[RFC][PATCH 0/4] rcu/tracing: Export the RCU tracepoint string pointers

[RFC][PATCH 0/4] rcu/tracing: Export the RCU tracepoint string pointers

[Steven Rostedt]

This has been on my todo list for a while. I've been promising Paul
to get a way to have his tracepoints work for user tools, and I finally
got around to it :-)

As his tracepoints just save a pointer to a string in the ring buffer
and have the output use that pointer with a "%s" printf field, which
works great in the kernel and is fast and efficient, it sucks for tools
that read the binary data from the kernel but has no access to the strings
that those pointers point to.

The trace_printk() had a similar problem when Frederic Weisbecker optimized
it to just save the pointer to the format string along with the parameters,
but that was solved by exporting a table of pointers with the strings
in the /sys/kernel/debug/tracing/printk_formats file.

I've done the same thing for tracepoints using a separate section but the
same file. I needed a seperate section as the trace_printk() section is
used to know if a trace_printk() was added to the kernel (they should
never be added to Linus's tree, unless its for a strict debugging option).
If a trace_printk() is used, several temporary buffers are allocated, which
we don't want in normal use.

Now tracepoints get a section that has the pointers to the strings and
this gets added to the printk_format file. Doing this means that none
of the tools (perf or trace-cmd) require any changes as they already
handle reading the printk_formats via the event_parse library.

-- Steve

Steven Rostedt (Red Hat) (4):
      rcu: Add const annotation to char * for RCU tracepoints and functions
      rcu: Simplify RCU_STATE_INITIALIZER() macro
      tracing: Add __tracepoint_string() to export string pointers
      rcu: Have the RCU tracepoints use the tracepoint_string infrastructure

----
 include/asm-generic/vmlinux.lds.h |    7 ++-
 include/linux/ftrace_event.h      |    7 +++
 include/linux/rcupdate.h          |    4 +-
 include/trace/events/rcu.h        |   82 +++++++++++++++---------------
 kernel/rcu.h                      |    2 +-
 kernel/rcupdate.c                 |    2 +-
 kernel/rcutiny_plugin.h           |    2 +-
 kernel/rcutorture.c               |   14 +++---
 kernel/rcutree.c                  |  100 +++++++++++++++++++++----------------
 kernel/rcutree.h                  |    2 +-
 kernel/rcutree_plugin.h           |   44 ++++++++--------
 kernel/trace/trace.h              |    3 ++
 kernel/trace/trace_printk.c       |    7 +++
 13 files changed, 154 insertions(+), 122 deletions(-)

[RFC][PATCH 1/4] rcu: Add const annotation to char * for RCU tracepoints and functions

[Steven Rostedt]

[-- Attachment #1: 0001-rcu-Add-const-annotation-to-char-for-RCU-tracepoints.patch --]
[-- Type: text/plain, Size: 16692 bytes --]

From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>

All the RCU tracepoints and functions that reference char pointers do
so with just 'char *' even though they do not modify the contents of
the string itself. This will cause warnings if a const char * is used
in one of these functions.

The RCU tracepoints store the pointer to the string to refer back to them
when the trace output is displayed. As this can be minutes, hours or
even days later, those strings had better be constant.

This change also opens the door to allow the RCU tracepoint strings and
their addresses to be exported so that userspace tracing tools can
translate the contents of the pointers of the RCU tracepoints.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 include/linux/rcupdate.h   |    4 +--
 include/trace/events/rcu.h |   82 ++++++++++++++++++++++----------------------
 kernel/rcu.h               |    2 +-
 kernel/rcupdate.c          |    2 +-
 kernel/rcutiny_plugin.h    |    2 +-
 kernel/rcutorture.c        |   14 ++++----
 kernel/rcutree.c           |    4 +--
 kernel/rcutree.h           |    2 +-
 kernel/rcutree_plugin.h    |    8 ++---
 9 files changed, 60 insertions(+), 60 deletions(-)

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 4b14bdc..0c38abb 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -52,7 +52,7 @@ extern int rcutorture_runnable; /* for sysctl */
 #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU)
 extern void rcutorture_record_test_transition(void);
 extern void rcutorture_record_progress(unsigned long vernum);
-extern void do_trace_rcu_torture_read(char *rcutorturename,
+extern void do_trace_rcu_torture_read(const char *rcutorturename,
 				      struct rcu_head *rhp,
 				      unsigned long secs,
 				      unsigned long c_old,
@@ -65,7 +65,7 @@ static inline void rcutorture_record_progress(unsigned long vernum)
 {
 }
 #ifdef CONFIG_RCU_TRACE
-extern void do_trace_rcu_torture_read(char *rcutorturename,
+extern void do_trace_rcu_torture_read(const char *rcutorturename,
 				      struct rcu_head *rhp,
 				      unsigned long secs,
 				      unsigned long c_old,
diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h
index 59ebcc8..ee2376c 100644
--- a/include/trace/events/rcu.h
+++ b/include/trace/events/rcu.h
@@ -19,12 +19,12 @@
  */
 TRACE_EVENT(rcu_utilization,
 
-	TP_PROTO(char *s),
+	TP_PROTO(const char *s),
 
 	TP_ARGS(s),
 
 	TP_STRUCT__entry(
-		__field(char *, s)
+		__field(const char *, s)
 	),
 
 	TP_fast_assign(
@@ -51,14 +51,14 @@ TRACE_EVENT(rcu_utilization,
  */
 TRACE_EVENT(rcu_grace_period,
 
-	TP_PROTO(char *rcuname, unsigned long gpnum, char *gpevent),
+	TP_PROTO(const char *rcuname, unsigned long gpnum, const char *gpevent),
 
 	TP_ARGS(rcuname, gpnum, gpevent),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(unsigned long, gpnum)
-		__field(char *, gpevent)
+		__field(const char *, gpevent)
 	),
 
 	TP_fast_assign(
@@ -89,21 +89,21 @@ TRACE_EVENT(rcu_grace_period,
  */
 TRACE_EVENT(rcu_future_grace_period,
 
-	TP_PROTO(char *rcuname, unsigned long gpnum, unsigned long completed,
+	TP_PROTO(const char *rcuname, unsigned long gpnum, unsigned long completed,
 		 unsigned long c, u8 level, int grplo, int grphi,
-		 char *gpevent),
+		 const char *gpevent),
 
 	TP_ARGS(rcuname, gpnum, completed, c, level, grplo, grphi, gpevent),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(unsigned long, gpnum)
 		__field(unsigned long, completed)
 		__field(unsigned long, c)
 		__field(u8, level)
 		__field(int, grplo)
 		__field(int, grphi)
-		__field(char *, gpevent)
+		__field(const char *, gpevent)
 	),
 
 	TP_fast_assign(
@@ -132,13 +132,13 @@ TRACE_EVENT(rcu_future_grace_period,
  */
 TRACE_EVENT(rcu_grace_period_init,
 
-	TP_PROTO(char *rcuname, unsigned long gpnum, u8 level,
+	TP_PROTO(const char *rcuname, unsigned long gpnum, u8 level,
 		 int grplo, int grphi, unsigned long qsmask),
 
