summaryrefslogtreecommitdiff
path: root/kernel/time
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2020-12-14 18:21:14 -0800
committerLinus Torvalds <torvalds@linux-foundation.org>2020-12-14 18:21:14 -0800
commit533369b145d8d1bc44b8ed7f0dd0ecffb16384cc (patch)
treea5ca74a94dc256bd5f57d518600d9ea9c0aa659a /kernel/time
parent76d4acf22b4847f6c7b2f9042366fbdc3d20f578 (diff)
parent3cabca87b329cbcbdf295be0094adbd72c7b1f67 (diff)
Merge tag 'timers-core-2020-12-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull timers and timekeeping updates from Thomas Gleixner: "Core: - Robustness improvements for the NOHZ tick management - Fixes and consolidation of the NTP/RTC synchronization code - Small fixes and improvements in various places - A set of function documentation udpates and fixes Drivers: - Cleanups and improvements in various clocksoure/event drivers - Removal of the EZChip NPS clocksource driver as the platfrom support was removed from ARC - The usual set of new device tree binding and json conversions - The RTC driver which have been acked by the RTC maintainer: * fix a long standing bug in the MC146818 library code which can cause reading garbage during the RTC internal update. * changes related to the NTP/RTC consolidation work" * tag 'timers-core-2020-12-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (46 commits) ntp: Fix prototype in the !CONFIG_GENERIC_CMOS_UPDATE case tick/sched: Make jiffies update quick check more robust ntp: Consolidate the RTC update implementation ntp: Make the RTC sync offset less obscure ntp, rtc: Move rtc_set_ntp_time() to ntp code ntp: Make the RTC synchronization more reliable rtc: core: Make the sync offset default more realistic rtc: cmos: Make rtc_cmos sync offset correct rtc: mc146818: Reduce spinlock section in mc146818_set_time() rtc: mc146818: Prevent reading garbage clocksource/drivers/sh_cmt: Fix potential deadlock when calling runtime PM clocksource/drivers/arm_arch_timer: Correct fault programming of CNTKCTL_EL1.EVNTI clocksource/drivers/arm_arch_timer: Use stable count reader in erratum sne clocksource/drivers/dw_apb_timer_of: Add error handling if no clock available clocksource/drivers/riscv: Make RISCV_TIMER depends on RISCV_SBI clocksource/drivers/ingenic: Fix section mismatch clocksource/drivers/cadence_ttc: Fix memory leak in ttc_setup_clockevent() dt-bindings: timer: renesas: tmu: Convert to json-schema dt-bindings: timer: renesas: tmu: Document r8a774e1 bindings clocksource/drivers/orion: Add missing clk_disable_unprepare() on error path ...
Diffstat (limited to 'kernel/time')
-rw-r--r--kernel/time/hrtimer.c2
-rw-r--r--kernel/time/jiffies.c3
-rw-r--r--kernel/time/ntp.c229
-rw-r--r--kernel/time/ntp_internal.h7
-rw-r--r--kernel/time/tick-broadcast.c25
-rw-r--r--kernel/time/tick-common.c12
-rw-r--r--kernel/time/tick-internal.h1
-rw-r--r--kernel/time/tick-sched.c122
-rw-r--r--kernel/time/timeconv.c6
-rw-r--r--kernel/time/timekeeping.c85
-rw-r--r--kernel/time/timekeeping.h2
-rw-r--r--kernel/time/timer.c57
-rw-r--r--kernel/time/timer_list.c66
13 files changed, 366 insertions, 251 deletions
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 387b4bef7dd1..743c852e10f2 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -1284,7 +1284,7 @@ int hrtimer_cancel(struct hrtimer *timer)
EXPORT_SYMBOL_GPL(hrtimer_cancel);
/**
- * hrtimer_get_remaining - get remaining time for the timer
+ * __hrtimer_get_remaining - get remaining time for the timer
* @timer: the timer to read
* @adjust: adjust relative timers when CONFIG_TIME_LOW_RES=y
*/
diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c
index eddcf4970444..a5cffe2a1770 100644
--- a/kernel/time/jiffies.c
+++ b/kernel/time/jiffies.c
@@ -59,7 +59,8 @@ static struct clocksource clocksource_jiffies = {
};
__cacheline_aligned_in_smp DEFINE_RAW_SPINLOCK(jiffies_lock);
-__cacheline_aligned_in_smp seqcount_t jiffies_seq;
+__cacheline_aligned_in_smp seqcount_raw_spinlock_t jiffies_seq =
+ SEQCNT_RAW_SPINLOCK_ZERO(jiffies_seq, &jiffies_lock);
#if (BITS_PER_LONG < 64)
u64 get_jiffies_64(void)
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index 069ca78fb0bf..7404d3831527 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -494,65 +494,74 @@ out:
return leap;
}
+#if defined(CONFIG_GENERIC_CMOS_UPDATE) || defined(CONFIG_RTC_SYSTOHC)
static void sync_hw_clock(struct work_struct *work);
-static DECLARE_DELAYED_WORK(sync_work, sync_hw_clock);
-
-static void sched_sync_hw_clock(struct timespec64 now,
- unsigned long target_nsec, bool fail)
+static DECLARE_WORK(sync_work, sync_hw_clock);
+static struct hrtimer sync_hrtimer;
+#define SYNC_PERIOD_NS (11UL * 60 * NSEC_PER_SEC)
+static enum hrtimer_restart sync_timer_callback(struct hrtimer *timer)
{
- struct timespec64 next;
-
- ktime_get_real_ts64(&next);
- if (!fail)
- next.tv_sec = 659;
- else {
- /*
- * Try again as soon as possible. Delaying long periods
- * decreases the accuracy of the work queue timer. Due to this
- * the algorithm is very likely to require a short-sleep retry
- * after the above long sleep to synchronize ts_nsec.
