[deliverable/linux.git] / include / linux / tracepoint.h

#ifndef _LINUX_TRACEPOINT_H
#define _LINUX_TRACEPOINT_H

/*
 * Kernel Tracepoint API.
 *
 * See Documentation/trace/tracepoints.txt.
 *
 * Copyright (C) 2008-2014 Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
 *
 * Heavily inspired from the Linux Kernel Markers.
 *
 * This file is released under the GPLv2.
 * See the file COPYING for more details.
 */

#include <linux/errno.h>
#include <linux/types.h>
#include <linux/rcupdate.h>
#include <linux/static_key.h>

struct module;
struct tracepoint;
struct notifier_block;

struct tracepoint_func {
	void *func;
	void *data;
	int prio;
};

struct tracepoint {
	const char *name;		/* Tracepoint name */
	struct static_key key;
	void (*regfunc)(void);
	void (*unregfunc)(void);
	struct tracepoint_func __rcu *funcs;
};

struct trace_enum_map {
	const char		*system;
	const char		*enum_string;
	unsigned long		enum_value;
};

#define TRACEPOINT_DEFAULT_PRIO	10

extern int
tracepoint_probe_register(struct tracepoint *tp, void *probe, void *data);
extern int
tracepoint_probe_register_prio(struct tracepoint *tp, void *probe, void *data,
			       int prio);
extern int
tracepoint_probe_unregister(struct tracepoint *tp, void *probe, void *data);
extern void
for_each_kernel_tracepoint(void (*fct)(struct tracepoint *tp, void *priv),
		void *priv);

#ifdef CONFIG_MODULES
struct tp_module {
	struct list_head list;
	struct module *mod;
};

bool trace_module_has_bad_taint(struct module *mod);
extern int register_tracepoint_module_notifier(struct notifier_block *nb);
extern int unregister_tracepoint_module_notifier(struct notifier_block *nb);
#else
static inline bool trace_module_has_bad_taint(struct module *mod)
{
	return false;
}
static inline
int register_tracepoint_module_notifier(struct notifier_block *nb)
{
	return 0;
}
static inline
int unregister_tracepoint_module_notifier(struct notifier_block *nb)
{
	return 0;
}
#endif /* CONFIG_MODULES */

/*
 * tracepoint_synchronize_unregister must be called between the last tracepoint
 * probe unregistration and the end of module exit to make sure there is no
 * caller executing a probe when it is freed.
 */
static inline void tracepoint_synchronize_unregister(void)
{
	synchronize_sched();
}

#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS
extern void syscall_regfunc(void);
extern void syscall_unregfunc(void);
#endif /* CONFIG_HAVE_SYSCALL_TRACEPOINTS */

#define PARAMS(args...) args

#define TRACE_DEFINE_ENUM(x)

#endif /* _LINUX_TRACEPOINT_H */

/*
 * Note: we keep the TRACE_EVENT and DECLARE_TRACE outside the include
 *  file ifdef protection.
 *  This is due to the way trace events work. If a file includes two
 *  trace event headers under one "CREATE_TRACE_POINTS" the first include
 *  will override the TRACE_EVENT and break the second include.
 */

#ifndef DECLARE_TRACE

#define TP_PROTO(args...)	args
#define TP_ARGS(args...)	args
#define TP_CONDITION(args...)	args

/*
 * Individual subsystem my have a separate configuration to
 * enable their tracepoints. By default, this file will create
 * the tracepoints if CONFIG_TRACEPOINT is defined. If a subsystem
 * wants to be able to disable its tracepoints from being created
 * it can define NOTRACE before including the tracepoint headers.
 */
#if defined(CONFIG_TRACEPOINTS) && !defined(NOTRACE)
#define TRACEPOINTS_ENABLED
#endif

#ifdef TRACEPOINTS_ENABLED

/*
 * it_func[0] is never NULL because there is at least one element in the array
 * when the array itself is non NULL.
 *
 * Note, the proto and args passed in includes "__data" as the first parameter.
 * The reason for this is to handle the "void" prototype. If a tracepoint
 * has a "void" prototype, then it is invalid to declare a function
 * as "(void *, void)". The DECLARE_TRACE_NOARGS() will pass in just
 * "void *data", where as the DECLARE_TRACE() will pass in "void *data, proto".
 */
#define __DO_TRACE(tp, proto, args, cond, prercu, postrcu)		\
	do {								\
		struct tracepoint_func *it_func_ptr;			\
		void *it_func;						\
		void *__data;						\
									\
		if (!(cond))						\
			return;						\
		prercu;							\
		rcu_read_lock_sched_notrace();				\
		it_func_ptr = rcu_dereference_sched((tp)->funcs);	\
		if (it_func_ptr) {					\
			do {						\
				it_func = (it_func_ptr)->func;		\
				__data = (it_func_ptr)->data;		\
				((void(*)(proto))(it_func))(args);	\
			} while ((++it_func_ptr)->func);		\
		}							\
		rcu_read_unlock_sched_notrace();			\
		postrcu;						\
	} while (0)