 	TP_ARGS(rcuname, gpnum, level, grplo, grphi, qsmask),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(unsigned long, gpnum)
 		__field(u8, level)
 		__field(int, grplo)
@@ -168,12 +168,12 @@ TRACE_EVENT(rcu_grace_period_init,
  */
 TRACE_EVENT(rcu_preempt_task,
 
-	TP_PROTO(char *rcuname, int pid, unsigned long gpnum),
+	TP_PROTO(const char *rcuname, int pid, unsigned long gpnum),
 
 	TP_ARGS(rcuname, pid, gpnum),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(unsigned long, gpnum)
 		__field(int, pid)
 	),
@@ -195,12 +195,12 @@ TRACE_EVENT(rcu_preempt_task,
  */
 TRACE_EVENT(rcu_unlock_preempted_task,
 
-	TP_PROTO(char *rcuname, unsigned long gpnum, int pid),
+	TP_PROTO(const char *rcuname, unsigned long gpnum, int pid),
 
 	TP_ARGS(rcuname, gpnum, pid),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(unsigned long, gpnum)
 		__field(int, pid)
 	),
@@ -224,14 +224,14 @@ TRACE_EVENT(rcu_unlock_preempted_task,
  */
 TRACE_EVENT(rcu_quiescent_state_report,
 
-	TP_PROTO(char *rcuname, unsigned long gpnum,
+	TP_PROTO(const char *rcuname, unsigned long gpnum,
 		 unsigned long mask, unsigned long qsmask,
 		 u8 level, int grplo, int grphi, int gp_tasks),
 
 	TP_ARGS(rcuname, gpnum, mask, qsmask, level, grplo, grphi, gp_tasks),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(unsigned long, gpnum)
 		__field(unsigned long, mask)
 		__field(unsigned long, qsmask)
@@ -268,15 +268,15 @@ TRACE_EVENT(rcu_quiescent_state_report,
  */
 TRACE_EVENT(rcu_fqs,
 
-	TP_PROTO(char *rcuname, unsigned long gpnum, int cpu, char *qsevent),
+	TP_PROTO(const char *rcuname, unsigned long gpnum, int cpu, const char *qsevent),
 
 	TP_ARGS(rcuname, gpnum, cpu, qsevent),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(unsigned long, gpnum)
 		__field(int, cpu)
-		__field(char *, qsevent)
+		__field(const char *, qsevent)
 	),
 
 	TP_fast_assign(
@@ -308,12 +308,12 @@ TRACE_EVENT(rcu_fqs,
  */
 TRACE_EVENT(rcu_dyntick,
 
-	TP_PROTO(char *polarity, long long oldnesting, long long newnesting),
+	TP_PROTO(const char *polarity, long long oldnesting, long long newnesting),
 
 	TP_ARGS(polarity, oldnesting, newnesting),
 
 	TP_STRUCT__entry(
-		__field(char *, polarity)
+		__field(const char *, polarity)
 		__field(long long, oldnesting)
 		__field(long long, newnesting)
 	),
@@ -352,12 +352,12 @@ TRACE_EVENT(rcu_dyntick,
  */
 TRACE_EVENT(rcu_prep_idle,
 
-	TP_PROTO(char *reason),
+	TP_PROTO(const char *reason),
 
 	TP_ARGS(reason),
 
 	TP_STRUCT__entry(
-		__field(char *, reason)
+		__field(const char *, reason)
 	),
 
 	TP_fast_assign(
@@ -376,13 +376,13 @@ TRACE_EVENT(rcu_prep_idle,
  */
 TRACE_EVENT(rcu_callback,
 
-	TP_PROTO(char *rcuname, struct rcu_head *rhp, long qlen_lazy,
+	TP_PROTO(const char *rcuname, struct rcu_head *rhp, long qlen_lazy,
 		 long qlen),
 
 	TP_ARGS(rcuname, rhp, qlen_lazy, qlen),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(void *, rhp)
 		__field(void *, func)
 		__field(long, qlen_lazy)
@@ -412,13 +412,13 @@ TRACE_EVENT(rcu_callback,
  */
 TRACE_EVENT(rcu_kfree_callback,
 
-	TP_PROTO(char *rcuname, struct rcu_head *rhp, unsigned long offset,
+	TP_PROTO(const char *rcuname, struct rcu_head *rhp, unsigned long offset,
 		 long qlen_lazy, long qlen),
 
 	TP_ARGS(rcuname, rhp, offset, qlen_lazy, qlen),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(void *, rhp)
 		__field(unsigned long, offset)
 		__field(long, qlen_lazy)
@@ -447,12 +447,12 @@ TRACE_EVENT(rcu_kfree_callback,
  */
 TRACE_EVENT(rcu_batch_start,
 
-	TP_PROTO(char *rcuname, long qlen_lazy, long qlen, long blimit),
+	TP_PROTO(const char *rcuname, long qlen_lazy, long qlen, long blimit),
 
 	TP_ARGS(rcuname, qlen_lazy, qlen, blimit),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(long, qlen_lazy)
 		__field(long, qlen)
 		__field(long, blimit)
@@ -477,12 +477,12 @@ TRACE_EVENT(rcu_batch_start,
  */
 TRACE_EVENT(rcu_invoke_callback,
 
-	TP_PROTO(char *rcuname, struct rcu_head *rhp),
+	TP_PROTO(const char *rcuname, struct rcu_head *rhp),
 
 	TP_ARGS(rcuname, rhp),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(void *, rhp)
 		__field(void *, func)
 	),
@@ -506,12 +506,12 @@ TRACE_EVENT(rcu_invoke_callback,
  */
 TRACE_EVENT(rcu_invoke_kfree_callback,
 
-	TP_PROTO(char *rcuname, struct rcu_head *rhp, unsigned long offset),
+	TP_PROTO(const char *rcuname, struct rcu_head *rhp, unsigned long offset),
 
 	TP_ARGS(rcuname, rhp, offset),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(void *, rhp)
 		__field(unsigned long, offset)
 	),
@@ -539,13 +539,13 @@ TRACE_EVENT(rcu_invoke_kfree_callback,
  */
 TRACE_EVENT(rcu_batch_end,
 
-	TP_PROTO(char *rcuname, int callbacks_invoked,
+	TP_PROTO(const char *rcuname, int callbacks_invoked,
 		 bool cb, bool nr, bool iit, bool risk),
 
 	TP_ARGS(rcuname, callbacks_invoked, cb, nr, iit, risk),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
+		__field(const char *, rcuname)
 		__field(int, callbacks_invoked)
 		__field(bool, cb)
 		__field(bool, nr)
@@ -577,13 +577,13 @@ TRACE_EVENT(rcu_batch_end,
  */
 TRACE_EVENT(rcu_torture_read,
 
-	TP_PROTO(char *rcutorturename, struct rcu_head *rhp,
+	TP_PROTO(const char *rcutorturename, struct rcu_head *rhp,
 		 unsigned long secs, unsigned long c_old, unsigned long c),
 
