rcu: Add event-tracing for RCU callback invocation

There was recently some controversy about the overhead of invoking RCU
callbacks.  Add TRACE_EVENT()s to obtain fine-grained timings for the
start and stop of a batch of callbacks and also for each callback invoked.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

2 files changed, 98 insertions, 50 deletions

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index ae5327de41aa..dd2bc2c6a285 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -794,44 +794,6 @@ static inline notrace void rcu_read_unlock_sched_notrace(void)
 #define RCU_INIT_POINTER(p, v) \
 		p = (typeof(*v) __force __rcu *)(v)
 
-/*
- * debug_rcu_head_queue()/debug_rcu_head_unqueue() are used internally
- * by call_rcu() and rcu callback execution, and are therefore not part of the
- * RCU API. Leaving in rcupdate.h because they are used by all RCU flavors.
- */
-
-#ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD
-# define STATE_RCU_HEAD_READY	0
-# define STATE_RCU_HEAD_QUEUED	1
-
-extern struct debug_obj_descr rcuhead_debug_descr;
-
-static inline void debug_rcu_head_queue(struct rcu_head *head)
-{
-	WARN_ON_ONCE((unsigned long)head & 0x3);
-	debug_object_activate(head, &rcuhead_debug_descr);
-	debug_object_active_state(head, &rcuhead_debug_descr,
-				  STATE_RCU_HEAD_READY,
-				  STATE_RCU_HEAD_QUEUED);
-}
-
-static inline void debug_rcu_head_unqueue(struct rcu_head *head)
-{
-	debug_object_active_state(head, &rcuhead_debug_descr,
-				  STATE_RCU_HEAD_QUEUED,
-				  STATE_RCU_HEAD_READY);
-	debug_object_deactivate(head, &rcuhead_debug_descr);
-}
-#else	/* !CONFIG_DEBUG_OBJECTS_RCU_HEAD */
-static inline void debug_rcu_head_queue(struct rcu_head *head)
-{
-}
-
-static inline void debug_rcu_head_unqueue(struct rcu_head *head)
-{
-}
-#endif	/* #else !CONFIG_DEBUG_OBJECTS_RCU_HEAD */
-
 static __always_inline bool __is_kfree_rcu_offset(unsigned long offset)
 {
 	return offset < 4096;
@@ -850,18 +812,6 @@ void __kfree_rcu(struct rcu_head *head, unsigned long offset)
 	call_rcu(head, (rcu_callback)offset);
 }
 
-extern void kfree(const void *);
-
-static inline void __rcu_reclaim(struct rcu_head *head)
-{
-	unsigned long offset = (unsigned long)head->func;
-
-	if (__is_kfree_rcu_offset(offset))
-		kfree((void *)head - offset);
-	else
-		head->func(head);
-}
-
 /**
  * kfree_rcu() - kfree an object after a grace period.
  * @ptr:	pointer to kfree
diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h
new file mode 100644
index 000000000000..db3f6e9e63e6
--- /dev/null
+++ b/include/trace/events/rcu.h
@@ -0,0 +1,98 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rcu
+
+#if !defined(_TRACE_RCU_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_RCU_H
+
+#include <linux/tracepoint.h>
+
+/*
+ * Tracepoint for calling rcu_do_batch, performed to start callback invocation:
+ */
+TRACE_EVENT(rcu_batch_start,
+
+	TP_PROTO(long callbacks_ready, int blimit),
+
+	TP_ARGS(callbacks_ready, blimit),
+
+	TP_STRUCT__entry(
+		__field(	long,	callbacks_ready		)
+		__field(	int,	blimit			)
+	),
+
+	TP_fast_assign(
+		__entry->callbacks_ready	= callbacks_ready;
+		__entry->blimit			= blimit;
+	),
+
+	TP_printk("CBs=%ld bl=%d", __entry->callbacks_ready, __entry->blimit)
+);
+
+/*
+ * Tracepoint for the invocation of a single RCU callback
+ */
+TRACE_EVENT(rcu_invoke_callback,
+
+	TP_PROTO(struct rcu_head *rhp),
+
+	TP_ARGS(rhp),
+
+	TP_STRUCT__entry(
+		__field(	void *,	rhp	)
+		__field(	void *,	func	)
+	),
+
+	TP_fast_assign(
+		__entry->rhp	= rhp;
+		__entry->func	= rhp->func;
+	),
+
+	TP_printk("rhp=%p func=%pf", __entry->rhp, __entry->func)
+);
+
+/*
+ * Tracepoint for the invocation of a single RCU kfree callback
+ */
+TRACE_EVENT(rcu_invoke_kfree_callback,
+
+	TP_PROTO(struct rcu_head *rhp, unsigned long offset),
+
+	TP_ARGS(rhp, offset),
+
+	TP_STRUCT__entry(
+		__field(void *,	rhp	)
+		__field(unsigned long,	offset	)
+	),
+
+	TP_fast_assign(
+		__entry->rhp	= rhp;
+		__entry->offset	= offset;
+	),
+
+	TP_printk("rhp=%p func=%ld", __entry->rhp, __entry->offset)
+);
+
+/*
+ * Tracepoint for leaving rcu_do_batch, performed after callback invocation:
+ */
+TRACE_EVENT(rcu_batch_end,
+
+	TP_PROTO(int callbacks_invoked),
+
+	TP_ARGS(callbacks_invoked),
+
+	TP_STRUCT__entry(
+		__field(	int,	callbacks_invoked		)
+	),
+
+	TP_fast_assign(
+		__entry->callbacks_invoked	= callbacks_invoked;
+	),
+
+	TP_printk("CBs-invoked=%d", __entry->callbacks_invoked)
+);
+
+#endif /* _TRACE_RCU_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>