- */
- next.tv_sec = 0;
- }
-
- /* Compute the needed delay that will get to tv_nsec == target_nsec */
- next.tv_nsec = target_nsec - next.tv_nsec;
- if (next.tv_nsec <= 0)
- next.tv_nsec += NSEC_PER_SEC;
- if (next.tv_nsec >= NSEC_PER_SEC) {
- next.tv_sec++;
- next.tv_nsec -= NSEC_PER_SEC;
- }
+ queue_work(system_power_efficient_wq, &sync_work);
- queue_delayed_work(system_power_efficient_wq, &sync_work,
- timespec64_to_jiffies(&next));
+ return HRTIMER_NORESTART;
}
-static void sync_rtc_clock(void)
+static void sched_sync_hw_clock(unsigned long offset_nsec, bool retry)
{
- unsigned long target_nsec;
- struct timespec64 adjust, now;
- int rc;
+ ktime_t exp = ktime_set(ktime_get_real_seconds(), 0);
- if (!IS_ENABLED(CONFIG_RTC_SYSTOHC))
- return;
+ if (retry)
+ exp = ktime_add_ns(exp, 2 * NSEC_PER_SEC - offset_nsec);
+ else
+ exp = ktime_add_ns(exp, SYNC_PERIOD_NS - offset_nsec);
- ktime_get_real_ts64(&now);
+ hrtimer_start(&sync_hrtimer, exp, HRTIMER_MODE_ABS);
+}
- adjust = now;
- if (persistent_clock_is_local)
- adjust.tv_sec -= (sys_tz.tz_minuteswest * 60);
+/*
+ * Check whether @now is correct versus the required time to update the RTC
+ * and calculate the value which needs to be written to the RTC so that the
+ * next seconds increment of the RTC after the write is aligned with the next
+ * seconds increment of clock REALTIME.
+ *
+ * tsched t1 write(t2.tv_sec - 1sec)) t2 RTC increments seconds
+ *
+ * t2.tv_nsec == 0
+ * tsched = t2 - set_offset_nsec
+ * newval = t2 - NSEC_PER_SEC
+ *
+ * ==> neval = tsched + set_offset_nsec - NSEC_PER_SEC
+ *
+ * As the execution of this code is not guaranteed to happen exactly at
+ * tsched this allows it to happen within a fuzzy region:
+ *
+ * abs(now - tsched) < FUZZ
+ *
+ * If @now is not inside the allowed window the function returns false.
+ */
+static inline bool rtc_tv_nsec_ok(unsigned long set_offset_nsec,
+ struct timespec64 *to_set,
+ const struct timespec64 *now)
+{
+ /* Allowed error in tv_nsec, arbitarily set to 5 jiffies in ns. */
+ const unsigned long TIME_SET_NSEC_FUZZ = TICK_NSEC * 5;
+ struct timespec64 delay = {.tv_sec = -1,
+ .tv_nsec = set_offset_nsec};
- /*
- * The current RTC in use will provide the target_nsec it wants to be
- * called at, and does rtc_tv_nsec_ok internally.
- */
- rc = rtc_set_ntp_time(adjust, &target_nsec);
- if (rc == -ENODEV)
- return;
+ *to_set = timespec64_add(*now, delay);
+
+ if (to_set->tv_nsec < TIME_SET_NSEC_FUZZ) {
+ to_set->tv_nsec = 0;
+ return true;
+ }
- sched_sync_hw_clock(now, target_nsec, rc);
+ if (to_set->tv_nsec > NSEC_PER_SEC - TIME_SET_NSEC_FUZZ) {
+ to_set->tv_sec++;
+ to_set->tv_nsec = 0;
+ return true;
+ }
+ return false;
}
#ifdef CONFIG_GENERIC_CMOS_UPDATE
@@ -560,48 +569,47 @@ int __weak update_persistent_clock64(struct timespec64 now64)
{
return -ENODEV;
}
+#else
+static inline int update_persistent_clock64(struct timespec64 now64)
+{
+ return -ENODEV;
+}
#endif
-static bool sync_cmos_clock(void)
+#ifdef CONFIG_RTC_SYSTOHC
+/* Save NTP synchronized time to the RTC */
+static int update_rtc(struct timespec64 *to_set, unsigned long *offset_nsec)
{
- static bool no_cmos;
- struct timespec64 now;
- struct timespec64 adjust;
- int rc = -EPROTO;
- long target_nsec = NSEC_PER_SEC / 2;
+ struct rtc_device *rtc;
+ struct rtc_time tm;
+ int err = -ENODEV;
- if (!IS_ENABLED(CONFIG_GENERIC_CMOS_UPDATE))
- return false;
+ rtc = rtc_class_open(CONFIG_RTC_SYSTOHC_DEVICE);
+ if (!rtc)
+ return -ENODEV;
- if (no_cmos)
- return false;
+ if (!rtc->ops || !rtc->ops->set_time)
+ goto out_close;
- /*
- * Historically update_persistent_clock64() has followed x86
- * semantics, which match the MC146818A/etc RTC. This RTC will store
- * 'adjust' and then in .5s it will advance once second.
- *
- * Architectures are strongly encouraged to use rtclib and not
- * implement this legacy API.
- */
- ktime_get_real_ts64(&now);
- if (rtc_tv_nsec_ok(-1 * target_nsec, &adjust, &now)) {
- if (persistent_clock_is_local)
- adjust.tv_sec -= (sys_tz.tz_minuteswest * 60);
- rc = update_persistent_clock64(adjust);
- /*
- * The machine does not support update_persistent_clock64 even
- * though it defines CONFIG_GENERIC_CMOS_UPDATE.
