Merge branch 'timers/core' of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks into timers/core

Pull tick/NOHZ updates from Frederic Weisbecker:

 - A fix for rare jiffies update stalls that were reported by Paul McKenney

 - Tick side cleanups after RCU_FAST_NO_HZ removal

 - Handle softirqs on idle more gracefully

Link: https://lore.kernel.org/all/20220307233034.34550-1-frederic@kernel.org

Author: KAGA-KOKO

include/linux/interrupt.h

@@ -579,7 +579,16 @@ enum
 	NR_SOFTIRQS
 };
 
-#define SOFTIRQ_STOP_IDLE_MASK (~(1 << RCU_SOFTIRQ))
+/*
+ * The following vectors can be safely ignored after ksoftirqd is parked:
+ *
+ * _ RCU:
+ * 	1) rcutree_migrate_callbacks() migrates the queue.
+ * 	2) rcu_report_dead() reports the final quiescent states.
+ *
+ * _ IRQ_POLL: irq_poll_cpu_dead() migrates the queue
+ */
+#define SOFTIRQ_HOTPLUG_SAFE_MASK (BIT(RCU_SOFTIRQ) | BIT(IRQ_POLL_SOFTIRQ))
 
 /* map softirq index to softirq name. update 'softirq_to_name' in
  * kernel/softirq.c when adding a new softirq.

include/linux/rcutiny.h

@@ -64,9 +64,8 @@ static inline void rcu_softirq_qs(void)
 		rcu_tasks_qs(current, (preempt)); \
 	} while (0)
 
-static inline int rcu_needs_cpu(u64 basemono, u64 *nextevt)
+static inline int rcu_needs_cpu(void)
 {
-	*nextevt = KTIME_MAX;
 	return 0;
 }
 

include/linux/rcutree.h

@@ -19,7 +19,7 @@
 
 void rcu_softirq_qs(void);
 void rcu_note_context_switch(bool preempt);
-int rcu_needs_cpu(u64 basem, u64 *nextevt);
+int rcu_needs_cpu(void);
 void rcu_cpu_stall_reset(void);
 
 /*

kernel/rcu/tree.c

@@ -1086,9 +1086,8 @@ void rcu_irq_enter_irqson(void)
  * Just check whether or not this CPU has non-offloaded RCU callbacks
  * queued.
  */
-int rcu_needs_cpu(u64 basemono, u64 *nextevt)
+int rcu_needs_cpu(void)
 {
-	*nextevt = KTIME_MAX;
 	return !rcu_segcblist_empty(&this_cpu_ptr(&rcu_data)->cblist) &&
 		!rcu_rdp_is_offloaded(this_cpu_ptr(&rcu_data));
 }

kernel/time/tick-sched.c

@@ -169,6 +169,8 @@ static ktime_t tick_init_jiffy_update(void)
 	return period;
 }
 
+#define MAX_STALLED_JIFFIES 5
+
 static void tick_sched_do_timer(struct tick_sched *ts, ktime_t now)
 {
 	int cpu = smp_processor_id();
@@ -196,6 +198,21 @@ static void tick_sched_do_timer(struct tick_sched *ts, ktime_t now)
 	if (tick_do_timer_cpu == cpu)
 		tick_do_update_jiffies64(now);
 
+	/*
+	 * If jiffies update stalled for too long (timekeeper in stop_machine()
+	 * or VMEXIT'ed for several msecs), force an update.
+	 */
+	if (ts->last_tick_jiffies != jiffies) {
+		ts->stalled_jiffies = 0;
+		ts->last_tick_jiffies = READ_ONCE(jiffies);
+	} else {
+		if (++ts->stalled_jiffies == MAX_STALLED_JIFFIES) {
+			tick_do_update_jiffies64(now);
+			ts->stalled_jiffies = 0;
+			ts->last_tick_jiffies = READ_ONCE(jiffies);
+		}
+	}
+
 	if (ts->inidle)
 		ts->got_idle_tick = 1;
 }
@@ -768,7 +785,7 @@ static inline bool local_timer_softirq_pending(void)
 
 static ktime_t tick_nohz_next_event(struct tick_sched *ts, int cpu)
 {
-	u64 basemono, next_tick, next_tmr, next_rcu, delta, expires;
+	u64 basemono, next_tick, delta, expires;
 	unsigned long basejiff;
 	unsigned int seq;
 