 	TP_ARGS(rcutorturename, rhp, secs, c_old, c),
 
 	TP_STRUCT__entry(
-		__field(char *, rcutorturename)
+		__field(const char *, rcutorturename)
 		__field(struct rcu_head *, rhp)
 		__field(unsigned long, secs)
 		__field(unsigned long, c_old)
@@ -623,13 +623,13 @@ TRACE_EVENT(rcu_torture_read,
  */
 TRACE_EVENT(rcu_barrier,
 
-	TP_PROTO(char *rcuname, char *s, int cpu, int cnt, unsigned long done),
+	TP_PROTO(const char *rcuname, const char *s, int cpu, int cnt, unsigned long done),
 
 	TP_ARGS(rcuname, s, cpu, cnt, done),
 
 	TP_STRUCT__entry(
-		__field(char *, rcuname)
-		__field(char *, s)
+		__field(const char *, rcuname)
+		__field(const char *, s)
 		__field(int, cpu)
 		__field(int, cnt)
 		__field(unsigned long, done)
diff --git a/kernel/rcu.h b/kernel/rcu.h
index 7f8e759..0a90ccc 100644
--- a/kernel/rcu.h
+++ b/kernel/rcu.h
@@ -94,7 +94,7 @@ static inline void debug_rcu_head_unqueue(struct rcu_head *head)
 
 extern void kfree(const void *);
 
-static inline bool __rcu_reclaim(char *rn, struct rcu_head *head)
+static inline bool __rcu_reclaim(const char *rn, struct rcu_head *head)
 {
 	unsigned long offset = (unsigned long)head->func;
 
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index cce6ba8..14994d4 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -377,7 +377,7 @@ EXPORT_SYMBOL_GPL(rcuhead_debug_descr);
 #endif /* #ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD */
 
 #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU) || defined(CONFIG_RCU_TRACE)
-void do_trace_rcu_torture_read(char *rcutorturename, struct rcu_head *rhp,
+void do_trace_rcu_torture_read(const char *rcutorturename, struct rcu_head *rhp,
 			       unsigned long secs,
 			       unsigned long c_old, unsigned long c)
 {
diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h
index 0cd385a..280d06c 100644
--- a/kernel/rcutiny_plugin.h
+++ b/kernel/rcutiny_plugin.h
@@ -36,7 +36,7 @@ struct rcu_ctrlblk {
 	RCU_TRACE(unsigned long gp_start); /* Start time for stalls. */
 	RCU_TRACE(unsigned long ticks_this_gp); /* Statistic for stalls. */
 	RCU_TRACE(unsigned long jiffies_stall); /* Jiffies at next stall. */
-	RCU_TRACE(char *name);		/* Name of RCU type. */
+	RCU_TRACE(const char *name);	/* Name of RCU type. */
 };
 
 /* Definition for rcupdate control block. */
diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index b1fa551..7afa001 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -267,7 +267,7 @@ rcutorture_shutdown_notify(struct notifier_block *unused1,
  * Absorb kthreads into a kernel function that won't return, so that
  * they won't ever access module text or data again.
  */
-static void rcutorture_shutdown_absorb(char *title)
+static void rcutorture_shutdown_absorb(const char *title)
 {
 	if (ACCESS_ONCE(fullstop) == FULLSTOP_SHUTDOWN) {
 		pr_notice(
@@ -337,7 +337,7 @@ rcu_random(struct rcu_random_state *rrsp)
 }
 
 static void
-rcu_stutter_wait(char *title)
+rcu_stutter_wait(const char *title)
 {
 	while (stutter_pause_test || !rcutorture_runnable) {
 		if (rcutorture_runnable)
@@ -363,10 +363,10 @@ struct rcu_torture_ops {
 	void (*call)(struct rcu_head *head, void (*func)(struct rcu_head *rcu));
 	void (*cb_barrier)(void);
 	void (*fqs)(void);
-	int (*stats)(char *page);
+	int (*stats)(const char *page);
 	int irq_capable;
 	int can_boost;
-	char *name;
+	const char *name;
 };
 
 static struct rcu_torture_ops *cur_ops;
@@ -650,7 +650,7 @@ static void srcu_torture_barrier(void)
 	srcu_barrier(&srcu_ctl);
 }
 
-static int srcu_torture_stats(char *page)
+static int srcu_torture_stats(const char *page)
 {
 	int cnt = 0;
 	int cpu;
@@ -1147,7 +1147,7 @@ rcu_torture_reader(void *arg)
  * Create an RCU-torture statistics message in the specified buffer.
  */
 static int
-rcu_torture_printk(char *page)
+rcu_torture_printk(const char *page)
 {
 	int cnt = 0;
 	int cpu;
@@ -1364,7 +1364,7 @@ rcu_torture_stutter(void *arg)
 }
 
 static inline void
-rcu_torture_print_module_parms(struct rcu_torture_ops *cur_ops, char *tag)
+rcu_torture_print_module_parms(struct rcu_torture_ops *cur_ops, const char *tag)
 {
 	pr_alert("%s" TORTURE_FLAG
 		 "--- %s: nreaders=%d nfakewriters=%d "
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index e08abb9..f560eb3 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -1032,7 +1032,7 @@ static unsigned long rcu_cbs_completed(struct rcu_state *rsp,
  * rcu_nocb_wait_gp().
  */
 static void trace_rcu_future_gp(struct rcu_node *rnp, struct rcu_data *rdp,
-				unsigned long c, char *s)
+				unsigned long c, const char *s)
 {
 	trace_rcu_future_grace_period(rdp->rsp->name, rnp->gpnum,
 				      rnp->completed, c, rnp->level,
@@ -2720,7 +2720,7 @@ static int rcu_cpu_has_callbacks(int cpu, bool *all_lazy)
  * Helper function for _rcu_barrier() tracing.  If tracing is disabled,
  * the compiler is expected to optimize this away.
  */
-static void _rcu_barrier_trace(struct rcu_state *rsp, char *s,
+static void _rcu_barrier_trace(struct rcu_state *rsp, const char *s,
 			       int cpu, unsigned long done)
 {
 	trace_rcu_barrier(rsp->name, s, cpu,
diff --git a/kernel/rcutree.h b/kernel/rcutree.h
index 4a39d36..ac0964d 100644
--- a/kernel/rcutree.h
+++ b/kernel/rcutree.h
@@ -445,7 +445,7 @@ struct rcu_state {
 						/*  for CPU stalls. */
 	unsigned long gp_max;			/* Maximum GP duration in */
 						/*  jiffies. */
-	char *name;				/* Name of structure. */
+	const char *name;			/* Name of structure. */
 	char abbr;				/* Abbreviated name. */
 	struct list_head flavors;		/* List of RCU flavors. */
 	struct irq_work wakeup_work;		/* Postponed wakeups */
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 63098a5..49df554 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -1884,7 +1884,7 @@ early_initcall(rcu_register_oom_notifier);
 
