Move the generic timekeeping code from kernel/timer.c to kernel/time/timekeeping.c . This requires some glue code which is added to the include/linux/timekeeping.h header. I tried to be as careful as possible in picking up recent changes to the timekeeping code. This patches is on top of -mm , and moves all the changes included in -mm. This is also moving do_timer and the load calculation code which was connect to the timekeeping code. Moving it provided for slightly better compiler optimization. Signed-Off-By: Daniel Walker --- include/linux/clocksource.h | 3 include/linux/timekeeping.h | 13 kernel/time/Makefile | 2 kernel/time/clocksource.c | 3 kernel/time/timekeeping.c | 711 ++++++++++++++++++++++++++++++++++++++++++++ kernel/timer.c | 694 ------------------------------------------ 6 files changed, 729 insertions(+), 697 deletions(-) Index: linux-2.6.19/include/linux/clocksource.h =================================================================== --- linux-2.6.19.orig/include/linux/clocksource.h +++ linux-2.6.19/include/linux/clocksource.h @@ -18,6 +18,9 @@ /* clocksource cycle base type */ typedef u64 cycle_t; +/* XXX - Would like a better way for initializing curr_clocksource */ +extern struct clocksource clocksource_jiffies; + /** * struct clocksource - hardware abstraction for a free running counter * Provides mostly state-free accessors to the underlying hardware. Index: linux-2.6.19/include/linux/timekeeping.h =================================================================== --- /dev/null +++ linux-2.6.19/include/linux/timekeeping.h @@ -0,0 +1,13 @@ +#ifndef _LINUX_TIMEKEEPING_H +#define _LINUX_TIMEKEEPING_H + +#include + +#ifndef CONFIG_GENERIC_TIME +static inline int change_clocksource(void) +{ + return 0; +} +#endif /* !CONFIG_GENERIC_TIME */ + +#endif /* _LINUX_TIMEKEEPING_H */ Index: linux-2.6.19/kernel/time/Makefile =================================================================== --- linux-2.6.19.orig/kernel/time/Makefile +++ linux-2.6.19/kernel/time/Makefile @@ -1,4 +1,4 @@ -obj-y += ntp.o clocksource.o jiffies.o timer_list.o +obj-y += ntp.o clocksource.o jiffies.o timer_list.o timekeeping.o obj-$(CONFIG_GENERIC_CLOCKEVENTS) += clockevents.o obj-$(CONFIG_GENERIC_CLOCKEVENTS) += tick-common.o Index: linux-2.6.19/kernel/time/clocksource.c =================================================================== --- linux-2.6.19.orig/kernel/time/clocksource.c +++ linux-2.6.19/kernel/time/clocksource.c @@ -29,9 +29,6 @@ #include #include -/* XXX - Would like a better way for initializing curr_clocksource */ -extern struct clocksource clocksource_jiffies; - /*[Clocksource internal variables]--------- * curr_clocksource: * currently selected clocksource. Initialized to clocksource_jiffies. Index: linux-2.6.19/kernel/time/timekeeping.c =================================================================== --- /dev/null +++ linux-2.6.19/kernel/time/timekeeping.c @@ -0,0 +1,711 @@ +/* + * linux/kernel/time/timekeeping.c + * + * timekeeping functions + * + */ + +#include +#include +#include +#include +#include +#include + +/* + * The current time + * 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 + * at zero at system boot time, so wall_to_monotonic will be negative, + * however, we will ALWAYS keep the tv_nsec part positive so we can use + * the usual normalization. + */ +struct timespec xtime __attribute__ ((aligned (16))); +struct timespec wall_to_monotonic __attribute__ ((aligned (16))); + +EXPORT_SYMBOL(xtime); + +/* + * flag for if timekeeping is suspended + */ +static int timekeeping_suspended; + +/* + * time in seconds when suspend began + */ +static unsigned long timekeeping_suspend_time; + +/* + * Clock used for timekeeping + */ +struct clocksource *clock = &clocksource_jiffies; + +#ifdef CONFIG_GENERIC_TIME +/** + * __get_nsec_offset - Returns nanoseconds since last call to periodic_hook + * + * private function, must hold xtime_lock lock when being + * called. Returns the number of nanoseconds since the + * last call to update_wall_time() (adjusted by NTP scaling) + */ +static inline s64 __get_nsec_offset(void) +{ + cycle_t cycle_now, cycle_delta; + s64 ns_offset; + + /* read clocksource: */ + cycle_now = clocksource_read(clock); + + /* calculate the delta since the last update_wall_time: */ + cycle_delta = (cycle_now - clock->cycle_last) & clock->mask; + + /* convert to nanoseconds: */ + ns_offset = cyc2ns(clock, cycle_delta); + + return ns_offset; +} + +/** + * __get_realtime_clock_ts - Returns the time of day in a timespec + * @ts: pointer to the timespec to be set + * + * Returns the time of day in a timespec. Used by + * do_gettimeofday() and get_realtime_clock_ts(). + */ +static inline void __get_realtime_clock_ts(struct timespec *ts) +{ + unsigned long seq; + s64 nsecs; + + do { + seq = read_seqbegin(&xtime_lock); + + *ts = xtime; + nsecs = __get_nsec_offset(); + + } while (read_seqretry(&xtime_lock, seq)); + + timespec_add_ns(ts, nsecs); +} + +/** + * getnstimeofday - Returns the time of day in a timespec + * @ts: pointer to the timespec to be set + * + * Returns the time of day in a timespec. + */ +void getnstimeofday(struct timespec *ts) +{ + __get_realtime_clock_ts(ts); +} + +EXPORT_SYMBOL(getnstimeofday); + +/** + * do_gettimeofday - Returns the time of day in a timeval + * @tv: pointer to the timeval to be set + * + * NOTE: Users should be converted to using get_realtime_clock_ts() + */ +void do_gettimeofday(struct timeval *tv) +{ + struct timespec now; + + __get_realtime_clock_ts(&now); + tv->tv_sec = now.tv_sec; + tv->tv_usec = now.tv_nsec/1000; +} + +EXPORT_SYMBOL(do_gettimeofday); +/** + * do_settimeofday - Sets the time of day + * @tv: pointer to the timespec variable containing the new time + * + * Sets the time of day to the new time and update NTP and notify hrtimers + */ +int do_settimeofday(struct timespec *tv) +{ + unsigned long flags; + time_t wtm_sec, sec = tv->tv_sec; + long wtm_nsec, nsec = tv->tv_nsec; + + if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) + return -EINVAL; + + write_seqlock_irqsave(&xtime_lock, flags); + + nsec -= __get_nsec_offset(); + + wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec); + wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec); + + set_normalized_timespec(&xtime, sec, nsec); + set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec); + + clock->error = 0; + ntp_clear(); + + write_sequnlock_irqrestore(&xtime_lock, flags); + + /* signal hrtimers about time change */ + clock_was_set(); + + return 0; +} + +EXPORT_SYMBOL(do_settimeofday); + +/** + * change_clocksource - Swaps clocksources if a new one is available + * + * Accumulates current time interval and initializes new clocksource + */ +static int change_clocksource(void) +{ + struct clocksource *new; + cycle_t now; + u64 nsec; + new = clocksource_get_next(); + if (clock != new) { + now = clocksource_read(new); + nsec = __get_nsec_offset(); + timespec_add_ns(&xtime, nsec); + + clock = new; + clock->cycle_last = now; + printk(KERN_INFO "Time: %s clocksource has been installed.\n", + clock->name); + return 1; + } else if (clock->update_callback) { + return clock->update_callback(); + } + return 0; +} + +/** + * timeofday_is_continuous - check to see if timekeeping is free running + */ +int timekeeping_is_continuous(void) +{ + unsigned long seq; + int ret; + + do { + seq = read_seqbegin(&xtime_lock); + + ret = clock->flags & CLOCK_SOURCE_IS_CONTINUOUS; + + } while (read_seqretry(&xtime_lock, seq)); + + return ret; +} +#endif /* CONFIG_GENERIC_TIME */ + +/** + * read_persistent_clock - Return time in seconds from the persistent clock. + * + * Weak dummy function for arches that do not yet support it. + * Returns seconds from epoch using the battery backed persistent clock. + * Returns zero if unsupported. + * + * XXX - Do be sure to remove it once all arches implement it. + */ +unsigned long __attribute__((weak)) read_persistent_clock(void) +{ + return 0; +} + +/* + * timekeeping_init - Initializes the clocksource and common timekeeping values + */ +void __init timekeeping_init(void) +{ + unsigned long flags; + unsigned long sec = read_persistent_clock(); + + write_seqlock_irqsave(&xtime_lock, flags); + + ntp_clear(); + + clock = clocksource_get_next(); + clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH); + clock->cycle_last = clocksource_read(clock); + + xtime.tv_sec = sec; + xtime.tv_nsec = 0; + set_normalized_timespec(&wall_to_monotonic, + -xtime.tv_sec, -xtime.tv_nsec); + + write_sequnlock_irqrestore(&xtime_lock, flags); +} + + +/* flag for if timekeeping is suspended */ +static int timekeeping_suspended; +/* time in seconds when suspend began */ +static unsigned long timekeeping_suspend_time; + +/** + * timekeeping_resume - Resumes the generic timekeeping subsystem. + * @dev: unused + * + * This is for the generic clocksource timekeeping. + * xtime/wall_to_monotonic/jiffies/etc are + * still managed by arch specific suspend/resume code. + */ +static int timekeeping_resume(struct sys_device *dev) +{ + unsigned long flags; + unsigned long now = read_persistent_clock(); + + write_seqlock_irqsave(&xtime_lock, flags); + + if (now && (now > timekeeping_suspend_time)) { + unsigned long sleep_length = now - timekeeping_suspend_time; + xtime.tv_sec += sleep_length; + jiffies_64 += (u64)sleep_length * HZ; + } + /* re-base the last cycle value */ + clock->cycle_last = clocksource_read(clock); + clock->error = 0; + timekeeping_suspended = 0; + write_sequnlock_irqrestore(&xtime_lock, flags); + + clockevents_notify(CLOCK_EVT_NOTIFY_RESUME, NULL); + /* Resume hrtimers */ + clock_was_set(); + + return 0; +} + +static int timekeeping_suspend(struct sys_device *dev, pm_message_t state) +{ + unsigned long flags; + + write_seqlock_irqsave(&xtime_lock, flags); + timekeeping_suspended = 1; + timekeeping_suspend_time = read_persistent_clock(); + write_sequnlock_irqrestore(&xtime_lock, flags); + return 0; +} + +/* sysfs resume/suspend bits for timekeeping */ +static struct sysdev_class timekeeping_sysclass = { + .resume = timekeeping_resume, + .suspend = timekeeping_suspend, + set_kset_name("timekeeping"), +}; + +static struct sys_device device_timer = { + .id = 0, + .cls = &timekeeping_sysclass, +}; + +static int __init timekeeping_init_device(void) +{ + int error = sysdev_class_register(&timekeeping_sysclass); + if (!error) + error = sysdev_register(&device_timer); + return error; +} + +device_initcall(timekeeping_init_device); + +/* + * If the error is already larger, we look ahead even further + * to compensate for late or lost adjustments. + */ +static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, + s64 *offset) +{ + s64 tick_error, i; + u32 look_ahead, adj; + s32 error2, mult; + + /* + * Use the current error value to determine how much to look ahead. + * The larger the error the slower we adjust for it to avoid problems + * with losing too many ticks, otherwise we would overadjust and + * produce an even larger error. The smaller the adjustment the + * faster we try to adjust for it, as lost ticks can do less harm + * here. This is tuned so that an error of about 1 msec is adusted + * within about 1 sec (or 2^20 nsec in 2^SHIFT_HZ ticks). + */ + error2 = clock->error >> (TICK_LENGTH_SHIFT + 22 - 2 * SHIFT_HZ); + error2 = abs(error2); + for (look_ahead = 0; error2 > 0; look_ahead++) + error2 >>= 2; + + /* + * Now calculate the error in (1 << look_ahead) ticks, but first + * remove the single look ahead already included in the error. + */ + tick_error = current_tick_length() >> + (TICK_LENGTH_SHIFT - clock->shift + 1); + tick_error -= clock->xtime_interval >> 1; + error = ((error - tick_error) >> look_ahead) + tick_error; + + /* Finally calculate the adjustment shift value. */ + i = *interval; + mult = 1; + if (error < 0) { + error = -error; + *interval = -*interval; + *offset = -*offset; + mult = -1; + } + for (adj = 0; error > i; adj++) + error >>= 1; + + *interval <<= adj; + *offset <<= adj; + return mult << adj; +} + +/* + * Adjust the multiplier to reduce the error value, + * this is optimized for the most common adjustments of -1,0,1, + * for other values we can do a bit more work. + */ +static void clocksource_adjust(struct clocksource *clock, s64 offset) +{ + s64 error, interval = clock->cycle_interval; + int adj; + + error = clock->error >> (TICK_LENGTH_SHIFT - clock->shift - 1); + if (error > interval) { + error >>= 2; + if (likely(error <= interval)) + adj = 1; + else + adj = clocksource_bigadjust(error, &interval, &offset); + } else if (error < -interval) { + error >>= 2; + if (likely(error >= -interval)) { + adj = -1; + interval = -interval; + offset = -offset; + } else + adj = clocksource_bigadjust(error, &interval, &offset); + } else + return; + + clock->mult += adj; + clock->xtime_interval += interval; + clock->xtime_nsec -= offset; + clock->error -= (interval - offset) << + (TICK_LENGTH_SHIFT - clock->shift); +} + +/** + * update_wall_time - Uses the current clocksource to increment the wall time + * + * Called from the timer interrupt, must hold a write on xtime_lock. + */ +static inline void update_wall_time(void) +{ + cycle_t offset; + + /* Make sure we're fully resumed: */ + if (unlikely(timekeeping_suspended)) + return; + +#ifdef CONFIG_GENERIC_TIME + offset = (clocksource_read(clock) - clock->cycle_last) & clock->mask; +#else + offset = clock->cycle_interval; +#endif + clock->xtime_nsec += (s64)xtime.tv_nsec << clock->shift; + + /* normally this loop will run just once, however in the + * case of lost or late ticks, it will accumulate correctly. + */ + while (offset >= clock->cycle_interval) { + /* accumulate one interval */ + clock->xtime_nsec += clock->xtime_interval; + clock->cycle_last += clock->cycle_interval; + offset -= clock->cycle_interval; + + if (clock->xtime_nsec >= (u64)NSEC_PER_SEC << clock->shift) { + clock->xtime_nsec -= (u64)NSEC_PER_SEC << clock->shift; + xtime.tv_sec++; + second_overflow(); + } + + /* interpolator bits */ + time_interpolator_update(clock->xtime_interval + >> clock->shift); + + /* accumulate error between NTP and clock interval */ + clock->error += current_tick_length(); + clock->error -= clock->xtime_interval << (TICK_LENGTH_SHIFT - clock->shift); + } + + /* correct the clock when NTP error is too big */ + clocksource_adjust(clock, offset); + + /* store full nanoseconds into xtime */ + xtime.tv_nsec = (s64)clock->xtime_nsec >> clock->shift; + clock->xtime_nsec -= (s64)xtime.tv_nsec << clock->shift; + + /* check to see if there is a new clocksource to use */ + if (change_clocksource()) { + clock->error = 0; + clock->xtime_nsec = 0; + tick_clock_notify(); + clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH); + } + update_vsyscall(&xtime, clock); +} + +/* + * Nr of active tasks - counted in fixed-point numbers + */ +static unsigned long count_active_tasks(void) +{ + return nr_active() * FIXED_1; +} + +/* + * Hmm.. Changed this, as the GNU make sources (load.c) seems to + * imply that avenrun[] is the standard name for this kind of thing. + * Nothing else seems to be standardized: the fractional size etc + * all seem to differ on different machines. + * + * Requires xtime_lock to access. + */ +unsigned long avenrun[3]; + +EXPORT_SYMBOL(avenrun); + +/* + * calc_load - given tick count, update the avenrun load estimates. + * This is called while holding a write_lock on xtime_lock. + */ +static inline void calc_load(unsigned long ticks) +{ + unsigned long active_tasks; /* fixed-point */ + static int count = LOAD_FREQ; + + count -= ticks; + if (unlikely(count < 0)) { + active_tasks = count_active_tasks(); + do { + CALC_LOAD(avenrun[0], EXP_1, active_tasks); + CALC_LOAD(avenrun[1], EXP_5, active_tasks); + CALC_LOAD(avenrun[2], EXP_15, active_tasks); + count += LOAD_FREQ; + } while (count < 0); + } +} + +/* + * Called by the timer interrupt. xtime_lock must already be taken + * by the timer IRQ! + */ +static inline void update_times(unsigned long ticks) +{ + update_wall_time(); + calc_load(ticks); +} + +/* + * The 64-bit jiffies value is not atomic - you MUST NOT read it + * without sampling the sequence number in xtime_lock. + * jiffies is defined in the linker script... + */ + +void do_timer(unsigned long ticks) +{ + jiffies_64 += ticks; + update_times(ticks); +} + +#ifdef CONFIG_TIME_INTERPOLATION + +struct time_interpolator *time_interpolator __read_mostly; +static struct time_interpolator *time_interpolator_list __read_mostly; +static DEFINE_SPINLOCK(time_interpolator_lock); + +static inline cycles_t time_interpolator_get_cycles(unsigned int src) +{ + unsigned long (*x)(void); + + switch (src) + { + case TIME_SOURCE_FUNCTION: + x = time_interpolator->addr; + return x(); + + case TIME_SOURCE_MMIO64 : + return readq_relaxed((void __iomem *)time_interpolator->addr); + + case TIME_SOURCE_MMIO32 : + return readl_relaxed((void __iomem *)time_interpolator->addr); + + default: return get_cycles(); + } +} + +static inline u64 time_interpolator_get_counter(int writelock) +{ + unsigned int src = time_interpolator->source; + + if (time_interpolator->jitter) + { + cycles_t lcycle; + cycles_t now; + + do { + lcycle = time_interpolator->last_cycle; + now = time_interpolator_get_cycles(src); + if (lcycle && time_after(lcycle, now)) + return lcycle; + + /* When holding the xtime write lock, there's no need + * to add the overhead of the cmpxchg. Readers are + * force to retry until the write lock is released. + */ + if (writelock) { + time_interpolator->last_cycle = now; + return now; + } + /* Keep track of the last timer value returned. The use of cmpxchg here + * will cause contention in an SMP environment. + */ + } while (unlikely(cmpxchg(&time_interpolator->last_cycle, lcycle, now) != lcycle)); + return now; + } + else + return time_interpolator_get_cycles(src); +} + +void time_interpolator_reset(void) +{ + time_interpolator->offset = 0; + time_interpolator->last_counter = time_interpolator_get_counter(1); +} + +#define GET_TI_NSECS(count,i) (((((count) - i->last_counter) & (i)->mask) * (i)->nsec_per_cyc) >> (i)->shift) + +unsigned long time_interpolator_get_offset(void) +{ + /* If we do not have a time interpolator set up then just return zero */ + if (!time_interpolator) + return 0; + + return time_interpolator->offset + + GET_TI_NSECS(time_interpolator_get_counter(0), time_interpolator); +} + +#define INTERPOLATOR_ADJUST 65536 +#define INTERPOLATOR_MAX_SKIP 10*INTERPOLATOR_ADJUST + +void time_interpolator_update(long delta_nsec) +{ + u64 counter; + unsigned long offset; + + /* If there is no time interpolator set up then do nothing */ + if (!time_interpolator) + return; + + /* + * The interpolator compensates for late ticks by accumulating the late + * time in time_interpolator->offset. A tick earlier than expected will + * lead to a reset of the offset and a corresponding jump of the clock + * forward. Again this only works if the interpolator clock is running + * slightly slower than the regular clock and the tuning logic insures + * that. + */ + + counter = time_interpolator_get_counter(1); + offset = time_interpolator->offset + + GET_TI_NSECS(counter, time_interpolator); + + if (delta_nsec < 0 || (unsigned long) delta_nsec < offset) + time_interpolator->offset = offset - delta_nsec; + else { + time_interpolator->skips++; + time_interpolator->ns_skipped += delta_nsec - offset; + time_interpolator->offset = 0; + } + time_interpolator->last_counter = counter; + + /* Tuning logic for time interpolator invoked every minute or so. + * Decrease interpolator clock speed if no skips occurred and an offset is carried. + * Increase interpolator clock speed if we skip too much time. + */ + if (jiffies % INTERPOLATOR_ADJUST == 0) + { + if (time_interpolator->skips == 0 && time_interpolator->offset > tick_nsec) + time_interpolator->nsec_per_cyc--; + if (time_interpolator->ns_skipped > INTERPOLATOR_MAX_SKIP && time_interpolator->offset == 0) + time_interpolator->nsec_per_cyc++; + time_interpolator->skips = 0; + time_interpolator->ns_skipped = 0; + } +} + +static inline int +is_better_time_interpolator(struct time_interpolator *new) +{ + if (!time_interpolator) + return 1; + return new->frequency > 2*time_interpolator->frequency || + (unsigned long)new->drift < (unsigned long)time_interpolator->drift; +} + +void +register_time_interpolator(struct time_interpolator *ti) +{ + unsigned long flags; + + /* Sanity check */ + BUG_ON(ti->frequency == 0 || ti->mask == 0); + + ti->nsec_per_cyc = ((u64)NSEC_PER_SEC << ti->shift) / ti->frequency; + spin_lock(&time_interpolator_lock); + write_seqlock_irqsave(&xtime_lock, flags); + if (is_better_time_interpolator(ti)) { + time_interpolator = ti; + time_interpolator_reset(); + } + write_sequnlock_irqrestore(&xtime_lock, flags); + + ti->next = time_interpolator_list; + time_interpolator_list = ti; + spin_unlock(&time_interpolator_lock); +} + +void +unregister_time_interpolator(struct time_interpolator *ti) +{ + struct time_interpolator *curr, **prev; + unsigned long flags; + + spin_lock(&time_interpolator_lock); + prev = &time_interpolator_list; + for (curr = *prev; curr; curr = curr->next) { + if (curr == ti) { + *prev = curr->next; + break; + } + prev = &curr->next; + } + + write_seqlock_irqsave(&xtime_lock, flags); + if (ti == time_interpolator) { + /* we lost the best time-interpolator: */ + time_interpolator = NULL; + /* find the next-best interpolator */ + for (curr = time_interpolator_list; curr; curr = curr->next) + if (is_better_time_interpolator(curr)) + time_interpolator = curr; + time_interpolator_reset(); + } + write_sequnlock_irqrestore(&xtime_lock, flags); + spin_unlock(&time_interpolator_lock); +} +#endif /* CONFIG_TIME_INTERPOLATION */ + Index: linux-2.6.19/kernel/timer.c =================================================================== --- linux-2.6.19.orig/kernel/timer.c +++ linux-2.6.19/kernel/timer.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include @@ -736,449 +737,6 @@ unsigned long next_timer_interrupt(void) #endif -/******************************************************************/ - -/* - * The current time - * 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 - * at zero at system boot time, so wall_to_monotonic will be negative, - * however, we will ALWAYS keep the tv_nsec part positive so we can use - * the usual normalization. - */ -struct timespec xtime __attribute__ ((aligned (16))); -struct timespec wall_to_monotonic __attribute__ ((aligned (16))); - -EXPORT_SYMBOL(xtime); - - -/* XXX - all of this timekeeping code should be later moved to time.c */ -#include -static struct clocksource *clock; /* pointer to current clocksource */ - -#ifdef CONFIG_GENERIC_TIME -/** - * __get_nsec_offset - Returns nanoseconds since last call to periodic_hook - * - * private function, must hold xtime_lock lock when being - * called. Returns the number of nanoseconds since the - * last call to update_wall_time() (adjusted by NTP scaling) - */ -static inline s64 __get_nsec_offset(void) -{ - cycle_t cycle_now, cycle_delta; - s64 ns_offset; - - /* read clocksource: */ - cycle_now = clocksource_read(clock); - - /* calculate the delta since the last update_wall_time: */ - cycle_delta = (cycle_now - clock->cycle_last) & clock->mask; - - /* convert to nanoseconds: */ - ns_offset = cyc2ns(clock, cycle_delta); - - return ns_offset; -} - -/** - * __get_realtime_clock_ts - Returns the time of day in a timespec - * @ts: pointer to the timespec to be set - * - * Returns the time of day in a timespec. Used by - * do_gettimeofday() and get_realtime_clock_ts(). - */ -static inline void __get_realtime_clock_ts(struct timespec *ts) -{ - unsigned long seq; - s64 nsecs; - - do { - seq = read_seqbegin(&xtime_lock); - - *ts = xtime; - nsecs = __get_nsec_offset(); - - } while (read_seqretry(&xtime_lock, seq)); - - timespec_add_ns(ts, nsecs); -} - -/** - * getnstimeofday - Returns the time of day in a timespec - * @ts: pointer to the timespec to be set - * - * Returns the time of day in a timespec. - */ -void getnstimeofday(struct timespec *ts) -{ - __get_realtime_clock_ts(ts); -} - -EXPORT_SYMBOL(getnstimeofday); - -/** - * do_gettimeofday - Returns the time of day in a timeval - * @tv: pointer to the timeval to be set - * - * NOTE: Users should be converted to using get_realtime_clock_ts() - */ -void do_gettimeofday(struct timeval *tv) -{ - struct timespec now; - - __get_realtime_clock_ts(&now); - tv->tv_sec = now.tv_sec; - tv->tv_usec = now.tv_nsec/1000; -} - -EXPORT_SYMBOL(do_gettimeofday); -/** - * do_settimeofday - Sets the time of day - * @tv: pointer to the timespec variable containing the new time - * - * Sets the time of day to the new time and update NTP and notify hrtimers - */ -int do_settimeofday(struct timespec *tv) -{ - unsigned long flags; - time_t wtm_sec, sec = tv->tv_sec; - long wtm_nsec, nsec = tv->tv_nsec; - - if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) - return -EINVAL; - - write_seqlock_irqsave(&xtime_lock, flags); - - nsec -= __get_nsec_offset(); - - wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec); - wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec); - - set_normalized_timespec(&xtime, sec, nsec); - set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec); - - clock->error = 0; - ntp_clear(); - - write_sequnlock_irqrestore(&xtime_lock, flags); - - /* signal hrtimers about time change */ - clock_was_set(); - - return 0; -} - -EXPORT_SYMBOL(do_settimeofday); - -/** - * change_clocksource - Swaps clocksources if a new one is available - * - * Accumulates current time interval and initializes new clocksource - */ -static int change_clocksource(void) -{ - struct clocksource *new; - cycle_t now; - u64 nsec; - new = clocksource_get_next(); - if (clock != new) { - now = clocksource_read(new); - nsec = __get_nsec_offset(); - timespec_add_ns(&xtime, nsec); - - clock = new; - clock->cycle_last = now; - tick_clock_notify(); - printk(KERN_INFO "Time: %s clocksource has been installed.