Merge tag 'timers-core-2024-11-18' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull timer updates from Thomas Gleixner:
 "A rather large update for timekeeping and timers:

   - The final step to get rid of auto-rearming posix-timers

     posix-timers are currently auto-rearmed by the kernel when the
     signal of the timer is ignored so that the timer signal can be
     delivered once the corresponding signal is unignored.

     This requires to throttle the timer to prevent a DoS by small
     intervals and keeps the system pointlessly out of low power states
     for no value. This is a long standing non-trivial problem due to
     the lock order of posix-timer lock and the sighand lock along with
     life time issues as the timer and the sigqueue have different life
     time rules.

     Cure this by:

       - Embedding the sigqueue into the timer struct to have the same
         life time rules. Aside of that this also avoids the lookup of
         the timer in the signal delivery and rearm path as it's just a
         always valid container_of() now.

       - Queuing ignored timer signals onto a seperate ignored list.

       - Moving queued timer signals onto the ignored list when the
         signal is switched to SIG_IGN before it could be delivered.

       - Walking the ignored list when SIG_IGN is lifted and requeue the
         signals to the actual signal lists. This allows the signal
         delivery code to rearm the timer.

     This also required to consolidate the signal delivery rules so they
     are consistent across all situations. With that all self test
     scenarios finally succeed.

   - Core infrastructure for VFS multigrain timestamping

     This is required to allow the kernel to use coarse grained time
     stamps by default and switch to fine grained time stamps when inode
     attributes are actively observed via getattr().

     These changes have been provided to the VFS tree as well, so that
     the VFS specific infrastructure could be built on top.

   - Cleanup and consolidation of the sleep() infrastructure

       - Move all sleep and timeout functions into one file

       - Rework udelay() and ndelay() into proper documented inline
         functions and replace the hardcoded magic numbers by proper
         defines.

       - Rework the fsleep() implementation to take the reality of the
         timer wheel granularity on different HZ values into account.
         Right now the boundaries are hard coded time ranges which fail
         to provide the requested accuracy on different HZ settings.

       - Update documentation for all sleep/timeout related functions
         and fix up stale documentation links all over the place

       - Fixup a few usage sites

   - Rework of timekeeping and adjtimex(2) to prepare for multiple PTP
     clocks

     A system can have multiple PTP clocks which are participating in
     seperate and independent PTP clock domains. So far the kernel only
     considers the PTP clock which is based on CLOCK TAI relevant as
     that's the clock which drives the timekeeping adjustments via the
     various user space daemons through adjtimex(2).

     The non TAI based clock domains are accessible via the file
     descriptor based posix clocks, but their usability is very limited.
     They can't be accessed fast as they always go all the way out to
     the hardware and they cannot be utilized in the kernel itself.

     As Time Sensitive Networking (TSN) gains traction it is required to
     provide fast user and kernel space access to these clocks.

     The approach taken is to utilize the timekeeping and adjtimex(2)
     infrastructure to provide this access in a similar way how the
     kernel provides access to clock MONOTONIC, REALTIME etc.

     Instead of creating a duplicated infrastructure this rework
     converts timekeeping and adjtimex(2) into generic functionality
     which operates on pointers to data structures instead of using
     static variables.

     This allows to provide time accessors and adjtimex(2) functionality
     for the independent PTP clocks in a subsequent step.

   - Consolidate hrtimer initialization

     hrtimers are set up by initializing the data structure and then
     seperately setting the callback function for historical reasons.

     That's an extra unnecessary step and makes Rust support less
     straight forward than it should be.

     Provide a new set of hrtimer_setup*() functions and convert the
     core code and a few usage sites of the less frequently used
     interfaces over.

     The bulk of the htimer_init() to hrtimer_setup() conversion is
     already prepared and scheduled for the next merge window.

   - Drivers:

       - Ensure that the global timekeeping clocksource is utilizing the
         cluster 0 timer on MIPS multi-cluster systems.

         Otherwise CPUs on different clusters use their cluster specific
         clocksource which is not guaranteed to be synchronized with
         other clusters.

       - Mostly boring cleanups, fixes, improvements and code movement"

* tag 'timers-core-2024-11-18' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (140 commits)
  posix-timers: Fix spurious warning on double enqueue versus do_exit()
  clocksource/drivers/arm_arch_timer: Use of_property_present() for non-boolean properties
  clocksource/drivers/gpx: Remove redundant casts
  clocksource/drivers/timer-ti-dm: Fix child node refcount handling
  dt-bindings: timer: actions,owl-timer: convert to YAML
  clocksource/drivers/ralink: Add Ralink System Tick Counter driver
  clocksource/drivers/mips-gic-timer: Always use cluster 0 counter as clocksource
  clocksource/drivers/timer-ti-dm: Don't fail probe if int not found
  clocksource/drivers:sp804: Make user selectable
  clocksource/drivers/dw_apb: Remove unused dw_apb_clockevent functions
  hrtimers: Delete hrtimer_init_on_stack()
  alarmtimer: Switch to use hrtimer_setup() and hrtimer_setup_on_stack()
  io_uring: Switch to use hrtimer_setup_on_stack()
  sched/idle: Switch to use hrtimer_setup_on_stack()
  hrtimers: Delete hrtimer_init_sleeper_on_stack()
  wait: Switch to use hrtimer_setup_sleeper_on_stack()
  timers: Switch to use hrtimer_setup_sleeper_on_stack()
  net: pktgen: Switch to use hrtimer_setup_sleeper_on_stack()
  futex: Switch to use hrtimer_setup_sleeper_on_stack()
  fs/aio: Switch to use hrtimer_setup_sleeper_on_stack()
  ...