#ifndef MODULE
#define __DECLARE_TRACE_RCU(name, proto, args, cond, data_proto, data_args)	\
	static inline void trace_##name##_rcuidle(proto)		\
	{								\
		if (static_key_false(&__tracepoint_##name.key))		\
			__DO_TRACE(&__tracepoint_##name,		\
				TP_PROTO(data_proto),			\
				TP_ARGS(data_args),			\
				TP_CONDITION(cond),			\
				rcu_irq_enter(),			\
				rcu_irq_exit());			\
	}
#else
#define __DECLARE_TRACE_RCU(name, proto, args, cond, data_proto, data_args)
#endif

/*
 * Make sure the alignment of the structure in the __tracepoints section will
 * not add unwanted padding between the beginning of the section and the
 * structure. Force alignment to the same alignment as the section start.
 *
 * When lockdep is enabled, we make sure to always do the RCU portions of
 * the tracepoint code, regardless of whether tracing is on. However,
 * don't check if the condition is false, due to interaction with idle
 * instrumentation. This lets us find RCU issues triggered with tracepoints
 * even when this tracepoint is off. This code has no purpose other than
 * poking RCU a bit.
 */
#define __DECLARE_TRACE(name, proto, args, cond, data_proto, data_args) \
	extern struct tracepoint __tracepoint_##name;			\
	static inline void trace_##name(proto)				\
	{								\
		if (static_key_false(&__tracepoint_##name.key))		\
			__DO_TRACE(&__tracepoint_##name,		\
				TP_PROTO(data_proto),			\
				TP_ARGS(data_args),			\
				TP_CONDITION(cond),,);			\
		if (IS_ENABLED(CONFIG_LOCKDEP) && (cond)) {		\
			rcu_read_lock_sched_notrace();			\
			rcu_dereference_sched(__tracepoint_##name.funcs);\
			rcu_read_unlock_sched_notrace();		\
		}							\
	}								\
	__DECLARE_TRACE_RCU(name, PARAMS(proto), PARAMS(args),		\
		PARAMS(cond), PARAMS(data_proto), PARAMS(data_args))	\
	static inline int						\
	register_trace_##name(void (*probe)(data_proto), void *data)	\
	{								\
		return tracepoint_probe_register(&__tracepoint_##name,	\
						(void *)probe, data);	\
	}								\
	static inline int						\
	register_trace_prio_##name(void (*probe)(data_proto), void *data,\
				   int prio)				\
	{								\
		return tracepoint_probe_register_prio(&__tracepoint_##name, \
					      (void *)probe, data, prio); \
	}								\
	static inline int						\
	unregister_trace_##name(void (*probe)(data_proto), void *data)	\
	{								\
		return tracepoint_probe_unregister(&__tracepoint_##name,\
						(void *)probe, data);	\
	}								\
	static inline void						\
	check_trace_callback_type_##name(void (*cb)(data_proto))	\
	{								\
	}								\
	static inline bool						\
	trace_##name##_enabled(void)					\
	{								\
		return static_key_false(&__tracepoint_##name.key);	\
	}

/*
 * We have no guarantee that gcc and the linker won't up-align the tracepoint
 * structures, so we create an array of pointers that will be used for iteration
 * on the tracepoints.
 */
#define DEFINE_TRACE_FN(name, reg, unreg)				 \
	static const char __tpstrtab_##name[]				 \
	__attribute__((section("__tracepoints_strings"))) = #name;	 \
	struct tracepoint __tracepoint_##name				 \
	__attribute__((section("__tracepoints"))) =			 \
		{ __tpstrtab_##name, STATIC_KEY_INIT_FALSE, reg, unreg, NULL };\
	static struct tracepoint * const __tracepoint_ptr_##name __used	 \
	__attribute__((section("__tracepoints_ptrs"))) =		 \
		&__tracepoint_##name;

#define DEFINE_TRACE(name)						\
	DEFINE_TRACE_FN(name, NULL, NULL);

#define EXPORT_TRACEPOINT_SYMBOL_GPL(name)				\
	EXPORT_SYMBOL_GPL(__tracepoint_##name)
#define EXPORT_TRACEPOINT_SYMBOL(name)					\
	EXPORT_SYMBOL(__tracepoint_##name)

#else /* !TRACEPOINTS_ENABLED */
#define __DECLARE_TRACE(name, proto, args, cond, data_proto, data_args) \
	static inline void trace_##name(proto)				\
	{ }								\
	static inline void trace_##name##_rcuidle(proto)		\
	{ }								\
	static inline int						\
	register_trace_##name(void (*probe)(data_proto),		\
			      void *data)				\
	{								\
		return -ENOSYS;						\
	}								\
	static inline int						\
	unregister_trace_##name(void (*probe)(data_proto),		\
				void *data)				\
	{								\
		return -ENOSYS;						\
	}								\
	static inline void check_trace_callback_type_##name(void (*cb)(data_proto)) \
	{								\
	}								\
	static inline bool						\
	trace_##name##_enabled(void)					\
	{								\
		return false;						\
	}

#define DEFINE_TRACE_FN(name, reg, unreg)
#define DEFINE_TRACE(name)
#define EXPORT_TRACEPOINT_SYMBOL_GPL(name)
#define EXPORT_TRACEPOINT_SYMBOL(name)