\n", - clock->name); - return 1; - } else if (clock->update_callback) { - return clock->update_callback(); - } - return 0; -} -#else -static inline int change_clocksource(void) -{ - return 0; -} -#endif - -/** - * timeofday_is_continuous - check to see if timekeeping is free running - */ -int timekeeping_is_continuous(void) -{ - unsigned long seq; - int ret; - - do { - seq = read_seqbegin(&xtime_lock); - - ret = clock->flags & CLOCK_SOURCE_IS_CONTINUOUS; - - } while (read_seqretry(&xtime_lock, seq)); - - return ret; -} - -/** - * read_persistent_clock - Return time in seconds from the persistent clock. - * - * Weak dummy function for arches that do not yet support it. - * Returns seconds from epoch using the battery backed persistent clock. - * Returns zero if unsupported. - * - * XXX - Do be sure to remove it once all arches implement it. - */ -unsigned long __attribute__((weak)) read_persistent_clock(void) -{ - return 0; -} - -/* - * timekeeping_init - Initializes the clocksource and common timekeeping values - */ -void __init timekeeping_init(void) -{ - unsigned long flags; - unsigned long sec = read_persistent_clock(); - - write_seqlock_irqsave(&xtime_lock, flags); - - ntp_clear(); - - clock = clocksource_get_next(); - clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH); - clock->cycle_last = clocksource_read(clock); - - xtime.tv_sec = sec; - xtime.tv_nsec = 0; - set_normalized_timespec(&wall_to_monotonic, - -xtime.tv_sec, -xtime.tv_nsec); - - write_sequnlock_irqrestore(&xtime_lock, flags); -} - - -/* flag for if timekeeping is suspended */ -static int timekeeping_suspended; -/* time in seconds when suspend began */ -static unsigned long timekeeping_suspend_time; - -/** - * timekeeping_resume - Resumes the generic timekeeping subsystem. - * @dev: unused - * - * This is for the generic clocksource timekeeping. - * xtime/wall_to_monotonic/jiffies/etc are - * still managed by arch specific suspend/resume code. - */ -static int timekeeping_resume(struct sys_device *dev) -{ - unsigned long flags; - unsigned long now = read_persistent_clock(); - - write_seqlock_irqsave(&xtime_lock, flags); - - if (now && (now > timekeeping_suspend_time)) { - unsigned long sleep_length = now - timekeeping_suspend_time; - xtime.tv_sec += sleep_length; - jiffies_64 += (u64)sleep_length * HZ; - } - /* re-base the last cycle value */ - clock->cycle_last = clocksource_read(clock); - clock->error = 0; - timekeeping_suspended = 0; - write_sequnlock_irqrestore(&xtime_lock, flags); - - clockevents_notify(CLOCK_EVT_NOTIFY_RESUME, NULL); - /* Resume hrtimers */ - clock_was_set(); - - return 0; -} - -static int timekeeping_suspend(struct sys_device *dev, pm_message_t state) -{ - unsigned long flags; - - write_seqlock_irqsave(&xtime_lock, flags); - timekeeping_suspended = 1; - timekeeping_suspend_time = read_persistent_clock(); - write_sequnlock_irqrestore(&xtime_lock, flags); - return 0; -} - -/* sysfs resume/suspend bits for timekeeping */ -static struct sysdev_class timekeeping_sysclass = { - .resume = timekeeping_resume, - .suspend = timekeeping_suspend, - set_kset_name("timekeeping"), -}; - -static struct sys_device device_timer = { - .id = 0, - .cls = &timekeeping_sysclass, -}; - -static int __init timekeeping_init_device(void) -{ - int error = sysdev_class_register(&timekeeping_sysclass); - if (!error) - error = sysdev_register(&device_timer); - return error; -} - -device_initcall(timekeeping_init_device); - -/* - * If the error is already larger, we look ahead even further - * to compensate for late or lost adjustments. - */ -static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, - s64 *offset) -{ - s64 tick_error, i; - u32 look_ahead, adj; - s32 error2, mult; - - /* - * Use the current error value to determine how much to look ahead. - * The larger the error the slower we adjust for it to avoid problems - * with losing too many ticks, otherwise we would overadjust and - * produce an even larger error. The smaller the adjustment the - * faster we try to adjust for it, as lost ticks can do less harm - * here. This is tuned so that an error of about 1 msec is adusted - * within about 1 sec (or 2^20 nsec in 2^SHIFT_HZ ticks). - */ - error2 = clock->error >> (TICK_LENGTH_SHIFT + 22 - 2 * SHIFT_HZ); - error2 = abs(error2); - for (look_ahead = 0; error2 > 0; look_ahead++) - error2 >>= 2; - - /* - * Now calculate the error in (1 << look_ahead) ticks, but first - * remove the single look ahead already included in the error. - */ - tick_error = current_tick_length() >> - (TICK_LENGTH_SHIFT - clock->shift + 1); - tick_error -= clock->xtime_interval >> 1; - error = ((error - tick_error) >> look_ahead) + tick_error; - - /* Finally calculate the adjustment shift value. */ - i = *interval; - mult = 1; - if (error < 0) { - error = -error; - *interval = -*interval; - *offset = -*offset; - mult = -1; - } - for (adj = 0; error > i; adj++) - error >>= 1; - - *interval <<= adj; - *offset <<= adj; - return mult << adj; -} - -/* - * Adjust the multiplier