1 files changed, 211 insertions, 321 deletions

diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index cdd61990cd14..0ca85ff4fbb4 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -30,8 +30,9 @@
 #include "timekeeping_internal.h"
 
 #define TK_CLEAR_NTP		(1 << 0)
-#define TK_MIRROR		(1 << 1)
-#define TK_CLOCK_WAS_SET	(1 << 2)
+#define TK_CLOCK_WAS_SET	(1 << 1)
+
+#define TK_UPDATE_ALL		(TK_CLEAR_NTP | TK_CLOCK_WAS_SET)
 
 enum timekeeping_adv_mode {
 	/* Update timekeeper when a tick has passed */
@@ -41,20 +42,18 @@ enum timekeeping_adv_mode {
 	TK_ADV_FREQ
 };
 
-DEFINE_RAW_SPINLOCK(timekeeper_lock);
-
 /*
  * The most important data for readout fits into a single 64 byte
  * cache line.
  */
-static struct {
+struct tk_data {
 	seqcount_raw_spinlock_t	seq;
 	struct timekeeper	timekeeper;
-} tk_core ____cacheline_aligned = {
-	.seq = SEQCNT_RAW_SPINLOCK_ZERO(tk_core.seq, &timekeeper_lock),
-};
+	struct timekeeper	shadow_timekeeper;
+	raw_spinlock_t		lock;
+} ____cacheline_aligned;
 
-static struct timekeeper shadow_timekeeper;
+static struct tk_data tk_core;
 
 /* flag for if timekeeping is suspended */
 int __read_mostly timekeeping_suspended;
@@ -114,6 +113,19 @@ static struct tk_fast tk_fast_raw  ____cacheline_aligned = {
 	.base[1] = FAST_TK_INIT,
 };
 
+unsigned long timekeeper_lock_irqsave(void)
+{
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&tk_core.lock, flags);
+	return flags;
+}
+
+void timekeeper_unlock_irqrestore(unsigned long flags)
+{
+	raw_spin_unlock_irqrestore(&tk_core.lock, flags);
+}
+
 /*
  * Multigrain timestamps require tracking the latest fine-grained timestamp
  * that has been issued, and never returning a coarse-grained timestamp that is
@@ -178,13 +190,15 @@ static void tk_set_wall_to_mono(struct timekeeper *tk, struct timespec64 wtm)
 	WARN_ON_ONCE(tk->offs_real != timespec64_to_ktime(tmp));
 	tk->wall_to_monotonic = wtm;
 	set_normalized_timespec64(&tmp, -wtm.tv_sec, -wtm.tv_nsec);
-	tk->offs_real = timespec64_to_ktime(tmp);
-	tk->offs_tai = ktime_add(tk->offs_real, ktime_set(tk->tai_offset, 0));
+	/* Paired with READ_ONCE() in ktime_mono_to_any() */
+	WRITE_ONCE(tk->offs_real, timespec64_to_ktime(tmp));
+	WRITE_ONCE(tk->offs_tai, ktime_add(tk->offs_real, ktime_set(tk->tai_offset, 0)));
 }
 
 static inline void tk_update_sleep_time(struct timekeeper *tk, ktime_t delta)
 {
-	tk->offs_boot = ktime_add(tk->offs_boot, delta);
+	/* Paired with READ_ONCE() in ktime_mono_to_any() */
+	WRITE_ONCE(tk->offs_boot, ktime_add(tk->offs_boot, delta));
 	/*
 	 * Timespec representation for VDSO update to avoid 64bit division
 	 * on every update.
@@ -201,7 +215,7 @@ static inline void tk_update_sleep_time(struct timekeeper *tk, ktime_t delta)
  * the tkr's clocksource may change between the read reference, and the
  * clock reference passed to the read function.  This can cause crashes if
  * the wrong clocksource is passed to the wrong read function.
- * This isn't necessary to use when holding the timekeeper_lock or doing
+ * This isn't necessary to use when holding the tk_core.lock or doing
  * a read of the fast-timekeeper tkrs (which is protected by its own locking
  * and update logic).
  */
@@ -212,97 +226,6 @@ static inline u64 tk_clock_read(const struct tk_read_base *tkr)
 	return clock->read(clock);
 }
 