#endif /* TRACEPOINTS_ENABLED */

#ifdef CONFIG_TRACING
/**
 * tracepoint_string - register constant persistent string to trace system
 * @str - a constant persistent string that will be referenced in tracepoints
 *
 * If constant strings are being used in tracepoints, it is faster and
 * more efficient to just save the pointer to the string and reference
 * that with a printf "%s" instead of saving the string in the ring buffer
 * and wasting space and time.
 *
 * The problem with the above approach is that userspace tools that read
 * the binary output of the trace buffers do not have access to the string.
 * Instead they just show the address of the string which is not very
 * useful to users.
 *
 * With tracepoint_string(), the string will be registered to the tracing
 * system and exported to userspace via the debugfs/tracing/printk_formats
 * file that maps the string address to the string text. This way userspace
 * tools that read the binary buffers have a way to map the pointers to
 * the ASCII strings they represent.
 *
 * The @str used must be a constant string and persistent as it would not
 * make sense to show a string that no longer exists. But it is still fine
 * to be used with modules, because when modules are unloaded, if they
 * had tracepoints, the ring buffers are cleared too. As long as the string
 * does not change during the life of the module, it is fine to use
 * tracepoint_string() within a module.
 */
#define tracepoint_string(str)						\
	({								\
		static const char *___tp_str __tracepoint_string = str; \
		___tp_str;						\
	})
#define __tracepoint_string	__attribute__((section("__tracepoint_str")))
#else
/*
 * tracepoint_string() is used to save the string address for userspace
 * tracing tools. When tracing isn't configured, there's no need to save
 * anything.
 */
# define tracepoint_string(str) str
# define __tracepoint_string
#endif

/*
 * The need for the DECLARE_TRACE_NOARGS() is to handle the prototype
 * (void). "void" is a special value in a function prototype and can
 * not be combined with other arguments. Since the DECLARE_TRACE()
 * macro adds a data element at the beginning of the prototype,
 * we need a way to differentiate "(void *data, proto)" from
 * "(void *data, void)". The second prototype is invalid.
 *
 * DECLARE_TRACE_NOARGS() passes "void" as the tracepoint prototype
 * and "void *__data" as the callback prototype.
 *
 * DECLARE_TRACE() passes "proto" as the tracepoint protoype and
 * "void *__data, proto" as the callback prototype.
 */
#define DECLARE_TRACE_NOARGS(name)					\
		__DECLARE_TRACE(name, void, , 1, void *__data, __data)

#define DECLARE_TRACE(name, proto, args)				\
		__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), 1,	\
				PARAMS(void *__data, proto),		\
				PARAMS(__data, args))

#define DECLARE_TRACE_CONDITION(name, proto, args, cond)		\
	__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), PARAMS(cond), \
			PARAMS(void *__data, proto),			\
			PARAMS(__data, args))

#define TRACE_EVENT_FLAGS(event, flag)

#define TRACE_EVENT_PERF_PERM(event, expr...)

#endif /* DECLARE_TRACE */

#ifndef TRACE_EVENT
/*
 * For use with the TRACE_EVENT macro:
 *
 * We define a tracepoint, its arguments, its printk format
 * and its 'fast binary record' layout.
 *
 * Firstly, name your tracepoint via TRACE_EVENT(name : the
 * 'subsystem_event' notation is fine.
 *
 * Think about this whole construct as the
 * 'trace_sched_switch() function' from now on.
 *
 *
 *  TRACE_EVENT(sched_switch,
 *
 *	*
 *	* A function has a regular function arguments
 *	* prototype, declare it via TP_PROTO():
 *	*
 *
 *	TP_PROTO(struct rq *rq, struct task_struct *prev,
 *		 struct task_struct *next),
 *
 *	*
 *	* Define the call signature of the 'function'.
 *	* (Design sidenote: we use this instead of a
 *	*  TP_PROTO1/TP_PROTO2/TP_PROTO3 ugliness.)
 *	*
 *
 *	TP_ARGS(rq, prev, next),
 *
 *	*
 *	* Fast binary tracing: define the trace record via
 *	* TP_STRUCT__entry(). You can think about it like a
 *	* regular C structure local variable definition.
 *	*
 *	* This is how the trace record is structured and will
 *	* be saved into the ring buffer. These are the fields
 *	* that will be exposed to user-space in
 *	* /sys/kernel/debug/tracing/events/<*>/format.
 *	*
 *	* The declared 'local variable' is called '__entry'
 *	*
 *	* __field(pid_t, prev_prid) is equivalent to a standard declariton:
 *	*
 *	*	pid_t	prev_pid;
 *	*
 *	* __array(char, prev_comm, TASK_COMM_LEN) is equivalent to:
 *	*
 *	*	char	prev_comm[TASK_COMM_LEN];
 *	*
 *
 *	TP_STRUCT__entry(
 *		__array(	char,	prev_comm,	TASK_COMM_LEN	)
 *		__field(	pid_t,	prev_pid			)
 *		__field(	int,	prev_prio			)
 *		__array(	char,	next_comm,	TASK_COMM_LEN	)
 *		__field(	pid_t,	next_pid			)
 *		__field(	int,	next_prio			)
 *	),
 *
 *	*
 *	* Assign the entry into the trace record, by embedding
 *	* a full C statement block into TP_fast_assign(). You
 *	* can refer to the trace record as '__entry' -
 *	* otherwise you can put arbitrary C code in here.
 *	*
 *	* Note: this C code will execute every time a trace event
 *	* happens, on an active tracepoint.
 *	*
 *
 *	TP_fast_assign(
 *		memcpy(__entry->next_comm, next->comm, TASK_COMM_LEN);
 *		__entry->prev_pid	= prev->pid;
 *		__entry->prev_prio	= prev->prio;
 *		memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
 *		__entry->next_pid	= next->pid;
 *		__entry->next_prio	= next->prio;
 *	),
 *
 *	*
 *	* Formatted output of a trace record via TP_printk().
 *	* This is how the tracepoint will appear under ftrace
 *	* plugins that make use of this tracepoint.
 *	*
 *	* (raw-binary tracing wont actually perform this step.)
 *	*
 *
 *	TP_printk("task %s:%d [%d] ==> %s:%d [%d]",
 *		__entry->prev_comm, __entry->prev_pid, __entry->prev_prio,
 *		__entry->next_comm, __entry->next_pid, __entry->next_prio),
 *
 * );
 *
 * This macro construct is thus used for the regular printk format
 * tracing setup, it is used to construct a function pointer based
 * tracepoint callback (this is used by programmatic plugins and
 * can also by used by generic instrumentation like SystemTap), and
 * it is also used to expose a structured trace record in
 * /sys/kernel/debug/tracing/events/.
 *
 * A set of (un)registration functions can be passed to the variant
 * TRACE_EVENT_FN to perform any (un)registration work.
 */

