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

/* SPDX-License-Identifier: GPL-2.0-only */
#ifndef _LINUX_TRACEPOINT_H
#define _LINUX_TRACEPOINT_H

/*
 * Kernel Tracepoint API.
 *
 * See Documentation/trace/tracepoints.rst.
 *
 * Copyright (C) 2008-2014 Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
 *
 * Heavily inspired from the Linux Kernel Markers.
 */

#include <linux/smp.h>
#include <linux/srcu.h>
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/cpumask.h>
#include <linux/rcupdate.h>
#include <linux/tracepoint-defs.h>

struct module;
struct tracepoint;
struct notifier_block;

struct trace_eval_map {
	const char		*system;
	const char		*eval_string;
	unsigned long		eval_value;
};

#define TRACEPOINT_DEFAULT_PRIO	10

extern struct srcu_struct tracepoint_srcu;

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.
 */
#ifdef CONFIG_TRACEPOINTS
static inline void tracepoint_synchronize_unregister(void)
{
	synchronize_srcu(&tracepoint_srcu);
	synchronize_rcu();
}
#else
static inline void tracepoint_synchronize_unregister(void)
{ }
#endif

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

#define PARAMS(args...) args

#define TRACE_DEFINE_ENUM(x)
#define TRACE_DEFINE_SIZEOF(x)

#ifdef CONFIG_HAVE_ARCH_PREL32_RELOCATIONS
static inline struct tracepoint *tracepoint_ptr_deref(tracepoint_ptr_t *p)
{
	return offset_to_ptr(p);
}

#define __TRACEPOINT_ENTRY(name)					\
	asm("	.section \"__tracepoints_ptrs\", \"a\"		\n"	\
	    "	.balign 4					\n"	\
	    "	.long 	__tracepoint_" #name " - .		\n"	\
	    "	.previous					\n")
#else
static inline struct tracepoint *tracepoint_ptr_deref(tracepoint_ptr_t *p)
{
	return *p;
}

#define __TRACEPOINT_ENTRY(name)					 \
	static tracepoint_ptr_t __tracepoint_ptr_##name __used		 \
	__attribute__((section("__tracepoints_ptrs"))) =		 \
		&__tracepoint_##name
#endif

#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)".
 */
#define __DO_TRACE(tp, proto, args, cond, rcuidle)			\
	do {								\
		struct tracepoint_func *it_func_ptr;			\
		void *it_func;						\
		void *__data;						\
		int __maybe_unused __idx = 0;				\
									\
		if (!(cond))						\
			return;						\
									\
		/* srcu can't be used from NMI */			\
		WARN_ON_ONCE(rcuidle && in_nmi());			\
									\
		/* keep srcu and sched-rcu usage consistent */		\
		preempt_disable_notrace();				\
									\
		/*							\
		 * For rcuidle callers, use srcu since sched-rcu	\
		 * doesn't work from the idle path.			\
		 */							\
		if (rcuidle) {						\
			__idx = srcu_read_lock_notrace(&tracepoint_srcu);\
			rcu_irq_enter_irqson();				\
		}							\
									\
		it_func_ptr = rcu_dereference_raw((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);		\
		}							\
									\
		if (rcuidle) {						\
			rcu_irq_exit_irqson();				\
			srcu_read_unlock_notrace(&tracepoint_srcu, __idx);\
		}							\
									\
		preempt_enable_notrace();				\
	} 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), 1);			\
	}
#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), 0);			\
		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"), used)) =		 \
		{ __tpstrtab_##name, STATIC_KEY_INIT_FALSE, reg, unreg, NULL };\
	__TRACEPOINT_ENTRY(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

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

#define DECLARE_TRACE_CONDITION(name, proto, args, cond)		\
	__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args),		\
			cpu_online(raw_smp_processor_id()) && (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_FN_COND(name, proto, args, cond, struct,		\
		assign, print, reg, unreg)			\
	DECLARE_TRACE_CONDITION(name, PARAMS(proto),	\
			PARAMS(args), PARAMS(cond))
#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...)