-#ifdef CONFIG_DEBUG_TIMEKEEPING
-#define WARNING_FREQ (HZ*300) /* 5 minute rate-limiting */
-
-static void timekeeping_check_update(struct timekeeper *tk, u64 offset)
-{
-
-	u64 max_cycles = tk->tkr_mono.clock->max_cycles;
-	const char *name = tk->tkr_mono.clock->name;
-
-	if (offset > max_cycles) {
-		printk_deferred("WARNING: timekeeping: Cycle offset (%lld) is larger than allowed by the '%s' clock's max_cycles value (%lld): time overflow danger\n",
-				offset, name, max_cycles);
-		printk_deferred("         timekeeping: Your kernel is sick, but tries to cope by capping time updates\n");
-	} else {
-		if (offset > (max_cycles >> 1)) {
-			printk_deferred("INFO: timekeeping: Cycle offset (%lld) is larger than the '%s' clock's 50%% safety margin (%lld)\n",
-					offset, name, max_cycles >> 1);
-			printk_deferred("      timekeeping: Your kernel is still fine, but is feeling a bit nervous\n");
-		}
-	}
-
-	if (tk->underflow_seen) {
-		if (jiffies - tk->last_warning > WARNING_FREQ) {
-			printk_deferred("WARNING: Underflow in clocksource '%s' observed, time update ignored.\n", name);
-			printk_deferred("         Please report this, consider using a different clocksource, if possible.\n");
-			printk_deferred("         Your kernel is probably still fine.\n");
-			tk->last_warning = jiffies;
-		}
-		tk->underflow_seen = 0;
-	}
-
-	if (tk->overflow_seen) {
-		if (jiffies - tk->last_warning > WARNING_FREQ) {
-			printk_deferred("WARNING: Overflow in clocksource '%s' observed, time update capped.\n", name);
-			printk_deferred("         Please report this, consider using a different clocksource, if possible.\n");
-			printk_deferred("         Your kernel is probably still fine.\n");
-			tk->last_warning = jiffies;
-		}
-		tk->overflow_seen = 0;
-	}
-}
-
-static inline u64 timekeeping_cycles_to_ns(const struct tk_read_base *tkr, u64 cycles);
-
-static inline u64 timekeeping_debug_get_ns(const struct tk_read_base *tkr)
-{
-	struct timekeeper *tk = &tk_core.timekeeper;
-	u64 now, last, mask, max, delta;
-	unsigned int seq;
-
-	/*
-	 * Since we're called holding a seqcount, the data may shift
-	 * under us while we're doing the calculation. This can cause
-	 * false positives, since we'd note a problem but throw the
-	 * results away. So nest another seqcount here to atomically
-	 * grab the points we are checking with.
-	 */
-	do {
-		seq = read_seqcount_begin(&tk_core.seq);
-		now = tk_clock_read(tkr);
-		last = tkr->cycle_last;
-		mask = tkr->mask;
-		max = tkr->clock->max_cycles;
-	} while (read_seqcount_retry(&tk_core.seq, seq));
-
-	delta = clocksource_delta(now, last, mask);
-
-	/*
-	 * Try to catch underflows by checking if we are seeing small
-	 * mask-relative negative values.
-	 */
-	if (unlikely((~delta & mask) < (mask >> 3)))
-		tk->underflow_seen = 1;
-
-	/* Check for multiplication overflows */
-	if (unlikely(delta > max))
-		tk->overflow_seen = 1;
-
-	/* timekeeping_cycles_to_ns() handles both under and overflow */
-	return timekeeping_cycles_to_ns(tkr, now);
-}
-#else
-static inline void timekeeping_check_update(struct timekeeper *tk, u64 offset)
-{
-}
-static inline u64 timekeeping_debug_get_ns(const struct tk_read_base *tkr)
-{
-	BUG();
-}
-#endif
-
 /**
  * tk_setup_internals - Set up internals to use clocksource clock.
  *
@@ -407,19 +330,11 @@ static inline u64 timekeeping_cycles_to_ns(const struct tk_read_base *tkr, u64 c
 	return ((delta * tkr->mult) + tkr->xtime_nsec) >> tkr->shift;
 }
 
-static __always_inline u64 __timekeeping_get_ns(const struct tk_read_base *tkr)
+static __always_inline u64 timekeeping_get_ns(const struct tk_read_base *tkr)
 {
 	return timekeeping_cycles_to_ns(tkr, tk_clock_read(tkr));
 }
 
-static inline u64 timekeeping_get_ns(const struct tk_read_base *tkr)
-{
-	if (IS_ENABLED(CONFIG_DEBUG_TIMEKEEPING))
-		return timekeeping_debug_get_ns(tkr);
-
-	return __timekeeping_get_ns(tkr);
-}
-
 /**
  * update_fast_timekeeper - Update the fast and NMI safe monotonic timekeeper.
  * @tkr: Timekeeping readout base from which we take the update
@@ -465,7 +380,7 @@ static __always_inline u64 __ktime_get_fast_ns(struct tk_fast *tkf)
 		seq = read_seqcount_latch(&tkf->seq);
 		tkr = tkf->base + (seq & 0x01);
 		now = ktime_to_ns(tkr->base);
-		now += __timekeeping_get_ns(tkr);
+		now += timekeeping_get_ns(tkr);
 	} while (read_seqcount_latch_retry(&tkf->seq, seq));
 
 	return now;
@@ -536,7 +451,7 @@ EXPORT_SYMBOL_GPL(ktime_get_raw_fast_ns);
  *    timekeeping_inject_sleeptime64()
  *    __timekeeping_inject_sleeptime(tk, delta);
  *                                                 timestamp();
- *    timekeeping_update(tk, TK_CLEAR_NTP...);
+ *    timekeeping_update_staged(tkd, TK_CLEAR_NTP...);
  *
  * (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
@@ -581,7 +496,7 @@ static __always_inline u64 __ktime_get_real_fast(struct tk_fast *tkf, u64 *mono)
 		tkr = tkf->base + (seq & 0x01);
 		basem = ktime_to_ns(tkr->base);
 		baser = ktime_to_ns(tkr->base_real);
-		delta = __timekeeping_get_ns(tkr);
+		delta = timekeeping_get_ns(tkr);
 	} while (raw_read_seqcount_latch_retry(&tkf->seq, seq));
 
 	if (mono)
@@ -695,13 +610,11 @@ static void update_pvclock_gtod(struct timekeeper *tk, bool was_set)
 int pvclock_gtod_register_notifier(struct notifier_block *nb)
 {
 	struct timekeeper *tk = &tk_core.timekeeper;
-	unsigned long flags;
 	int ret;
 
