Merge branch 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull core timer updates from Thomas Gleixner:
 "Timers and timekeeping updates:

   - A large overhaul of the posix CPU timer code which is a preparation
     for moving the CPU timer expiry out into task work so it can be
     properly accounted on the task/process.

     An update to the bogus permission checks will come later during the
     merge window as feedback was not complete before heading of for
     travel.

   - Switch the timerqueue code to use cached rbtrees and get rid of the
     homebrewn caching of the leftmost node.

   - Consolidate hrtimer_init() + hrtimer_init_sleeper() calls into a
     single function

   - Implement the separation of hrtimers to be forced to expire in hard
     interrupt context even when PREEMPT_RT is enabled and mark the
     affected timers accordingly.

   - Implement a mechanism for hrtimers and the timer wheel to protect
     RT against priority inversion and live lock issues when a (hr)timer
     which should be canceled is currently executing the callback.
     Instead of infinitely spinning, the task which tries to cancel the
     timer blocks on a per cpu base expiry lock which is held and
     released by the (hr)timer expiry code.

   - Enable the Hyper-V TSC page based sched_clock for Hyper-V guests
     resulting in faster access to timekeeping functions.

   - Updates to various clocksource/clockevent drivers and their device
     tree bindings.

   - The usual small improvements all over the place"

* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (101 commits)
  posix-cpu-timers: Fix permission check regression
  posix-cpu-timers: Always clear head pointer on dequeue
  hrtimer: Add a missing bracket and hide `migration_base' on !SMP
  posix-cpu-timers: Make expiry_active check actually work correctly
  posix-timers: Unbreak CONFIG_POSIX_TIMERS=n build
  tick: Mark sched_timer to expire in hard interrupt context
  hrtimer: Add kernel doc annotation for HRTIMER_MODE_HARD
  x86/hyperv: Hide pv_ops access for CONFIG_PARAVIRT=n
  posix-cpu-timers: Utilize timerqueue for storage
  posix-cpu-timers: Move state tracking to struct posix_cputimers
  posix-cpu-timers: Deduplicate rlimit handling
  posix-cpu-timers: Remove pointless comparisons
  posix-cpu-timers: Get rid of 64bit divisions
  posix-cpu-timers: Consolidate timer expiry further
  posix-cpu-timers: Get rid of zero checks
  rlimit: Rewrite non-sensical RLIMIT_CPU comment
  posix-cpu-timers: Respect INFINITY for hard RTTIME limit
  posix-cpu-timers: Switch thread group sampling to array
  posix-cpu-timers: Restructure expiry array
  posix-cpu-timers: Remove cputime_expires
  ...

9 files changed, 889 insertions, 581 deletions

diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c
index b7d75a9e8ccf..271ce6c12907 100644
--- a/kernel/time/alarmtimer.c
+++ b/kernel/time/alarmtimer.c
@@ -432,7 +432,7 @@ int alarm_cancel(struct alarm *alarm)
 		int ret = alarm_try_to_cancel(alarm);
 		if (ret >= 0)
 			return ret;
-		cpu_relax();
+		hrtimer_cancel_wait_running(&alarm->timer);
 	}
 }
 EXPORT_SYMBOL_GPL(alarm_cancel);
@@ -606,6 +606,19 @@ static int alarm_timer_try_to_cancel(struct k_itimer *timr)
 }
 
 /**
+ * alarm_timer_wait_running - Posix timer callback to wait for a timer
+ * @timr:	Pointer to the posixtimer data struct
+ *
+ * Called from the core code when timer cancel detected that the callback
+ * is running. @timr is unlocked and rcu read lock is held to prevent it
+ * from being freed.
+ */
+static void alarm_timer_wait_running(struct k_itimer *timr)
+{
+	hrtimer_cancel_wait_running(&timr->it.alarm.alarmtimer.timer);
+}
+
+/**
  * alarm_timer_arm - Posix timer callback to arm a timer
  * @timr:	Pointer to the posixtimer data struct
  * @expires:	The new expiry time
@@ -834,6 +847,7 @@ const struct k_clock alarm_clock = {
 	.timer_forward		= alarm_timer_forward,
 	.timer_remaining	= alarm_timer_remaining,
 	.timer_try_to_cancel	= alarm_timer_try_to_cancel,
+	.timer_wait_running	= alarm_timer_wait_running,
 	.nsleep			= alarm_timer_nsleep,
 };
 #endif /* CONFIG_POSIX_TIMERS */
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 5ee77f1a8a92..0d4dc241c0fb 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -140,6 +140,11 @@ static struct hrtimer_cpu_base migration_cpu_base = {
 
 #define migration_base	migration_cpu_base.clock_base[0]
 
+static inline bool is_migration_base(struct hrtimer_clock_base *base)
+{
+	return base == &migration_base;
+}
+
 /*
  * We are using hashed locking: holding per_cpu(hrtimer_bases)[n].lock
  * means that all timers which are tied to this base via timer->base are
@@ -264,6 +269,11 @@ again:
 
 #else /* CONFIG_SMP */
 
+static inline bool is_migration_base(struct hrtimer_clock_base *base)
+{
+	return false;
+}
+
 static inline struct hrtimer_clock_base *
 lock_hrtimer_base(const struct hrtimer *timer, unsigned long *flags)
 {
@@ -427,6 +437,17 @@ void hrtimer_init_on_stack(struct hrtimer *timer, clockid_t clock_id,
 }
 EXPORT_SYMBOL_GPL(hrtimer_init_on_stack);
 
+static void __hrtimer_init_sleeper(struct hrtimer_sleeper *sl,
+				   clockid_t clock_id, enum hrtimer_mode mode);
+
+void hrtimer_init_sleeper_on_stack(struct hrtimer_sleeper *sl,
+				   clockid_t clock_id, enum hrtimer_mode mode)
+{
+	debug_object_init_on_stack(&sl->timer, &hrtimer_debug_descr);
+	__hrtimer_init_sleeper(sl, clock_id, mode);
+}
+EXPORT_SYMBOL_GPL(hrtimer_init_sleeper_on_stack);
+
 void destroy_hrtimer_on_stack(struct hrtimer *timer)
 {
 	debug_object_free(timer, &hrtimer_debug_descr);
@@ -1096,9 +1117,13 @@ void hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
 
 	/*
 	 * Check whether the HRTIMER_MODE_SOFT bit and hrtimer.is_soft
-	 * match.
+	 * match on CONFIG_PREEMPT_RT = n. With PREEMPT_RT check the hard
+	 * expiry mode because unmarked timers are moved to softirq expiry.
 	 */
-	WARN_ON_ONCE(!(mode & HRTIMER_MODE_SOFT) ^ !timer->is_soft);
+	if (!IS_ENABLED(CONFIG_PREEMPT_RT))
+		WARN_ON_ONCE(!(mode & HRTIMER_MODE_SOFT) ^ !timer->is_soft);
+	else
+		WARN_ON_ONCE(!(mode & HRTIMER_MODE_HARD) ^ !timer->is_hard);
 
 	base = lock_hrtimer_base(timer, &flags);
 
@@ -1147,6 +1172,93 @@ int hrtimer_try_to_cancel(struct hrtimer *timer)
 }
 EXPORT_SYMBOL_GPL(hrtimer_try_to_cancel);
 
+#ifdef CONFIG_PREEMPT_RT
+static void hrtimer_cpu_base_init_expiry_lock(struct hrtimer_cpu_base *base)
+{
+	spin_lock_init(&base->softirq_expiry_lock);
+}
+
+static void hrtimer_cpu_base_lock_expiry(struct hrtimer_cpu_base *base)
+{
+	spin_lock(&base->softirq_expiry_lock);
+}
+
+static void hrtimer_cpu_base_unlock_expiry(struct hrtimer_cpu_base *base)
+{
+	spin_unlock(&base->softirq_expiry_lock);
+}
+
+/*
+ * The counterpart to hrtimer_cancel_wait_running().
+ *
+ * If there is a waiter for cpu_base->expiry_lock, then it was waiting for
+ * the timer callback to finish. Drop expiry_lock and reaquire it. That
+ * allows the waiter to acquire the lock and make progress.
+ */
+static void hrtimer_sync_wait_running(struct hrtimer_cpu_base *cpu_base,
+				      unsigned long flags)
+{
+	if (atomic_read(&cpu_base->timer_waiters)) {
+		raw_spin_unlock_irqrestore(&cpu_base->lock, flags);
+		spin_unlock(&cpu_base->softirq_expiry_lock);
+		spin_lock(&cpu_base->softirq_expiry_lock);
+		raw_spin_lock_irq(&cpu_base->lock);
+	}
+}
+
+/*
+ * This function is called on PREEMPT_RT kernels when the fast path
+ * deletion of a timer failed because the timer callback function was
+ * running.
+ *
+ * This prevents priority inversion: if the soft irq thread is preempted
+ * in the middle of a timer callback, then calling del_timer_sync() can
+ * lead to two issues:
+ *
+ *  - If the caller is on a remote CPU then it has to spin wait for the timer
+ *    handler to complete. This can result in unbound priority inversion.
+ *
+ *  - If the caller originates from the task which preempted the timer
+ *    handler on the same CPU, then spin waiting for the timer handler to
+ *    complete is never going to end.
+ */
+void hrtimer_cancel_wait_running(const struct hrtimer *timer)
+{
+	/* Lockless read. Prevent the compiler from reloading it below */
+	struct hrtimer_clock_base *base = READ_ONCE(timer->base);
+
+	/*
+	 * Just relax if the timer expires in hard interrupt context or if
+	 * it is currently on the migration base.
+	 */
+	if (!timer->is_soft || is_migration_base(base)) {
+		cpu_relax();
+		return;
+	}
+
+	/*
+	 * Mark the base as contended and grab the expiry lock, which is
+	 * held by the softirq across the timer callback. Drop the lock
+	 * immediately so the softirq can expire the next timer. In theory
+	 * the timer could already be running again, but that's more than
+	 * unlikely and just causes another wait loop.
+	 */
+	atomic_inc(&base->cpu_base->timer_waiters);
+	spin_lock_bh(&base->cpu_base->softirq_expiry_lock);
+	atomic_dec(&base->cpu_base->timer_waiters);
+	spin_unlock_bh(&base->cpu_base->softirq_expiry_lock);
+}
+#else
+static inline void
+hrtimer_cpu_base_init_expiry_lock(struct hrtimer_cpu_base *base) { }
+static inline void
+hrtimer_cpu_base_lock_expiry(struct hrtimer_cpu_base *base) { }
+static inline void
+hrtimer_cpu_base_unlock_expiry(struct hrtimer_cpu_base *base) { }
+static inline void hrtimer_sync_wait_running(struct hrtimer_cpu_base *base,
+					     unsigned long flags) { }
+#endif
+
 /**
  * hrtimer_cancel - cancel a timer and wait for the handler to finish.
  * @timer:	the timer to be cancelled
@@ -1157,13 +1269,15 @@ EXPORT_SYMBOL_GPL(hrtimer_try_to_cancel);
  */
 int hrtimer_cancel(struct hrtimer *timer)
 {
-	for (;;) {
-		int ret = hrtimer_try_to_cancel(timer);
+	int ret;
 
-		if (ret >= 0)
-			return ret;
-		cpu_relax();
-	}
+	do {
+		ret = hrtimer_try_to_cancel(timer);
+
+		if (ret < 0)
+			hrtimer_cancel_wait_running(timer);
+	} while (ret < 0);
+	return ret;
 }
 EXPORT_SYMBOL_GPL(hrtimer_cancel);
 
@@ -1260,8 +1374,17 @@ static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id,
 			   enum hrtimer_mode mode)
 {
 	bool softtimer = !!(mode & HRTIMER_MODE_SOFT);
-	int base = softtimer ? HRTIMER_MAX_CLOCK_BASES / 2 : 0;
 	struct hrtimer_cpu_base *cpu_base;
+	int base;
+
+	/*
+	 * On PREEMPT_RT enabled kernels hrtimers which are not explicitely
+	 * marked for hard interrupt expiry mode are moved into soft
+	 * interrupt context for latency reasons and because the callbacks
+	 * can invoke functions which might sleep on RT, e.g. spin_lock().
+	 */
+	if (IS_ENABLED(CONFIG_PREEMPT_RT) && !(mode & HRTIMER_MODE_HARD))
+		softtimer = true;
 
 	memset(timer, 0, sizeof(struct hrtimer));
 