 #ifdef CONFIG_RCU_FAST_NO_HZ
 
-static void print_cpu_stall_fast_no_hz(char *cp, int cpu)
+static void print_cpu_stall_fast_no_hz(const char *cp, int cpu)
 {
 	struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu);
 	unsigned long nlpd = rdtp->nonlazy_posted - rdtp->nonlazy_posted_snap;
@@ -1898,7 +1898,7 @@ static void print_cpu_stall_fast_no_hz(char *cp, int cpu)
 
 #else /* #ifdef CONFIG_RCU_FAST_NO_HZ */
 
-static void print_cpu_stall_fast_no_hz(char *cp, int cpu)
+static void print_cpu_stall_fast_no_hz(const char *cp, int cpu)
 {
 	*cp = '\0';
 }
@@ -2020,7 +2020,7 @@ static void increment_cpu_stall_ticks(void)
 
 
 /* Parse the boot-time rcu_nocb_mask CPU list from the kernel parameters. */
-static int __init rcu_nocb_setup(char *str)
+static int __init rcu_nocb_setup(const char *str)
 {
 	alloc_bootmem_cpumask_var(&rcu_nocb_mask);
 	have_rcu_nocb_mask = true;
@@ -2029,7 +2029,7 @@ static int __init rcu_nocb_setup(char *str)
 }
 __setup("rcu_nocbs=", rcu_nocb_setup);
 
-static int __init parse_rcu_nocb_poll(char *arg)
+static int __init parse_rcu_nocb_poll(const char *arg)
 {
 	rcu_nocb_poll = 1;
 	return 0;
-- 
1.7.10.4

[RFC][PATCH 2/4] rcu: Simplify RCU_STATE_INITIALIZER() macro

[Steven Rostedt]

[-- Attachment #1: 0002-rcu-Simplify-RCU_STATE_INITIALIZER-macro.patch --]
[-- Type: text/plain, Size: 3089 bytes --]

From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>

The RCU_STATE_INITIALIZER() macro is used only in the rcutree.c file
as well as the rcutree_plugin.h file. It is passed as a rvalue to
a variable of a similar name. A per_cpu variable is also created
with a similar name as well.

The uses of RCU_STATE_INITIALIZER() can be simplified to remove some
of the duplicate code that is done. Currently the three users of this
macro has this format:

struct rcu_state rcu_sched_state =
	RCU_STATE_INITIALIZER(rcu_sched, call_rcu_sched);
DEFINE_PER_CPU(struct rcu_data, rcu_sched_data);

Notice that "rcu_sched" is called three times. This is the same with
the other two users. This can be condensed to just:

RCU_STATE_INITIALIZER(rcu_sched, call_rcu_sched);

by moving the rest into the macro itself.

This also opens the door to allow the RCU tracepoint strings and
their addresses to be exported so that userspace tracing tools can
translate the contents of the pointers of the RCU tracepoints.
The change will allow for helper code to be placed in the
RCU_STATE_INITIALIZER() macro to export the name that is used.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 kernel/rcutree.c        |   14 ++++++--------
 kernel/rcutree_plugin.h |    4 +---
 2 files changed, 7 insertions(+), 11 deletions(-)

diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index f560eb3..89a8b10 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -64,7 +64,8 @@
 static struct lock_class_key rcu_node_class[RCU_NUM_LVLS];
 static struct lock_class_key rcu_fqs_class[RCU_NUM_LVLS];
 
-#define RCU_STATE_INITIALIZER(sname, sabbr, cr) { \
+#define RCU_STATE_INITIALIZER(sname, sabbr, cr) \
+struct rcu_state sname##_state = { \
 	.level = { &sname##_state.node[0] }, \
 	.call = cr, \
 	.fqs_state = RCU_GP_IDLE, \
@@ -77,14 +78,11 @@ static struct lock_class_key rcu_fqs_class[RCU_NUM_LVLS];
 	.onoff_mutex = __MUTEX_INITIALIZER(sname##_state.onoff_mutex), \
 	.name = #sname, \
 	.abbr = sabbr, \
-}
-
-struct rcu_state rcu_sched_state =
-	RCU_STATE_INITIALIZER(rcu_sched, 's', call_rcu_sched);
-DEFINE_PER_CPU(struct rcu_data, rcu_sched_data);
+}; \
+DEFINE_PER_CPU(struct rcu_data, sname##_data)
 
-struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh, 'b', call_rcu_bh);
-DEFINE_PER_CPU(struct rcu_data, rcu_bh_data);
+RCU_STATE_INITIALIZER(rcu_sched, 's', call_rcu_sched);
+RCU_STATE_INITIALIZER(rcu_bh, 'b', call_rcu_bh);
 
 static struct rcu_state *rcu_state;
 LIST_HEAD(rcu_struct_flavors);
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 49df554..5cdb559 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -110,9 +110,7 @@ static void __init rcu_bootup_announce_oddness(void)
 
 #ifdef CONFIG_TREE_PREEMPT_RCU
 
-struct rcu_state rcu_preempt_state =
-	RCU_STATE_INITIALIZER(rcu_preempt, 'p', call_rcu);
-DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data);
+RCU_STATE_INITIALIZER(rcu_preempt, 'p', call_rcu);
 static struct rcu_state *rcu_state = &rcu_preempt_state;
 
 static int rcu_preempted_readers_exp(struct rcu_node *rnp);
-- 
1.7.10.4

[RFC][PATCH 3/4] tracing: Add __tracepoint_string() to export string pointers

[Steven Rostedt]

[-- Attachment #1: 0003-tracing-Add-__tracepoint_string-to-export-string-poi.patch --]
[-- Type: text/plain, Size: 5102 bytes --]

From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>

There are several tracepoints (mostly in RCU), that reference a string
pointer and uses the print format of "%s" to display the string that
exists in the kernel, instead of copying the actual string to the
ring buffer (saves time and ring buffer space).

But this has an issue with userspace tools that read the binary buffers
that has the address of the string but has no access to what the string
itself is. The end result is just output that looks like:

 rcu_dyntick:          ffffffff818adeaa 1 0
 rcu_dyntick:          ffffffff818adeb5 0 140000000000000
 rcu_dyntick:          ffffffff818adeb5 0 140000000000000
 rcu_utilization:      ffffffff8184333b
 rcu_utilization:      ffffffff8184333b

The above is pretty useless when read by the userspace tools. Ideally
we would want something that looks like this:

 rcu_dyntick:          Start 1 0
 rcu_dyntick:          End 0 140000000000000
 rcu_dyntick:          Start 140000000000000 0
 rcu_callback:         rcu_preempt rhp=0xffff880037aff710 func=put_cred_rcu 0/4
 rcu_callback:         rcu_preempt rhp=0xffff880078961980 func=file_free_rcu 0/5
 rcu_dyntick:          End 0 1

The trace_printk() which also only stores the address of the string
format instead of recording the string into the buffer itself, exports
the mapping of kernel addresses to format strings via the printk_format
file in the debugfs tracing directory.

The tracepoint strings can use this same method and output the format
to the same file and the userspace tools will be able to decipher
the address without any modification.