- */
- if (rc == -ENODEV) {
- no_cmos = true;
- return false;
- }
+ /* First call might not have the correct offset */
+ if (*offset_nsec == rtc->set_offset_nsec) {
+ rtc_time64_to_tm(to_set->tv_sec, &tm);
+ err = rtc_set_time(rtc, &tm);
+ } else {
+ /* Store the update offset and let the caller try again */
+ *offset_nsec = rtc->set_offset_nsec;
+ err = -EAGAIN;
}
-
- sched_sync_hw_clock(now, target_nsec, rc);
- return true;
+out_close:
+ rtc_class_close(rtc);
+ return err;
+}
+#else
+static inline int update_rtc(struct timespec64 *to_set, unsigned long *offset_nsec)
+{
+ return -ENODEV;
}
+#endif
/*
* If we have an externally synchronized Linux clock, then update RTC clock
@@ -613,24 +621,64 @@ static bool sync_cmos_clock(void)
*/
static void sync_hw_clock(struct work_struct *work)
{
- if (!ntp_synced())
- return;
+ /*
+ * The default synchronization offset is 500ms for the deprecated
+ * update_persistent_clock64() under the assumption that it uses
+ * the infamous CMOS clock (MC146818).
+ */
+ static unsigned long offset_nsec = NSEC_PER_SEC / 2;
+ struct timespec64 now, to_set;
+ int res = -EAGAIN;
- if (sync_cmos_clock())
+ /*
+ * Don't update if STA_UNSYNC is set and if ntp_notify_cmos_timer()
+ * managed to schedule the work between the timer firing and the
+ * work being able to rearm the timer. Wait for the timer to expire.
+ */
+ if (!ntp_synced() || hrtimer_is_queued(&sync_hrtimer))
return;
- sync_rtc_clock();
+ ktime_get_real_ts64(&now);
+ /* If @now is not in the allowed window, try again */
+ if (!rtc_tv_nsec_ok(offset_nsec, &to_set, &now))
+ goto rearm;
+
+ /* Take timezone adjusted RTCs into account */
+ if (persistent_clock_is_local)
+ to_set.tv_sec -= (sys_tz.tz_minuteswest * 60);
+
+ /* Try the legacy RTC first. */
+ res = update_persistent_clock64(to_set);
+ if (res != -ENODEV)
+ goto rearm;
+
+ /* Try the RTC class */
+ res = update_rtc(&to_set, &offset_nsec);
+ if (res == -ENODEV)
+ return;
+rearm:
+ sched_sync_hw_clock(offset_nsec, res != 0);
}
void ntp_notify_cmos_timer(void)
{
- if (!ntp_synced())
- return;
+ /*
+ * When the work is currently executed but has not yet the timer
+ * rearmed this queues the work immediately again. No big issue,
+ * just a pointless work scheduled.
+ */
+ if (ntp_synced() && !hrtimer_is_queued(&sync_hrtimer))
+ queue_work(system_power_efficient_wq, &sync_work);
+}
- if (IS_ENABLED(CONFIG_GENERIC_CMOS_UPDATE) ||
- IS_ENABLED(CONFIG_RTC_SYSTOHC))
- queue_delayed_work(system_power_efficient_wq, &sync_work, 0);
+static void __init ntp_init_cmos_sync(void)
+{
+ hrtimer_init(&sync_hrtimer, CLOCK_REALTIME, HRTIMER_MODE_ABS);
+ sync_hrtimer.function = sync_timer_callback;
}
+#else /* CONFIG_GENERIC_CMOS_UPDATE) || defined(CONFIG_RTC_SYSTOHC) */
+static inline void __init ntp_init_cmos_sync(void) { }
+#endif /* !CONFIG_GENERIC_CMOS_UPDATE) || defined(CONFIG_RTC_SYSTOHC) */
/*
* Propagate a new txc->status value into the NTP state:
@@ -1044,4 +1092,5 @@ __setup("ntp_tick_adj=", ntp_tick_adj_setup);
void __init ntp_init(void)
{
ntp_clear();
+ ntp_init_cmos_sync();
}
diff --git a/kernel/time/ntp_internal.h b/kernel/time/ntp_internal.h
index 908ecaa65fc3..23d1b74c3065 100644
--- a/kernel/time/ntp_internal.h
+++ b/kernel/time/ntp_internal.h
@@ -12,4 +12,11 @@ extern int __do_adjtimex(struct __kernel_timex *txc,
const struct timespec64 *ts,
s32 *time_tai, struct audit_ntp_data *ad);
extern void __hardpps(const struct timespec64 *phase_ts, const struct timespec64 *raw_ts);
+
+#if defined(CONFIG_GENERIC_CMOS_UPDATE) || defined(CONFIG_RTC_SYSTOHC)
+extern void ntp_notify_cmos_timer(void);
+#else
+static inline void ntp_notify_cmos_timer(void) { }
+#endif
+
#endif /* _LINUX_NTP_INTERNAL_H */
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 36d7464c8962..5a23829372c7 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -331,7 +331,7 @@ static void tick_handle_periodic_broadcast(struct clock_event_device *dev)
bc_local = tick_do_periodic_broadcast();
if (clockevent_state_oneshot(dev)) {
- ktime_t next = ktime_add(dev->next_event, tick_period);
+ ktime_t next = ktime_add_ns(dev->next_event, TICK_NSEC);
clockevents_program_event(dev, next, true);
}
@@ -877,6 +877,22 @@ static void tick_broadcast_init_next_event(struct cpumask *mask,
}
}
+static inline ktime_t tick_get_next_period(void)
+{
+ ktime_t next;
+
+ /*
+ * Protect against concurrent updates (store /load tearing on
+ * 32bit). It does not matter if the time is already in the
+ * past. The broadcast device which is about to be programmed will
+ * fire in any case.