@@ -1275,8 +1398,10 @@ static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id,
 	if (clock_id == CLOCK_REALTIME && mode & HRTIMER_MODE_REL)
 		clock_id = CLOCK_MONOTONIC;
 
+	base = softtimer ? HRTIMER_MAX_CLOCK_BASES / 2 : 0;
 	base += hrtimer_clockid_to_base(clock_id);
 	timer->is_soft = softtimer;
+	timer->is_hard = !softtimer;
 	timer->base = &cpu_base->clock_base[base];
 	timerqueue_init(&timer->node);
 }
@@ -1449,6 +1574,8 @@ static void __hrtimer_run_queues(struct hrtimer_cpu_base *cpu_base, ktime_t now,
 				break;
 
 			__run_hrtimer(cpu_base, base, timer, &basenow, flags);
+			if (active_mask == HRTIMER_ACTIVE_SOFT)
+				hrtimer_sync_wait_running(cpu_base, flags);
 		}
 	}
 }
@@ -1459,6 +1586,7 @@ static __latent_entropy void hrtimer_run_softirq(struct softirq_action *h)
 	unsigned long flags;
 	ktime_t now;
 
+	hrtimer_cpu_base_lock_expiry(cpu_base);
 	raw_spin_lock_irqsave(&cpu_base->lock, flags);
 
 	now = hrtimer_update_base(cpu_base);
@@ -1468,6 +1596,7 @@ static __latent_entropy void hrtimer_run_softirq(struct softirq_action *h)
 	hrtimer_update_softirq_timer(cpu_base, true);
 
 	raw_spin_unlock_irqrestore(&cpu_base->lock, flags);
+	hrtimer_cpu_base_unlock_expiry(cpu_base);
 }
 
 #ifdef CONFIG_HIGH_RES_TIMERS
@@ -1639,10 +1768,75 @@ static enum hrtimer_restart hrtimer_wakeup(struct hrtimer *timer)
 	return HRTIMER_NORESTART;
 }
 
-void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, struct task_struct *task)
+/**
+ * hrtimer_sleeper_start_expires - Start a hrtimer sleeper timer
+ * @sl:		sleeper to be started
+ * @mode:	timer mode abs/rel
+ *
+ * Wrapper around hrtimer_start_expires() for hrtimer_sleeper based timers
+ * to allow PREEMPT_RT to tweak the delivery mode (soft/hardirq context)
+ */
+void hrtimer_sleeper_start_expires(struct hrtimer_sleeper *sl,
+				   enum hrtimer_mode mode)
+{
+	/*
+	 * Make the enqueue delivery mode check work on RT. If the sleeper
+	 * was initialized for hard interrupt delivery, force the mode bit.
+	 * This is a special case for hrtimer_sleepers because
+	 * hrtimer_init_sleeper() determines the delivery mode on RT so the
+	 * fiddling with this decision is avoided at the call sites.
+	 */
+	if (IS_ENABLED(CONFIG_PREEMPT_RT) && sl->timer.is_hard)
+		mode |= HRTIMER_MODE_HARD;
+
+	hrtimer_start_expires(&sl->timer, mode);
+}
+EXPORT_SYMBOL_GPL(hrtimer_sleeper_start_expires);
+
+static void __hrtimer_init_sleeper(struct hrtimer_sleeper *sl,
+				   clockid_t clock_id, enum hrtimer_mode mode)
 {
+	/*
+	 * On PREEMPT_RT enabled kernels hrtimers which are not explicitely
+	 * marked for hard interrupt expiry mode are moved into soft
+	 * interrupt context either for latency reasons or because the
+	 * hrtimer callback takes regular spinlocks or invokes other
+	 * functions which are not suitable for hard interrupt context on
+	 * PREEMPT_RT.
+	 *
+	 * The hrtimer_sleeper callback is RT compatible in hard interrupt
+	 * context, but there is a latency concern: Untrusted userspace can
+	 * spawn many threads which arm timers for the same expiry time on
+	 * the same CPU. That causes a latency spike due to the wakeup of
+	 * a gazillion threads.
+	 *
+	 * OTOH, priviledged real-time user space applications rely on the
+	 * low latency of hard interrupt wakeups. If the current task is in
+	 * a real-time scheduling class, mark the mode for hard interrupt
+	 * expiry.
+	 */
+	if (IS_ENABLED(CONFIG_PREEMPT_RT)) {
+		if (task_is_realtime(current) && !(mode & HRTIMER_MODE_SOFT))
+			mode |= HRTIMER_MODE_HARD;
+	}
+
+	__hrtimer_init(&sl->timer, clock_id, mode);
 	sl->timer.function = hrtimer_wakeup;
-	sl->task = task;
+	sl->task = current;
+}
+
+/**
+ * hrtimer_init_sleeper - initialize sleeper to the given clock
+ * @sl:		sleeper to be initialized
+ * @clock_id:	the clock to be used
+ * @mode:	timer mode abs/rel
+ */
+void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, clockid_t clock_id,
+			  enum hrtimer_mode mode)
+{
+	debug_init(&sl->timer, clock_id, mode);
+	__hrtimer_init_sleeper(sl, clock_id, mode);
+
 }
 EXPORT_SYMBOL_GPL(hrtimer_init_sleeper);
 
@@ -1669,11 +1863,9 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod
 {
 	struct restart_block *restart;
 
-	hrtimer_init_sleeper(t, current);
-
 	do {
 		set_current_state(TASK_INTERRUPTIBLE);
-		hrtimer_start_expires(&t->timer, mode);
+		hrtimer_sleeper_start_expires(t, mode);
 
 		if (likely(t->task))
 			freezable_schedule();
@@ -1707,10 +1899,9 @@ static long __sched hrtimer_nanosleep_restart(struct restart_block *restart)
 	struct hrtimer_sleeper t;
 	int ret;
 
-	hrtimer_init_on_stack(&t.timer, restart->nanosleep.clockid,
-				HRTIMER_MODE_ABS);
+	hrtimer_init_sleeper_on_stack(&t, restart->nanosleep.clockid,
+				      HRTIMER_MODE_ABS);
 	hrtimer_set_expires_tv64(&t.timer, restart->nanosleep.expires);
-
 	ret = do_nanosleep(&t, HRTIMER_MODE_ABS);
 	destroy_hrtimer_on_stack(&t.timer);
 	return ret;
@@ -1728,7 +1919,7 @@ long hrtimer_nanosleep(const struct timespec64 *rqtp,
 	if (dl_task(current) || rt_task(current))
 		slack = 0;
 
-	hrtimer_init_on_stack(&t.timer, clockid, mode);
+	hrtimer_init_sleeper_on_stack(&t, clockid, mode);
 	hrtimer_set_expires_range_ns(&t.timer, timespec64_to_ktime(*rqtp), slack);
 	ret = do_nanosleep(&t, mode);
 	if (ret != -ERESTART_RESTARTBLOCK)
@@ -1809,6 +2000,7 @@ int hrtimers_prepare_cpu(unsigned int cpu)
 	cpu_base->softirq_next_timer = NULL;
 	cpu_base->expires_next = KTIME_MAX;
 	cpu_base->softirq_expires_next = KTIME_MAX;
+	hrtimer_cpu_base_init_expiry_lock(cpu_base);
 	return 0;
 }
 
@@ -1927,12 +2119,9 @@ schedule_hrtimeout_range_clock(ktime_t *expires, u64 delta,
 		return -EINTR;
 	}
 
-	hrtimer_init_on_stack(&t.timer, clock_id, mode);
+	hrtimer_init_sleeper_on_stack(&t, clock_id, mode);
 	hrtimer_set_expires_range_ns(&t.timer, *expires, delta);
-
-	hrtimer_init_sleeper(&t, current);
-
-	hrtimer_start_expires(&t.timer, mode);
+	hrtimer_sleeper_start_expires(&t, mode);
 
 	if (likely(t.task))
 		schedule();
diff --git a/kernel/time/itimer.c b/kernel/time/itimer.c
index 02068b2d5862..77f1e5635cc1 100644
--- a/kernel/time/itimer.c
+++ b/kernel/time/itimer.c
@@ -55,15 +55,10 @@ static void get_cpu_itimer(struct task_struct *tsk, unsigned int clock_id,
 	val = it->expires;
 	interval = it->incr;
 	if (val) {
-		struct task_cputime cputime;
-		u64 t;
+		u64 t, samples[CPUCLOCK_MAX];
 
-		thread_group_cputimer(tsk, &cputime);
-		if (clock_id == CPUCLOCK_PROF)
-			t = cputime.utime + cputime.stime;
-		else
-			/* CPUCLOCK_VIRT */
-			t = cputime.utime;
+		thread_group_sample_cputime(tsk, samples);
+		t = samples[clock_id];
 
 		if (val < t)
 			/* about to fire */
@@ -213,6 +208,7 @@ again:
 		/* We are sharing ->siglock with it_real_fn() */
 		if (hrtimer_try_to_cancel(timer) < 0) {
 			spin_unlock_irq(&tsk->sighand->siglock);
+			hrtimer_cancel_wait_running(timer);
 			goto again;
 		}
 		expires = timeval_to_ktime(value->it_value);
diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c
index 0a426f4e3125..92a431981b1c 100644
--- a/kernel/time/posix-cpu-timers.c
+++ b/kernel/time/posix-cpu-timers.c
@@ -20,11 +20,20 @@
 
 static void posix_cpu_timer_rearm(struct k_itimer *timer);
 
+void posix_cputimers_group_init(struct posix_cputimers *pct, u64 cpu_limit)
+{
+	posix_cputimers_init(pct);
+	if (cpu_limit != RLIM_INFINITY) {
+		pct->bases[CPUCLOCK_PROF].nextevt = cpu_limit * NSEC_PER_SEC;
+		pct->timers_active = true;
+	}
+}
+
 /*
  * Called after updating RLIMIT_CPU to run cpu timer and update
- * tsk->signal->cputime_expires expiration cache if necessary. Needs
- * siglock protection since other code may update expiration cache as
- * well.
+ * tsk->signal->posix_cputimers.bases[clock].nextevt expiration cache if
+ * necessary. Needs siglock protection since other code may update the
+ * expiration cache as well.
  */
 void update_rlimit_cpu(struct task_struct *task, unsigned long rlim_new)
 {
@@ -35,46 +44,97 @@ void update_rlimit_cpu(struct task_struct *task, unsigned long rlim_new)
 	spin_unlock_irq(&task->sighand->siglock);
 }
 
-static int check_clock(const clockid_t which_clock)
+/*
+ * Functions for validating access to tasks.
+ */
+static struct task_struct *lookup_task(const pid_t pid, bool thread,
+				       bool gettime)
 {
-	int error = 0;
 	struct task_struct *p;
-	const pid_t pid = CPUCLOCK_PID(which_clock);
-
-	if (CPUCLOCK_WHICH(which_clock) >= CPUCLOCK_MAX)
-		return -EINVAL;
 
-	if (pid == 0)
-		return 0;
+	/*
+	 * If the encoded PID is 0, then the timer is targeted at current
+	 * or the process to which current belongs.
+	 */
+	if (!pid)
+		return thread ? current : current->group_leader;
 
