespec __current_kernel_time(void)
 {
-        return timekeeper.xtime;
+        return tk_xtime(&timekeeper);
 }
 
 struct timespec current_kernel_time(void)
@@ -1190,7 +1254,7 @@ struct timespec current_kernel_time(void)
         do {
                 seq = read_seqbegin(&timekeeper.lock);
 
-                now = timekeeper.xtime;
+                now = tk_xtime(&timekeeper);
         } while (read_seqretry(&timekeeper.lock, seq));
 
         return now;
@@ -1205,7 +1269,7 @@ struct timespec get_monotonic_coarse(void)
         do {
                 seq = read_seqbegin(&timekeeper.lock);
 
-                now = timekeeper.xtime;
+                now = tk_xtime(&timekeeper);
                 mono = timekeeper.wall_to_monotonic;
         } while (read_seqretry(&timekeeper.lock, seq));
 
@@ -1240,12 +1304,43 @@ void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim,
 
         do {
                 seq = read_seqbegin(&timekeeper.lock);
-                *xtim = timekeeper.xtime;
+                *xtim = tk_xtime(&timekeeper);
                 *wtom = timekeeper.wall_to_monotonic;
                 *sleep = timekeeper.total_sleep_time;
         } while (read_seqretry(&timekeeper.lock, seq));
 }
 
+#ifdef CONFIG_HIGH_RES_TIMERS
+/**
+ * ktime_get_update_offsets - hrtimer helper
+ * @offs_real:  pointer to storage for monotonic -> realtime offset
+ * @offs_boot:  pointer to storage for monotonic -> boottime offset
+ *
+ * Returns current monotonic time and updates the offsets
+ * Called from hrtimer_interupt() or retrigger_next_event()
+ */
+ktime_t ktime_get_update_offsets(ktime_t *offs_real, ktime_t *offs_boot)
+{
+        ktime_t now;
+        unsigned int seq;
+        u64 secs, nsecs;
+
+        do {
+                seq = read_seqbegin(&timekeeper.lock);
+
+                secs = timekeeper.xtime_sec;
+                nsecs = timekeeping_get_ns(&timekeeper);
+
+                *offs_real = timekeeper.offs_real;
+                *offs_boot = timekeeper.offs_boot;
+        } while (read_seqretry(&timekeeper.lock, seq));
+
+        now = ktime_add_ns(ktime_set(secs, 0), nsecs);
+        now = ktime_sub(now, *offs_real);
+        return now;
+}
+#endif
+
 /**
  * ktime_get_monotonic_offset() - get wall_to_monotonic in ktime_t format
  */