#define DECLARE_EVENT_CLASS(name, proto, args, tstruct, assign, print)
#define DEFINE_EVENT(template, name, proto, args)		\
	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
#define DEFINE_EVENT_FN(template, name, proto, args, reg, unreg)\
	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
#define DEFINE_EVENT_PRINT(template, name, proto, args, print)	\
	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
#define DEFINE_EVENT_CONDITION(template, name, proto,		\
			       args, cond)			\
	DECLARE_TRACE_CONDITION(name, PARAMS(proto),		\
				PARAMS(args), PARAMS(cond))

#define TRACE_EVENT(name, proto, args, struct, assign, print)	\
	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
#define TRACE_EVENT_FN(name, proto, args, struct,		\
		assign, print, reg, unreg)			\
	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
#define TRACE_EVENT_CONDITION(name, proto, args, cond,		\
			      struct, assign, print)		\
	DECLARE_TRACE_CONDITION(name, PARAMS(proto),		\
				PARAMS(args), PARAMS(cond))

#define TRACE_EVENT_FLAGS(event, flag)

#define TRACE_EVENT_PERF_PERM(event, expr...)

#endif /* ifdef TRACE_EVENT (see note above) */
Commit	Line	Data
97e1c18e MD	1	#ifndef _LINUX_TRACEPOINT_H
	2	#define _LINUX_TRACEPOINT_H
	3
	4	/*
	5	* Kernel Tracepoint API.
	6	*
8cd09a59	7	* See Documentation/trace/tracepoints.txt.
97e1c18e	8	*
de7b2973	9	* Copyright (C) 2008-2014 Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
97e1c18e MD	10	*
	11	* Heavily inspired from the Linux Kernel Markers.
	12	*
	13	* This file is released under the GPLv2.
	14	* See the file COPYING for more details.
	15	*/
	16
b70e4f05	17	#include <linux/errno.h>
97e1c18e MD	18	#include <linux/types.h>
97e1c18e MD	19	#include <linux/rcupdate.h>
c5905afb	20	#include <linux/static_key.h>
97e1c18e MD	21
	22	struct module;
	23	struct tracepoint;
de7b2973	24	struct notifier_block;
97e1c18e	25
38516ab5 SR	26	struct tracepoint_func {
	27	void *func;
	28	void *data;
7904b5c4	29	int prio;
38516ab5 SR	30	};
38516ab5 SR	31
97e1c18e MD	32	struct tracepoint {
97e1c18e MD	33	const char name; / Tracepoint name */
c5905afb	34	struct static_key key;
97419875 JS	35	void (*regfunc)(void);
97419875 JS	36	void (*unregfunc)(void);
bd1c8b22	37	struct tracepoint_func __rcu *funcs;
65498646	38	};
97e1c18e	39
0c564a53 SRRH	40	struct trace_enum_map {
	41	const char *system;
	42	const char *enum_string;
	43	unsigned long enum_value;
	44	};
	45
7904b5c4 SRRH	46	#define TRACEPOINT_DEFAULT_PRIO 10
7904b5c4 SRRH	47
38516ab5	48	extern int
de7b2973 MD	49	tracepoint_probe_register(struct tracepoint tp, void probe, void *data);
de7b2973 MD	50	extern int
7904b5c4 SRRH	51	tracepoint_probe_register_prio(struct tracepoint tp, void probe, void *data,
	52	int prio);
	53	extern int
de7b2973 MD	54	tracepoint_probe_unregister(struct tracepoint tp, void probe, void *data);
	55	extern void
	56	for_each_kernel_tracepoint(void (fct)(struct tracepoint tp, void *priv),
	57	void *priv);
2e26ca71	58
b75ef8b4 MD	59	#ifdef CONFIG_MODULES
	60	struct tp_module {
	61	struct list_head list;
eb7d035c	62	struct module *mod;
b75ef8b4	63	};
de7b2973	64
45ab2813	65	bool trace_module_has_bad_taint(struct module *mod);
de7b2973 MD	66	extern int register_tracepoint_module_notifier(struct notifier_block *nb);
de7b2973 MD	67	extern int unregister_tracepoint_module_notifier(struct notifier_block *nb);
45ab2813 SRRH	68	#else
	69	static inline bool trace_module_has_bad_taint(struct module *mod)
	70	{
	71	return false;
	72	}
de7b2973 MD	73	static inline
	74	int register_tracepoint_module_notifier(struct notifier_block *nb)
	75	{
	76	return 0;
	77	}
	78	static inline
	79	int unregister_tracepoint_module_notifier(struct notifier_block *nb)
	80	{
	81	return 0;
	82	}
b75ef8b4 MD	83	#endif /* CONFIG_MODULES */
b75ef8b4 MD	84
2e26ca71 SR	85	/*
	86	* tracepoint_synchronize_unregister must be called between the last tracepoint
	87	* probe unregistration and the end of module exit to make sure there is no
	88	* caller executing a probe when it is freed.
	89	*/
	90	static inline void tracepoint_synchronize_unregister(void)
	91	{
	92	synchronize_sched();
	93	}
	94
b725dfea MD	95	#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS
	96	extern void syscall_regfunc(void);
	97	extern void syscall_unregfunc(void);
	98	#endif /* CONFIG_HAVE_SYSCALL_TRACEPOINTS */
	99
2e26ca71 SR	100	#define PARAMS(args...) args
2e26ca71 SR	101
0c564a53 SRRH	102	#define TRACE_DEFINE_ENUM(x)
0c564a53 SRRH	103
2e26ca71 SR	104	#endif /* _LINUX_TRACEPOINT_H */
	105
	106	/*
	107	* Note: we keep the TRACE_EVENT and DECLARE_TRACE outside the include
	108	* file ifdef protection.
	109	* This is due to the way trace events work. If a file includes two
	110	* trace event headers under one "CREATE_TRACE_POINTS" the first include
	111	* will override the TRACE_EVENT and break the second include.
	112	*/
	113
ea20d929 SR	114	#ifndef DECLARE_TRACE
ea20d929 SR	115
2939b046	116	#define TP_PROTO(args...) args
8cd09a59	117	#define TP_ARGS(args...) args
287050d3	118	#define TP_CONDITION(args...) args
97e1c18e	119
c63b7682 TS	120	/*
	121	* Individual subsystem my have a separate configuration to
	122	* enable their tracepoints. By default, this file will create
	123	* the tracepoints if CONFIG_TRACEPOINT is defined. If a subsystem
	124	* wants to be able to disable its tracepoints from being created
	125	* it can define NOTRACE before including the tracepoint headers.
	126	*/
	127	#if defined(CONFIG_TRACEPOINTS) && !defined(NOTRACE)
	128	#define TRACEPOINTS_ENABLED
	129	#endif
	130
	131	#ifdef TRACEPOINTS_ENABLED
97e1c18e MD	132
	133	/*
	134	* it_func[0] is never NULL because there is at least one element in the array
	135	* when the array itself is non NULL.
38516ab5 SR	136	*
	137	* Note, the proto and args passed in includes "__data" as the first parameter.
	138	* The reason for this is to handle the "void" prototype. If a tracepoint
	139	* has a "void" prototype, then it is invalid to declare a function
	140	* as "(void *, void)". The DECLARE_TRACE_NOARGS() will pass in just
	141	* "void data", where as the DECLARE_TRACE() will pass in "void data, proto".
97e1c18e	142	*/
2fbb90db	143	#define __DO_TRACE(tp, proto, args, cond, prercu, postrcu) \
97e1c18e	144	do { \
38516ab5 SR	145	struct tracepoint_func *it_func_ptr; \
	146	void *it_func; \
	147	void *__data; \
97e1c18e	148	\
287050d3 SR	149	if (!(cond)) \
287050d3 SR	150	return; \
2fbb90db	151	prercu; \
da7b3eab	152	rcu_read_lock_sched_notrace(); \
38516ab5 SR	153	it_func_ptr = rcu_dereference_sched((tp)->funcs); \
38516ab5 SR	154	if (it_func_ptr) { \
97e1c18e	155	do { \
38516ab5 SR	156	it_func = (it_func_ptr)->func; \
	157	__data = (it_func_ptr)->data; \
	158	((void(*)(proto))(it_func))(args); \
	159	} while ((++it_func_ptr)->func); \
97e1c18e	160	} \
da7b3eab	161	rcu_read_unlock_sched_notrace(); \
2fbb90db	162	postrcu; \
97e1c18e MD	163	} while (0)
97e1c18e MD	164
7ece55a4 JT	165	#ifndef MODULE
	166	#define __DECLARE_TRACE_RCU(name, proto, args, cond, data_proto, data_args) \
	167	static inline void trace_##name##_rcuidle(proto) \
	168	{ \
	169	if (static_key_false(&__tracepoint_##name.key)) \
	170	__DO_TRACE(&__tracepoint_##name, \
	171	TP_PROTO(data_proto), \
	172	TP_ARGS(data_args), \
	173	TP_CONDITION(cond), \
d6284099 PM	174	rcu_irq_enter(), \
d6284099 PM	175	rcu_irq_exit()); \
7ece55a4 JT	176	}
	177	#else
	178	#define __DECLARE_TRACE_RCU(name, proto, args, cond, data_proto, data_args)
	179	#endif
	180
97e1c18e MD	181	/*
	182	* Make sure the alignment of the structure in the __tracepoints section will
	183	* not add unwanted padding between the beginning of the section and the
	184	* structure. Force alignment to the same alignment as the section start.
3a630178 DH	185	*
3a630178 DH	186	* When lockdep is enabled, we make sure to always do the RCU portions of
a15920be MD	187	* the tracepoint code, regardless of whether tracing is on. However,
	188	* don't check if the condition is false, due to interaction with idle
	189	* instrumentation. This lets us find RCU issues triggered with tracepoints
	190	* even when this tracepoint is off. This code has no purpose other than
	191	* poking RCU a bit.
97e1c18e	192	*/
2fbb90db	193	#define __DECLARE_TRACE(name, proto, args, cond, data_proto, data_args) \
7e066fb8	194	extern struct tracepoint __tracepoint_##name; \
97e1c18e MD	195	static inline void trace_##name(proto) \
97e1c18e MD	196	{ \