-	rcu_read_lock();
 	p = find_task_by_vpid(pid);
-	if (!p || !(CPUCLOCK_PERTHREAD(which_clock) ?
-		   same_thread_group(p, current) : has_group_leader_pid(p))) {
-		error = -EINVAL;
+	if (!p)
+		return p;
+
+	if (thread)
+		return same_thread_group(p, current) ? p : NULL;
+
+	if (gettime) {
+		/*
+		 * For clock_gettime(PROCESS) the task does not need to be
+		 * the actual group leader. tsk->sighand gives
+		 * access to the group's clock.
+		 *
+		 * Timers need the group leader because they take a
+		 * reference on it and store the task pointer until the
+		 * timer is destroyed.
+		 */
+		return (p == current || thread_group_leader(p)) ? p : NULL;
 	}
+
+	/*
+	 * For processes require that p is group leader.
+	 */
+	return has_group_leader_pid(p) ? p : NULL;
+}
+
+static struct task_struct *__get_task_for_clock(const clockid_t clock,
+						bool getref, bool gettime)
+{
+	const bool thread = !!CPUCLOCK_PERTHREAD(clock);
+	const pid_t pid = CPUCLOCK_PID(clock);
+	struct task_struct *p;
+
+	if (CPUCLOCK_WHICH(clock) >= CPUCLOCK_MAX)
+		return NULL;
+
+	rcu_read_lock();
+	p = lookup_task(pid, thread, gettime);
+	if (p && getref)
+		get_task_struct(p);
 	rcu_read_unlock();
+	return p;
+}
 
-	return error;
+static inline struct task_struct *get_task_for_clock(const clockid_t clock)
+{
+	return __get_task_for_clock(clock, true, false);
+}
+
+static inline struct task_struct *get_task_for_clock_get(const clockid_t clock)
+{
+	return __get_task_for_clock(clock, true, true);
+}
+
+static inline int validate_clock_permissions(const clockid_t clock)
+{
+	return __get_task_for_clock(clock, false, false) ? 0 : -EINVAL;
 }
 
 /*
  * Update expiry time from increment, and increase overrun count,
  * given the current clock sample.
  */
-static void bump_cpu_timer(struct k_itimer *timer, u64 now)
+static u64 bump_cpu_timer(struct k_itimer *timer, u64 now)
 {
+	u64 delta, incr, expires = timer->it.cpu.node.expires;
 	int i;
-	u64 delta, incr;
 
 	if (!timer->it_interval)
-		return;
+		return expires;
 
-	if (now < timer->it.cpu.expires)
-		return;
+	if (now < expires)
+		return expires;
 
 	incr = timer->it_interval;
-	delta = now + incr - timer->it.cpu.expires;
+	delta = now + incr - expires;
 
 	/* Don't use (incr*2 < delta), incr*2 might overflow. */
 	for (i = 0; incr < delta - incr; i++)
@@ -84,48 +144,26 @@ static void bump_cpu_timer(struct k_itimer *timer, u64 now)
 		if (delta < incr)
 			continue;
 
-		timer->it.cpu.expires += incr;
+		timer->it.cpu.node.expires += incr;
 		timer->it_overrun += 1LL << i;
 		delta -= incr;
 	}
+	return timer->it.cpu.node.expires;
 }
 
-/**
- * task_cputime_zero - Check a task_cputime struct for all zero fields.
- *
- * @cputime:	The struct to compare.
- *
- * Checks @cputime to see if all fields are zero.  Returns true if all fields
- * are zero, false if any field is nonzero.
- */
-static inline int task_cputime_zero(const struct task_cputime *cputime)
+/* Check whether all cache entries contain U64_MAX, i.e. eternal expiry time */
+static inline bool expiry_cache_is_inactive(const struct posix_cputimers *pct)
 {
-	if (!cputime->utime && !cputime->stime && !cputime->sum_exec_runtime)
-		return 1;
-	return 0;
-}
-
-static inline u64 prof_ticks(struct task_struct *p)
-{
-	u64 utime, stime;
-
-	task_cputime(p, &utime, &stime);
-
-	return utime + stime;
-}
-static inline u64 virt_ticks(struct task_struct *p)
-{
-	u64 utime, stime;
-
-	task_cputime(p, &utime, &stime);
-
-	return utime;
+	return !(~pct->bases[CPUCLOCK_PROF].nextevt |
+		 ~pct->bases[CPUCLOCK_VIRT].nextevt |
+		 ~pct->bases[CPUCLOCK_SCHED].nextevt);
 }
 
 static int
 posix_cpu_clock_getres(const clockid_t which_clock, struct timespec64 *tp)
 {
-	int error = check_clock(which_clock);
+	int error = validate_clock_permissions(which_clock);
+
 	if (!error) {
 		tp->tv_sec = 0;
 		tp->tv_nsec = ((NSEC_PER_SEC + HZ - 1) / HZ);
@@ -142,42 +180,66 @@ posix_cpu_clock_getres(const clockid_t which_clock, struct timespec64 *tp)
 }
 
 static int
-posix_cpu_clock_set(const clockid_t which_clock, const struct timespec64 *tp)
+posix_cpu_clock_set(const clockid_t clock, const struct timespec64 *tp)
 {
+	int error = validate_clock_permissions(clock);
+
 	/*
 	 * You can never reset a CPU clock, but we check for other errors
 	 * in the call before failing with EPERM.
 	 */
-	int error = check_clock(which_clock);
-	if (error == 0) {
-		error = -EPERM;
-	}
-	return error;
+	return error ? : -EPERM;
 }
 
-
 /*
- * Sample a per-thread clock for the given task.
+ * Sample a per-thread clock for the given task. clkid is validated.
  */
-static int cpu_clock_sample(const clockid_t which_clock,
-			    struct task_struct *p, u64 *sample)
+static u64 cpu_clock_sample(const clockid_t clkid, struct task_struct *p)
 {
-	switch (CPUCLOCK_WHICH(which_clock)) {
-	default:
-		return -EINVAL;
+	u64 utime, stime;
+
+	if (clkid == CPUCLOCK_SCHED)
+		return task_sched_runtime(p);
+
+	task_cputime(p, &utime, &stime);
+
+	switch (clkid) {
 	case CPUCLOCK_PROF:
-		*sample = prof_ticks(p);
-		break;
+		return utime + stime;
 	case CPUCLOCK_VIRT:
-		*sample = virt_ticks(p);
-		break;
-	case CPUCLOCK_SCHED:
-		*sample = task_sched_runtime(p);
-		break;
+		return utime;
+	default:
+		WARN_ON_ONCE(1);
 	}
 	return 0;
 }
 
+static inline void store_samples(u64 *samples, u64 stime, u64 utime, u64 rtime)
+{
+	samples[CPUCLOCK_PROF] = stime + utime;
+	samples[CPUCLOCK_VIRT] = utime;
+	samples[CPUCLOCK_SCHED] = rtime;
+}
+
+static void task_sample_cputime(struct task_struct *p, u64 *samples)
+{
+	u64 stime, utime;
+
+	task_cputime(p, &utime, &stime);
+	store_samples(samples, stime, utime, p->se.sum_exec_runtime);
+}
+
+static void proc_sample_cputime_atomic(struct task_cputime_atomic *at,
+				       u64 *samples)
+{
+	u64 stime, utime, rtime;
+
+	utime = atomic64_read(&at->utime);
+	stime = atomic64_read(&at->stime);
+	rtime = atomic64_read(&at->sum_exec_runtime);
+	store_samples(samples, stime, utime, rtime);
+}
+
 /*
  * Set cputime to sum_cputime if sum_cputime > cputime. Use cmpxchg
  * to avoid race conditions with concurrent updates to cputime.
@@ -193,29 +255,56 @@ retry:
 	}
 }
 
-static void update_gt_cputime(struct task_cputime_atomic *cputime_atomic, struct task_cputime *sum)
+static void update_gt_cputime(struct task_cputime_atomic *cputime_atomic,
+			      struct task_cputime *sum)
 {
 	__update_gt_cputime(&cputime_atomic->utime, sum->utime);
 	__update_gt_cputime(&cputime_atomic->stime, sum->stime);
 	__update_gt_cputime(&cputime_atomic->sum_exec_runtime, sum->sum_exec_runtime);
 }
 
-/* Sample task_cputime_atomic values in "atomic_timers", store results in "times". */
-static inline void sample_cputime_atomic(struct task_cputime *times,
-					 struct task_cputime_atomic *atomic_times)
+/**
+ * thread_group_sample_cputime - Sample cputime for a given task
+ * @tsk:	Task for which cputime needs to be started
+ * @iimes:	Storage for time samples
+ *
+ * Called from sys_getitimer() to calculate the expiry time of an active
+ * timer. That means group cputime accounting is already active. Called
+ * with task sighand lock held.
+ *
+ * Updates @times with an uptodate sample of the thread group cputimes.
+ */
+void thread_group_sample_cputime(struct task_struct *tsk, u64 *samples)
 {
-	times->utime = atomic64_read(&atomic_times->utime);
-	times->stime = atomic64_read(&atomic_times->stime);
-	times->sum_exec_runtime = atomic64_read(&atomic_times->sum_exec_runtime);
+	struct thread_group_cputimer *cputimer = &tsk->signal->cputimer;
+	struct posix_cputimers *pct = &tsk->signal->posix_cputimers;
+
+	WARN_ON_ONCE(!pct->timers_active);
+
+	proc_sample_cputime_atomic(&cputimer->cputime_atomic, samples);
 }
 
-void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times)
+/**
+ * thread_group_start_cputime - Start cputime and return a sample
+ * @tsk:	Task for which cputime needs to be started
+ * @samples:	Storage for time samples
+ *
+ * The thread group cputime accouting is avoided when there are no posix
+ * CPU timers armed. Before starting a timer it's required to check whether
+ * the time accounting is active. If not, a full update of the atomic
+ * accounting store needs to be done and the accounting enabled.
+ *
+ * Updates @times with an uptodate sample of the thread group cputimes.
+ */
+static void thread_group_start_cputime(struct task_struct *tsk, u64 *samples)
 {
 	struct thread_group_cputimer *cputimer = &tsk->signal->cputimer;
-	struct task_cputime sum;
+	struct posix_cputimers *pct = &tsk->signal->posix_cputimers;
 
 	/* Check if cputimer isn't running. This is accessed without locking. */
-	if (!READ_ONCE(cputimer->running)) {
+	if (!READ_ONCE(pct->timers_active)) {
+		struct task_cputime sum;
+
 		/*
 		 * The POSIX timer interface allows for absolute time expiry
 		 * values through the TIMER_ABSTIME flag, therefore we have
@@ -225,94 +314,69 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times)
 		update_gt_cputime(&cputimer->cputime_atomic, &sum);
 
 		/*
-		 * We're setting cputimer->running without a lock. Ensure
-		 * this only gets written to in one operation. We set
-		 * running after update_gt_cputime() as a small optimization,
-		 * but barriers are not required because update_gt_cputime()
+		 * We're setting timers_active without a lock. Ensure this
+		 * only gets written to in one operation. We set it after
+		 * update_gt_cputime() as a small optimization, but
+		 * barriers are not required because update_gt_cputime()
 		 * can handle concurrent updates.
 		 */
-		WRITE_ONCE(cputimer->running, true);
+		WRITE_ONCE(pct->timers_active, true);
 	}
-	sample_cputime_atomic(times, &cputimer->cputime_atomic);
+	proc_sample_cputime_atomic(&cputimer->cputime_atomic, samples);
 }
 
-/*
- * Sample a process (thread group) clock for the given group_leader task.
- * Must be called with task sighand lock held for safe while_each_thread()
- * traversal.
- */
-static int cpu_clock_sample_group(const clockid_t which_clock,
-				  struct task_struct *p,
-				  u64 *sample)
+static void __thread_group_cputime(struct task_struct *tsk, u64 *samples)
 {
-	struct task_cputime cputime;
+	struct task_cputime ct;
 
-	switch (CPUCLOCK_WHICH(which_clock)) {
-	default:
-		return -EINVAL;
-	case CPUCLOCK_PROF:
-		thread_group_cputime(p, &cputime);
-		*sample = cputime.utime + cputime.stime;
-		break;
-	case CPUCLOCK_VIRT:
-		thread_group_cputime(p, &cputime);
-		*sample = cputime.utime;
-		break;
-	case CPUCLOCK_SCHED:
-		thread_group_cputime(p, &cputime);
-		*sample = cputime.sum_exec_runtime;
-		break;
-	}
-	return 0;
+	thread_group_cputime(tsk, &ct);
+	store_samples(samples, ct.stime, ct.utime, ct.sum_exec_runtime);
 }
 