to reduce the error value, - * this is optimized for the most common adjustments of -1,0,1, - * for other values we can do a bit more work. - */ -static void clocksource_adjust(struct clocksource *clock, s64 offset) -{ - s64 error, interval = clock->cycle_interval; - int adj; - - error = clock->error >> (TICK_LENGTH_SHIFT - clock->shift - 1); - if (error > interval) { - error >>= 2; - if (likely(error <= interval)) - adj = 1; - else - adj = clocksource_bigadjust(error, &interval, &offset); - } else if (error < -interval) { - error >>= 2; - if (likely(error >= -interval)) { - adj = -1; - interval = -interval; - offset = -offset; - } else - adj = clocksource_bigadjust(error, &interval, &offset); - } else - return; - - clock->mult += adj; - clock->xtime_interval += interval; - clock->xtime_nsec -= offset; - clock->error -= (interval - offset) << - (TICK_LENGTH_SHIFT - clock->shift); -} - -/** - * update_wall_time - Uses the current clocksource to increment the wall time - * - * Called from the timer interrupt, must hold a write on xtime_lock. - */ -static void update_wall_time(void) -{ - cycle_t offset; - - /* Make sure we're fully resumed: */ - if (unlikely(timekeeping_suspended)) - return; - -#ifdef CONFIG_GENERIC_TIME - offset = (clocksource_read(clock) - clock->cycle_last) & clock->mask; -#else - offset = clock->cycle_interval; -#endif - clock->xtime_nsec += (s64)xtime.tv_nsec << clock->shift; - - /* normally this loop will run just once, however in the - * case of lost or late ticks, it will accumulate correctly. - */ - while (offset >= clock->cycle_interval) { - /* accumulate one interval */ - clock->xtime_nsec += clock->xtime_interval; - clock->cycle_last += clock->cycle_interval; - offset -= clock->cycle_interval; - - if (clock->xtime_nsec >= (u64)NSEC_PER_SEC << clock->shift) { - clock->xtime_nsec -= (u64)NSEC_PER_SEC << clock->shift; - xtime.tv_sec++; - second_overflow(); - } - - /* interpolator bits */ - time_interpolator_update(clock->xtime_interval - >> clock->shift); - - /* accumulate error between NTP and clock interval */ - clock->error += current_tick_length(); - clock->error -= clock->xtime_interval << (TICK_LENGTH_SHIFT - clock->shift); - } - - /* correct the clock when NTP error is too big */ - clocksource_adjust(clock, offset); - - /* store full nanoseconds into xtime */ - xtime.tv_nsec = (s64)clock->xtime_nsec >> clock->shift; - clock->xtime_nsec -= (s64)xtime.tv_nsec << clock->shift; - - /* check to see if there is a new clocksource to use */ - if (change_clocksource()) { - clock->error = 0; - clock->xtime_nsec = 0; - clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH); - } - update_vsyscall(&xtime, clock); -} - /* * Called from the timer interrupt handler to charge one tick to the current * process. user_tick is 1 if the tick is user time, 0 for system. @@ -1201,47 +759,6 @@ void update_process_times(int user_tick) } /* - * Nr of active tasks - counted in fixed-point numbers - */ -static unsigned long count_active_tasks(void) -{ - return nr_active() * FIXED_1; -} - -/* - * Hmm.. Changed this, as the GNU make sources (load.c) seems to - * imply that avenrun[] is the standard name for this kind of thing. - * Nothing else seems to be standardized: the fractional size etc - * all seem to differ on different machines. - * - * Requires xtime_lock to access. - */ -unsigned long avenrun[3]; - -EXPORT_SYMBOL(avenrun); - -/* - * calc_load - given tick count, update the avenrun load estimates. - * This is called while holding a write_lock on xtime_lock. - */ -static inline void calc_load(unsigned long ticks) -{ - unsigned long active_tasks; /* fixed-point */ - static int count = LOAD_FREQ; - - count -= ticks; - if (unlikely(count < 0)) { - active_tasks = count_active_tasks(); - do { - CALC_LOAD(avenrun[0], EXP_1, active_tasks); - CALC_LOAD(avenrun[1], EXP_5, active_tasks); - CALC_LOAD(avenrun[2], EXP_15, active_tasks); - count += LOAD_FREQ; - } while (count < 0); - } -} - -/* * This read-write spinlock protects us from races in SMP while * playing with xtime and avenrun. */ @@ -1271,27 +788,6 @@ void run_local_timers(void) softlockup_tick(); } -/* - * Called by the timer interrupt. xtime_lock must already be taken - * by the timer IRQ! - */ -static inline void update_times(unsigned long ticks) -{ - update_wall_time(); - calc_load(ticks); -} - -/* - * The 64-bit jiffies value is not atomic - you MUST NOT read it - * without sampling the sequence number in xtime_lock. - * jiffies is defined in the linker script... - */ - -void do_timer(unsigned long ticks) -{ - jiffies_64 += ticks; - update_times(ticks); -} #ifdef __ARCH_WANT_SYS_ALARM @@ -1710,194 +1206,6 @@ void __init init_timers(void) open_softirq(TIMER_SOFTIRQ, run_timer_softirq, NULL); } -#ifdef CONFIG_TIME_INTERPOLATION - -struct time_interpolator *time_interpolator __read_mostly; -static struct time_interpolator *time_interpolator_list __read_mostly; -static DEFINE_SPINLOCK(time_interpolator_lock); - -static inline cycles_t time_interpolator_get_cycles(unsigned int src) -{ - unsigned long (*x)(void); - - switch (src) - { - case TIME_SOURCE_FUNCTION: - x = time_interpolator->addr; - return