@@ -791,7 +808,7 @@ static ktime_t tick_nohz_next_event(struct tick_sched *ts, int cpu)
 	 * minimal delta which brings us back to this place
 	 * immediately. Lather, rinse and repeat...
 	 */
-	if (rcu_needs_cpu(basemono, &next_rcu) || arch_needs_cpu() ||
+	if (rcu_needs_cpu() || arch_needs_cpu() ||
 	    irq_work_needs_cpu() || local_timer_softirq_pending()) {
 		next_tick = basemono + TICK_NSEC;
 	} else {
@@ -802,10 +819,8 @@ static ktime_t tick_nohz_next_event(struct tick_sched *ts, int cpu)
 		 * disabled this also looks at the next expiring
 		 * hrtimer.
 		 */
-		next_tmr = get_next_timer_interrupt(basejiff, basemono);
-		ts->next_timer = next_tmr;
-		/* Take the next rcu event into account */
-		next_tick = next_rcu < next_tmr ? next_rcu : next_tmr;
+		next_tick = get_next_timer_interrupt(basejiff, basemono);
+		ts->next_timer = next_tick;
 	}
 
 	/*
@@ -984,6 +999,45 @@ static void tick_nohz_full_update_tick(struct tick_sched *ts)
 	__tick_nohz_full_update_tick(ts, ktime_get());
 }
 
+/*
+ * A pending softirq outside an IRQ (or softirq disabled section) context
+ * should be waiting for ksoftirqd to handle it. Therefore we shouldn't
+ * reach here due to the need_resched() early check in can_stop_idle_tick().
+ *
+ * However if we are between CPUHP_AP_SMPBOOT_THREADS and CPU_TEARDOWN_CPU on the
+ * cpu_down() process, softirqs can still be raised while ksoftirqd is parked,
+ * triggering the below since wakep_softirqd() is ignored.
+ *
+ */
+static bool report_idle_softirq(void)
+{
+	static int ratelimit;
+	unsigned int pending = local_softirq_pending();
+
+	if (likely(!pending))
+		return false;
+
+	/* Some softirqs claim to be safe against hotplug and ksoftirqd parking */
+	if (!cpu_active(smp_processor_id())) {
+		pending &= ~SOFTIRQ_HOTPLUG_SAFE_MASK;
+		if (!pending)
+			return false;
+	}
+
+	if (ratelimit < 10)
+		return false;
+
+	/* On RT, softirqs handling may be waiting on some lock */
+	if (!local_bh_blocked())
+		return false;
+
+	pr_warn("NOHZ tick-stop error: local softirq work is pending, handler #%02x!!!\n",
+		pending);
+	ratelimit++;
+
+	return true;
+}
+
 static bool can_stop_idle_tick(int cpu, struct tick_sched *ts)
 {
 	/*
@@ -1010,17 +1064,8 @@ static bool can_stop_idle_tick(int cpu, struct tick_sched *ts)
 	if (need_resched())
 		return false;
 
-	if (unlikely(local_softirq_pending())) {
-		static int ratelimit;
-
-		if (ratelimit < 10 && !local_bh_blocked() &&
-		    (local_softirq_pending() & SOFTIRQ_STOP_IDLE_MASK)) {
-			pr_warn("NOHZ tick-stop error: Non-RCU local softirq work is pending, handler #%02x!!!\n",
-				(unsigned int) local_softirq_pending());
-			ratelimit++;
-		}
+	if (unlikely(report_idle_softirq()))
 		return false;
-	}
 
 	if (tick_nohz_full_enabled()) {
 		/*

kernel/time/tick-sched.h

@@ -49,6 +49,8 @@ enum tick_nohz_mode {
  * @timer_expires_base:	Base time clock monotonic for @timer_expires
  * @next_timer:		Expiry time of next expiring timer for debugging purpose only
  * @tick_dep_mask:	Tick dependency mask - is set, if someone needs the tick
+ * @last_tick_jiffies:	Value of jiffies seen on last tick
+ * @stalled_jiffies:	Number of stalled jiffies detected across ticks
  */
 struct tick_sched {
 	struct hrtimer			sched_timer;
@@ -77,6 +79,8 @@ struct tick_sched {
 	u64				next_timer;
 	ktime_t				idle_expires;
 	atomic_t			tick_dep_mask;
+	unsigned long			last_tick_jiffies;
+	unsigned int			stalled_jiffies;
 };
 
 extern struct tick_sched *tick_get_tick_sched(int cpu);