c5905afb	197	if (static_key_false(&__tracepoint_##name.key)) \
97e1c18e	198	__DO_TRACE(&__tracepoint_##name, \
38516ab5	199	TP_PROTO(data_proto), \
287050d3	200	TP_ARGS(data_args), \
2fbb90db	201	TP_CONDITION(cond),,); \
a05d59a5	202	if (IS_ENABLED(CONFIG_LOCKDEP) && (cond)) { \
3a630178 DH	203	rcu_read_lock_sched_notrace(); \
	204	rcu_dereference_sched(__tracepoint_##name.funcs);\
	205	rcu_read_unlock_sched_notrace(); \
	206	} \
2fbb90db	207	} \
7ece55a4 JT	208	__DECLARE_TRACE_RCU(name, PARAMS(proto), PARAMS(args), \
7ece55a4 JT	209	PARAMS(cond), PARAMS(data_proto), PARAMS(data_args)) \
38516ab5 SR	210	static inline int \
38516ab5 SR	211	register_trace_##name(void (probe)(data_proto), void data) \
97e1c18e	212	{ \
de7b2973 MD	213	return tracepoint_probe_register(&__tracepoint_##name, \
de7b2973 MD	214	(void *)probe, data); \
97e1c18e	215	} \
38516ab5	216	static inline int \
7904b5c4 SRRH	217	register_trace_prio_##name(void (probe)(data_proto), void data,\
	218	int prio) \
	219	{ \
	220	return tracepoint_probe_register_prio(&__tracepoint_##name, \
	221	(void *)probe, data, prio); \
	222	} \
	223	static inline int \
38516ab5	224	unregister_trace_##name(void (probe)(data_proto), void data) \
97e1c18e	225	{ \
de7b2973 MD	226	return tracepoint_probe_unregister(&__tracepoint_##name,\
de7b2973 MD	227	(void *)probe, data); \
53da59aa	228	} \
38516ab5 SR	229	static inline void \
38516ab5 SR	230	check_trace_callback_type_##name(void (*cb)(data_proto)) \
53da59aa	231	{ \
7c65bbc7 SRRH	232	} \
	233	static inline bool \
	234	trace_##name##_enabled(void) \
	235	{ \
	236	return static_key_false(&__tracepoint_##name.key); \
97e1c18e MD	237	}
97e1c18e MD	238
65498646 MD	239	/*
	240	* We have no guarantee that gcc and the linker won't up-align the tracepoint
	241	* structures, so we create an array of pointers that will be used for iteration
	242	* on the tracepoints.
	243	*/
d430d3d7 JB	244	#define DEFINE_TRACE_FN(name, reg, unreg) \
	245	static const char __tpstrtab_##name[] \
	246	__attribute__((section("__tracepoints_strings"))) = #name; \
	247	struct tracepoint __tracepoint_##name \
	248	__attribute__((section("__tracepoints"))) = \
c5905afb	249	{ __tpstrtab_##name, STATIC_KEY_INIT_FALSE, reg, unreg, NULL };\
d430d3d7 JB	250	static struct tracepoint * const __tracepoint_ptr_##name __used \
d430d3d7 JB	251	__attribute__((section("__tracepoints_ptrs"))) = \
65498646	252	&__tracepoint_##name;
97419875 JS	253
	254	#define DEFINE_TRACE(name) \
	255	DEFINE_TRACE_FN(name, NULL, NULL);
7e066fb8 MD	256
	257	#define EXPORT_TRACEPOINT_SYMBOL_GPL(name) \
	258	EXPORT_SYMBOL_GPL(__tracepoint_##name)
	259	#define EXPORT_TRACEPOINT_SYMBOL(name) \
	260	EXPORT_SYMBOL(__tracepoint_##name)
	261
c63b7682	262	#else /* !TRACEPOINTS_ENABLED */
2fbb90db	263	#define __DECLARE_TRACE(name, proto, args, cond, data_proto, data_args) \
97e1c18e MD	264	static inline void trace_##name(proto) \
97e1c18e MD	265	{ } \
2fbb90db SR	266	static inline void trace_##name##_rcuidle(proto) \
2fbb90db SR	267	{ } \
38516ab5 SR	268	static inline int \
	269	register_trace_##name(void (*probe)(data_proto), \
	270	void *data) \
97e1c18e MD	271	{ \
	272	return -ENOSYS; \
	273	} \
38516ab5 SR	274	static inline int \
	275	unregister_trace_##name(void (*probe)(data_proto), \
	276	void *data) \
c420970e MD	277	{ \
c420970e MD	278	return -ENOSYS; \
53da59aa	279	} \
38516ab5	280	static inline void check_trace_callback_type_##name(void (*cb)(data_proto)) \
53da59aa	281	{ \
7c65bbc7 SRRH	282	} \
	283	static inline bool \
	284	trace_##name##_enabled(void) \
	285	{ \
	286	return false; \
c420970e	287	}
97e1c18e	288
97419875	289	#define DEFINE_TRACE_FN(name, reg, unreg)
7e066fb8 MD	290	#define DEFINE_TRACE(name)
	291	#define EXPORT_TRACEPOINT_SYMBOL_GPL(name)
	292	#define EXPORT_TRACEPOINT_SYMBOL(name)
	293
c63b7682	294	#endif /* TRACEPOINTS_ENABLED */
38516ab5	295
3c49b52b SR	296	#ifdef CONFIG_TRACING
	297	/**
	298	* tracepoint_string - register constant persistent string to trace system
	299	* @str - a constant persistent string that will be referenced in tracepoints
	300	*
	301	* If constant strings are being used in tracepoints, it is faster and
	302	* more efficient to just save the pointer to the string and reference