The tracepoint strings need its own section to save the strings because
the trace_printk section will cause the trace_printk() buffers to be
allocated if anything exists within the section. trace_printk() is only
used for debugging and should never exist in the kernel, we can not use
the trace_printk sections.

Add a new tracepoint_str section that will also be examined by the output
of the printk_format file.

Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 include/asm-generic/vmlinux.lds.h |    7 ++++++-
 include/linux/ftrace_event.h      |    7 +++++++
 kernel/trace/trace.h              |    3 +++
 kernel/trace/trace_printk.c       |    7 +++++++
 4 files changed, 23 insertions(+), 1 deletion(-)

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 69732d2..83e2c31 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -122,8 +122,12 @@
 #define TRACE_PRINTKS() VMLINUX_SYMBOL(__start___trace_bprintk_fmt) = .;      \
 			 *(__trace_printk_fmt) /* Trace_printk fmt' pointer */ \
 			 VMLINUX_SYMBOL(__stop___trace_bprintk_fmt) = .;
+#define TRACEPOINT_STR() VMLINUX_SYMBOL(__start___tracepoint_str) = .;	\
+			 *(__tracepoint_str) /* Trace_printk fmt' pointer */ \
+			 VMLINUX_SYMBOL(__stop___tracepoint_str) = .;
 #else
 #define TRACE_PRINTKS()
+#define TRACEPOINT_STR()
 #endif
 
 #ifdef CONFIG_FTRACE_SYSCALLS
@@ -190,7 +194,8 @@
 	VMLINUX_SYMBOL(__stop___verbose) = .;				\
 	LIKELY_PROFILE()		       				\
 	BRANCH_PROFILE()						\
-	TRACE_PRINTKS()
+	TRACE_PRINTKS()							\
+	TRACEPOINT_STR()
 
 /*
  * Data section helpers
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 4372658..cead2e9 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -357,6 +357,13 @@ do {									\
 		__trace_printk(ip, fmt, ##args);			\
 } while (0)
 
+#define __tracepoint_string	__attribute__((section("__tracepoint_str")))
+#define tracepoint_string(str)						\
+	({								\
+		static const char *___tp_str __tracepoint_string = str; \
+		___tp_str;						\
+	})
+
 #ifdef CONFIG_PERF_EVENTS
 struct perf_event;
 
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 4a4f6e1..ba321f1 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -1022,6 +1022,9 @@ extern struct list_head ftrace_events;
 extern const char *__start___trace_bprintk_fmt[];
 extern const char *__stop___trace_bprintk_fmt[];
 
+extern const char *__start___tracepoint_str[];
+extern const char *__stop___tracepoint_str[];
+
 void trace_printk_init_buffers(void);
 void trace_printk_start_comm(void);
 int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set);
diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c
index a9077c1..fa8d070 100644
--- a/kernel/trace/trace_printk.c
+++ b/kernel/trace/trace_printk.c
@@ -244,12 +244,19 @@ static const char **find_next(void *v, loff_t *pos)
 {
 	const char **fmt = v;
 	int start_index;
+	int last_index;
 
 	start_index = __stop___trace_bprintk_fmt - __start___trace_bprintk_fmt;
 
 	if (*pos < start_index)
 		return __start___trace_bprintk_fmt + *pos;
 
+	last_index = start_index;
+	start_index = __stop___tracepoint_str - __start___tracepoint_str;
+
+	if (*pos < last_index + start_index)
+		return __start___tracepoint_str + (*pos - last_index);
+
 	return find_next_mod_format(start_index, v, fmt, pos);
 }
 
-- 
1.7.10.4

[RFC][PATCH 4/4] rcu: Have the RCU tracepoints use the tracepoint_string infrastructure

[Steven Rostedt]

[-- Attachment #1: 0004-rcu-Have-the-RCU-tracepoints-use-the-tracepoint_stri.patch --]
[-- Type: text/plain, Size: 19163 bytes --]

From: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>

Currently, RCU tracepoints save only a pointer to strings in the
ring buffer. When displayed via the /sys/kernel/debug/tracing/trace file
they are referenced like the printf "%s" that looks at the address
in the ring buffer and prints out the string it points too. This requires
that the strings are constand and presistent in the kernel.

The problem with this is for tools like trace-cmd and perf that read the
binary data from the buffers but have no access to the kernel memory to
find out what string is represented by the address in the buffer.

By using the tracepoint_string infrastructure, the RCU tracepoint strings
can be exported such that userspace tools can map the addresses to
the strings.

 # cat /sys/kernel/debug/tracing/printk_formats
0xffffffff81a4a0e8 : "rcu_preempt"
0xffffffff81a4a0f4 : "rcu_bh"
0xffffffff81a4a100 : "rcu_sched"
0xffffffff818437a0 : "cpuqs"
0xffffffff818437a6 : "rcu_sched"
0xffffffff818437a0 : "cpuqs"
0xffffffff818437b0 : "rcu_bh"
0xffffffff818437b7 : "Start context switch"
0xffffffff818437cc : "End context switch"
0xffffffff818437a0 : "cpuqs"
[...]

Now userspaces tools can display:

 rcu_utilization:      Start context switch
 rcu_dyntick:          Start 1 0
 rcu_utilization:      End context switch
 rcu_batch_start:      rcu_preempt CBs=0/5 bl=10
 rcu_dyntick:          End 0 140000000000000
 rcu_invoke_callback:  rcu_preempt rhp=0xffff880071c0d600 func=proc_i_callback
 rcu_invoke_callback:  rcu_preempt rhp=0xffff880077b5b230 func=__d_free
 rcu_dyntick:          Start 140000000000000 0
 rcu_invoke_callback:  rcu_preempt rhp=0xffff880077563980 func=file_free_rcu
 rcu_batch_end:        rcu_preempt CBs-invoked=3 idle=>c<>c<>c<>c<
 rcu_utilization:      End RCU core
 rcu_grace_period:     rcu_preempt 9741 start
 rcu_dyntick:          Start 1 0
 rcu_dyntick:          End 0 140000000000000
 rcu_dyntick:          Start 140000000000000 0

Instead of:

 rcu_utilization:      ffffffff81843110
 rcu_future_grace_period: ffffffff81842f1d 9939 9939 9940 0 0 3 ffffffff81842f32
 rcu_batch_start:      ffffffff81842f1d CBs=0/4 bl=10
 rcu_future_grace_period: ffffffff81842f1d 9939 9939 9940 0 0 3 ffffffff81842f3c
 rcu_grace_period:     ffffffff81842f1d 9939 ffffffff81842f80
 rcu_invoke_callback:  ffffffff81842f1d rhp=0xffff88007888aac0 func=file_free_rcu
 rcu_grace_period:     ffffffff81842f1d 9939 ffffffff81842f95
 rcu_invoke_callback:  ffffffff81842f1d rhp=0xffff88006aeb4600 func=proc_i_callback
 rcu_future_grace_period: ffffffff81842f1d 9939 9939 9940 0 0 3 ffffffff81842f32
 rcu_future_grace_period: ffffffff81842f1d 9939 9939 9940 0 0 3 ffffffff81842f3c
 rcu_invoke_callback:  ffffffff81842f1d rhp=0xffff880071cb9fc0 func=__d_free
 rcu_grace_period:     ffffffff81842f1d 9939 ffffffff81842f80
 rcu_invoke_callback:  ffffffff81842f1d rhp=0xffff88007888ae80 func=file_free_rcu
 rcu_batch_end:        ffffffff81842f1d CBs-invoked=4 idle=>c<>c<>c<>c<
 rcu_utilization:      ffffffff8184311f