+ */
+ raw_spin_lock(&jiffies_lock);
+ next = tick_next_period;
+ raw_spin_unlock(&jiffies_lock);
+ return next;
+}
+
/**
* tick_broadcast_setup_oneshot - setup the broadcast device
*/
@@ -905,10 +921,11 @@ static void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
tick_broadcast_oneshot_mask, tmpmask);
if (was_periodic && !cpumask_empty(tmpmask)) {
+ ktime_t nextevt = tick_get_next_period();
+
clockevents_switch_state(bc, CLOCK_EVT_STATE_ONESHOT);
- tick_broadcast_init_next_event(tmpmask,
- tick_next_period);
- tick_broadcast_set_event(bc, cpu, tick_next_period);
+ tick_broadcast_init_next_event(tmpmask, nextevt);
+ tick_broadcast_set_event(bc, cpu, nextevt);
} else
bc->next_event = KTIME_MAX;
} else {
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index 6c9c342dd0e5..a03764df5366 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -27,10 +27,11 @@
*/
DEFINE_PER_CPU(struct tick_device, tick_cpu_device);
/*
- * Tick next event: keeps track of the tick time
+ * Tick next event: keeps track of the tick time. It's updated by the
+ * CPU which handles the tick and protected by jiffies_lock. There is
+ * no requirement to write hold the jiffies seqcount for it.
*/
ktime_t tick_next_period;
-ktime_t tick_period;
/*
* tick_do_timer_cpu is a timer core internal variable which holds the CPU NR
@@ -88,7 +89,7 @@ static void tick_periodic(int cpu)
write_seqcount_begin(&jiffies_seq);
/* Keep track of the next tick event */
- tick_next_period = ktime_add(tick_next_period, tick_period);
+ tick_next_period = ktime_add_ns(tick_next_period, TICK_NSEC);
do_timer(1);
write_seqcount_end(&jiffies_seq);
@@ -127,7 +128,7 @@ void tick_handle_periodic(struct clock_event_device *dev)
* Setup the next period for devices, which do not have
* periodic mode:
*/
- next = ktime_add(next, tick_period);
+ next = ktime_add_ns(next, TICK_NSEC);
if (!clockevents_program_event(dev, next, false))
return;
@@ -173,7 +174,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast)
for (;;) {
if (!clockevents_program_event(dev, next, false))
return;
- next = ktime_add(next, tick_period);
+ next = ktime_add_ns(next, TICK_NSEC);
}
}
}
@@ -220,7 +221,6 @@ static void tick_setup_device(struct tick_device *td,
tick_do_timer_cpu = cpu;
tick_next_period = ktime_get();
- tick_period = NSEC_PER_SEC / HZ;
#ifdef CONFIG_NO_HZ_FULL
/*
* The boot CPU may be nohz_full, in which case set
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index 7b2496136729..7a981c9e87a4 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -15,7 +15,6 @@
DECLARE_PER_CPU(struct tick_device, tick_cpu_device);
extern ktime_t tick_next_period;
-extern ktime_t tick_period;
extern int tick_do_timer_cpu __read_mostly;
extern void tick_setup_periodic(struct clock_event_device *dev, int broadcast);
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 81632cd5e3b7..a9e68936822d 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -20,6 +20,7 @@
#include <linux/sched/clock.h>
#include <linux/sched/stat.h>
#include <linux/sched/nohz.h>
+#include <linux/sched/loadavg.h>
#include <linux/module.h>
#include <linux/irq_work.h>
#include <linux/posix-timers.h>
@@ -44,7 +45,9 @@ struct tick_sched *tick_get_tick_sched(int cpu)
#if defined(CONFIG_NO_HZ_COMMON) || defined(CONFIG_HIGH_RES_TIMERS)
/*
- * The time, when the last jiffy update happened. Protected by jiffies_lock.
+ * The time, when the last jiffy update happened. Write access must hold
+ * jiffies_lock and jiffies_seq. tick_nohz_next_event() needs to get a
+ * consistent view of jiffies and last_jiffies_update.
*/
static ktime_t last_jiffies_update;
@@ -53,50 +56,97 @@ static ktime_t last_jiffies_update;
*/
static void tick_do_update_jiffies64(ktime_t now)
{
- unsigned long ticks = 0;
- ktime_t delta;
+ unsigned long ticks = 1;
+ ktime_t delta, nextp;
/*
- * Do a quick check without holding jiffies_lock:
- * The READ_ONCE() pairs with two updates done later in this function.
+ * 64bit can do a quick check without holding jiffies lock and
+ * without looking at the sequence count. The smp_load_acquire()
+ * pairs with the update done later in this function.
+ *
+ * 32bit cannot do that because the store of tick_next_period
+ * consists of two 32bit stores and the first store could move it
+ * to a random point in the future.
*/
- delta = ktime_sub(now, READ_ONCE(last_jiffies_update));
- if (delta < tick_period)
- return;
+ if (IS_ENABLED(CONFIG_64BIT)) {
+ if (ktime_before(now, smp_load_acquire(&tick_next_period)))
+ return;
+ } else {
+ unsigned int seq;
- /* Reevaluate with jiffies_lock held */
+ /*
+ * Avoid contention on jiffies_lock and protect the quick
+ * check with the sequence count.
+ */
+ do {
+ seq = read_seqcount_begin(&jiffies_seq);
+ nextp = tick_next_period;
+ } while (read_seqcount_retry(&jiffies_seq, seq));
+
+ if (ktime_before(now, nextp))
+ return;
+ }
+
+ /* Quick check failed, i.e. update is required. */
raw_spin_lock(&jiffies_lock);
+ /*
+ * Reevaluate with the lock held. Another CPU might have done the
+ * update already.