	303	* that with a printf "%s" instead of saving the string in the ring buffer
	304	* and wasting space and time.
	305	*
	306	* The problem with the above approach is that userspace tools that read
	307	* the binary output of the trace buffers do not have access to the string.
	308	* Instead they just show the address of the string which is not very
	309	* useful to users.
	310	*
	311	* With tracepoint_string(), the string will be registered to the tracing
	312	* system and exported to userspace via the debugfs/tracing/printk_formats
	313	* file that maps the string address to the string text. This way userspace
	314	* tools that read the binary buffers have a way to map the pointers to
	315	* the ASCII strings they represent.
	316	*
	317	* The @str used must be a constant string and persistent as it would not
	318	* make sense to show a string that no longer exists. But it is still fine
	319	* to be used with modules, because when modules are unloaded, if they
	320	* had tracepoints, the ring buffers are cleared too. As long as the string
	321	* does not change during the life of the module, it is fine to use
	322	* tracepoint_string() within a module.
	323	*/
	324	#define tracepoint_string(str) \
	325	({ \
	326	static const char *___tp_str __tracepoint_string = str; \
	327	___tp_str; \
	328	})
	329	#define __tracepoint_string __attribute__((section("__tracepoint_str")))
	330	#else
	331	/*
	332	* tracepoint_string() is used to save the string address for userspace
	333	* tracing tools. When tracing isn't configured, there's no need to save
	334	* anything.
	335	*/
	336	# define tracepoint_string(str) str
	337	# define __tracepoint_string
	338	#endif
	339
38516ab5 SR	340	/*
	341	* The need for the DECLARE_TRACE_NOARGS() is to handle the prototype
	342	* (void). "void" is a special value in a function prototype and can
	343	* not be combined with other arguments. Since the DECLARE_TRACE()
	344	* macro adds a data element at the beginning of the prototype,
	345	* we need a way to differentiate "(void *data, proto)" from
	346	* "(void *data, void)". The second prototype is invalid.
	347	*
	348	* DECLARE_TRACE_NOARGS() passes "void" as the tracepoint prototype
	349	* and "void *__data" as the callback prototype.
	350	*
	351	* DECLARE_TRACE() passes "proto" as the tracepoint protoype and
	352	* "void *__data, proto" as the callback prototype.
	353	*/
	354	#define DECLARE_TRACE_NOARGS(name) \
287050d3	355	__DECLARE_TRACE(name, void, , 1, void *__data, __data)
38516ab5 SR	356
38516ab5 SR	357	#define DECLARE_TRACE(name, proto, args) \
287050d3	358	__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), 1, \
38516ab5 SR	359	PARAMS(void *__data, proto), \
	360	PARAMS(__data, args))
	361
287050d3 SR	362	#define DECLARE_TRACE_CONDITION(name, proto, args, cond) \
	363	__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), PARAMS(cond), \
	364	PARAMS(void *__data, proto), \
	365	PARAMS(__data, args))
	366
1ed0c597 FW	367	#define TRACE_EVENT_FLAGS(event, flag)
1ed0c597 FW	368
d5b5f391 PZ	369	#define TRACE_EVENT_PERF_PERM(event, expr...)
d5b5f391 PZ	370
ea20d929	371	#endif /* DECLARE_TRACE */
97e1c18e	372
ea20d929	373	#ifndef TRACE_EVENT
823f9124 SR	374	/*
	375	* For use with the TRACE_EVENT macro:
	376	*
	377	* We define a tracepoint, its arguments, its printk format
2621bca8	378	* and its 'fast binary record' layout.
823f9124 SR	379	*
	380	* Firstly, name your tracepoint via TRACE_EVENT(name : the
	381	* 'subsystem_event' notation is fine.
	382	*
	383	* Think about this whole construct as the
	384	* 'trace_sched_switch() function' from now on.
	385	*
	386	*
	387	* TRACE_EVENT(sched_switch,
	388	*
	389	* *
	390	* * A function has a regular function arguments
	391	* * prototype, declare it via TP_PROTO():
	392	* *
	393	*
ef18012b SR	394	* TP_PROTO(struct rq rq, struct task_struct prev,
ef18012b SR	395	* struct task_struct *next),
823f9124 SR	396	*
	397	* *
	398	* * Define the call signature of the 'function'.
	399	* * (Design sidenote: we use this instead of a
	400	* * TP_PROTO1/TP_PROTO2/TP_PROTO3 ugliness.)
	401	* *
	402	*
ef18012b	403	* TP_ARGS(rq, prev, next),
823f9124 SR	404	*
	405	* *
	406	* * Fast binary tracing: define the trace record via