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 kernel/rcutree.c        |   82 +++++++++++++++++++++++++++--------------------
 kernel/rcutree_plugin.h |   32 +++++++++---------
 2 files changed, 64 insertions(+), 50 deletions(-)

diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index 89a8b10..32d403a 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -53,18 +53,31 @@
 #include <linux/delay.h>
 #include <linux/stop_machine.h>
 #include <linux/random.h>
+#include <linux/ftrace_event.h>
 
 #include "rcutree.h"
 #include <trace/events/rcu.h>
 
 #include "rcu.h"
 
+#define TPS(x)	tracepoint_string(x)
+
 /* Data structures. */
 
 static struct lock_class_key rcu_node_class[RCU_NUM_LVLS];
 static struct lock_class_key rcu_fqs_class[RCU_NUM_LVLS];
 
+/*
+ * In order to export the rcu_state name to the tracing tools, it
+ * needs to be added in the __tracepoint_string section.
+ * This requires defining a separate variable tp_<sname>_varname
+ * that points to the string being used, and this will allow
+ * the tracing userspace tools to be able to decipher the string
+ * address to the matching string.
+ */
 #define RCU_STATE_INITIALIZER(sname, sabbr, cr) \
+static char sname##_varname[] = #sname; \
+static const char *tp_##sname##_varname __used __tracepoint_string = sname##_varname; \
 struct rcu_state sname##_state = { \
 	.level = { &sname##_state.node[0] }, \
 	.call = cr, \
@@ -76,7 +89,7 @@ struct rcu_state sname##_state = { \
 	.orphan_donetail = &sname##_state.orphan_donelist, \
 	.barrier_mutex = __MUTEX_INITIALIZER(sname##_state.barrier_mutex), \
 	.onoff_mutex = __MUTEX_INITIALIZER(sname##_state.onoff_mutex), \
-	.name = #sname, \
+	.name = sname##_varname, \
 	.abbr = sabbr, \
 }; \
 DEFINE_PER_CPU(struct rcu_data, sname##_data)
@@ -176,7 +189,7 @@ void rcu_sched_qs(int cpu)
 	struct rcu_data *rdp = &per_cpu(rcu_sched_data, cpu);
 
 	if (rdp->passed_quiesce == 0)
-		trace_rcu_grace_period("rcu_sched", rdp->gpnum, "cpuqs");
+		trace_rcu_grace_period(TPS("rcu_sched"), rdp->gpnum, TPS("cpuqs"));
 	rdp->passed_quiesce = 1;
 }
 
@@ -185,7 +198,7 @@ void rcu_bh_qs(int cpu)
 	struct rcu_data *rdp = &per_cpu(rcu_bh_data, cpu);
 
 	if (rdp->passed_quiesce == 0)
-		trace_rcu_grace_period("rcu_bh", rdp->gpnum, "cpuqs");
+		trace_rcu_grace_period(TPS("rcu_bh"), rdp->gpnum, TPS("cpuqs"));
 	rdp->passed_quiesce = 1;
 }
 
@@ -196,10 +209,10 @@ void rcu_bh_qs(int cpu)
  */
 void rcu_note_context_switch(int cpu)
 {
-	trace_rcu_utilization("Start context switch");
+	trace_rcu_utilization(TPS("Start context switch"));
 	rcu_sched_qs(cpu);
 	rcu_preempt_note_context_switch(cpu);
-	trace_rcu_utilization("End context switch");
+	trace_rcu_utilization(TPS("End context switch"));
 }
 EXPORT_SYMBOL_GPL(rcu_note_context_switch);
 
@@ -343,11 +356,11 @@ static struct rcu_node *rcu_get_root(struct rcu_state *rsp)
 static void rcu_eqs_enter_common(struct rcu_dynticks *rdtp, long long oldval,
 				bool user)
 {
-	trace_rcu_dyntick("Start", oldval, rdtp->dynticks_nesting);
+	trace_rcu_dyntick(TPS("Start"), oldval, rdtp->dynticks_nesting);
 	if (!user && !is_idle_task(current)) {
 		struct task_struct *idle = idle_task(smp_processor_id());
 
-		trace_rcu_dyntick("Error on entry: not idle task", oldval, 0);
+		trace_rcu_dyntick(TPS("Error on entry: not idle task"), oldval, 0);
 		ftrace_dump(DUMP_ORIG);
 		WARN_ONCE(1, "Current pid: %d comm: %s / Idle pid: %d comm: %s",
 			  current->pid, current->comm,
@@ -477,7 +490,7 @@ void rcu_irq_exit(void)
 	rdtp->dynticks_nesting--;
 	WARN_ON_ONCE(rdtp->dynticks_nesting < 0);
 	if (rdtp->dynticks_nesting)
-		trace_rcu_dyntick("--=", oldval, rdtp->dynticks_nesting);
+		trace_rcu_dyntick(TPS("--="), oldval, rdtp->dynticks_nesting);
 	else
 		rcu_eqs_enter_common(rdtp, oldval, true);
 	local_irq_restore(flags);
@@ -499,11 +512,11 @@ static void rcu_eqs_exit_common(struct rcu_dynticks *rdtp, long long oldval,
 	smp_mb__after_atomic_inc();  /* See above. */
 	WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1));
 	rcu_cleanup_after_idle(smp_processor_id());
-	trace_rcu_dyntick("End", oldval, rdtp->dynticks_nesting);
+	trace_rcu_dyntick(TPS("End"), oldval, rdtp->dynticks_nesting);
 	if (!user && !is_idle_task(current)) {
 		struct task_struct *idle = idle_task(smp_processor_id());
 