-static int posix_cpu_clock_get_task(struct task_struct *tsk,
-				    const clockid_t which_clock,
-				    struct timespec64 *tp)
+/*
+ * Sample a process (thread group) clock for the given task clkid. If the
+ * group's cputime accounting is already enabled, read the atomic
+ * store. Otherwise a full update is required.  Task's sighand lock must be
+ * held to protect the task traversal on a full update. clkid is already
+ * validated.
+ */
+static u64 cpu_clock_sample_group(const clockid_t clkid, struct task_struct *p,
+				  bool start)
 {
-	int err = -EINVAL;
-	u64 rtn;
+	struct thread_group_cputimer *cputimer = &p->signal->cputimer;
+	struct posix_cputimers *pct = &p->signal->posix_cputimers;
+	u64 samples[CPUCLOCK_MAX];
 
-	if (CPUCLOCK_PERTHREAD(which_clock)) {
-		if (same_thread_group(tsk, current))
-			err = cpu_clock_sample(which_clock, tsk, &rtn);
+	if (!READ_ONCE(pct->timers_active)) {
+		if (start)
+			thread_group_start_cputime(p, samples);
+		else
+			__thread_group_cputime(p, samples);
 	} else {
-		if (tsk == current || thread_group_leader(tsk))
-			err = cpu_clock_sample_group(which_clock, tsk, &rtn);
+		proc_sample_cputime_atomic(&cputimer->cputime_atomic, samples);
 	}
 
-	if (!err)
-		*tp = ns_to_timespec64(rtn);
-
-	return err;
+	return samples[clkid];
 }
 
-
-static int posix_cpu_clock_get(const clockid_t which_clock, struct timespec64 *tp)
+static int posix_cpu_clock_get(const clockid_t clock, struct timespec64 *tp)
 {
-	const pid_t pid = CPUCLOCK_PID(which_clock);
-	int err = -EINVAL;
+	const clockid_t clkid = CPUCLOCK_WHICH(clock);
+	struct task_struct *tsk;
+	u64 t;
 
-	if (pid == 0) {
-		/*
-		 * Special case constant value for our own clocks.
-		 * We don't have to do any lookup to find ourselves.
-		 */
-		err = posix_cpu_clock_get_task(current, which_clock, tp);
-	} else {
-		/*
-		 * Find the given PID, and validate that the caller
-		 * should be able to see it.
-		 */
-		struct task_struct *p;
-		rcu_read_lock();
-		p = find_task_by_vpid(pid);
-		if (p)
-			err = posix_cpu_clock_get_task(p, which_clock, tp);
-		rcu_read_unlock();
-	}
+	tsk = get_task_for_clock_get(clock);
+	if (!tsk)
+		return -EINVAL;
 
-	return err;
+	if (CPUCLOCK_PERTHREAD(clock))
+		t = cpu_clock_sample(clkid, tsk);
+	else
+		t = cpu_clock_sample_group(clkid, tsk, false);
+	put_task_struct(tsk);
+
+	*tp = ns_to_timespec64(t);
+	return 0;
 }
 
 /*
@@ -322,44 +386,15 @@ static int posix_cpu_clock_get(const clockid_t which_clock, struct timespec64 *t
  */
 static int posix_cpu_timer_create(struct k_itimer *new_timer)
 {
-	int ret = 0;
-	const pid_t pid = CPUCLOCK_PID(new_timer->it_clock);
-	struct task_struct *p;
+	struct task_struct *p = get_task_for_clock(new_timer->it_clock);
 
-	if (CPUCLOCK_WHICH(new_timer->it_clock) >= CPUCLOCK_MAX)
+	if (!p)
 		return -EINVAL;
 
 	new_timer->kclock = &clock_posix_cpu;
-
-	INIT_LIST_HEAD(&new_timer->it.cpu.entry);
-
-	rcu_read_lock();
-	if (CPUCLOCK_PERTHREAD(new_timer->it_clock)) {
-		if (pid == 0) {
-			p = current;
-		} else {
-			p = find_task_by_vpid(pid);
-			if (p && !same_thread_group(p, current))
-				p = NULL;
-		}
-	} else {
-		if (pid == 0) {
-			p = current->group_leader;
-		} else {
-			p = find_task_by_vpid(pid);
-			if (p && !has_group_leader_pid(p))
-				p = NULL;
-		}
-	}
+	timerqueue_init(&new_timer->it.cpu.node);
 	new_timer->it.cpu.task = p;
-	if (p) {
-		get_task_struct(p);
-	} else {
-		ret = -EINVAL;
-	}
-	rcu_read_unlock();
-
-	return ret;
+	return 0;
 }
 
 /*
@@ -370,12 +405,14 @@ static int posix_cpu_timer_create(struct k_itimer *new_timer)
  */
 static int posix_cpu_timer_del(struct k_itimer *timer)
 {
-	int ret = 0;
-	unsigned long flags;
+	struct cpu_timer *ctmr = &timer->it.cpu;
+	struct task_struct *p = ctmr->task;
 	struct sighand_struct *sighand;
-	struct task_struct *p = timer->it.cpu.task;
+	unsigned long flags;
+	int ret = 0;
 
-	WARN_ON_ONCE(p == NULL);
+	if (WARN_ON_ONCE(!p))
+		return -EINVAL;
 
 	/*
 	 * Protect against sighand release/switch in exit/exec and process/
@@ -384,15 +421,15 @@ static int posix_cpu_timer_del(struct k_itimer *timer)
 	sighand = lock_task_sighand(p, &flags);
 	if (unlikely(sighand == NULL)) {
 		/*
-		 * We raced with the reaping of the task.
-		 * The deletion should have cleared us off the list.
+		 * This raced with the reaping of the task. The exit cleanup
+		 * should have removed this timer from the timer queue.
 		 */
-		WARN_ON_ONCE(!list_empty(&timer->it.cpu.entry));
+		WARN_ON_ONCE(ctmr->head || timerqueue_node_queued(&ctmr->node));
 	} else {
 		if (timer->it.cpu.firing)
 			ret = TIMER_RETRY;
 		else
-			list_del(&timer->it.cpu.entry);
+			cpu_timer_dequeue(ctmr);
 
 		unlock_task_sighand(p, &flags);
 	}
@@ -403,25 +440,30 @@ static int posix_cpu_timer_del(struct k_itimer *timer)
 	return ret;
 }
 
-static void cleanup_timers_list(struct list_head *head)
+static void cleanup_timerqueue(struct timerqueue_head *head)
 {
-	struct cpu_timer_list *timer, *next;
+	struct timerqueue_node *node;
+	struct cpu_timer *ctmr;
 
-	list_for_each_entry_safe(timer, next, head, entry)
-		list_del_init(&timer->entry);
+	while ((node = timerqueue_getnext(head))) {
+		timerqueue_del(head, node);
+		ctmr = container_of(node, struct cpu_timer, node);
+		ctmr->head = NULL;
+	}
 }
 
 /*
- * Clean out CPU timers still ticking when a thread exited.  The task
- * pointer is cleared, and the expiry time is replaced with the residual
- * time for later timer_gettime calls to return.
+ * Clean out CPU timers which are still armed when a thread exits. The
+ * timers are only removed from the list. No other updates are done. The
+ * corresponding posix timers are still accessible, but cannot be rearmed.
+ *
  * This must be called with the siglock held.
  */
-static void cleanup_timers(struct list_head *head)
+static void cleanup_timers(struct posix_cputimers *pct)
 {
-	cleanup_timers_list(head);
-	cleanup_timers_list(++head);
-	cleanup_timers_list(++head);
+	cleanup_timerqueue(&pct->bases[CPUCLOCK_PROF].tqhead);
+	cleanup_timerqueue(&pct->bases[CPUCLOCK_VIRT].tqhead);
+	cleanup_timerqueue(&pct->bases[CPUCLOCK_SCHED].tqhead);
 }
 
 /*
@@ -431,16 +473,11 @@ static void cleanup_timers(struct list_head *head)
  */
 void posix_cpu_timers_exit(struct task_struct *tsk)
 {
-	cleanup_timers(tsk->cpu_timers);
+	cleanup_timers(&tsk->posix_cputimers);
 }
 void posix_cpu_timers_exit_group(struct task_struct *tsk)
 {
-	cleanup_timers(tsk->signal->cpu_timers);
-}
-
-static inline int expires_gt(u64 expires, u64 new_exp)
-{
-	return expires == 0 || expires > new_exp;
+	cleanup_timers(&tsk->signal->posix_cputimers);
 }
 
 /*
@@ -449,58 +486,33 @@ static inline int expires_gt(u64 expires, u64 new_exp)
  */
 static void arm_timer(struct k_itimer *timer)
 {
-	struct task_struct *p = timer->it.cpu.task;
-	struct list_head *head, *listpos;
-	struct task_cputime *cputime_expires;
-	struct cpu_timer_list *const nt = &timer->it.cpu;
-	struct cpu_timer_list *next;
-
-	if (CPUCLOCK_PERTHREAD(timer->it_clock)) {
-		head = p->cpu_timers;
-		cputime_expires = &p->cputime_expires;
-	} else {
-		head = p->signal->cpu_timers;
-		cputime_expires = &p->signal->cputime_expires;
-	}
-	head += CPUCLOCK_WHICH(timer->it_clock);
-
-	listpos = head;
-	list_for_each_entry(next, head, entry) {
-		if (nt->expires < next->expires)
-			break;
-		listpos = &next->entry;
-	}
-	list_add(&nt->entry, listpos);
-
-	if (listpos == head) {
-		u64 exp = nt->expires;
+	int clkidx = CPUCLOCK_WHICH(timer->it_clock);
+	struct cpu_timer *ctmr = &timer->it.cpu;
+	u64 newexp = cpu_timer_getexpires(ctmr);
+	struct task_struct *p = ctmr->task;
+	struct posix_cputimer_base *base;
+
+	if (CPUCLOCK_PERTHREAD(timer->it_clock))
+		base = p->posix_cputimers.bases + clkidx;
+	else
+		base = p->signal->posix_cputimers.bases + clkidx;
+
+	if (!cpu_timer_enqueue(&base->tqhead, ctmr))
+		return;
 
-		/*
-		 * We are the new earliest-expiring POSIX 1.b timer, hence
-		 * need to update expiration cache. Take into account that
-		 * for process timers we share expiration cache with itimers
-		 * and RLIMIT_CPU and for thread timers with RLIMIT_RTTIME.
-		 */
+	/*
+	 * We are the new earliest-expiring POSIX 1.b timer, hence
+	 * need to update expiration cache. Take into account that
+	 * for process timers we share expiration cache with itimers
+	 * and RLIMIT_CPU and for thread timers with RLIMIT_RTTIME.
+	 */
+	if (newexp < base->nextevt)
+		base->nextevt = newexp;
 
-		switch (CPUCLOCK_WHICH(timer->it_clock)) {
-		case CPUCLOCK_PROF:
-			if (expires_gt(cputime_expires->prof_exp, exp))
-				cputime_expires->prof_exp = exp;
-			break;
-		case CPUCLOCK_VIRT:
-			if (expires_gt(cputime_expires->virt_exp, exp))
-				cputime_expires->virt_exp = exp;
-			break;
-		case CPUCLOCK_SCHED:
-			if (expires_gt(cputime_expires->sched_exp, exp))
-				cputime_expires->sched_exp = exp;
-			break;
-		}
-		if (CPUCLOCK_PERTHREAD(timer->it_clock))
-			tick_dep_set_task(p, TICK_DEP_BIT_POSIX_TIMER);
-		else
-			tick_dep_set_signal(p->signal, TICK_DEP_BIT_POSIX_TIMER);
-	}
+	if (CPUCLOCK_PERTHREAD(timer->it_clock))
+		tick_dep_set_task(p, TICK_DEP_BIT_POSIX_TIMER);
+	else
+		tick_dep_set_signal(p->signal, TICK_DEP_BIT_POSIX_TIMER);
 }
 