+ */
+ if (ktime_before(now, tick_next_period)) {
+ raw_spin_unlock(&jiffies_lock);
+ return;
+ }
+
write_seqcount_begin(&jiffies_seq);
- delta = ktime_sub(now, last_jiffies_update);
- if (delta >= tick_period) {
+ delta = ktime_sub(now, tick_next_period);
+ if (unlikely(delta >= TICK_NSEC)) {
+ /* Slow path for long idle sleep times */
+ s64 incr = TICK_NSEC;
- delta = ktime_sub(delta, tick_period);
- /* Pairs with the lockless read in this function. */
- WRITE_ONCE(last_jiffies_update,
- ktime_add(last_jiffies_update, tick_period));
+ ticks += ktime_divns(delta, incr);
- /* Slow path for long timeouts */
- if (unlikely(delta >= tick_period)) {
- s64 incr = ktime_to_ns(tick_period);
+ last_jiffies_update = ktime_add_ns(last_jiffies_update,
+ incr * ticks);
+ } else {
+ last_jiffies_update = ktime_add_ns(last_jiffies_update,
+ TICK_NSEC);
+ }
- ticks = ktime_divns(delta, incr);
+ /* Advance jiffies to complete the jiffies_seq protected job */
+ jiffies_64 += ticks;
- /* Pairs with the lockless read in this function. */
- WRITE_ONCE(last_jiffies_update,
- ktime_add_ns(last_jiffies_update,
- incr * ticks));
- }
- do_timer(++ticks);
+ /*
+ * Keep the tick_next_period variable up to date.
+ */
+ nextp = ktime_add_ns(last_jiffies_update, TICK_NSEC);
- /* Keep the tick_next_period variable up to date */
- tick_next_period = ktime_add(last_jiffies_update, tick_period);
+ if (IS_ENABLED(CONFIG_64BIT)) {
+ /*
+ * Pairs with smp_load_acquire() in the lockless quick
+ * check above and ensures that the update to jiffies_64 is
+ * not reordered vs. the store to tick_next_period, neither
+ * by the compiler nor by the CPU.
+ */
+ smp_store_release(&tick_next_period, nextp);
} else {
- write_seqcount_end(&jiffies_seq);
- raw_spin_unlock(&jiffies_lock);
- return;
+ /*
+ * A plain store is good enough on 32bit as the quick check
+ * above is protected by the sequence count.
+ */
+ tick_next_period = nextp;
}
+
+ /*
+ * Release the sequence count. calc_global_load() below is not
+ * protected by it, but jiffies_lock needs to be held to prevent
+ * concurrent invocations.
+ */
write_seqcount_end(&jiffies_seq);
+
+ calc_global_load();
+
raw_spin_unlock(&jiffies_lock);
update_wall_time();
}
@@ -661,7 +711,7 @@ static void tick_nohz_restart(struct tick_sched *ts, ktime_t now)
hrtimer_set_expires(&ts->sched_timer, ts->last_tick);
/* Forward the time to expire in the future */
- hrtimer_forward(&ts->sched_timer, now, tick_period);
+ hrtimer_forward(&ts->sched_timer, now, TICK_NSEC);
if (ts->nohz_mode == NOHZ_MODE_HIGHRES) {
hrtimer_start_expires(&ts->sched_timer,
@@ -1230,7 +1280,7 @@ static void tick_nohz_handler(struct clock_event_device *dev)
if (unlikely(ts->tick_stopped))
return;
- hrtimer_forward(&ts->sched_timer, now, tick_period);
+ hrtimer_forward(&ts->sched_timer, now, TICK_NSEC);
tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1);
}
@@ -1267,7 +1317,7 @@ static void tick_nohz_switch_to_nohz(void)
next = tick_init_jiffy_update();
hrtimer_set_expires(&ts->sched_timer, next);
- hrtimer_forward_now(&ts->sched_timer, tick_period);
+ hrtimer_forward_now(&ts->sched_timer, TICK_NSEC);
tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1);
tick_nohz_activate(ts, NOHZ_MODE_LOWRES);
}
@@ -1333,7 +1383,7 @@ static enum hrtimer_restart tick_sched_timer(struct hrtimer *timer)
if (unlikely(ts->tick_stopped))
return HRTIMER_NORESTART;
- hrtimer_forward(timer, now, tick_period);
+ hrtimer_forward(timer, now, TICK_NSEC);
return HRTIMER_RESTART;
}
@@ -1367,13 +1417,13 @@ void tick_setup_sched_timer(void)
/* Offset the tick to avert jiffies_lock contention. */
if (sched_skew_tick) {
- u64 offset = ktime_to_ns(tick_period) >> 1;
+ u64 offset = TICK_NSEC >> 1;
do_div(offset, num_possible_cpus());
offset *= smp_processor_id();
hrtimer_add_expires_ns(&ts->sched_timer, offset);
}
- hrtimer_forward(&ts->sched_timer, now, tick_period);
+ hrtimer_forward(&ts->sched_timer, now, TICK_NSEC);
hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED_HARD);
tick_nohz_activate(ts, NOHZ_MODE_HIGHRES);
}
diff --git a/kernel/time/timeconv.c b/kernel/time/timeconv.c
index 589e0a552129..62e3b46717a6 100644
--- a/kernel/time/timeconv.c
+++ b/kernel/time/timeconv.c
@@ -70,10 +70,10 @@ static const unsigned short __mon_yday[2][13] = {
/**
* time64_to_tm - converts the calendar time to local broken-down time
*
- * @totalsecs the number of seconds elapsed since 00:00:00 on January 1, 1970,
+ * @totalsecs: the number of seconds elapsed since 00:00:00 on January 1, 1970,
* Coordinated Universal Time (UTC).