	407	* * TP_STRUCT__entry(). You can think about it like a
	408	* * regular C structure local variable definition.
	409	* *
	410	* * This is how the trace record is structured and will
	411	* * be saved into the ring buffer. These are the fields
	412	* * that will be exposed to user-space in
156f5a78	413	* * /sys/kernel/debug/tracing/events/<*>/format.
823f9124 SR	414	* *
	415	* * The declared 'local variable' is called '__entry'
	416	* *
	417	* * __field(pid_t, prev_prid) is equivalent to a standard declariton:
	418	* *
	419	* * pid_t prev_pid;
	420	* *
	421	* * __array(char, prev_comm, TASK_COMM_LEN) is equivalent to:
	422	* *
	423	* * char prev_comm[TASK_COMM_LEN];
	424	* *
	425	*
	426	* TP_STRUCT__entry(
	427	* __array( char, prev_comm, TASK_COMM_LEN )
	428	* __field( pid_t, prev_pid )
	429	* __field( int, prev_prio )
	430	* __array( char, next_comm, TASK_COMM_LEN )
	431	* __field( pid_t, next_pid )
	432	* __field( int, next_prio )
	433	* ),
	434	*
	435	* *
	436	* * Assign the entry into the trace record, by embedding
	437	* * a full C statement block into TP_fast_assign(). You
	438	* * can refer to the trace record as '__entry' -
	439	* * otherwise you can put arbitrary C code in here.
	440	* *
	441	* * Note: this C code will execute every time a trace event
	442	* * happens, on an active tracepoint.
	443	* *
	444	*
ef18012b SR	445	* TP_fast_assign(
	446	* memcpy(__entry->next_comm, next->comm, TASK_COMM_LEN);
	447	* __entry->prev_pid = prev->pid;
	448	* __entry->prev_prio = prev->prio;
823f9124 SR	449	* memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
823f9124 SR	450	* __entry->next_pid = next->pid;
ef18012b	451	* __entry->next_prio = next->prio;
ec6e7c3a	452	* ),
823f9124 SR	453	*
	454	* *
	455	* * Formatted output of a trace record via TP_printk().
	456	* * This is how the tracepoint will appear under ftrace
	457	* * plugins that make use of this tracepoint.
	458	* *
	459	* * (raw-binary tracing wont actually perform this step.)
	460	* *
	461	*
	462	* TP_printk("task %s:%d [%d] ==> %s:%d [%d]",
	463	* __entry->prev_comm, __entry->prev_pid, __entry->prev_prio,
	464	* __entry->next_comm, __entry->next_pid, __entry->next_prio),
	465	*
	466	* );
	467	*
	468	* This macro construct is thus used for the regular printk format
	469	* tracing setup, it is used to construct a function pointer based
	470	* tracepoint callback (this is used by programmatic plugins and
	471	* can also by used by generic instrumentation like SystemTap), and
	472	* it is also used to expose a structured trace record in
156f5a78	473	* /sys/kernel/debug/tracing/events/.
97419875 JS	474	*
	475	* A set of (un)registration functions can be passed to the variant
	476	* TRACE_EVENT_FN to perform any (un)registration work.
823f9124 SR	477	*/
823f9124 SR	478
091ad365	479	#define DECLARE_EVENT_CLASS(name, proto, args, tstruct, assign, print)
ff038f5c SR	480	#define DEFINE_EVENT(template, name, proto, args) \
ff038f5c SR	481	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
f5abaa1b SR	482	#define DEFINE_EVENT_FN(template, name, proto, args, reg, unreg)\
f5abaa1b SR	483	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
e5bc9721 SR	484	#define DEFINE_EVENT_PRINT(template, name, proto, args, print) \
e5bc9721 SR	485	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
287050d3 SR	486	#define DEFINE_EVENT_CONDITION(template, name, proto, \
	487	args, cond) \
	488	DECLARE_TRACE_CONDITION(name, PARAMS(proto), \
	489	PARAMS(args), PARAMS(cond))
ff038f5c	490
30a8fecc	491	#define TRACE_EVENT(name, proto, args, struct, assign, print) \
da4d0302	492	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
97419875 JS	493	#define TRACE_EVENT_FN(name, proto, args, struct, \
	494	assign, print, reg, unreg) \
	495	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
287050d3 SR	496	#define TRACE_EVENT_CONDITION(name, proto, args, cond, \
	497	struct, assign, print) \
	498	DECLARE_TRACE_CONDITION(name, PARAMS(proto), \
	499	PARAMS(args), PARAMS(cond))
7cb2e3ee	500
1ed0c597 FW	501	#define TRACE_EVENT_FLAGS(event, flag)
1ed0c597 FW	502
d5b5f391 PZ	503	#define TRACE_EVENT_PERF_PERM(event, expr...)
d5b5f391 PZ	504
7cb2e3ee	505	#endif /* ifdef TRACE_EVENT (see note above) */