 /*
@@ -508,24 +520,26 @@ static void arm_timer(struct k_itimer *timer)
  */
 static void cpu_timer_fire(struct k_itimer *timer)
 {
+	struct cpu_timer *ctmr = &timer->it.cpu;
+
 	if ((timer->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE) {
 		/*
 		 * User don't want any signal.
 		 */
-		timer->it.cpu.expires = 0;
+		cpu_timer_setexpires(ctmr, 0);
 	} else if (unlikely(timer->sigq == NULL)) {
 		/*
 		 * This a special case for clock_nanosleep,
 		 * not a normal timer from sys_timer_create.
 		 */
 		wake_up_process(timer->it_process);
-		timer->it.cpu.expires = 0;
+		cpu_timer_setexpires(ctmr, 0);
 	} else if (!timer->it_interval) {
 		/*
 		 * One-shot timer.  Clear it as soon as it's fired.
 		 */
 		posix_timer_event(timer, 0);
-		timer->it.cpu.expires = 0;
+		cpu_timer_setexpires(ctmr, 0);
 	} else if (posix_timer_event(timer, ++timer->it_requeue_pending)) {
 		/*
 		 * The signal did not get queued because the signal
@@ -539,33 +553,6 @@ static void cpu_timer_fire(struct k_itimer *timer)
 }
 
 /*
- * Sample a process (thread group) timer for the given group_leader task.
- * Must be called with task sighand lock held for safe while_each_thread()
- * traversal.
- */
-static int cpu_timer_sample_group(const clockid_t which_clock,
-				  struct task_struct *p, u64 *sample)
-{
-	struct task_cputime cputime;
-
-	thread_group_cputimer(p, &cputime);
-	switch (CPUCLOCK_WHICH(which_clock)) {
-	default:
-		return -EINVAL;
-	case CPUCLOCK_PROF:
-		*sample = cputime.utime + cputime.stime;
-		break;
-	case CPUCLOCK_VIRT:
-		*sample = cputime.utime;
-		break;
-	case CPUCLOCK_SCHED:
-		*sample = cputime.sum_exec_runtime;
-		break;
-	}
-	return 0;
-}
-
-/*
  * Guts of sys_timer_settime for CPU timers.
  * This is called with the timer locked and interrupts disabled.
  * If we return TIMER_RETRY, it's necessary to release the timer's lock
@@ -574,13 +561,16 @@ static int cpu_timer_sample_group(const clockid_t which_clock,
 static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags,
 			       struct itimerspec64 *new, struct itimerspec64 *old)
 {
-	unsigned long flags;
-	struct sighand_struct *sighand;
-	struct task_struct *p = timer->it.cpu.task;
+	clockid_t clkid = CPUCLOCK_WHICH(timer->it_clock);
 	u64 old_expires, new_expires, old_incr, val;
-	int ret;
+	struct cpu_timer *ctmr = &timer->it.cpu;
+	struct task_struct *p = ctmr->task;
+	struct sighand_struct *sighand;
+	unsigned long flags;
+	int ret = 0;
 
-	WARN_ON_ONCE(p == NULL);
+	if (WARN_ON_ONCE(!p))
+		return -EINVAL;
 
 	/*
 	 * Use the to_ktime conversion because that clamps the maximum
@@ -597,22 +587,21 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags,
 	 * If p has just been reaped, we can no
 	 * longer get any information about it at all.
 	 */
-	if (unlikely(sighand == NULL)) {
+	if (unlikely(sighand == NULL))
 		return -ESRCH;
-	}
 
 	/*
 	 * Disarm any old timer after extracting its expiry time.
 	 */
-
-	ret = 0;
 	old_incr = timer->it_interval;
-	old_expires = timer->it.cpu.expires;
+	old_expires = cpu_timer_getexpires(ctmr);
+
 	if (unlikely(timer->it.cpu.firing)) {
 		timer->it.cpu.firing = -1;
 		ret = TIMER_RETRY;
-	} else
-		list_del_init(&timer->it.cpu.entry);
+	} else {
+		cpu_timer_dequeue(ctmr);
+	}
 
 	/*
 	 * We need to sample the current value to convert the new
@@ -622,11 +611,10 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags,
 	 * times (in arm_timer).  With an absolute time, we must
 	 * check if it's already passed.  In short, we need a sample.
 	 */
-	if (CPUCLOCK_PERTHREAD(timer->it_clock)) {
-		cpu_clock_sample(timer->it_clock, p, &val);
-	} else {
-		cpu_timer_sample_group(timer->it_clock, p, &val);
-	}
+	if (CPUCLOCK_PERTHREAD(timer->it_clock))
+		val = cpu_clock_sample(clkid, p);
+	else
+		val = cpu_clock_sample_group(clkid, p, true);
 
 	if (old) {
 		if (old_expires == 0) {
@@ -634,18 +622,16 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags,
 			old->it_value.tv_nsec = 0;
 		} else {
 			/*
-			 * Update the timer in case it has
-			 * overrun already.  If it has,
-			 * we'll report it as having overrun
-			 * and with the next reloaded timer
-			 * already ticking, though we are
-			 * swallowing that pending
-			 * notification here to install the
-			 * new setting.
+			 * Update the timer in case it has overrun already.
+			 * If it has, we'll report it as having overrun and
+			 * with the next reloaded timer already ticking,
+			 * though we are swallowing that pending
+			 * notification here to install the new setting.
 			 */
-			bump_cpu_timer(timer, val);
-			if (val < timer->it.cpu.expires) {
-				old_expires = timer->it.cpu.expires - val;
+			u64 exp = bump_cpu_timer(timer, val);
+
+			if (val < exp) {
+				old_expires = exp - val;
 				old->it_value = ns_to_timespec64(old_expires);
 			} else {
 				old->it_value.tv_nsec = 1;
@@ -674,7 +660,7 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags,
 	 * For a timer with no notification action, we don't actually
 	 * arm the timer (we'll just fake it for timer_gettime).
 	 */
-	timer->it.cpu.expires = new_expires;
+	cpu_timer_setexpires(ctmr, new_expires);
 	if (new_expires != 0 && val < new_expires) {
 		arm_timer(timer);
 	}
@@ -715,24 +701,27 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags,
 
 static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec64 *itp)
 {
-	u64 now;
-	struct task_struct *p = timer->it.cpu.task;
+	clockid_t clkid = CPUCLOCK_WHICH(timer->it_clock);
+	struct cpu_timer *ctmr = &timer->it.cpu;
+	u64 now, expires = cpu_timer_getexpires(ctmr);
+	struct task_struct *p = ctmr->task;
 
-	WARN_ON_ONCE(p == NULL);
+	if (WARN_ON_ONCE(!p))
+		return;
 
 	/*
 	 * Easy part: convert the reload time.
 	 */
 	itp->it_interval = ktime_to_timespec64(timer->it_interval);
 
-	if (!timer->it.cpu.expires)
+	if (!expires)
 		return;
 
 	/*
 	 * Sample the clock to take the difference with the expiry time.
 	 */
 	if (CPUCLOCK_PERTHREAD(timer->it_clock)) {
-		cpu_clock_sample(timer->it_clock, p, &now);
+		now = cpu_clock_sample(clkid, p);
 	} else {
 		struct sighand_struct *sighand;
 		unsigned long flags;
@@ -747,18 +736,18 @@ static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec64 *itp
 			/*
 			 * The process has been reaped.
 			 * We can't even collect a sample any more.
-			 * Call the timer disarmed, nothing else to do.
+			 * Disarm the timer, nothing else to do.
 			 */
-			timer->it.cpu.expires = 0;
+			cpu_timer_setexpires(ctmr, 0);
 			return;
 		} else {
-			cpu_timer_sample_group(timer->it_clock, p, &now);
+			now = cpu_clock_sample_group(clkid, p, false);
 			unlock_task_sighand(p, &flags);
 		}
 	}
 
-	if (now < timer->it.cpu.expires) {
-		itp->it_value = ns_to_timespec64(timer->it.cpu.expires - now);
+	if (now < expires) {
+		itp->it_value = ns_to_timespec64(expires - now);
 	} else {
 		/*
 		 * The timer should have expired already, but the firing
@@ -769,26 +758,42 @@ static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec64 *itp
 	}
 }
 
-static unsigned long long
-check_timers_list(struct list_head *timers,
-		  struct list_head *firing,
-		  unsigned long long curr)
-{
-	int maxfire = 20;
+#define MAX_COLLECTED	20
 
-	while (!list_empty(timers)) {
-		struct cpu_timer_list *t;
+static u64 collect_timerqueue(struct timerqueue_head *head,
+			      struct list_head *firing, u64 now)
+{
+	struct timerqueue_node *next;
+	int i = 0;
+
+	while ((next = timerqueue_getnext(head))) {
+		struct cpu_timer *ctmr;
+		u64 expires;
+
+		ctmr = container_of(next, struct cpu_timer, node);
+		expires = cpu_timer_getexpires(ctmr);
+		/* Limit the number of timers to expire at once */
+		if (++i == MAX_COLLECTED || now < expires)
+			return expires;
+
+		ctmr->firing = 1;
+		cpu_timer_dequeue(ctmr);
+		list_add_tail(&ctmr->elist, firing);
+	}
 
-		t = list_first_entry(timers, struct cpu_timer_list, entry);
+	return U64_MAX;
+}
 
-		if (!--maxfire || curr < t->expires)
-			return t->expires;
+static void collect_posix_cputimers(struct posix_cputimers *pct, u64 *samples,
+				    struct list_head *firing)
+{
+	struct posix_cputimer_base *base = pct->bases;
+	int i;
 
-		t->firing = 1;
-		list_move_tail(&t->entry, firing);
+	for (i = 0; i < CPUCLOCK_MAX; i++, base++) {
+		base->nextevt = collect_timerqueue(&base->tqhead, firing,
+						    samples[i]);
 	}
-
-	return 0;
 }
 
 static inline void check_dl_overrun(struct task_struct *tsk)
@@ -799,6 +804,20 @@ static inline void check_dl_overrun(struct task_struct *tsk)
 	}
 }
 
+static bool check_rlimit(u64 time, u64 limit, int signo, bool rt, bool hard)
+{
+	if (time < limit)
+		return false;
+
+	if (print_fatal_signals) {
+		pr_info("%s Watchdog Timeout (%s): %s[%d]\n",
+			rt ? "RT" : "CPU", hard ? "hard" : "soft",
+			current->comm, task_pid_nr(current));
+	}
+	__group_send_sig_info(signo, SEND_SIG_PRIV, current);
+	return true;
+}
+
 /*
  * Check for any per-thread CPU timers that have fired and move them off
  * the tsk->cpu_timers[N] list onto the firing list.  Here we update the
@@ -807,76 +826,50 @@ static inline void check_dl_overrun(struct task_struct *tsk)
 static void check_thread_timers(struct task_struct *tsk,
 				struct list_head *firing)
 {
-	struct list_head *timers = tsk->cpu_timers;
-	struct task_cputime *tsk_expires = &tsk->cputime_expires;
-	u64 expires;
+	struct posix_cputimers *pct = &tsk->posix_cputimers;
+	u64 samples[CPUCLOCK_MAX];
 	unsigned long soft;
 
 	if (dl_task(tsk))
 		check_dl_overrun(tsk);
 
-	/*
-	 * If cputime_expires is zero, then there are no active
-	 * per thread CPU timers.
-	 */
-	if (task_cputime_zero(&tsk->cputime_expires))
+	if (expiry_cache_is_inactive(pct))
 		return;
 
-	expires = check_timers_list(timers, firing, prof_ticks(tsk));
-	tsk_expires->prof_exp = expires;
-
-	expires = check_timers_list(++timers, firing, virt_ticks(tsk));
-	tsk_expires->virt_exp = expires;
-
-	tsk_expires->sched_exp = check_timers_list(++timers, firing,
-						   tsk->se.sum_exec_runtime);
+	task_sample_cputime(tsk, samples);
+	collect_posix_cputimers(pct, samples, firing);
 