-		trace_rcu_dyntick("Error on exit: not idle task",
+		trace_rcu_dyntick(TPS("Error on exit: not idle task"),
 				  oldval, rdtp->dynticks_nesting);
 		ftrace_dump(DUMP_ORIG);
 		WARN_ONCE(1, "Current pid: %d comm: %s / Idle pid: %d comm: %s",
@@ -618,7 +631,7 @@ void rcu_irq_enter(void)
 	rdtp->dynticks_nesting++;
 	WARN_ON_ONCE(rdtp->dynticks_nesting == 0);
 	if (oldval)
-		trace_rcu_dyntick("++=", oldval, rdtp->dynticks_nesting);
+		trace_rcu_dyntick(TPS("++="), oldval, rdtp->dynticks_nesting);
 	else
 		rcu_eqs_exit_common(rdtp, oldval, true);
 	local_irq_restore(flags);
@@ -773,7 +786,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp)
 	 * of the current RCU grace period.
 	 */
 	if ((curr & 0x1) == 0 || UINT_CMP_GE(curr, snap + 2)) {
-		trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, "dti");
+		trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("dti"));
 		rdp->dynticks_fqs++;
 		return 1;
 	}
@@ -793,7 +806,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp)
 		return 0;  /* Grace period is not old enough. */
 	barrier();
 	if (cpu_is_offline(rdp->cpu)) {
-		trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, "ofl");
+		trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("ofl"));
 		rdp->offline_fqs++;
 		return 1;
 	}
@@ -1056,9 +1069,9 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp)
 	 * grace period is already marked as needed, return to the caller.
 	 */
 	c = rcu_cbs_completed(rdp->rsp, rnp);
-	trace_rcu_future_gp(rnp, rdp, c, "Startleaf");
+	trace_rcu_future_gp(rnp, rdp, c, TPS("Startleaf"));
 	if (rnp->need_future_gp[c & 0x1]) {
-		trace_rcu_future_gp(rnp, rdp, c, "Prestartleaf");
+		trace_rcu_future_gp(rnp, rdp, c, TPS("Prestartleaf"));
 		return c;
 	}
 
@@ -1072,7 +1085,7 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp)
 	if (rnp->gpnum != rnp->completed ||
 	    ACCESS_ONCE(rnp->gpnum) != ACCESS_ONCE(rnp->completed)) {
 		rnp->need_future_gp[c & 0x1]++;
-		trace_rcu_future_gp(rnp, rdp, c, "Startedleaf");
+		trace_rcu_future_gp(rnp, rdp, c, TPS("Startedleaf"));
 		return c;
 	}
 
@@ -1100,7 +1113,7 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp)
 	 * recorded, trace and leave.
 	 */
 	if (rnp_root->need_future_gp[c & 0x1]) {
-		trace_rcu_future_gp(rnp, rdp, c, "Prestartedroot");
+		trace_rcu_future_gp(rnp, rdp, c, TPS("Prestartedroot"));
 		goto unlock_out;
 	}
 
@@ -1109,9 +1122,9 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp)
 
 	/* If a grace period is not already in progress, start one. */
 	if (rnp_root->gpnum != rnp_root->completed) {
-		trace_rcu_future_gp(rnp, rdp, c, "Startedleafroot");
+		trace_rcu_future_gp(rnp, rdp, c, TPS("Startedleafroot"));
 	} else {
-		trace_rcu_future_gp(rnp, rdp, c, "Startedroot");
+		trace_rcu_future_gp(rnp, rdp, c, TPS("Startedroot"));
 		rcu_start_gp_advanced(rdp->rsp, rnp_root, rdp);
 	}
 unlock_out:
@@ -1135,7 +1148,8 @@ static int rcu_future_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp)
 	rcu_nocb_gp_cleanup(rsp, rnp);
 	rnp->need_future_gp[c & 0x1] = 0;
 	needmore = rnp->need_future_gp[(c + 1) & 0x1];
-	trace_rcu_future_gp(rnp, rdp, c, needmore ? "CleanupMore" : "Cleanup");
+	trace_rcu_future_gp(rnp, rdp, c,
+			    needmore ? TPS("CleanupMore") : TPS("Cleanup"));
 	return needmore;
 }
 
@@ -1203,9 +1217,9 @@ static void rcu_accelerate_cbs(struct rcu_state *rsp, struct rcu_node *rnp,
 
 	/* Trace depending on how much we were able to accelerate. */
 	if (!*rdp->nxttail[RCU_WAIT_TAIL])
-		trace_rcu_grace_period(rsp->name, rdp->gpnum, "AccWaitCB");
+		trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("AccWaitCB"));
 	else
-		trace_rcu_grace_period(rsp->name, rdp->gpnum, "AccReadyCB");
+		trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("AccReadyCB"));
 }
 
 /*
@@ -1271,7 +1285,7 @@ static void __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, struc
 
 		/* Remember that we saw this grace-period completion. */
 		rdp->completed = rnp->completed;
-		trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpuend");
+		trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuend"));
 	}
 
 	if (rdp->gpnum != rnp->gpnum) {
@@ -1281,7 +1295,7 @@ static void __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, struc
 		 * go looking for one.
 		 */
 		rdp->gpnum = rnp->gpnum;
-		trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpustart");
+		trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpustart"));
 		rdp->passed_quiesce = 0;
 		rdp->qs_pending = !!(rnp->qsmask & rdp->grpmask);
 		zero_cpu_stall_ticks(rdp);
@@ -1324,7 +1338,7 @@ static int rcu_gp_init(struct rcu_state *rsp)
 
 	/* Advance to a new grace period and initialize state. */
 	rsp->gpnum++;
-	trace_rcu_grace_period(rsp->name, rsp->gpnum, "start");
+	trace_rcu_grace_period(rsp->name, rsp->gpnum, TPS("start"));
 	record_gp_stall_check_time(rsp);
 	raw_spin_unlock_irq(&rnp->lock);
 
@@ -1446,7 +1460,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp)
 	rcu_nocb_gp_set(rnp, nocb);
 
 	rsp->completed = rsp->gpnum; /* Declare grace period done. */
-	trace_rcu_grace_period(rsp->name, rsp->completed, "end");
+	trace_rcu_grace_period(rsp->name, rsp->completed, TPS("end"));
 	rsp->fqs_state = RCU_GP_IDLE;
 	rdp = this_cpu_ptr(rsp->rda);
 	rcu_advance_cbs(rsp, rnp, rdp);  /* Reduce false positives below. */
@@ -1855,7 +1869,7 @@ static void rcu_cleanup_dying_cpu(struct rcu_state *rsp)
 	RCU_TRACE(mask = rdp->grpmask);
 	trace_rcu_grace_period(rsp->name,
 			       rnp->gpnum + 1 - !!(rnp->qsmask & mask),
-			       "cpuofl");
+			       TPS("cpuofl"));
 }
 
 /*
@@ -2042,7 +2056,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp)
  */
 void rcu_check_callbacks(int cpu, int user)
 {
-	trace_rcu_utilization("Start scheduler-tick");
+	trace_rcu_utilization(TPS("Start scheduler-tick"));
 	increment_cpu_stall_ticks();
 	if (user || rcu_is_cpu_rrupt_from_idle()) {
 
@@ -2075,7 +2089,7 @@ void rcu_check_callbacks(int cpu, int user)
 	rcu_preempt_check_callbacks(cpu);
 	if (rcu_pending(cpu))
 		invoke_rcu_core();
-	trace_rcu_utilization("End scheduler-tick");
+	trace_rcu_utilization(TPS("End scheduler-tick"));
 }
 
 /*
@@ -2206,10 +2220,10 @@ static void rcu_process_callbacks(struct softirq_action *unused)
 
 	if (cpu_is_offline(smp_processor_id()))
 		return;
-	trace_rcu_utilization("Start RCU core");
+	trace_rcu_utilization(TPS("Start RCU core"));
 	for_each_rcu_flavor(rsp)
 		__rcu_process_callbacks(rsp);
-	trace_rcu_utilization("End RCU core");
+	trace_rcu_utilization(TPS("End RCU core"));
 }
 