- * @offset offset seconds adding to totalsecs.
- * @result pointer to struct tm variable to receive broken-down time
+ * @offset: offset seconds adding to totalsecs.
+ * @result: pointer to struct tm variable to receive broken-down time
*/
void time64_to_tm(time64_t totalsecs, int offset, struct tm *result)
{
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 6858a31364b6..74503c0151e5 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -407,6 +407,7 @@ static inline u64 timekeeping_cycles_to_ns(const struct tk_read_base *tkr, u64 c
/**
* update_fast_timekeeper - Update the fast and NMI safe monotonic timekeeper.
* @tkr: Timekeeping readout base from which we take the update
+ * @tkf: Pointer to NMI safe timekeeper
*
* We want to use this from any context including NMI and tracing /
* instrumenting the timekeeping code itself.
@@ -436,6 +437,27 @@ static void update_fast_timekeeper(const struct tk_read_base *tkr,
memcpy(base + 1, base, sizeof(*base));
}
+static __always_inline u64 __ktime_get_fast_ns(struct tk_fast *tkf)
+{
+ struct tk_read_base *tkr;
+ unsigned int seq;
+ u64 now;
+
+ do {
+ seq = raw_read_seqcount_latch(&tkf->seq);
+ tkr = tkf->base + (seq & 0x01);
+ now = ktime_to_ns(tkr->base);
+
+ now += timekeeping_delta_to_ns(tkr,
+ clocksource_delta(
+ tk_clock_read(tkr),
+ tkr->cycle_last,
+ tkr->mask));
+ } while (read_seqcount_latch_retry(&tkf->seq, seq));
+
+ return now;
+}
+
/**
* ktime_get_mono_fast_ns - Fast NMI safe access to clock monotonic
*
@@ -462,39 +484,24 @@ static void update_fast_timekeeper(const struct tk_read_base *tkr,
*
* So reader 6 will observe time going backwards versus reader 5.
*
- * While other CPUs are likely to be able observe that, the only way
+ * While other CPUs are likely to be able to observe that, the only way
* for a CPU local observation is when an NMI hits in the middle of
* the update. Timestamps taken from that NMI context might be ahead
* of the following timestamps. Callers need to be aware of that and
* deal with it.
*/
-static __always_inline u64 __ktime_get_fast_ns(struct tk_fast *tkf)
-{
- struct tk_read_base *tkr;
- unsigned int seq;
- u64 now;
-
- do {
- seq = raw_read_seqcount_latch(&tkf->seq);
- tkr = tkf->base + (seq & 0x01);
- now = ktime_to_ns(tkr->base);
-
- now += timekeeping_delta_to_ns(tkr,
- clocksource_delta(
- tk_clock_read(tkr),
- tkr->cycle_last,
- tkr->mask));
- } while (read_seqcount_latch_retry(&tkf->seq, seq));
-
- return now;
-}
-
u64 ktime_get_mono_fast_ns(void)
{
return __ktime_get_fast_ns(&tk_fast_mono);
}
EXPORT_SYMBOL_GPL(ktime_get_mono_fast_ns);
+/**
+ * ktime_get_raw_fast_ns - Fast NMI safe access to clock monotonic raw
+ *
+ * Contrary to ktime_get_mono_fast_ns() this is always correct because the
+ * conversion factor is not affected by NTP/PTP correction.
+ */
u64 ktime_get_raw_fast_ns(void)
{
return __ktime_get_fast_ns(&tk_fast_raw);
@@ -521,6 +528,9 @@ EXPORT_SYMBOL_GPL(ktime_get_raw_fast_ns);
* (2) On 32-bit systems, the 64-bit boot offset (tk->offs_boot) may be
* partially updated. Since the tk->offs_boot update is a rare event, this
* should be a rare occurrence which postprocessing should be able to handle.
+ *
+ * The caveats vs. timestamp ordering as documented for ktime_get_fast_ns()
+ * apply as well.
*/
u64 notrace ktime_get_boot_fast_ns(void)
{
@@ -530,9 +540,6 @@ u64 notrace ktime_get_boot_fast_ns(void)
}
EXPORT_SYMBOL_GPL(ktime_get_boot_fast_ns);
-/*
- * See comment for __ktime_get_fast_ns() vs. timestamp ordering
- */
static __always_inline u64 __ktime_get_real_fast(struct tk_fast *tkf, u64 *mono)
{
struct tk_read_base *tkr;
@@ -557,6 +564,8 @@ static __always_inline u64 __ktime_get_real_fast(struct tk_fast *tkf, u64 *mono)
/**
* ktime_get_real_fast_ns: - NMI safe and fast access to clock realtime.
+ *
+ * See ktime_get_fast_ns() for documentation of the time stamp ordering.
*/
u64 ktime_get_real_fast_ns(void)
{
@@ -654,6 +663,7 @@ static void update_pvclock_gtod(struct timekeeper *tk, bool was_set)
/**
* pvclock_gtod_register_notifier - register a pvclock timedata update listener
+ * @nb: Pointer to the notifier block to register
*/
int pvclock_gtod_register_notifier(struct notifier_block *nb)
{
@@ -673,6 +683,7 @@ EXPORT_SYMBOL_GPL(pvclock_gtod_register_notifier);
/**
* pvclock_gtod_unregister_notifier - unregister a pvclock
* timedata update listener
+ * @nb: Pointer to the notifier block to unregister
*/
int pvclock_gtod_unregister_notifier(struct notifier_block *nb)
{
@@ -763,6 +774,7 @@ static void timekeeping_update(struct timekeeper *tk, unsigned int action)
/**
* timekeeping_forward_now - update clock to the current time
+ * @tk: Pointer to the timekeeper to update
*
* Forward the current clock to update its state since the last call to
* update_wall_time(). This is useful before significant clock changes,
@@ -1339,7 +1351,7 @@ EXPORT_SYMBOL(do_settimeofday64);
/**
* timekeeping_inject_offset - Adds or subtracts from the current time.