 	/*
 	 * Check for the special case thread timers.
 	 */
 	soft = task_rlimit(tsk, RLIMIT_RTTIME);
 	if (soft != RLIM_INFINITY) {
+		/* Task RT timeout is accounted in jiffies. RTTIME is usec */
+		unsigned long rttime = tsk->rt.timeout * (USEC_PER_SEC / HZ);
 		unsigned long hard = task_rlimit_max(tsk, RLIMIT_RTTIME);
 
+		/* At the hard limit, send SIGKILL. No further action. */
 		if (hard != RLIM_INFINITY &&
-		    tsk->rt.timeout > DIV_ROUND_UP(hard, USEC_PER_SEC/HZ)) {
-			/*
-			 * At the hard limit, we just die.
-			 * No need to calculate anything else now.
-			 */
-			if (print_fatal_signals) {
-				pr_info("CPU Watchdog Timeout (hard): %s[%d]\n",
-					tsk->comm, task_pid_nr(tsk));
-			}
-			__group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk);
+		    check_rlimit(rttime, hard, SIGKILL, true, true))
 			return;
-		}
-		if (tsk->rt.timeout > DIV_ROUND_UP(soft, USEC_PER_SEC/HZ)) {
-			/*
-			 * At the soft limit, send a SIGXCPU every second.
-			 */
-			if (soft < hard) {
-				soft += USEC_PER_SEC;
-				tsk->signal->rlim[RLIMIT_RTTIME].rlim_cur =
-					soft;
-			}
-			if (print_fatal_signals) {
-				pr_info("RT Watchdog Timeout (soft): %s[%d]\n",
-					tsk->comm, task_pid_nr(tsk));
-			}
-			__group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk);
+
+		/* At the soft limit, send a SIGXCPU every second */
+		if (check_rlimit(rttime, soft, SIGXCPU, true, false)) {
+			soft += USEC_PER_SEC;
+			tsk->signal->rlim[RLIMIT_RTTIME].rlim_cur = soft;
 		}
 	}
-	if (task_cputime_zero(tsk_expires))
+
+	if (expiry_cache_is_inactive(pct))
 		tick_dep_clear_task(tsk, TICK_DEP_BIT_POSIX_TIMER);
 }
 
 static inline void stop_process_timers(struct signal_struct *sig)
 {
-	struct thread_group_cputimer *cputimer = &sig->cputimer;
+	struct posix_cputimers *pct = &sig->posix_cputimers;
 
-	/* Turn off cputimer->running. This is done without locking. */
-	WRITE_ONCE(cputimer->running, false);
+	/* Turn off the active flag. This is done without locking. */
+	WRITE_ONCE(pct->timers_active, false);
 	tick_dep_clear_signal(sig, TICK_DEP_BIT_POSIX_TIMER);
 }
 
@@ -898,7 +891,7 @@ static void check_cpu_itimer(struct task_struct *tsk, struct cpu_itimer *it,
 		__group_send_sig_info(signo, SEND_SIG_PRIV, tsk);
 	}
 
-	if (it->expires && (!*expires || it->expires < *expires))
+	if (it->expires && it->expires < *expires)
 		*expires = it->expires;
 }
 
@@ -911,87 +904,69 @@ static void check_process_timers(struct task_struct *tsk,
 				 struct list_head *firing)
 {
 	struct signal_struct *const sig = tsk->signal;
-	u64 utime, ptime, virt_expires, prof_expires;
-	u64 sum_sched_runtime, sched_expires;
-	struct list_head *timers = sig->cpu_timers;
-	struct task_cputime cputime;
+	struct posix_cputimers *pct = &sig->posix_cputimers;
+	u64 samples[CPUCLOCK_MAX];
 	unsigned long soft;
 
 	/*
-	 * If cputimer is not running, then there are no active
-	 * process wide timers (POSIX 1.b, itimers, RLIMIT_CPU).
+	 * If there are no active process wide timers (POSIX 1.b, itimers,
+	 * RLIMIT_CPU) nothing to check. Also skip the process wide timer
+	 * processing when there is already another task handling them.
 	 */
-	if (!READ_ONCE(tsk->signal->cputimer.running))
+	if (!READ_ONCE(pct->timers_active) || pct->expiry_active)
 		return;
 
-        /*
+	/*
 	 * Signify that a thread is checking for process timers.
 	 * Write access to this field is protected by the sighand lock.
 	 */
-	sig->cputimer.checking_timer = true;
+	pct->expiry_active = true;
 
 	/*
-	 * Collect the current process totals.
+	 * Collect the current process totals. Group accounting is active
+	 * so the sample can be taken directly.
 	 */
-	thread_group_cputimer(tsk, &cputime);
-	utime = cputime.utime;
-	ptime = utime + cputime.stime;
-	sum_sched_runtime = cputime.sum_exec_runtime;
-
-	prof_expires = check_timers_list(timers, firing, ptime);
-	virt_expires = check_timers_list(++timers, firing, utime);
-	sched_expires = check_timers_list(++timers, firing, sum_sched_runtime);
+	proc_sample_cputime_atomic(&sig->cputimer.cputime_atomic, samples);
+	collect_posix_cputimers(pct, samples, firing);
 
 	/*
 	 * Check for the special case process timers.
 	 */
-	check_cpu_itimer(tsk, &sig->it[CPUCLOCK_PROF], &prof_expires, ptime,
-			 SIGPROF);
-	check_cpu_itimer(tsk, &sig->it[CPUCLOCK_VIRT], &virt_expires, utime,
-			 SIGVTALRM);
+	check_cpu_itimer(tsk, &sig->it[CPUCLOCK_PROF],
+			 &pct->bases[CPUCLOCK_PROF].nextevt,
+			 samples[CPUCLOCK_PROF], SIGPROF);
+	check_cpu_itimer(tsk, &sig->it[CPUCLOCK_VIRT],
+			 &pct->bases[CPUCLOCK_VIRT].nextevt,
+			 samples[CPUCLOCK_VIRT], SIGVTALRM);
+
 	soft = task_rlimit(tsk, RLIMIT_CPU);
 	if (soft != RLIM_INFINITY) {
-		unsigned long psecs = div_u64(ptime, NSEC_PER_SEC);
+		/* RLIMIT_CPU is in seconds. Samples are nanoseconds */
 		unsigned long hard = task_rlimit_max(tsk, RLIMIT_CPU);
-		u64 x;
-		if (psecs >= hard) {
-			/*
-			 * At the hard limit, we just die.
-			 * No need to calculate anything else now.
-			 */
-			if (print_fatal_signals) {
-				pr_info("RT Watchdog Timeout (hard): %s[%d]\n",
-					tsk->comm, task_pid_nr(tsk));
-			}
-			__group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk);
+		u64 ptime = samples[CPUCLOCK_PROF];
+		u64 softns = (u64)soft * NSEC_PER_SEC;
+		u64 hardns = (u64)hard * NSEC_PER_SEC;
+
+		/* At the hard limit, send SIGKILL. No further action. */
+		if (hard != RLIM_INFINITY &&
+		    check_rlimit(ptime, hardns, SIGKILL, false, true))
 			return;
+
+		/* At the soft limit, send a SIGXCPU every second */
+		if (check_rlimit(ptime, softns, SIGXCPU, false, false)) {
+			sig->rlim[RLIMIT_CPU].rlim_cur = soft + 1;
+			softns += NSEC_PER_SEC;
 		}
-		if (psecs >= soft) {
-			/*
-			 * At the soft limit, send a SIGXCPU every second.
-			 */
-			if (print_fatal_signals) {
-				pr_info("CPU Watchdog Timeout (soft): %s[%d]\n",
-					tsk->comm, task_pid_nr(tsk));
-			}
-			__group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk);
-			if (soft < hard) {
-				soft++;
-				sig->rlim[RLIMIT_CPU].rlim_cur = soft;
-			}
-		}
-		x = soft * NSEC_PER_SEC;
-		if (!prof_expires || x < prof_expires)
-			prof_expires = x;
+
+		/* Update the expiry cache */
+		if (softns < pct->bases[CPUCLOCK_PROF].nextevt)
+			pct->bases[CPUCLOCK_PROF].nextevt = softns;
 	}
 
-	sig->cputime_expires.prof_exp = prof_expires;
-	sig->cputime_expires.virt_exp = virt_expires;
-	sig->cputime_expires.sched_exp = sched_expires;
-	if (task_cputime_zero(&sig->cputime_expires))
+	if (expiry_cache_is_inactive(pct))
 		stop_process_timers(sig);
 
-	sig->cputimer.checking_timer = false;
+	pct->expiry_active = false;
 }
 
 /*
@@ -1000,18 +975,21 @@ static void check_process_timers(struct task_struct *tsk,
  */
 static void posix_cpu_timer_rearm(struct k_itimer *timer)
 {
+	clockid_t clkid = CPUCLOCK_WHICH(timer->it_clock);
+	struct cpu_timer *ctmr = &timer->it.cpu;
+	struct task_struct *p = ctmr->task;
 	struct sighand_struct *sighand;
 	unsigned long flags;
-	struct task_struct *p = timer->it.cpu.task;
 	u64 now;
 
-	WARN_ON_ONCE(p == NULL);
+	if (WARN_ON_ONCE(!p))
+		return;
 
 	/*
 	 * Fetch the current sample and update the timer's expiry time.
 	 */
 	if (CPUCLOCK_PERTHREAD(timer->it_clock)) {
-		cpu_clock_sample(timer->it_clock, p, &now);
+		now = cpu_clock_sample(clkid, p);
 		bump_cpu_timer(timer, now);
 		if (unlikely(p->exit_state))
 			return;
@@ -1031,13 +1009,13 @@ static void posix_cpu_timer_rearm(struct k_itimer *timer)
 			 * The process has been reaped.
 			 * We can't even collect a sample any more.
 			 */
-			timer->it.cpu.expires = 0;
+			cpu_timer_setexpires(ctmr, 0);
 			return;
 		} else if (unlikely(p->exit_state) && thread_group_empty(p)) {
 			/* If the process is dying, no need to rearm */
 			goto unlock;
 		}
-		cpu_timer_sample_group(timer->it_clock, p, &now);
+		now = cpu_clock_sample_group(clkid, p, true);
 		bump_cpu_timer(timer, now);
 		/* Leave the sighand locked for the call below.  */
 	}
@@ -1051,26 +1029,24 @@ unlock:
 }
 
 /**
- * task_cputime_expired - Compare two task_cputime entities.
+ * task_cputimers_expired - Check whether posix CPU timers are expired
  *
- * @sample:	The task_cputime structure to be checked for expiration.
- * @expires:	Expiration times, against which @sample will be checked.
+ * @samples:	Array of current samples for the CPUCLOCK clocks
+ * @pct:	Pointer to a posix_cputimers container
  *
- * Checks @sample against @expires to see if any field of @sample has expired.
- * Returns true if any field of the former is greater than the corresponding
- * field of the latter if the latter field is set.  Otherwise returns false.
+ * Returns true if any member of @samples is greater than the corresponding
+ * member of @pct->bases[CLK].nextevt. False otherwise
  */
-static inline int task_cputime_expired(const struct task_cputime *sample,
-					const struct task_cputime *expires)
+static inline bool
+task_cputimers_expired(const u64 *sample, struct posix_cputimers *pct)
 {
-	if (expires->utime && sample->utime >= expires->utime)
-		return 1;
-	if (expires->stime && sample->utime + sample->stime >= expires->stime)
-		return 1;
-	if (expires->sum_exec_runtime != 0 &&
-	    sample->sum_exec_runtime >= expires->sum_exec_runtime)
-		return 1;
-	return 0;
+	int i;
+
+	for (i = 0; i < CPUCLOCK_MAX; i++) {
+		if (sample[i] >= pct->bases[i].nextevt)
+			return true;
+	}
+	return false;
 }
 
 /**
@@ -1083,48 +1059,50 @@ static inline int task_cputime_expired(const struct task_cputime *sample,
  * timers and compare them with the corresponding expiration times.  Return
  * true if a timer has expired, else return false.
  */
-static inline int fastpath_timer_check(struct task_struct *tsk)
+static inline bool fastpath_timer_check(struct task_struct *tsk)
 {
+	struct posix_cputimers *pct = &tsk->posix_cputimers;
 	struct signal_struct *sig;
 