x(); - - case TIME_SOURCE_MMIO64 : - return readq_relaxed((void __iomem *)time_interpolator->addr); - - case TIME_SOURCE_MMIO32 : - return readl_relaxed((void __iomem *)time_interpolator->addr); - - default: return get_cycles(); - } -} - -static inline u64 time_interpolator_get_counter(int writelock) -{ - unsigned int src = time_interpolator->source; - - if (time_interpolator->jitter) - { - cycles_t lcycle; - cycles_t now; - - do { - lcycle = time_interpolator->last_cycle; - now = time_interpolator_get_cycles(src); - if (lcycle && time_after(lcycle, now)) - return lcycle; - - /* When holding the xtime write lock, there's no need - * to add the overhead of the cmpxchg. Readers are - * force to retry until the write lock is released. - */ - if (writelock) { - time_interpolator->last_cycle = now; - return now; - } - /* Keep track of the last timer value returned. The use of cmpxchg here - * will cause contention in an SMP environment. - */ - } while (unlikely(cmpxchg(&time_interpolator->last_cycle, lcycle, now) != lcycle)); - return now; - } - else - return time_interpolator_get_cycles(src); -} - -void time_interpolator_reset(void) -{ - time_interpolator->offset = 0; - time_interpolator->last_counter = time_interpolator_get_counter(1); -} - -#define GET_TI_NSECS(count,i) (((((count) - i->last_counter) & (i)->mask) * (i)->nsec_per_cyc) >> (i)->shift) - -unsigned long time_interpolator_get_offset(void) -{ - /* If we do not have a time interpolator set up then just return zero */ - if (!time_interpolator) - return 0; - - return time_interpolator->offset + - GET_TI_NSECS(time_interpolator_get_counter(0), time_interpolator); -} - -#define INTERPOLATOR_ADJUST 65536 -#define INTERPOLATOR_MAX_SKIP 10*INTERPOLATOR_ADJUST - -void time_interpolator_update(long delta_nsec) -{ - u64 counter; - unsigned long offset; - - /* If there is no time interpolator set up then do nothing */ - if (!time_interpolator) - return; - - /* - * The interpolator compensates for late ticks by accumulating the late - * time in time_interpolator->offset. A tick earlier than expected will - * lead to a reset of the offset and a corresponding jump of the clock - * forward. Again this only works if the interpolator clock is running - * slightly slower than the regular clock and the tuning logic insures - * that. - */ - - counter = time_interpolator_get_counter(1); - offset = time_interpolator->offset + - GET_TI_NSECS(counter, time_interpolator); - - if (delta_nsec < 0 || (unsigned long) delta_nsec < offset) - time_interpolator->offset = offset - delta_nsec; - else { - time_interpolator->skips++; - time_interpolator->ns_skipped += delta_nsec - offset; - time_interpolator->offset = 0; - } - time_interpolator->last_counter = counter; - - /* Tuning logic for time interpolator invoked every minute or so. - * Decrease interpolator clock speed if no skips occurred and an offset is carried. - * Increase interpolator clock speed if we skip too much time. - */ - if (jiffies % INTERPOLATOR_ADJUST == 0) - { - if (time_interpolator->skips == 0 && time_interpolator->offset > tick_nsec) - time_interpolator->nsec_per_cyc--; - if (time_interpolator->ns_skipped > INTERPOLATOR_MAX_SKIP && time_interpolator->offset == 0) - time_interpolator->nsec_per_cyc++; - time_interpolator->skips = 0; - time_interpolator->ns_skipped = 0; - } -} - -static inline int -is_better_time_interpolator(struct time_interpolator *new) -{ - if (!time_interpolator) - return 1; - return new->frequency > 2*time_interpolator->frequency || - (unsigned long)new->drift < (unsigned long)time_interpolator->drift; -} - -void -register_time_interpolator(struct time_interpolator *ti) -{ - unsigned long flags; - - /* Sanity check */ - BUG_ON(ti->frequency == 0 || ti->mask == 0); - - ti->nsec_per_cyc = ((u64)NSEC_PER_SEC << ti->shift) / ti->frequency; - spin_lock(&time_interpolator_lock); - write_seqlock_irqsave(&xtime_lock, flags); - if (is_better_time_interpolator(ti)) { - time_interpolator = ti; - time_interpolator_reset(); - } - write_sequnlock_irqrestore(&xtime_lock, flags); - - ti->next = time_interpolator_list; - time_interpolator_list = ti; - spin_unlock(&time_interpolator_lock); -} - -void -unregister_time_interpolator(struct time_interpolator *ti) -{ - struct time_interpolator *curr, **prev; - unsigned long flags; - - spin_lock(&time_interpolator_lock); - prev = &time_interpolator_list; - for (curr = *prev; curr; curr = curr->next) { - if (curr == ti) { - *prev = curr->next; - break; - } - prev = &curr->next; - } - - write_seqlock_irqsave(&xtime_lock, flags); - if (ti == time_interpolator) { - /* we lost the best time-interpolator: */ - time_interpolator = NULL; - /* find the next-best interpolator */ - for (curr = time_interpolator_list; curr; curr = curr->next) - if (is_better_time_interpolator(curr)) - time_interpolator = curr; - time_interpolator_reset(); - } - write_sequnlock_irqrestore(&xtime_lock, flags); - spin_unlock(&time_interpolator_lock); -} -#endif /* CONFIG_TIME_INTERPOLATION */ - /** * msleep - sleep safely even with waitqueue interruptions * @msecs: Time in milliseconds to sleep for -- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/