- * @tv: pointer to the timespec variable containing the offset
+ * @ts: Pointer to the timespec variable containing the offset
*
* Adds or subtracts an offset value from the current time.
*/
@@ -1415,9 +1427,8 @@ void timekeeping_warp_clock(void)
}
}
-/**
+/*
* __timekeeping_set_tai_offset - Sets the TAI offset from UTC and monotonic
- *
*/
static void __timekeeping_set_tai_offset(struct timekeeper *tk, s32 tai_offset)
{
@@ -1425,7 +1436,7 @@ static void __timekeeping_set_tai_offset(struct timekeeper *tk, s32 tai_offset)
tk->offs_tai = ktime_add(tk->offs_real, ktime_set(tai_offset, 0));
}
-/**
+/*
* change_clocksource - Swaps clocksources if a new one is available
*
* Accumulates current time interval and initializes new clocksource
@@ -1548,6 +1559,7 @@ u64 timekeeping_max_deferment(void)
/**
* read_persistent_clock64 - Return time from the persistent clock.
+ * @ts: Pointer to the storage for the readout value
*
* Weak dummy function for arches that do not yet support it.
* Reads the time from the battery backed persistent clock.
@@ -1566,8 +1578,9 @@ void __weak read_persistent_clock64(struct timespec64 *ts)
* from the boot.
*
* Weak dummy function for arches that do not yet support it.
- * wall_time - current time as returned by persistent clock
- * boot_offset - offset that is defined as wall_time - boot_time
+ * @wall_time: - current time as returned by persistent clock
+ * @boot_offset: - offset that is defined as wall_time - boot_time
+ *
* The default function calculates offset based on the current value of
* local_clock(). This way architectures that support sched_clock() but don't
* support dedicated boot time clock will provide the best estimate of the
@@ -1652,7 +1665,8 @@ static struct timespec64 timekeeping_suspend_time;
/**
* __timekeeping_inject_sleeptime - Internal function to add sleep interval
- * @delta: pointer to a timespec delta value
+ * @tk: Pointer to the timekeeper to be updated
+ * @delta: Pointer to the delta value in timespec64 format
*
* Takes a timespec offset measuring a suspend interval and properly
* adds the sleep offset to the timekeeping variables.
@@ -2023,13 +2037,12 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset)
}
}
-/**
+/*
* accumulate_nsecs_to_secs - Accumulates nsecs into secs
*
* Helper function that accumulates the nsecs greater than 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 unsigned int accumulate_nsecs_to_secs(struct timekeeper *tk)
{
@@ -2071,7 +2084,7 @@ static inline unsigned int accumulate_nsecs_to_secs(struct timekeeper *tk)
return clock_set;
}
-/**
+/*
* logarithmic_accumulation - shifted accumulation of cycles
*
* This functions accumulates a shifted interval of cycles into
@@ -2314,7 +2327,7 @@ ktime_t ktime_get_update_offsets_now(unsigned int *cwsseq, ktime_t *offs_real,
return base;
}
-/**
+/*
* timekeeping_validate_timex - Ensures the timex is ok for use in do_adjtimex
*/
static int timekeeping_validate_timex(const struct __kernel_timex *txc)
diff --git a/kernel/time/timekeeping.h b/kernel/time/timekeeping.h
index 099737f6f10c..6c2cbd9ef999 100644
--- a/kernel/time/timekeeping.h
+++ b/kernel/time/timekeeping.h
@@ -26,7 +26,7 @@ extern void do_timer(unsigned long ticks);
extern void update_wall_time(void);
extern raw_spinlock_t jiffies_lock;
-extern seqcount_t jiffies_seq;
+extern seqcount_raw_spinlock_t jiffies_seq;
#define CS_NAME_LEN 32
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index c3ad64fb9d8b..8dbc008f8942 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1283,7 +1283,7 @@ static void del_timer_wait_running(struct timer_list *timer)
u32 tf;
tf = READ_ONCE(timer->flags);
- if (!(tf & TIMER_MIGRATING)) {
+ if (!(tf & (TIMER_MIGRATING | TIMER_IRQSAFE))) {
struct timer_base *base = get_timer_base(tf);
/*
@@ -1367,6 +1367,13 @@ int del_timer_sync(struct timer_list *timer)
*/
WARN_ON(in_irq() && !(timer->flags & TIMER_IRQSAFE));
+ /*
+ * Must be able to sleep on PREEMPT_RT because of the slowpath in
+ * del_timer_wait_running().
+ */
+ if (IS_ENABLED(CONFIG_PREEMPT_RT) && !(timer->flags & TIMER_IRQSAFE))
+ lockdep_assert_preemption_enabled();
+
do {
ret = try_to_del_timer_sync(timer);
@@ -1693,29 +1700,6 @@ void timer_clear_idle(void)
}
#endif
-/*
- * 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.
- */
-void update_process_times(int user_tick)
-{
- struct task_struct *p = current;
-
- PRANDOM_ADD_NOISE(jiffies, user_tick, p, 0);
-
- /* Note: this timer irq context must be accounted for as well. */
- account_process_tick(p, user_tick);
- run_local_timers();
- rcu_sched_clock_irq(user_tick);
-#ifdef CONFIG_IRQ_WORK
- if (in_irq())
- irq_work_tick();
-#endif
- scheduler_tick();
- if (IS_ENABLED(CONFIG_POSIX_TIMERS))
- run_posix_cpu_timers();
-}
-
/**
* __run_timers - run all expired timers (if any) on this CPU.