-	if (!task_cputime_zero(&tsk->cputime_expires)) {
-		struct task_cputime task_sample;
+	if (!expiry_cache_is_inactive(pct)) {
+		u64 samples[CPUCLOCK_MAX];
 
-		task_cputime(tsk, &task_sample.utime, &task_sample.stime);
-		task_sample.sum_exec_runtime = tsk->se.sum_exec_runtime;
-		if (task_cputime_expired(&task_sample, &tsk->cputime_expires))
-			return 1;
+		task_sample_cputime(tsk, samples);
+		if (task_cputimers_expired(samples, pct))
+			return true;
 	}
 
 	sig = tsk->signal;
+	pct = &sig->posix_cputimers;
 	/*
-	 * Check if thread group timers expired when the cputimer is
-	 * running and no other thread in the group is already checking
-	 * for thread group cputimers. These fields are read without the
-	 * sighand lock. However, this is fine because this is meant to
-	 * be a fastpath heuristic to determine whether we should try to
-	 * acquire the sighand lock to check/handle timers.
+	 * Check if thread group timers expired when timers are active and
+	 * no other thread in the group is already handling expiry for
+	 * thread group cputimers. These fields are read without the
+	 * sighand lock. However, this is fine because this is meant to be
+	 * a fastpath heuristic to determine whether we should try to
+	 * acquire the sighand lock to handle timer expiry.
 	 *
-	 * In the worst case scenario, if 'running' or 'checking_timer' gets
-	 * set but the current thread doesn't see the change yet, we'll wait
-	 * until the next thread in the group gets a scheduler interrupt to
-	 * handle the timer. This isn't an issue in practice because these
-	 * types of delays with signals actually getting sent are expected.
+	 * In the worst case scenario, if concurrently timers_active is set
+	 * or expiry_active is cleared, but the current thread doesn't see
+	 * the change yet, the timer checks are delayed until the next
+	 * thread in the group gets a scheduler interrupt to handle the
+	 * timer. This isn't an issue in practice because these types of
+	 * delays with signals actually getting sent are expected.
 	 */
-	if (READ_ONCE(sig->cputimer.running) &&
-	    !READ_ONCE(sig->cputimer.checking_timer)) {
-		struct task_cputime group_sample;
+	if (READ_ONCE(pct->timers_active) && !READ_ONCE(pct->expiry_active)) {
+		u64 samples[CPUCLOCK_MAX];
 
-		sample_cputime_atomic(&group_sample, &sig->cputimer.cputime_atomic);
+		proc_sample_cputime_atomic(&sig->cputimer.cputime_atomic,
+					   samples);
 
-		if (task_cputime_expired(&group_sample, &sig->cputime_expires))
-			return 1;
+		if (task_cputimers_expired(samples, pct))
+			return true;
 	}
 
 	if (dl_task(tsk) && tsk->dl.dl_overrun)
-		return 1;
+		return true;
 
-	return 0;
+	return false;
 }
 
 /*
@@ -1132,11 +1110,12 @@ static inline int fastpath_timer_check(struct task_struct *tsk)
  * already updated our counts.  We need to check if any timers fire now.
  * Interrupts are disabled.
  */
-void run_posix_cpu_timers(struct task_struct *tsk)
+void run_posix_cpu_timers(void)
 {
-	LIST_HEAD(firing);
+	struct task_struct *tsk = current;
 	struct k_itimer *timer, *next;
 	unsigned long flags;
+	LIST_HEAD(firing);
 
 	lockdep_assert_irqs_disabled();
 
@@ -1174,11 +1153,11 @@ void run_posix_cpu_timers(struct task_struct *tsk)
 	 * each timer's lock before clearing its firing flag, so no
 	 * timer call will interfere.
 	 */
-	list_for_each_entry_safe(timer, next, &firing, it.cpu.entry) {
+	list_for_each_entry_safe(timer, next, &firing, it.cpu.elist) {
 		int cpu_firing;
 
 		spin_lock(&timer->it_lock);
-		list_del_init(&timer->it.cpu.entry);
+		list_del_init(&timer->it.cpu.elist);
 		cpu_firing = timer->it.cpu.firing;
 		timer->it.cpu.firing = 0;
 		/*
@@ -1196,16 +1175,18 @@ void run_posix_cpu_timers(struct task_struct *tsk)
  * Set one of the process-wide special case CPU timers or RLIMIT_CPU.
  * The tsk->sighand->siglock must be held by the caller.
  */
-void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx,
+void set_process_cpu_timer(struct task_struct *tsk, unsigned int clkid,
 			   u64 *newval, u64 *oldval)
 {
-	u64 now;
-	int ret;
+	u64 now, *nextevt;
+
+	if (WARN_ON_ONCE(clkid >= CPUCLOCK_SCHED))
+		return;
 
-	WARN_ON_ONCE(clock_idx == CPUCLOCK_SCHED);
-	ret = cpu_timer_sample_group(clock_idx, tsk, &now);
+	nextevt = &tsk->signal->posix_cputimers.bases[clkid].nextevt;
+	now = cpu_clock_sample_group(clkid, tsk, true);
 
-	if (oldval && ret != -EINVAL) {
+	if (oldval) {
 		/*
 		 * We are setting itimer. The *oldval is absolute and we update
 		 * it to be relative, *newval argument is relative and we update
@@ -1226,19 +1207,11 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx,
 	}
 
 	/*
-	 * Update expiration cache if we are the earliest timer, or eventually
-	 * RLIMIT_CPU limit is earlier than prof_exp cpu timer expire.
+	 * Update expiration cache if this is the earliest timer. CPUCLOCK_PROF
+	 * expiry cache is also used by RLIMIT_CPU!.
 	 */
-	switch (clock_idx) {
-	case CPUCLOCK_PROF:
-		if (expires_gt(tsk->signal->cputime_expires.prof_exp, *newval))
-			tsk->signal->cputime_expires.prof_exp = *newval;
-		break;
-	case CPUCLOCK_VIRT:
-		if (expires_gt(tsk->signal->cputime_expires.virt_exp, *newval))
-			tsk->signal->cputime_expires.virt_exp = *newval;
-		break;
-	}
+	if (*newval < *nextevt)
+		*nextevt = *newval;
 
 	tick_dep_set_signal(tsk->signal, TICK_DEP_BIT_POSIX_TIMER);
 }
@@ -1260,6 +1233,7 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags,
 	timer.it_overrun = -1;
 	error = posix_cpu_timer_create(&timer);
 	timer.it_process = current;
+
 	if (!error) {
 		static struct itimerspec64 zero_it;
 		struct restart_block *restart;
@@ -1275,7 +1249,7 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags,
 		}
 
 		while (!signal_pending(current)) {
-			if (timer.it.cpu.expires == 0) {
+			if (!cpu_timer_getexpires(&timer.it.cpu)) {
 				/*
 				 * Our timer fired and was reset, below
 				 * deletion can not fail.
@@ -1297,7 +1271,7 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags,
 		/*
 		 * We were interrupted by a signal.
 		 */
-		expires = timer.it.cpu.expires;
+		expires = cpu_timer_getexpires(&timer.it.cpu);
 		error = posix_cpu_timer_set(&timer, 0, &zero_it, &it);
 		if (!error) {
 			/*
diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c
index d7f2d91acdac..0ec5b7a1d769 100644
--- a/kernel/time/posix-timers.c
+++ b/kernel/time/posix-timers.c
@@ -442,7 +442,7 @@ static struct k_itimer * alloc_posix_timer(void)
 
 static void k_itimer_rcu_free(struct rcu_head *head)
 {
-	struct k_itimer *tmr = container_of(head, struct k_itimer, it.rcu);
+	struct k_itimer *tmr = container_of(head, struct k_itimer, rcu);
 
 	kmem_cache_free(posix_timers_cache, tmr);
 }
@@ -459,7 +459,7 @@ static void release_posix_timer(struct k_itimer *tmr, int it_id_set)
 	}
 	put_pid(tmr->it_pid);
 	sigqueue_free(tmr->sigq);
-	call_rcu(&tmr->it.rcu, k_itimer_rcu_free);
+	call_rcu(&tmr->rcu, k_itimer_rcu_free);
 }
 
 static int common_timer_create(struct k_itimer *new_timer)
@@ -805,6 +805,35 @@ static int common_hrtimer_try_to_cancel(struct k_itimer *timr)
 	return hrtimer_try_to_cancel(&timr->it.real.timer);
 }
 
+static void common_timer_wait_running(struct k_itimer *timer)
+{
+	hrtimer_cancel_wait_running(&timer->it.real.timer);
+}
+
+/*
+ * On PREEMPT_RT this prevent priority inversion against softirq kthread in
+ * case it gets preempted while executing a timer callback. See comments in
+ * hrtimer_cancel_wait_running. For PREEMPT_RT=n this just results in a
+ * cpu_relax().
+ */
+static struct k_itimer *timer_wait_running(struct k_itimer *timer,
+					   unsigned long *flags)
+{
+	const struct k_clock *kc = READ_ONCE(timer->kclock);
+	timer_t timer_id = READ_ONCE(timer->it_id);
+
+	/* Prevent kfree(timer) after dropping the lock */
+	rcu_read_lock();
+	unlock_timer(timer, *flags);
+
+	if (!WARN_ON_ONCE(!kc->timer_wait_running))
+		kc->timer_wait_running(timer);
+
+	rcu_read_unlock();
+	/* Relock the timer. It might be not longer hashed. */
+	return lock_timer(timer_id, flags);
+}
+
 /* Set a POSIX.1b interval timer. */
 int common_timer_set(struct k_itimer *timr, int flags,
 		     struct itimerspec64 *new_setting,
@@ -844,13 +873,13 @@ int common_timer_set(struct k_itimer *timr, int flags,
 	return 0;
 }
 
-static int do_timer_settime(timer_t timer_id, int flags,
+static int do_timer_settime(timer_t timer_id, int tmr_flags,
 			    struct itimerspec64 *new_spec64,
 			    struct itimerspec64 *old_spec64)
 {
 	const struct k_clock *kc;
 	struct k_itimer *timr;
-	unsigned long flag;
+	unsigned long flags;
 	int error = 0;
 
 	if (!timespec64_valid(&new_spec64->it_interval) ||
@@ -859,8 +888,9 @@ static int do_timer_settime(timer_t timer_id, int flags,
 
 	if (old_spec64)
 		memset(old_spec64, 0, sizeof(*old_spec64));
+
+	timr = lock_timer(timer_id, &flags);
 retry:
-	timr = lock_timer(timer_id, &flag);
 	if (!timr)
 		return -EINVAL;
 
@@ -868,13 +898,16 @@ retry:
 	if (WARN_ON_ONCE(!kc || !kc->timer_set))
 		error = -EINVAL;
 	else
-		error = kc->timer_set(timr, flags, new_spec64, old_spec64);
+		error = kc->timer_set(timr, tmr_flags, new_spec64, old_spec64);
 
-	unlock_timer(timr, flag);
 	if (error == TIMER_RETRY) {
-		old_spec64 = NULL;	// We already got the old time...
+		// We already got the old time...
+		old_spec64 = NULL;
+		/* Unlocks and relocks the timer if it still exists */
+		timr = timer_wait_running(timr, &flags);
 		goto retry;
 	}
+	unlock_timer(timr, flags);
 
 	return error;
 }
@@ -951,13 +984,15 @@ SYSCALL_DEFINE1(timer_delete, timer_t, timer_id)
 	struct k_itimer *timer;
 	unsigned long flags;
 
-retry_delete:
 	timer = lock_timer(timer_id, &flags);
+
+retry_delete:
 	if (!timer)
 		return -EINVAL;
 
-	if (timer_delete_hook(timer) == TIMER_RETRY) {
-		unlock_timer(timer, flags);
+	if (unlikely(timer_delete_hook(timer) == TIMER_RETRY)) {
+		/* Unlocks and relocks the timer if it still exists */
+		timer = timer_wait_running(timer, &flags);
 		goto retry_delete;
 	}
 
