timers: Fix excessive granularity of new timers after a nohz idle

When a timer base is idle, it is forwarded when a new timer is added
to ensure that granularity does not become excessive. When not idle,
the timer tick is expected to increment the base.

However there are several problems:

- If an existing timer is modified, the base is forwarded only after
  the index is calculated.

- The base is not forwarded by add_timer_on.

- There is a window after a timer is restarted from a nohz idle, after
  it is marked not-idle and before the timer tick on this CPU, where a
  timer may be added but the ancient base does not get forwarded.

These result in excessive granularity (a 1 jiffy timeout can blow out
to 100s of jiffies), which cause the rcu lockup detector to trigger,
among other things.

Fix this by keeping track of whether the timer base has been idle
since it was last run or forwarded, and if so then forward it before
adding a new timer.

There is still a case where mod_timer optimises the case of a pending
timer mod with the same expiry time, where the timer can see excessive
granularity relative to the new, shorter interval. A comment is added,
but it's not changed because it is an important fastpath for
networking.

This has been tested and found to fix the RCU softlockup messages.

Testing was also done with tracing to measure requested versus
achieved wakeup latencies for all non-deferrable timers in an idle
system (with no lockup watchdogs running). Wakeup latency relative to
absolute latency is calculated (note this suffers from round-up skew
at low absolute times) and analysed:

             max     avg      std
upstream   506.0    1.20     4.68
patched      2.0    1.08     0.15

The bug was noticed due to the lockup detector Kconfig changes
dropping it out of people's .configs and resulting in larger base
clk skew When the lockup detectors are enabled, no CPU can go idle for
longer than 4 seconds, which limits the granularity errors.
Sub-optimal timer behaviour is observable on a smaller scale in that
case:

	     max     avg      std
upstream     9.0    1.05     0.19
patched      2.0    1.04     0.11

Fixes: Fixes: a683f390 ("timers: Forward the wheel clock whenever possible")
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Tested-by: David Miller <davem@davemloft.net>
Cc: dzickus@redhat.com
Cc: sfr@canb.auug.org.au
Cc: mpe@ellerman.id.au
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: linuxarm@huawei.com
Cc: abdhalee@linux.vnet.ibm.com
Cc: John Stultz <john.stultz@linaro.org>
Cc: akpm@linux-foundation.org
Cc: paulmck@linux.vnet.ibm.com
Cc: torvalds@linux-foundation.org
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/20170822084348.21436-1-npiggin@gmail.com

kernel/time/timer.c

@@ -203,6 +203,7 @@ struct timer_base {
 	bool			migration_enabled;
 	bool			nohz_active;
 	bool			is_idle;
+	bool			must_forward_clk;
 	DECLARE_BITMAP(pending_map, WHEEL_SIZE);
 	struct hlist_head	vectors[WHEEL_SIZE];
 } ____cacheline_aligned;
@@ -856,13 +857,19 @@ get_target_base(struct timer_base *base, unsigned tflags)
 
 static inline void forward_timer_base(struct timer_base *base)
 {
-	unsigned long jnow = READ_ONCE(jiffies);
+	unsigned long jnow;
 
 	/*
-	 * We only forward the base when it's idle and we have a delta between
-	 * base clock and jiffies.
+	 * We only forward the base when we are idle or have just come out of
+	 * idle (must_forward_clk logic), and have a delta between base clock
+	 * and jiffies. In the common case, run_timers will take care of it.
 	 */
-	if (!base->is_idle || (long) (jnow - base->clk) < 2)
+	if (likely(!base->must_forward_clk))
+		return;
+
+	jnow = READ_ONCE(jiffies);
+	base->must_forward_clk = base->is_idle;
+	if ((long)(jnow - base->clk) < 2)
 		return;
 
 	/*
@@ -938,6 +945,11 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 	 * same array bucket then just return:
 	 */
 	if (timer_pending(timer)) {
+		/*
+		 * The downside of this optimization is that it can result in
+		 * larger granularity than you would get from adding a new
+		 * timer with this expiry.
+		 */
 		if (timer->expires == expires)
 			return 1;
 
@@ -948,6 +960,7 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 		 * dequeue/enqueue dance.
 		 */
 		base = lock_timer_base(timer, &flags);
+		forward_timer_base(base);
 
 		clk = base->clk;
 		idx = calc_wheel_index(expires, clk);
@@ -964,6 +977,7 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 		}
 	} else {
 		base = lock_timer_base(timer, &flags);
+		forward_timer_base(base);
 	}
 
 	ret = detach_if_pending(timer, base, false);
@@ -991,12 +1005,10 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 			raw_spin_lock(&base->lock);
 			WRITE_ONCE(timer->flags,
 				   (timer->flags & ~TIMER_BASEMASK) | base->cpu);
+			forward_timer_base(base);
 		}
 	}
 
-	/* Try to forward a stale timer base clock */
-	forward_timer_base(base);
-
 	timer->expires = expires;
 	/*
 	 * If 'idx' was calculated above and the base time did not advance
@@ -1112,6 +1124,7 @@ void add_timer_on(struct timer_list *timer, int cpu)
 		WRITE_ONCE(timer->flags,
 			   (timer->flags & ~TIMER_BASEMASK) | cpu);
 	}
+	forward_timer_base(base);
 
 	debug_activate(timer, timer->expires);
 	internal_add_timer(base, timer);
@@ -1497,11 +1510,17 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
 		if (!is_max_delta)
 			expires = basem + (u64)(nextevt - basej) * TICK_NSEC;
 		/*
-		 * If we expect to sleep more than a tick, mark the base idle:
-		 */
-		if ((expires - basem) > TICK_NSEC)
+		 * If we expect to sleep more than a tick, mark the base idle.
+		 * Also the tick is stopped so any added timer must forward
+		 * the base clk itself to keep granularity small. This idle
+		 * logic is only maintained for the BASE_STD base, deferrable
+		 * timers may still see large granularity skew (by design).
+		 */
+		if ((expires - basem) > TICK_NSEC) {
+			base->must_forward_clk = true;
 			base->is_idle = true;
 		}
+	}
 	raw_spin_unlock(&base->lock);
 
 	return cmp_next_hrtimer_event(basem, expires);
@@ -1611,6 +1630,19 @@ static __latent_entropy void run_timer_softirq(struct softirq_action *h)
 {
 	struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
 
+	/*
+	 * must_forward_clk must be cleared before running timers so that any
+	 * timer functions that call mod_timer will not try to forward the
+	 * base. idle trcking / clock forwarding logic is only used with
+	 * BASE_STD timers.
+	 *
+	 * The deferrable base does not do idle tracking at all, so we do
+	 * not forward it. This can result in very large variations in
+	 * granularity for deferrable timers, but they can be deferred for
+	 * long periods due to idle.
+	 */
+	base->must_forward_clk = false;
+
 	__run_timers(base);
 	if (IS_ENABLED(CONFIG_NO_HZ_COMMON) && base->nohz_active)
 		__run_timers(this_cpu_ptr(&timer_bases[BASE_DEF]));