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
+	guard(raw_spinlock_irqsave)(&tk_core.lock);
 	ret = raw_notifier_chain_register(&pvclock_gtod_chain, nb);
 	update_pvclock_gtod(tk, true);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
 
 	return ret;
 }
@@ -714,14 +627,8 @@ EXPORT_SYMBOL_GPL(pvclock_gtod_register_notifier);
  */
 int pvclock_gtod_unregister_notifier(struct notifier_block *nb)
 {
-	unsigned long flags;
-	int ret;
-
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	ret = raw_notifier_chain_unregister(&pvclock_gtod_chain, nb);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
-
-	return ret;
+	guard(raw_spinlock_irqsave)(&tk_core.lock);
+	return raw_notifier_chain_unregister(&pvclock_gtod_chain, nb);
 }
 EXPORT_SYMBOL_GPL(pvclock_gtod_unregister_notifier);
 
@@ -737,6 +644,18 @@ static inline void tk_update_leap_state(struct timekeeper *tk)
 }
 
 /*
+ * Leap state update for both shadow and the real timekeeper
+ * Separate to spare a full memcpy() of the timekeeper.
+ */
+static void tk_update_leap_state_all(struct tk_data *tkd)
+{
+	write_seqcount_begin(&tkd->seq);
+	tk_update_leap_state(&tkd->shadow_timekeeper);
+	tkd->timekeeper.next_leap_ktime = tkd->shadow_timekeeper.next_leap_ktime;
+	write_seqcount_end(&tkd->seq);
+}
+
+/*
  * Update the ktime_t based scalar nsec members of the timekeeper
  */
 static inline void tk_update_ktime_data(struct timekeeper *tk)
@@ -769,9 +688,30 @@ static inline void tk_update_ktime_data(struct timekeeper *tk)
 	tk->tkr_raw.base = ns_to_ktime(tk->raw_sec * NSEC_PER_SEC);
 }
 
-/* must hold timekeeper_lock */
-static void timekeeping_update(struct timekeeper *tk, unsigned int action)
+/*
+ * Restore the shadow timekeeper from the real timekeeper.
+ */
+static void timekeeping_restore_shadow(struct tk_data *tkd)
+{
+	lockdep_assert_held(&tkd->lock);
+	memcpy(&tkd->shadow_timekeeper, &tkd->timekeeper, sizeof(tkd->timekeeper));
+}
+
+static void timekeeping_update_from_shadow(struct tk_data *tkd, unsigned int action)
 {
+	struct timekeeper *tk = &tk_core.shadow_timekeeper;
+
+	lockdep_assert_held(&tkd->lock);
+
+	/*
+	 * Block out readers before running the updates below because that
+	 * updates VDSO and other time related infrastructure. Not blocking
+	 * the readers might let a reader see time going backwards when
+	 * reading from the VDSO after the VDSO update and then reading in
+	 * the kernel from the timekeeper before that got updated.
+	 */
+	write_seqcount_begin(&tkd->seq);
+
 	if (action & TK_CLEAR_NTP) {
 		tk->ntp_error = 0;
 		ntp_clear();
@@ -789,14 +729,17 @@ static void timekeeping_update(struct timekeeper *tk, unsigned int action)
 
 	if (action & TK_CLOCK_WAS_SET)
 		tk->clock_was_set_seq++;
+
 	/*
-	 * The mirroring of the data to the shadow-timekeeper needs
-	 * to happen last here to ensure we don't over-write the
-	 * timekeeper structure on the next update with stale data
+	 * Update the real timekeeper.
+	 *
+	 * We could avoid this memcpy() by switching pointers, but that has
+	 * the downside that the reader side does not longer benefit from
+	 * the cacheline optimized data layout of the timekeeper and requires
+	 * another indirection.
 	 */
-	if (action & TK_MIRROR)
-		memcpy(&shadow_timekeeper, &tk_core.timekeeper,
-		       sizeof(tk_core.timekeeper));
+	memcpy(&tkd->timekeeper, tk, sizeof(*tk));
+	write_seqcount_end(&tkd->seq);
 }
 
 /**
@@ -949,6 +892,14 @@ ktime_t ktime_mono_to_any(ktime_t tmono, enum tk_offsets offs)
 	unsigned int seq;
 	ktime_t tconv;
 
+	if (IS_ENABLED(CONFIG_64BIT)) {
+		/*
+		 * Paired with WRITE_ONCE()s in tk_set_wall_to_mono() and
+		 * tk_update_sleep_time().
+		 */
+		return ktime_add(tmono, READ_ONCE(*offset));
+	}
+
 	do {
 		seq = read_seqcount_begin(&tk_core.seq);
 		tconv = ktime_add(tmono, *offset);
@@ -1079,6 +1030,7 @@ void ktime_get_snapshot(struct system_time_snapshot *systime_snapshot)
 	unsigned int seq;
 	ktime_t base_raw;
 	ktime_t base_real;
+	ktime_t base_boot;
 	u64 nsec_raw;
 	u64 nsec_real;
 	u64 now;
@@ -1093,6 +1045,8 @@ void ktime_get_snapshot(struct system_time_snapshot *systime_snapshot)
 		systime_snapshot->clock_was_set_seq = tk->clock_was_set_seq;
 		base_real = ktime_add(tk->tkr_mono.base,
 				      tk_core.timekeeper.offs_real);
+		base_boot = ktime_add(tk->tkr_mono.base,
+				      tk_core.timekeeper.offs_boot);
 		base_raw = tk->tkr_raw.base;
 		nsec_real = timekeeping_cycles_to_ns(&tk->tkr_mono, now);
 		nsec_raw  = timekeeping_cycles_to_ns(&tk->tkr_raw, now);
@@ -1100,6 +1054,7 @@ void ktime_get_snapshot(struct system_time_snapshot *systime_snapshot)
 
 	systime_snapshot->cycles = now;
 	systime_snapshot->real = ktime_add_ns(base_real, nsec_real);
+	systime_snapshot->boot = ktime_add_ns(base_boot, nsec_real);
 	systime_snapshot->raw = ktime_add_ns(base_raw, nsec_raw);
 }
 EXPORT_SYMBOL_GPL(ktime_get_snapshot);
@@ -1459,45 +1414,35 @@ EXPORT_SYMBOL_GPL(timekeeping_clocksource_has_base);
  */
 int do_settimeofday64(const struct timespec64 *ts)
 {
-	struct timekeeper *tk = &tk_core.timekeeper;
 	struct timespec64 ts_delta, xt;
-	unsigned long flags;
-	int ret = 0;
 
 	if (!timespec64_valid_settod(ts))
 		return -EINVAL;
 