#define DECLARE_EVENT_NOP(name, proto, args)				\
	static inline void trace_##name(proto)				\
	{ }								\
	static inline bool trace_##name##_enabled(void)			\
	{								\
		return false;						\
	}

#define TRACE_EVENT_NOP(name, proto, args, struct, assign, print)	\
	DECLARE_EVENT_NOP(name, PARAMS(proto), PARAMS(args))

#define DECLARE_EVENT_CLASS_NOP(name, proto, args, tstruct, assign, print)
#define DEFINE_EVENT_NOP(template, name, proto, args)			\
	DECLARE_EVENT_NOP(name, PARAMS(proto), PARAMS(args))

#endif /* ifdef TRACE_EVENT (see note above) */
Commit	Line	Data
52a6e82a	1	/* SPDX-License-Identifier: GPL-2.0-only */
97e1c18e MD	2	#ifndef _LINUX_TRACEPOINT_H
	3	#define _LINUX_TRACEPOINT_H
	4
	5	/*
	6	* Kernel Tracepoint API.
	7	*
ec15872d	8	* See Documentation/trace/tracepoints.rst.
97e1c18e	9	*
de7b2973	10	* Copyright (C) 2008-2014 Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
97e1c18e MD	11	*
97e1c18e MD	12	* Heavily inspired from the Linux Kernel Markers.
97e1c18e MD	13	*/
97e1c18e MD	14
f3775549	15	#include <linux/smp.h>
e6753f23	16	#include <linux/srcu.h>
b70e4f05	17	#include <linux/errno.h>
97e1c18e	18	#include <linux/types.h>
f3775549	19	#include <linux/cpumask.h>
97e1c18e	20	#include <linux/rcupdate.h>
bd2a634d	21	#include <linux/tracepoint-defs.h>
97e1c18e MD	22
	23	struct module;
	24	struct tracepoint;
de7b2973	25	struct notifier_block;
97e1c18e	26
00f4b652	27	struct trace_eval_map {
0c564a53	28	const char *system;
00f4b652 JL	29	const char *eval_string;
00f4b652 JL	30	unsigned long eval_value;
0c564a53 SRRH	31	};
0c564a53 SRRH	32
7904b5c4 SRRH	33	#define TRACEPOINT_DEFAULT_PRIO 10
7904b5c4 SRRH	34
e6753f23 JFG	35	extern struct srcu_struct tracepoint_srcu;
e6753f23 JFG	36
38516ab5	37	extern int
de7b2973 MD	38	tracepoint_probe_register(struct tracepoint tp, void probe, void *data);
de7b2973 MD	39	extern int
7904b5c4 SRRH	40	tracepoint_probe_register_prio(struct tracepoint tp, void probe, void *data,
	41	int prio);
	42	extern int
de7b2973 MD	43	tracepoint_probe_unregister(struct tracepoint tp, void probe, void *data);
	44	extern void
	45	for_each_kernel_tracepoint(void (fct)(struct tracepoint tp, void *priv),
	46	void *priv);
2e26ca71	47
b75ef8b4 MD	48	#ifdef CONFIG_MODULES
	49	struct tp_module {
	50	struct list_head list;
eb7d035c	51	struct module *mod;
b75ef8b4	52	};
de7b2973	53
45ab2813	54	bool trace_module_has_bad_taint(struct module *mod);
de7b2973 MD	55	extern int register_tracepoint_module_notifier(struct notifier_block *nb);
de7b2973 MD	56	extern int unregister_tracepoint_module_notifier(struct notifier_block *nb);
45ab2813 SRRH	57	#else
	58	static inline bool trace_module_has_bad_taint(struct module *mod)
	59	{
	60	return false;
	61	}
de7b2973 MD	62	static inline
	63	int register_tracepoint_module_notifier(struct notifier_block *nb)
	64	{
	65	return 0;
	66	}
	67	static inline
	68	int unregister_tracepoint_module_notifier(struct notifier_block *nb)
	69	{
	70	return 0;
	71	}
b75ef8b4 MD	72	#endif /* CONFIG_MODULES */
b75ef8b4 MD	73
2e26ca71 SR	74	/*
	75	* tracepoint_synchronize_unregister must be called between the last tracepoint
	76	* probe unregistration and the end of module exit to make sure there is no