 /*
@@ -2950,7 +2964,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible)
 			rdp->completed = rnp->completed;
 			rdp->passed_quiesce = 0;
 			rdp->qs_pending = 0;
-			trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpuonl");
+			trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuonl"));
 		}
 		raw_spin_unlock(&rnp->lock); /* irqs already disabled. */
 		rnp = rnp->parent;
@@ -2980,7 +2994,7 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
 	struct rcu_node *rnp = rdp->mynode;
 	struct rcu_state *rsp;
 
-	trace_rcu_utilization("Start CPU hotplug");
+	trace_rcu_utilization(TPS("Start CPU hotplug"));
 	switch (action) {
 	case CPU_UP_PREPARE:
 	case CPU_UP_PREPARE_FROZEN:
@@ -3009,7 +3023,7 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
 	default:
 		break;
 	}
-	trace_rcu_utilization("End CPU hotplug");
+	trace_rcu_utilization(TPS("End CPU hotplug"));
 	return NOTIFY_OK;
 }
 
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 5cdb559..59fe9e2 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -167,7 +167,7 @@ static void rcu_preempt_qs(int cpu)
 	struct rcu_data *rdp = &per_cpu(rcu_preempt_data, cpu);
 
 	if (rdp->passed_quiesce == 0)
-		trace_rcu_grace_period("rcu_preempt", rdp->gpnum, "cpuqs");
+		trace_rcu_grace_period(TPS("rcu_preempt"), rdp->gpnum, TPS("cpuqs"));
 	rdp->passed_quiesce = 1;
 	current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS;
 }
@@ -386,7 +386,7 @@ void rcu_read_unlock_special(struct task_struct *t)
 		np = rcu_next_node_entry(t, rnp);
 		list_del_init(&t->rcu_node_entry);
 		t->rcu_blocked_node = NULL;
-		trace_rcu_unlock_preempted_task("rcu_preempt",
+		trace_rcu_unlock_preempted_task(TPS("rcu_preempt"),
 						rnp->gpnum, t->pid);
 		if (&t->rcu_node_entry == rnp->gp_tasks)
 			rnp->gp_tasks = np;
@@ -410,7 +410,7 @@ void rcu_read_unlock_special(struct task_struct *t)
 		 */
 		empty_exp_now = !rcu_preempted_readers_exp(rnp);
 		if (!empty && !rcu_preempt_blocked_readers_cgp(rnp)) {
-			trace_rcu_quiescent_state_report("preempt_rcu",
+			trace_rcu_quiescent_state_report(TPS("preempt_rcu"),
 							 rnp->gpnum,
 							 0, rnp->qsmask,
 							 rnp->level,
@@ -1248,12 +1248,12 @@ static int rcu_boost_kthread(void *arg)
 	int spincnt = 0;
 	int more2boost;
 
-	trace_rcu_utilization("Start boost kthread@init");
+	trace_rcu_utilization(TPS("Start boost kthread@init"));
 	for (;;) {
 		rnp->boost_kthread_status = RCU_KTHREAD_WAITING;
-		trace_rcu_utilization("End boost kthread@rcu_wait");
+		trace_rcu_utilization(TPS("End boost kthread@rcu_wait"));
 		rcu_wait(rnp->boost_tasks || rnp->exp_tasks);
-		trace_rcu_utilization("Start boost kthread@rcu_wait");
+		trace_rcu_utilization(TPS("Start boost kthread@rcu_wait"));
 		rnp->boost_kthread_status = RCU_KTHREAD_RUNNING;
 		more2boost = rcu_boost(rnp);
 		if (more2boost)
@@ -1262,14 +1262,14 @@ static int rcu_boost_kthread(void *arg)
 			spincnt = 0;
 		if (spincnt > 10) {
 			rnp->boost_kthread_status = RCU_KTHREAD_YIELDING;
-			trace_rcu_utilization("End boost kthread@rcu_yield");
+			trace_rcu_utilization(TPS("End boost kthread@rcu_yield"));
 			schedule_timeout_interruptible(2);
-			trace_rcu_utilization("Start boost kthread@rcu_yield");
+			trace_rcu_utilization(TPS("Start boost kthread@rcu_yield"));
 			spincnt = 0;
 		}
 	}
 	/* NOTREACHED */
-	trace_rcu_utilization("End boost kthread@notreached");
+	trace_rcu_utilization(TPS("End boost kthread@notreached"));
 	return 0;
 }
 
@@ -1417,7 +1417,7 @@ static void rcu_cpu_kthread(unsigned int cpu)
 	int spincnt;
 
 	for (spincnt = 0; spincnt < 10; spincnt++) {
-		trace_rcu_utilization("Start CPU kthread@rcu_wait");
+		trace_rcu_utilization(TPS("Start CPU kthread@rcu_wait"));
 		local_bh_disable();
 		*statusp = RCU_KTHREAD_RUNNING;
 		this_cpu_inc(rcu_cpu_kthread_loops);
@@ -1429,15 +1429,15 @@ static void rcu_cpu_kthread(unsigned int cpu)
 			rcu_kthread_do_work();
 		local_bh_enable();
 		if (*workp == 0) {
-			trace_rcu_utilization("End CPU kthread@rcu_wait");
+			trace_rcu_utilization(TPS("End CPU kthread@rcu_wait"));
 			*statusp = RCU_KTHREAD_WAITING;
 			return;
 		}
 	}
 	*statusp = RCU_KTHREAD_YIELDING;
-	trace_rcu_utilization("Start CPU kthread@rcu_yield");
+	trace_rcu_utilization(TPS("Start CPU kthread@rcu_yield"));
 	schedule_timeout_interruptible(2);
-	trace_rcu_utilization("End CPU kthread@rcu_yield");
+	trace_rcu_utilization(TPS("End CPU kthread@rcu_yield"));
 	*statusp = RCU_KTHREAD_WAITING;
 }
 
@@ -2200,7 +2200,7 @@ static void rcu_nocb_wait_gp(struct rcu_data *rdp)
 	 * Wait for the grace period.  Do so interruptibly to avoid messing
 	 * up the load average.
 	 */
-	trace_rcu_future_gp(rnp, rdp, c, "StartWait");
+	trace_rcu_future_gp(rnp, rdp, c, TPS("StartWait"));
 	for (;;) {
 		wait_event_interruptible(
 			rnp->nocb_gp_wq[c & 0x1],
@@ -2208,9 +2208,9 @@ static void rcu_nocb_wait_gp(struct rcu_data *rdp)
 		if (likely(d))
 			break;
 		flush_signals(current);
-		trace_rcu_future_gp(rnp, rdp, c, "ResumeWait");
+		trace_rcu_future_gp(rnp, rdp, c, TPS("ResumeWait"));
 	}
-	trace_rcu_future_gp(rnp, rdp, c, "EndWait");
+	trace_rcu_future_gp(rnp, rdp, c, TPS("EndWait"));
 	smp_mb(); /* Ensure that CB invocation happens after GP end. */
 }
 
-- 
1.7.10.4