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	write_seqcount_begin(&tk_core.seq);
-
-	timekeeping_forward_now(tk);
+	scoped_guard (raw_spinlock_irqsave, &tk_core.lock) {
+		struct timekeeper *tks = &tk_core.shadow_timekeeper;
 
-	xt = tk_xtime(tk);
-	ts_delta = timespec64_sub(*ts, xt);
+		timekeeping_forward_now(tks);
 
-	if (timespec64_compare(&tk->wall_to_monotonic, &ts_delta) > 0) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	tk_set_wall_to_mono(tk, timespec64_sub(tk->wall_to_monotonic, ts_delta));
+		xt = tk_xtime(tks);
+		ts_delta = timespec64_sub(*ts, xt);
 
-	tk_set_xtime(tk, ts);
-out:
-	timekeeping_update(tk, TK_CLEAR_NTP | TK_MIRROR | TK_CLOCK_WAS_SET);
+		if (timespec64_compare(&tks->wall_to_monotonic, &ts_delta) > 0) {
+			timekeeping_restore_shadow(&tk_core);
+			return -EINVAL;
+		}
 
-	write_seqcount_end(&tk_core.seq);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
+		tk_set_wall_to_mono(tks, timespec64_sub(tks->wall_to_monotonic, ts_delta));
+		tk_set_xtime(tks, ts);
+		timekeeping_update_from_shadow(&tk_core, TK_UPDATE_ALL);
+	}
 
 	/* Signal hrtimers about time change */
 	clock_was_set(CLOCK_SET_WALL);
 
-	if (!ret) {
-		audit_tk_injoffset(ts_delta);
-		add_device_randomness(ts, sizeof(*ts));
-	}
-
-	return ret;
+	audit_tk_injoffset(ts_delta);
+	add_device_randomness(ts, sizeof(*ts));
+	return 0;
 }
 EXPORT_SYMBOL(do_settimeofday64);
 
@@ -1509,40 +1454,31 @@ EXPORT_SYMBOL(do_settimeofday64);
  */
 static int timekeeping_inject_offset(const struct timespec64 *ts)
 {
-	struct timekeeper *tk = &tk_core.timekeeper;
-	unsigned long flags;
-	struct timespec64 tmp;
-	int ret = 0;
-
 	if (ts->tv_nsec < 0 || ts->tv_nsec >= NSEC_PER_SEC)
 		return -EINVAL;
 
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	write_seqcount_begin(&tk_core.seq);
-
-	timekeeping_forward_now(tk);
-
-	/* Make sure the proposed value is valid */
-	tmp = timespec64_add(tk_xtime(tk), *ts);
-	if (timespec64_compare(&tk->wall_to_monotonic, ts) > 0 ||
-	    !timespec64_valid_settod(&tmp)) {
-		ret = -EINVAL;
-		goto error;
-	}
+	scoped_guard (raw_spinlock_irqsave, &tk_core.lock) {
+		struct timekeeper *tks = &tk_core.shadow_timekeeper;
+		struct timespec64 tmp;
 
-	tk_xtime_add(tk, ts);
-	tk_set_wall_to_mono(tk, timespec64_sub(tk->wall_to_monotonic, *ts));
+		timekeeping_forward_now(tks);
 
-error: /* even if we error out, we forwarded the time, so call update */
-	timekeeping_update(tk, TK_CLEAR_NTP | TK_MIRROR | TK_CLOCK_WAS_SET);
+		/* Make sure the proposed value is valid */
+		tmp = timespec64_add(tk_xtime(tks), *ts);
+		if (timespec64_compare(&tks->wall_to_monotonic, ts) > 0 ||
+		    !timespec64_valid_settod(&tmp)) {
+			timekeeping_restore_shadow(&tk_core);
+			return -EINVAL;
+		}
 
-	write_seqcount_end(&tk_core.seq);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
+		tk_xtime_add(tks, ts);
+		tk_set_wall_to_mono(tks, timespec64_sub(tks->wall_to_monotonic, *ts));
+		timekeeping_update_from_shadow(&tk_core, TK_UPDATE_ALL);
+	}
 
 	/* Signal hrtimers about time change */
 	clock_was_set(CLOCK_SET_WALL);
-
-	return ret;
+	return 0;
 }
 
 /*
@@ -1595,43 +1531,34 @@ static void __timekeeping_set_tai_offset(struct timekeeper *tk, s32 tai_offset)
  */
 static int change_clocksource(void *data)
 {
-	struct timekeeper *tk = &tk_core.timekeeper;
-	struct clocksource *new, *old = NULL;
-	unsigned long flags;
-	bool change = false;
-
-	new = (struct clocksource *) data;
+	struct clocksource *new = data, *old = NULL;
 
 	/*
-	 * If the cs is in module, get a module reference. Succeeds
-	 * for built-in code (owner == NULL) as well.
+	 * If the clocksource is in a module, get a module reference.
+	 * Succeeds for built-in code (owner == NULL) as well. Abort if the
+	 * reference can't be acquired.
 	 */
-	if (try_module_get(new->owner)) {
-		if (!new->enable || new->enable(new) == 0)
-			change = true;
-		else
-			module_put(new->owner);
-	}
-
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	write_seqcount_begin(&tk_core.seq);
-
-	timekeeping_forward_now(tk);
+	if (!try_module_get(new->owner))
+		return 0;
 