	77	* caller executing a probe when it is freed.
	78	*/
e6753f23	79	#ifdef CONFIG_TRACEPOINTS
2e26ca71 SR	80	static inline void tracepoint_synchronize_unregister(void)
2e26ca71 SR	81	{
e6753f23	82	synchronize_srcu(&tracepoint_srcu);
74401729	83	synchronize_rcu();
2e26ca71	84	}
e6753f23 JFG	85	#else
	86	static inline void tracepoint_synchronize_unregister(void)
	87	{ }
	88	#endif
2e26ca71	89
b725dfea	90	#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS
8cf868af	91	extern int syscall_regfunc(void);
b725dfea MD	92	extern void syscall_unregfunc(void);
	93	#endif /* CONFIG_HAVE_SYSCALL_TRACEPOINTS */
	94
2e26ca71 SR	95	#define PARAMS(args...) args
2e26ca71 SR	96
0c564a53	97	#define TRACE_DEFINE_ENUM(x)
4f0dfd76	98	#define TRACE_DEFINE_SIZEOF(x)
0c564a53	99
9c0be3f6 MD	100	#ifdef CONFIG_HAVE_ARCH_PREL32_RELOCATIONS
	101	static inline struct tracepoint tracepoint_ptr_deref(tracepoint_ptr_t p)
	102	{
	103	return offset_to_ptr(p);
	104	}
	105
	106	#define __TRACEPOINT_ENTRY(name) \
	107	asm(" .section \"__tracepoints_ptrs\", \"a\" \n" \
	108	" .balign 4 \n" \
	109	" .long __tracepoint_" #name " - . \n" \
	110	" .previous \n")
	111	#else
	112	static inline struct tracepoint tracepoint_ptr_deref(tracepoint_ptr_t p)
	113	{
	114	return *p;
	115	}
	116
	117	#define __TRACEPOINT_ENTRY(name) \
	118	static tracepoint_ptr_t __tracepoint_ptr_##name __used \
	119	__attribute__((section("__tracepoints_ptrs"))) = \
	120	&__tracepoint_##name
	121	#endif
	122
2e26ca71 SR	123	#endif /* _LINUX_TRACEPOINT_H */
	124
	125	/*
	126	* Note: we keep the TRACE_EVENT and DECLARE_TRACE outside the include
	127	* file ifdef protection.
	128	* This is due to the way trace events work. If a file includes two
	129	* trace event headers under one "CREATE_TRACE_POINTS" the first include
	130	* will override the TRACE_EVENT and break the second include.
	131	*/
	132
ea20d929 SR	133	#ifndef DECLARE_TRACE
ea20d929 SR	134
2939b046	135	#define TP_PROTO(args...) args
8cd09a59	136	#define TP_ARGS(args...) args
287050d3	137	#define TP_CONDITION(args...) args
97e1c18e	138
c63b7682 TS	139	/*
	140	* Individual subsystem my have a separate configuration to
	141	* enable their tracepoints. By default, this file will create
	142	* the tracepoints if CONFIG_TRACEPOINT is defined. If a subsystem
	143	* wants to be able to disable its tracepoints from being created
	144	* it can define NOTRACE before including the tracepoint headers.
	145	*/
	146	#if defined(CONFIG_TRACEPOINTS) && !defined(NOTRACE)
	147	#define TRACEPOINTS_ENABLED
	148	#endif
	149
	150	#ifdef TRACEPOINTS_ENABLED
97e1c18e MD	151
	152	/*
	153	* it_func[0] is never NULL because there is at least one element in the array
	154	* when the array itself is non NULL.
38516ab5 SR	155	*
	156	* Note, the proto and args passed in includes "__data" as the first parameter.
	157	* The reason for this is to handle the "void" prototype. If a tracepoint
	158	* has a "void" prototype, then it is invalid to declare a function
a2806ef7	159	* as "(void *, void)".
97e1c18e	160	*/
e6753f23	161	#define __DO_TRACE(tp, proto, args, cond, rcuidle) \
97e1c18e	162	do { \
38516ab5 SR	163	struct tracepoint_func *it_func_ptr; \
	164	void *it_func; \