@@ -1238,6 +1273,7 @@ static const struct k_clock clock_realtime = {
 	.timer_forward		= common_hrtimer_forward,
 	.timer_remaining	= common_hrtimer_remaining,
 	.timer_try_to_cancel	= common_hrtimer_try_to_cancel,
+	.timer_wait_running	= common_timer_wait_running,
 	.timer_arm		= common_hrtimer_arm,
 };
 
@@ -1253,6 +1289,7 @@ static const struct k_clock clock_monotonic = {
 	.timer_forward		= common_hrtimer_forward,
 	.timer_remaining	= common_hrtimer_remaining,
 	.timer_try_to_cancel	= common_hrtimer_try_to_cancel,
+	.timer_wait_running	= common_timer_wait_running,
 	.timer_arm		= common_hrtimer_arm,
 };
 
@@ -1283,6 +1320,7 @@ static const struct k_clock clock_tai = {
 	.timer_forward		= common_hrtimer_forward,
 	.timer_remaining	= common_hrtimer_remaining,
 	.timer_try_to_cancel	= common_hrtimer_try_to_cancel,
+	.timer_wait_running	= common_timer_wait_running,
 	.timer_arm		= common_hrtimer_arm,
 };
 
@@ -1298,6 +1336,7 @@ static const struct k_clock clock_boottime = {
 	.timer_forward		= common_hrtimer_forward,
 	.timer_remaining	= common_hrtimer_remaining,
 	.timer_try_to_cancel	= common_hrtimer_try_to_cancel,
+	.timer_wait_running	= common_timer_wait_running,
 	.timer_arm		= common_hrtimer_arm,
 };
 
diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h
index de5daa6d975a..897c29e162b9 100644
--- a/kernel/time/posix-timers.h
+++ b/kernel/time/posix-timers.h
@@ -24,6 +24,7 @@ struct k_clock {
 	int	(*timer_try_to_cancel)(struct k_itimer *timr);
 	void	(*timer_arm)(struct k_itimer *timr, ktime_t expires,
 			     bool absolute, bool sigev_none);
+	void	(*timer_wait_running)(struct k_itimer *timr);
 };
 
 extern const struct k_clock clock_posix_cpu;
diff --git a/kernel/time/tick-broadcast-hrtimer.c b/kernel/time/tick-broadcast-hrtimer.c
index 5be6154e2fd2..c1f5bb590b5e 100644
--- a/kernel/time/tick-broadcast-hrtimer.c
+++ b/kernel/time/tick-broadcast-hrtimer.c
@@ -59,11 +59,16 @@ static int bc_set_next(ktime_t expires, struct clock_event_device *bc)
 	 * hrtimer_{start/cancel} functions call into tracing,
 	 * calls to these functions must be bound within RCU_NONIDLE.
 	 */
-	RCU_NONIDLE({
+	RCU_NONIDLE(
+		{
 			bc_moved = hrtimer_try_to_cancel(&bctimer) >= 0;
-			if (bc_moved)
+			if (bc_moved) {
 				hrtimer_start(&bctimer, expires,
-					      HRTIMER_MODE_ABS_PINNED);});
+					      HRTIMER_MODE_ABS_PINNED_HARD);
+			}
+		}
+	);
+
 	if (bc_moved) {
 		/* Bind the "device" to the cpu */
 		bc->bound_on = smp_processor_id();
@@ -104,7 +109,7 @@ static enum hrtimer_restart bc_handler(struct hrtimer *t)
 
 void tick_setup_hrtimer_broadcast(void)
 {
-	hrtimer_init(&bctimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+	hrtimer_init(&bctimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_HARD);
 	bctimer.function = bc_handler;
 	clockevents_register_device(&ce_broadcast_hrtimer);
 }
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index be9707f68024..955851748dc3 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -634,10 +634,12 @@ static void tick_nohz_restart(struct tick_sched *ts, ktime_t now)
 	/* Forward the time to expire in the future */
 	hrtimer_forward(&ts->sched_timer, now, tick_period);
 
-	if (ts->nohz_mode == NOHZ_MODE_HIGHRES)
-		hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED);
-	else
+	if (ts->nohz_mode == NOHZ_MODE_HIGHRES) {
+		hrtimer_start_expires(&ts->sched_timer,
+				      HRTIMER_MODE_ABS_PINNED_HARD);
+	} else {
 		tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1);
+	}
 
 	/*
 	 * Reset to make sure next tick stop doesn't get fooled by past
@@ -802,7 +804,8 @@ static void tick_nohz_stop_tick(struct tick_sched *ts, int cpu)
 	}
 
 	if (ts->nohz_mode == NOHZ_MODE_HIGHRES) {
-		hrtimer_start(&ts->sched_timer, tick, HRTIMER_MODE_ABS_PINNED);
+		hrtimer_start(&ts->sched_timer, tick,
+			      HRTIMER_MODE_ABS_PINNED_HARD);
 	} else {
 		hrtimer_set_expires(&ts->sched_timer, tick);
 		tick_program_event(tick, 1);
@@ -1230,7 +1233,7 @@ static void tick_nohz_switch_to_nohz(void)
 	 * Recycle the hrtimer in ts, so we can share the
 	 * hrtimer_forward with the highres code.
 	 */
-	hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+	hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_HARD);
 	/* Get the next period */
 	next = tick_init_jiffy_update();
 
@@ -1327,7 +1330,7 @@ void tick_setup_sched_timer(void)
 	/*
 	 * Emulate tick processing via per-CPU hrtimers:
 	 */
-	hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+	hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_HARD);
 	ts->sched_timer.function = tick_sched_timer;
 
 	/* Get the next period (per-CPU) */
@@ -1342,7 +1345,7 @@ void tick_setup_sched_timer(void)
 	}
 
 	hrtimer_forward(&ts->sched_timer, now, tick_period);
-	hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED);
+	hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED_HARD);
 	tick_nohz_activate(ts, NOHZ_MODE_HIGHRES);
 }
 #endif /* HIGH_RES_TIMERS */
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 343c7ba33b1c..0e315a2e77ae 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -196,6 +196,10 @@ EXPORT_SYMBOL(jiffies_64);
 struct timer_base {
 	raw_spinlock_t		lock;
 	struct timer_list	*running_timer;
+#ifdef CONFIG_PREEMPT_RT
+	spinlock_t		expiry_lock;
+	atomic_t		timer_waiters;
+#endif
 	unsigned long		clk;
 	unsigned long		next_expiry;
 	unsigned int		cpu;
@@ -1227,7 +1231,78 @@ int try_to_del_timer_sync(struct timer_list *timer)
 }
 EXPORT_SYMBOL(try_to_del_timer_sync);
 
-#ifdef CONFIG_SMP
+#ifdef CONFIG_PREEMPT_RT
+static __init void timer_base_init_expiry_lock(struct timer_base *base)
+{
+	spin_lock_init(&base->expiry_lock);
+}
+
+static inline void timer_base_lock_expiry(struct timer_base *base)
+{
+	spin_lock(&base->expiry_lock);
+}
+
+static inline void timer_base_unlock_expiry(struct timer_base *base)
+{
+	spin_unlock(&base->expiry_lock);
+}
+
+/*
+ * The counterpart to del_timer_wait_running().
+ *
+ * If there is a waiter for base->expiry_lock, then it was waiting for the
+ * timer callback to finish. Drop expiry_lock and reaquire it. That allows
+ * the waiter to acquire the lock and make progress.
+ */
+static void timer_sync_wait_running(struct timer_base *base)
+{
+	if (atomic_read(&base->timer_waiters)) {
+		spin_unlock(&base->expiry_lock);
+		spin_lock(&base->expiry_lock);
+	}
+}
+
+/*
+ * This function is called on PREEMPT_RT kernels when the fast path
+ * deletion of a timer failed because the timer callback function was
+ * running.
+ *
+ * This prevents priority inversion, if the softirq thread on a remote CPU
+ * got preempted, and it prevents a life lock when the task which tries to
+ * delete a timer preempted the softirq thread running the timer callback
+ * function.
+ */
+static void del_timer_wait_running(struct timer_list *timer)
+{
+	u32 tf;
+
+	tf = READ_ONCE(timer->flags);
+	if (!(tf & TIMER_MIGRATING)) {
+		struct timer_base *base = get_timer_base(tf);
+
+		/*
+		 * Mark the base as contended and grab the expiry lock,
+		 * which is held by the softirq across the timer
+		 * callback. Drop the lock immediately so the softirq can
+		 * expire the next timer. In theory the timer could already
+		 * be running again, but that's more than unlikely and just
+		 * causes another wait loop.
+		 */
+		atomic_inc(&base->timer_waiters);
+		spin_lock_bh(&base->expiry_lock);
+		atomic_dec(&base->timer_waiters);
+		spin_unlock_bh(&base->expiry_lock);
+	}
+}
+#else
+static inline void timer_base_init_expiry_lock(struct timer_base *base) { }
+static inline void timer_base_lock_expiry(struct timer_base *base) { }
+static inline void timer_base_unlock_expiry(struct timer_base *base) { }
+static inline void timer_sync_wait_running(struct timer_base *base) { }
+static inline void del_timer_wait_running(struct timer_list *timer) { }
+#endif
+
+#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT)
 /**
  * del_timer_sync - deactivate a timer and wait for the handler to finish.
  * @timer: the timer to be deactivated
@@ -1266,6 +1341,8 @@ EXPORT_SYMBOL(try_to_del_timer_sync);
  */
 int del_timer_sync(struct timer_list *timer)
 {
+	int ret;
+
 #ifdef CONFIG_LOCKDEP
 	unsigned long flags;
 
@@ -1283,12 +1360,17 @@ int del_timer_sync(struct timer_list *timer)
 	 * could lead to deadlock.
 	 */
 	WARN_ON(in_irq() && !(timer->flags & TIMER_IRQSAFE));
-	for (;;) {
-		int ret = try_to_del_timer_sync(timer);
-		if (ret >= 0)
-			return ret;
-		cpu_relax();
-	}
+
+	do {
+		ret = try_to_del_timer_sync(timer);
+
+		if (unlikely(ret < 0)) {
+			del_timer_wait_running(timer);
+			cpu_relax();
+		}
+	} while (ret < 0);
+
+	return ret;
 }
 EXPORT_SYMBOL(del_timer_sync);
 #endif
@@ -1360,10 +1442,13 @@ static void expire_timers(struct timer_base *base, struct hlist_head *head)
 		if (timer->flags & TIMER_IRQSAFE) {
 			raw_spin_unlock(&base->lock);
 			call_timer_fn(timer, fn, baseclk);
+			base->running_timer = NULL;
 			raw_spin_lock(&base->lock);
 		} else {
 			raw_spin_unlock_irq(&base->lock);
 			call_timer_fn(timer, fn, baseclk);
+			base->running_timer = NULL;
+			timer_sync_wait_running(base);
 			raw_spin_lock_irq(&base->lock);
 		}
 	}
@@ -1643,7 +1728,7 @@ void update_process_times(int user_tick)
 #endif
 	scheduler_tick();
 	if (IS_ENABLED(CONFIG_POSIX_TIMERS))
-		run_posix_cpu_timers(p);
+		run_posix_cpu_timers();
 }
 
 /**
@@ -1658,6 +1743,7 @@ static inline void __run_timers(struct timer_base *base)
 	if (!time_after_eq(jiffies, base->clk))
 		return;
 
+	timer_base_lock_expiry(base);
 	raw_spin_lock_irq(&base->lock);
 
 	/*
@@ -1684,8 +1770,8 @@ static inline void __run_timers(struct timer_base *base)
 		while (levels--)
 			expire_timers(base, heads + levels);
 	}
-	base->running_timer = NULL;
 	raw_spin_unlock_irq(&base->lock);
+	timer_base_unlock_expiry(base);
 }
 
 /*
@@ -1930,6 +2016,7 @@ static void __init init_timer_cpu(int cpu)
 		base->cpu = cpu;
 		raw_spin_lock_init(&base->lock);
 		base->clk = jiffies;
+		timer_base_init_expiry_lock(base);
 	}
 }