locking/lock_events: Use this_cpu_add() when necessary

The kernel test robot has reported that the use of __this_cpu_add()
causes bug messages like:

  BUG: using __this_cpu_add() in preemptible [00000000] code: ...

Given the imprecise nature of the count and the possibility of resetting
the count and doing the measurement again, this is not really a big
problem to use the unprotected __this_cpu_*() functions.

To make the preemption checking code happy, the this_cpu_*() functions
will be used if CONFIG_DEBUG_PREEMPT is defined.

The imprecise nature of the locking counts are also documented with
the suggestion that we should run the measurement a few times with the
counts reset in between to get a better picture of what is going on
under the hood.

Fixes: a8654596f0371 ("locking/rwsem: Enable lock event counting")
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 40 insertions, 2 deletions

diff --git a/kernel/locking/lock_events.h b/kernel/locking/lock_events.h
index feb1acc54611..46b71af8eef2 100644
--- a/kernel/locking/lock_events.h
+++ b/kernel/locking/lock_events.h
@@ -31,12 +31,50 @@ enum lock_events {
 DECLARE_PER_CPU(unsigned long, lockevents[lockevent_num]);
 
 /*
+ * The purpose of the lock event counting subsystem is to provide a low
+ * overhead way to record the number of specific locking events by using
+ * percpu counters. It is the percpu sum that matters, not specifically
+ * how many of them happens in each cpu.
+ *
+ * It is possible that the same percpu counter may be modified in both
+ * the process and interrupt contexts. For architectures that perform
+ * percpu operation with multiple instructions, it is possible to lose
+ * count if a process context percpu update is interrupted in the middle
+ * and the same counter is updated in the interrupt context. Therefore,
+ * the generated percpu sum may not be precise. The error, if any, should
+ * be small and insignificant.
+ *
+ * For those architectures that do multi-instruction percpu operation,
+ * preemption in the middle and moving the task to another cpu may cause
+ * a larger error in the count. Again, this will be few and far between.
+ * Given the imprecise nature of the count and the possibility of resetting
+ * the count and doing the measurement again, this is not really a big
+ * problem.
+ *
+ * To get a better picture of what is happening under the hood, it is
+ * suggested that a few measurements should be taken with the counts
+ * reset in between to stamp out outliner because of these possible
+ * error conditions.
+ *
+ * To minimize overhead, we use __this_cpu_*() in all cases except when
+ * CONFIG_DEBUG_PREEMPT is defined. In this particular case, this_cpu_*()
+ * will be used to avoid the appearance of unwanted BUG messages.
+ */
+#ifdef CONFIG_DEBUG_PREEMPT
+#define lockevent_percpu_inc(x)		this_cpu_inc(x)
+#define lockevent_percpu_add(x, v)	this_cpu_add(x, v)
+#else
+#define lockevent_percpu_inc(x)		__this_cpu_inc(x)
+#define lockevent_percpu_add(x, v)	__this_cpu_add(x, v)
+#endif
+
+/*
  * Increment the PV qspinlock statistical counters
  */
 static inline void __lockevent_inc(enum lock_events event, bool cond)
 {
 	if (cond)
-		__this_cpu_inc(lockevents[event]);
+		lockevent_percpu_inc(lockevents[event]);
 }
 
 #define lockevent_inc(ev)	  __lockevent_inc(LOCKEVENT_ ##ev, true)
@@ -44,7 +82,7 @@ static inline void __lockevent_inc(enum lock_events event, bool cond)
 
 static inline void __lockevent_add(enum lock_events event, int inc)
 {
-	__this_cpu_add(lockevents[event], inc);
+	lockevent_percpu_add(lockevents[event], inc);
 }
 
 #define lockevent_add(ev, c)	__lockevent_add(LOCKEVENT_ ##ev, c)