	165	void *__data; \
0c7a52e4	166	int __maybe_unused __idx = 0; \
97e1c18e	167	\
287050d3 SR	168	if (!(cond)) \
287050d3 SR	169	return; \
e6753f23 JFG	170	\
	171	/* srcu can't be used from NMI */ \
	172	WARN_ON_ONCE(rcuidle && in_nmi()); \
	173	\
	174	/* keep srcu and sched-rcu usage consistent */ \
	175	preempt_disable_notrace(); \
	176	\
	177	/* \
	178	* For rcuidle callers, use srcu since sched-rcu \
	179	* doesn't work from the idle path. \
	180	*/ \
865e63b0	181	if (rcuidle) { \
0c7a52e4	182	__idx = srcu_read_lock_notrace(&tracepoint_srcu);\
865e63b0 SRV	183	rcu_irq_enter_irqson(); \
865e63b0 SRV	184	} \
e6753f23 JFG	185	\
	186	it_func_ptr = rcu_dereference_raw((tp)->funcs); \
	187	\
38516ab5	188	if (it_func_ptr) { \
97e1c18e	189	do { \
38516ab5 SR	190	it_func = (it_func_ptr)->func; \
	191	__data = (it_func_ptr)->data; \
	192	((void(*)(proto))(it_func))(args); \
	193	} while ((++it_func_ptr)->func); \
97e1c18e	194	} \
e6753f23	195	\
865e63b0 SRV	196	if (rcuidle) { \
865e63b0 SRV	197	rcu_irq_exit_irqson(); \
0c7a52e4	198	srcu_read_unlock_notrace(&tracepoint_srcu, __idx);\
865e63b0	199	} \
e6753f23 JFG	200	\
e6753f23 JFG	201	preempt_enable_notrace(); \
97e1c18e MD	202	} while (0)
97e1c18e MD	203
7ece55a4	204	#ifndef MODULE
d54b6eeb	205	#define __DECLARE_TRACE_RCU(name, proto, args, cond, data_proto, data_args) \
7ece55a4 JT	206	static inline void trace_##name##_rcuidle(proto) \
	207	{ \
	208	if (static_key_false(&__tracepoint_##name.key)) \
	209	__DO_TRACE(&__tracepoint_##name, \
	210	TP_PROTO(data_proto), \
	211	TP_ARGS(data_args), \
d54b6eeb	212	TP_CONDITION(cond), 1); \
7ece55a4 JT	213	}
	214	#else
	215	#define __DECLARE_TRACE_RCU(name, proto, args, cond, data_proto, data_args)
	216	#endif
	217
97e1c18e MD	218	/*
	219	* Make sure the alignment of the structure in the __tracepoints section will
	220	* not add unwanted padding between the beginning of the section and the
	221	* structure. Force alignment to the same alignment as the section start.
3a630178 DH	222	*
3a630178 DH	223	* When lockdep is enabled, we make sure to always do the RCU portions of
a15920be MD	224	* the tracepoint code, regardless of whether tracing is on. However,
	225	* don't check if the condition is false, due to interaction with idle
	226	* instrumentation. This lets us find RCU issues triggered with tracepoints
	227	* even when this tracepoint is off. This code has no purpose other than
	228	* poking RCU a bit.
97e1c18e	229	*/
2fbb90db	230	#define __DECLARE_TRACE(name, proto, args, cond, data_proto, data_args) \
7e066fb8	231	extern struct tracepoint __tracepoint_##name; \
97e1c18e MD	232	static inline void trace_##name(proto) \
97e1c18e MD	233	{ \
c5905afb	234	if (static_key_false(&__tracepoint_##name.key)) \
97e1c18e	235	__DO_TRACE(&__tracepoint_##name, \
38516ab5	236	TP_PROTO(data_proto), \
287050d3	237	TP_ARGS(data_args), \
d54b6eeb	238	TP_CONDITION(cond), 0); \
a05d59a5	239	if (IS_ENABLED(CONFIG_LOCKDEP) && (cond)) { \
3a630178 DH	240	rcu_read_lock_sched_notrace(); \
	241	rcu_dereference_sched(__tracepoint_##name.funcs);\
	242	rcu_read_unlock_sched_notrace(); \
	243	} \
2fbb90db	244	} \
7ece55a4 JT	245	__DECLARE_TRACE_RCU(name, PARAMS(proto), PARAMS(args), \