-	if (change) {
-		old = tk->tkr_mono.clock;
-		tk_setup_internals(tk, new);
+	/* Abort if the device can't be enabled */
+	if (new->enable && new->enable(new) != 0) {
+		module_put(new->owner);
+		return 0;
 	}
 
-	timekeeping_update(tk, TK_CLEAR_NTP | TK_MIRROR | TK_CLOCK_WAS_SET);
+	scoped_guard (raw_spinlock_irqsave, &tk_core.lock) {
+		struct timekeeper *tks = &tk_core.shadow_timekeeper;
 
-	write_seqcount_end(&tk_core.seq);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
+		timekeeping_forward_now(tks);
+		old = tks->tkr_mono.clock;
+		tk_setup_internals(tks, new);
+		timekeeping_update_from_shadow(&tk_core, TK_UPDATE_ALL);
+	}
 
 	if (old) {
 		if (old->disable)
 			old->disable(old);
-
 		module_put(old->owner);
 	}
 
@@ -1756,6 +1683,12 @@ read_persistent_wall_and_boot_offset(struct timespec64 *wall_time,
 	*boot_offset = ns_to_timespec64(local_clock());
 }
 
+static __init void tkd_basic_setup(struct tk_data *tkd)
+{
+	raw_spin_lock_init(&tkd->lock);
+	seqcount_raw_spinlock_init(&tkd->seq, &tkd->lock);
+}
+
 /*
  * Flag reflecting whether timekeeping_resume() has injected sleeptime.
  *
@@ -1780,9 +1713,10 @@ static bool persistent_clock_exists;
 void __init timekeeping_init(void)
 {
 	struct timespec64 wall_time, boot_offset, wall_to_mono;
-	struct timekeeper *tk = &tk_core.timekeeper;
+	struct timekeeper *tks = &tk_core.shadow_timekeeper;
 	struct clocksource *clock;
-	unsigned long flags;
+
+	tkd_basic_setup(&tk_core);
 
 	read_persistent_wall_and_boot_offset(&wall_time, &boot_offset);
 	if (timespec64_valid_settod(&wall_time) &&
@@ -1802,24 +1736,21 @@ void __init timekeeping_init(void)
 	 */
 	wall_to_mono = timespec64_sub(boot_offset, wall_time);
 
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	write_seqcount_begin(&tk_core.seq);
+	guard(raw_spinlock_irqsave)(&tk_core.lock);
+
 	ntp_init();
 
 	clock = clocksource_default_clock();
 	if (clock->enable)
 		clock->enable(clock);
-	tk_setup_internals(tk, clock);
-
-	tk_set_xtime(tk, &wall_time);
-	tk->raw_sec = 0;
+	tk_setup_internals(tks, clock);
 
-	tk_set_wall_to_mono(tk, wall_to_mono);
+	tk_set_xtime(tks, &wall_time);
+	tks->raw_sec = 0;
 
-	timekeeping_update(tk, TK_MIRROR | TK_CLOCK_WAS_SET);
+	tk_set_wall_to_mono(tks, wall_to_mono);
 
-	write_seqcount_end(&tk_core.seq);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
+	timekeeping_update_from_shadow(&tk_core, TK_CLOCK_WAS_SET);
 }
 
 /* time in seconds when suspend began for persistent clock */
@@ -1897,22 +1828,14 @@ bool timekeeping_rtc_skipsuspend(void)
  */
 void timekeeping_inject_sleeptime64(const struct timespec64 *delta)
 {
-	struct timekeeper *tk = &tk_core.timekeeper;
-	unsigned long flags;
+	scoped_guard(raw_spinlock_irqsave, &tk_core.lock) {
+		struct timekeeper *tks = &tk_core.shadow_timekeeper;
 
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	write_seqcount_begin(&tk_core.seq);
-
-	suspend_timing_needed = false;
-
-	timekeeping_forward_now(tk);
-
-	__timekeeping_inject_sleeptime(tk, delta);
-
-	timekeeping_update(tk, TK_CLEAR_NTP | TK_MIRROR | TK_CLOCK_WAS_SET);
-
-	write_seqcount_end(&tk_core.seq);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
+		suspend_timing_needed = false;
+		timekeeping_forward_now(tks);
+		__timekeeping_inject_sleeptime(tks, delta);
+		timekeeping_update_from_shadow(&tk_core, TK_UPDATE_ALL);
+	}
 
 	/* Signal hrtimers about time change */
 	clock_was_set(CLOCK_SET_WALL | CLOCK_SET_BOOT);
@@ -1924,20 +1847,19 @@ void timekeeping_inject_sleeptime64(const struct timespec64 *delta)
  */
 void timekeeping_resume(void)
 {
-	struct timekeeper *tk = &tk_core.timekeeper;
-	struct clocksource *clock = tk->tkr_mono.clock;
-	unsigned long flags;
+	struct timekeeper *tks = &tk_core.shadow_timekeeper;
+	struct clocksource *clock = tks->tkr_mono.clock;
 	struct timespec64 ts_new, ts_delta;
-	u64 cycle_now, nsec;
 	bool inject_sleeptime = false;
+	u64 cycle_now, nsec;
+	unsigned long flags;
 
 	read_persistent_clock64(&ts_new);
 
 	clockevents_resume();
 	clocksource_resume();
 