* @base: the timer vector to be processed.
@@ -1765,7 +1749,7 @@ static __latent_entropy void run_timer_softirq(struct softirq_action *h)
/*
* Called by the local, per-CPU timer interrupt on SMP.
*/
-void run_local_timers(void)
+static void run_local_timers(void)
{
struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
@@ -1783,6 +1767,29 @@ void run_local_timers(void)
}
/*
+ * 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.
+ */
+void update_process_times(int user_tick)
+{
+ struct task_struct *p = current;
+
+ PRANDOM_ADD_NOISE(jiffies, user_tick, p, 0);
+
+ /* Note: this timer irq context must be accounted for as well. */
+ account_process_tick(p, user_tick);
+ run_local_timers();
+ rcu_sched_clock_irq(user_tick);
+#ifdef CONFIG_IRQ_WORK
+ if (in_irq())
+ irq_work_tick();
+#endif
+ scheduler_tick();
+ if (IS_ENABLED(CONFIG_POSIX_TIMERS))
+ run_posix_cpu_timers();
+}
+
+/*
* Since schedule_timeout()'s timer is defined on the stack, it must store
* the target task on the stack as well.
*/
diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c
index acb326f5f50a..6939140ab7c5 100644
--- a/kernel/time/timer_list.c
+++ b/kernel/time/timer_list.c
@@ -42,24 +42,11 @@ static void SEQ_printf(struct seq_file *m, const char *fmt, ...)
va_end(args);
}
-static void print_name_offset(struct seq_file *m, void *sym)
-{
- char symname[KSYM_NAME_LEN];
-
- if (lookup_symbol_name((unsigned long)sym, symname) < 0)
- SEQ_printf(m, "<%pK>", sym);
- else
- SEQ_printf(m, "%s", symname);
-}
-
static void
print_timer(struct seq_file *m, struct hrtimer *taddr, struct hrtimer *timer,
int idx, u64 now)
{
- SEQ_printf(m, " #%d: ", idx);
- print_name_offset(m, taddr);
- SEQ_printf(m, ", ");
- print_name_offset(m, timer->function);
+ SEQ_printf(m, " #%d: <%pK>, %ps", idx, taddr, timer->function);
SEQ_printf(m, ", S:%02x", timer->state);
SEQ_printf(m, "\n");
SEQ_printf(m, " # expires at %Lu-%Lu nsecs [in %Ld to %Ld nsecs]\n",
@@ -116,9 +103,7 @@ print_base(struct seq_file *m, struct hrtimer_clock_base *base, u64 now)
SEQ_printf(m, " .resolution: %u nsecs\n", hrtimer_resolution);
- SEQ_printf(m, " .get_time: ");
- print_name_offset(m, base->get_time);
- SEQ_printf(m, "\n");
+ SEQ_printf(m, " .get_time: %ps\n", base->get_time);
#ifdef CONFIG_HIGH_RES_TIMERS
SEQ_printf(m, " .offset: %Lu nsecs\n",
(unsigned long long) ktime_to_ns(base->offset));
@@ -218,42 +203,29 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu)
SEQ_printf(m, " next_event: %Ld nsecs\n",
(unsigned long long) ktime_to_ns(dev->next_event));
- SEQ_printf(m, " set_next_event: ");
- print_name_offset(m, dev->set_next_event);
- SEQ_printf(m, "\n");
+ SEQ_printf(m, " set_next_event: %ps\n", dev->set_next_event);
- if (dev->set_state_shutdown) {
- SEQ_printf(m, " shutdown: ");
- print_name_offset(m, dev->set_state_shutdown);
- SEQ_printf(m, "\n");
- }
+ if (dev->set_state_shutdown)
+ SEQ_printf(m, " shutdown: %ps\n",
+ dev->set_state_shutdown);
- if (dev->set_state_periodic) {
- SEQ_printf(m, " periodic: ");
- print_name_offset(m, dev->set_state_periodic);
- SEQ_printf(m, "\n");
- }
+ if (dev->set_state_periodic)
+ SEQ_printf(m, " periodic: %ps\n",
+ dev->set_state_periodic);
- if (dev->set_state_oneshot) {
- SEQ_printf(m, " oneshot: ");
- print_name_offset(m, dev->set_state_oneshot);
- SEQ_printf(m, "\n");
- }
+ if (dev->set_state_oneshot)
+ SEQ_printf(m, " oneshot: %ps\n",
+ dev->set_state_oneshot);
- if (dev->set_state_oneshot_stopped) {
- SEQ_printf(m, " oneshot stopped: ");
- print_name_offset(m, dev->set_state_oneshot_stopped);
- SEQ_printf(m, "\n");
- }
+ if (dev->set_state_oneshot_stopped)
+ SEQ_printf(m, " oneshot stopped: %ps\n",
+ dev->set_state_oneshot_stopped);
- if (dev->tick_resume) {
- SEQ_printf(m, " resume: ");
- print_name_offset(m, dev->tick_resume);
- SEQ_printf(m, "\n");
- }
+ if (dev->tick_resume)
+ SEQ_printf(m, " resume: %ps\n",
+ dev->tick_resume);
- SEQ_printf(m, " event_handler: ");
- print_name_offset(m, dev->event_handler);
+ SEQ_printf(m, " event_handler: %ps\n", dev->event_handler);
SEQ_printf(m, "\n");
SEQ_printf(m, " retries: %lu\n", dev->retries);
SEQ_printf(m, "\n");