7ece55a4 JT	246	PARAMS(cond), PARAMS(data_proto), PARAMS(data_args)) \
38516ab5 SR	247	static inline int \
38516ab5 SR	248	register_trace_##name(void (probe)(data_proto), void data) \
97e1c18e	249	{ \
de7b2973 MD	250	return tracepoint_probe_register(&__tracepoint_##name, \
de7b2973 MD	251	(void *)probe, data); \
97e1c18e	252	} \
38516ab5	253	static inline int \
7904b5c4 SRRH	254	register_trace_prio_##name(void (probe)(data_proto), void data,\
	255	int prio) \
	256	{ \
	257	return tracepoint_probe_register_prio(&__tracepoint_##name, \
	258	(void *)probe, data, prio); \
	259	} \
	260	static inline int \
38516ab5	261	unregister_trace_##name(void (probe)(data_proto), void data) \
97e1c18e	262	{ \
de7b2973 MD	263	return tracepoint_probe_unregister(&__tracepoint_##name,\
de7b2973 MD	264	(void *)probe, data); \
53da59aa	265	} \
38516ab5 SR	266	static inline void \
38516ab5 SR	267	check_trace_callback_type_##name(void (*cb)(data_proto)) \
53da59aa	268	{ \
7c65bbc7 SRRH	269	} \
	270	static inline bool \
	271	trace_##name##_enabled(void) \
	272	{ \
	273	return static_key_false(&__tracepoint_##name.key); \
97e1c18e MD	274	}
97e1c18e MD	275
65498646 MD	276	/*
	277	* We have no guarantee that gcc and the linker won't up-align the tracepoint
	278	* structures, so we create an array of pointers that will be used for iteration
	279	* on the tracepoints.
	280	*/
d430d3d7 JB	281	#define DEFINE_TRACE_FN(name, reg, unreg) \
	282	static const char __tpstrtab_##name[] \
	283	__attribute__((section("__tracepoints_strings"))) = #name; \
	284	struct tracepoint __tracepoint_##name \
46e0c9be	285	__attribute__((section("__tracepoints"), used)) = \
c5905afb	286	{ __tpstrtab_##name, STATIC_KEY_INIT_FALSE, reg, unreg, NULL };\
46e0c9be	287	__TRACEPOINT_ENTRY(name);
97419875 JS	288
	289	#define DEFINE_TRACE(name) \
	290	DEFINE_TRACE_FN(name, NULL, NULL);
7e066fb8 MD	291
	292	#define EXPORT_TRACEPOINT_SYMBOL_GPL(name) \
	293	EXPORT_SYMBOL_GPL(__tracepoint_##name)
	294	#define EXPORT_TRACEPOINT_SYMBOL(name) \
	295	EXPORT_SYMBOL(__tracepoint_##name)
	296
c63b7682	297	#else /* !TRACEPOINTS_ENABLED */
2fbb90db	298	#define __DECLARE_TRACE(name, proto, args, cond, data_proto, data_args) \
97e1c18e MD	299	static inline void trace_##name(proto) \
97e1c18e MD	300	{ } \
2fbb90db SR	301	static inline void trace_##name##_rcuidle(proto) \
2fbb90db SR	302	{ } \
38516ab5 SR	303	static inline int \
	304	register_trace_##name(void (*probe)(data_proto), \
	305	void *data) \
97e1c18e MD	306	{ \
	307	return -ENOSYS; \
	308	} \
38516ab5 SR	309	static inline int \
	310	unregister_trace_##name(void (*probe)(data_proto), \
	311	void *data) \
c420970e MD	312	{ \
c420970e MD	313	return -ENOSYS; \
53da59aa	314	} \
38516ab5	315	static inline void check_trace_callback_type_##name(void (*cb)(data_proto)) \
53da59aa	316	{ \
7c65bbc7 SRRH	317	} \
	318	static inline bool \
	319	trace_##name##_enabled(void) \
	320	{ \
	321	return false; \
c420970e	322	}
97e1c18e	323
97419875	324	#define DEFINE_TRACE_FN(name, reg, unreg)
7e066fb8 MD	325	#define DEFINE_TRACE(name)
	326	#define EXPORT_TRACEPOINT_SYMBOL_GPL(name)
	327	#define EXPORT_TRACEPOINT_SYMBOL(name)
	328
c63b7682	329	#endif /* TRACEPOINTS_ENABLED */