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	write_seqcount_begin(&tk_core.seq);
+	raw_spin_lock_irqsave(&tk_core.lock, flags);
 
 	/*
 	 * After system resumes, we need to calculate the suspended time and
@@ -1951,7 +1873,7 @@ void timekeeping_resume(void)
 	 * The less preferred source will only be tried if there is no better
 	 * usable source. The rtc part is handled separately in rtc core code.
 	 */
-	cycle_now = tk_clock_read(&tk->tkr_mono);
+	cycle_now = tk_clock_read(&tks->tkr_mono);
 	nsec = clocksource_stop_suspend_timing(clock, cycle_now);
 	if (nsec > 0) {
 		ts_delta = ns_to_timespec64(nsec);
@@ -1963,18 +1885,17 @@ void timekeeping_resume(void)
 
 	if (inject_sleeptime) {
 		suspend_timing_needed = false;
-		__timekeeping_inject_sleeptime(tk, &ts_delta);
+		__timekeeping_inject_sleeptime(tks, &ts_delta);
 	}
 
 	/* Re-base the last cycle value */
-	tk->tkr_mono.cycle_last = cycle_now;
-	tk->tkr_raw.cycle_last  = cycle_now;
+	tks->tkr_mono.cycle_last = cycle_now;
+	tks->tkr_raw.cycle_last  = cycle_now;
 
-	tk->ntp_error = 0;
+	tks->ntp_error = 0;
 	timekeeping_suspended = 0;
-	timekeeping_update(tk, TK_MIRROR | TK_CLOCK_WAS_SET);
-	write_seqcount_end(&tk_core.seq);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
+	timekeeping_update_from_shadow(&tk_core, TK_CLOCK_WAS_SET);
+	raw_spin_unlock_irqrestore(&tk_core.lock, flags);
 
 	touch_softlockup_watchdog();
 
@@ -1986,11 +1907,11 @@ void timekeeping_resume(void)
 
 int timekeeping_suspend(void)
 {
-	struct timekeeper *tk = &tk_core.timekeeper;
-	unsigned long flags;
-	struct timespec64		delta, delta_delta;
-	static struct timespec64	old_delta;
+	struct timekeeper *tks = &tk_core.shadow_timekeeper;
+	struct timespec64 delta, delta_delta;
+	static struct timespec64 old_delta;
 	struct clocksource *curr_clock;
+	unsigned long flags;
 	u64 cycle_now;
 
 	read_persistent_clock64(&timekeeping_suspend_time);
@@ -2005,9 +1926,8 @@ int timekeeping_suspend(void)
 
 	suspend_timing_needed = true;
 
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	write_seqcount_begin(&tk_core.seq);
-	timekeeping_forward_now(tk);
+	raw_spin_lock_irqsave(&tk_core.lock, flags);
+	timekeeping_forward_now(tks);
 	timekeeping_suspended = 1;
 
 	/*
@@ -2015,8 +1935,8 @@ int timekeeping_suspend(void)
 	 * just read from the current clocksource. Save this to potentially
 	 * use in suspend timing.
 	 */
-	curr_clock = tk->tkr_mono.clock;
-	cycle_now = tk->tkr_mono.cycle_last;
+	curr_clock = tks->tkr_mono.clock;
+	cycle_now = tks->tkr_mono.cycle_last;
 	clocksource_start_suspend_timing(curr_clock, cycle_now);
 
 	if (persistent_clock_exists) {
@@ -2026,7 +1946,7 @@ int timekeeping_suspend(void)
 		 * try to compensate so the difference in system time
 		 * and persistent_clock time stays close to constant.
 		 */
-		delta = timespec64_sub(tk_xtime(tk), timekeeping_suspend_time);
+		delta = timespec64_sub(tk_xtime(tks), timekeeping_suspend_time);
 		delta_delta = timespec64_sub(delta, old_delta);
 		if (abs(delta_delta.tv_sec) >= 2) {
 			/*
@@ -2041,10 +1961,9 @@ int timekeeping_suspend(void)
 		}
 	}
 
-	timekeeping_update(tk, TK_MIRROR);
-	halt_fast_timekeeper(tk);
-	write_seqcount_end(&tk_core.seq);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
+	timekeeping_update_from_shadow(&tk_core, 0);
+	halt_fast_timekeeper(tks);
+	raw_spin_unlock_irqrestore(&tk_core.lock, flags);
 
 	tick_suspend();
 	clocksource_suspend();
@@ -2149,16 +2068,17 @@ static __always_inline void timekeeping_apply_adjustment(struct timekeeper *tk,
  */
 static void timekeeping_adjust(struct timekeeper *tk, s64 offset)
 {
+	u64 ntp_tl = ntp_tick_length();
 	u32 mult;
 
 	/*
 	 * Determine the multiplier from the current NTP tick length.
 	 * Avoid expensive division when the tick length doesn't change.
 	 */
-	if (likely(tk->ntp_tick == ntp_tick_length())) {
+	if (likely(tk->ntp_tick == ntp_tl)) {
 		mult = tk->tkr_mono.mult - tk->ntp_err_mult;
 	} else {
-		tk->ntp_tick = ntp_tick_length();
+		tk->ntp_tick = ntp_tl;
 		mult = div64_u64((tk->ntp_tick >> tk->ntp_error_shift) -
 				 tk->xtime_remainder, tk->cycle_interval);
 	}
@@ -2297,28 +2217,24 @@ static u64 logarithmic_accumulation(struct timekeeper *tk, u64 offset,
  */
 static bool timekeeping_advance(enum timekeeping_adv_mode mode)
 {
+	struct timekeeper *tk = &tk_core.shadow_timekeeper;
 	struct timekeeper *real_tk = &tk_core.timekeeper;
-	struct timekeeper *tk = &shadow_timekeeper;
-	u64 offset;
-	int shift = 0, maxshift;
 	unsigned int clock_set = 0;
-	unsigned long flags;
+	int shift = 0, maxshift;
+	u64 offset;
 
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
+	guard(raw_spinlock_irqsave)(&tk_core.lock);
 
 	/* Make sure we're fully resumed: */
 	if (unlikely(timekeeping_suspended))
-		goto out;
+		return false;
 
 	offset = clocksource_delta(tk_clock_read(&tk->tkr_mono),
 				   tk->tkr_mono.cycle_last, tk->tkr_mono.mask);
 
 	/* Check if there's really nothing to do */
 	if (offset < real_tk->cycle_interval && mode == TK_ADV_TICK)
-		goto out;
-
-	/* Do some additional sanity checking */
-	timekeeping_check_update(tk, offset);
+		return false;
 
 	/*
 	 * With NO_HZ we may have to accumulate many cycle_intervals
@@ -2334,8 +2250,7 @@ static bool timekeeping_advance(enum timekeeping_adv_mode mode)
 	maxshift = (64 - (ilog2(ntp_tick_length())+1)) - 1;
 	shift = min(shift, maxshift);
 	while (offset >= tk->cycle_interval) {
-		offset = logarithmic_accumulation(tk, offset, shift,
-							&clock_set);
+		offset = logarithmic_accumulation(tk, offset, shift, &clock_set);
 		if (offset < tk->cycle_interval<<shift)
 			shift--;
 	}
@@ -2349,23 +2264,7 @@ static bool timekeeping_advance(enum timekeeping_adv_mode mode)
 	 */
 	clock_set |= accumulate_nsecs_to_secs(tk);
 
-	write_seqcount_begin(&tk_core.seq);
-	/*
-	 * Update the real timekeeper.
-	 *
-	 * We could avoid this memcpy by switching pointers, but that
-	 * requires changes to all other timekeeper usage sites as
-	 * well, i.e. move the timekeeper pointer getter into the
-	 * spinlocked/seqcount protected sections. And we trade this
-	 * memcpy under the tk_core.seq against one before we start
-	 * updating.
-	 */
-	timekeeping_update(tk, clock_set);
-	memcpy(real_tk, tk, sizeof(*tk));
-	/* The memcpy must come last. Do not put anything here! */
-	write_seqcount_end(&tk_core.seq);
-out:
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
+	timekeeping_update_from_shadow(&tk_core, clock_set);
 
 	return !!clock_set;
 }
@@ -2658,13 +2557,10 @@ EXPORT_SYMBOL_GPL(random_get_entropy_fallback);
  */
 int do_adjtimex(struct __kernel_timex *txc)
 {
-	struct timekeeper *tk = &tk_core.timekeeper;
 	struct audit_ntp_data ad;
 	bool offset_set = false;
 	bool clock_set = false;
 	struct timespec64 ts;
-	unsigned long flags;
-	s32 orig_tai, tai;
 	int ret;
 
 	/* Validate the data before disabling interrupts */
@@ -2675,6 +2571,7 @@ int do_adjtimex(struct __kernel_timex *txc)
 
 	if (txc->modes & ADJ_SETOFFSET) {
 		struct timespec64 delta;
+
 		delta.tv_sec  = txc->time.tv_sec;
 		delta.tv_nsec = txc->time.tv_usec;
 		if (!(txc->modes & ADJ_NANO))
@@ -2692,21 +2589,21 @@ int do_adjtimex(struct __kernel_timex *txc)
 	ktime_get_real_ts64(&ts);
 	add_device_randomness(&ts, sizeof(ts));
 
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	write_seqcount_begin(&tk_core.seq);
+	scoped_guard (raw_spinlock_irqsave, &tk_core.lock) {
+		struct timekeeper *tks = &tk_core.shadow_timekeeper;
+		s32 orig_tai, tai;
 
-	orig_tai = tai = tk->tai_offset;
-	ret = __do_adjtimex(txc, &ts, &tai, &ad);
+		orig_tai = tai = tks->tai_offset;
+		ret = __do_adjtimex(txc, &ts, &tai, &ad);
 
-	if (tai != orig_tai) {
-		__timekeeping_set_tai_offset(tk, tai);
-		timekeeping_update(tk, TK_MIRROR | TK_CLOCK_WAS_SET);
-		clock_set = true;
+		if (tai != orig_tai) {
+			__timekeeping_set_tai_offset(tks, tai);
+			timekeeping_update_from_shadow(&tk_core, TK_CLOCK_WAS_SET);
+			clock_set = true;
+		} else {
+			tk_update_leap_state_all(&tk_core);
+		}
 	}
-	tk_update_leap_state(tk);
-
-	write_seqcount_end(&tk_core.seq);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
 
 	audit_ntp_log(&ad);
 
@@ -2730,15 +2627,8 @@ int do_adjtimex(struct __kernel_timex *txc)
  */
 void hardpps(const struct timespec64 *phase_ts, const struct timespec64 *raw_ts)
 {
-	unsigned long flags;
-
-	raw_spin_lock_irqsave(&timekeeper_lock, flags);
-	write_seqcount_begin(&tk_core.seq);
-
+	guard(raw_spinlock_irqsave)(&tk_core.lock);
 	__hardpps(phase_ts, raw_ts);
-
-	write_seqcount_end(&tk_core.seq);
-	raw_spin_unlock_irqrestore(&timekeeper_lock, flags);
 }
 EXPORT_SYMBOL(hardpps);
 #endif /* CONFIG_NTP_PPS */