38516ab5	330
3c49b52b SR	331	#ifdef CONFIG_TRACING
	332	/**
	333	* tracepoint_string - register constant persistent string to trace system
	334	* @str - a constant persistent string that will be referenced in tracepoints
	335	*
	336	* If constant strings are being used in tracepoints, it is faster and
	337	* more efficient to just save the pointer to the string and reference
	338	* that with a printf "%s" instead of saving the string in the ring buffer
	339	* and wasting space and time.
	340	*
	341	* The problem with the above approach is that userspace tools that read
	342	* the binary output of the trace buffers do not have access to the string.
	343	* Instead they just show the address of the string which is not very
	344	* useful to users.
	345	*
	346	* With tracepoint_string(), the string will be registered to the tracing
	347	* system and exported to userspace via the debugfs/tracing/printk_formats
	348	* file that maps the string address to the string text. This way userspace
	349	* tools that read the binary buffers have a way to map the pointers to
	350	* the ASCII strings they represent.
	351	*
	352	* The @str used must be a constant string and persistent as it would not
	353	* make sense to show a string that no longer exists. But it is still fine
	354	* to be used with modules, because when modules are unloaded, if they
	355	* had tracepoints, the ring buffers are cleared too. As long as the string
	356	* does not change during the life of the module, it is fine to use
	357	* tracepoint_string() within a module.
	358	*/
	359	#define tracepoint_string(str) \
	360	({ \
	361	static const char *___tp_str __tracepoint_string = str; \
	362	___tp_str; \
	363	})
	364	#define __tracepoint_string __attribute__((section("__tracepoint_str")))
	365	#else
	366	/*
	367	* tracepoint_string() is used to save the string address for userspace
	368	* tracing tools. When tracing isn't configured, there's no need to save
	369	* anything.
	370	*/
	371	# define tracepoint_string(str) str
	372	# define __tracepoint_string
	373	#endif
	374
38516ab5	375	#define DECLARE_TRACE(name, proto, args) \
dc17147d SRRH	376	__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), \
	377	cpu_online(raw_smp_processor_id()), \
	378	PARAMS(void *__data, proto), \
	379	PARAMS(__data, args))
38516ab5	380
287050d3	381	#define DECLARE_TRACE_CONDITION(name, proto, args, cond) \
dc17147d SRRH	382	__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), \
dc17147d SRRH	383	cpu_online(raw_smp_processor_id()) && (PARAMS(cond)), \
287050d3 SR	384	PARAMS(void *__data, proto), \
	385	PARAMS(__data, args))
	386
1ed0c597 FW	387	#define TRACE_EVENT_FLAGS(event, flag)
1ed0c597 FW	388
d5b5f391 PZ	389	#define TRACE_EVENT_PERF_PERM(event, expr...)
d5b5f391 PZ	390
ea20d929	391	#endif /* DECLARE_TRACE */
97e1c18e	392
ea20d929	393	#ifndef TRACE_EVENT
823f9124 SR	394	/*
	395	* For use with the TRACE_EVENT macro:
	396	*
	397	* We define a tracepoint, its arguments, its printk format
2621bca8	398	* and its 'fast binary record' layout.
823f9124 SR	399	*
	400	* Firstly, name your tracepoint via TRACE_EVENT(name : the
	401	* 'subsystem_event' notation is fine.
	402	*
	403	* Think about this whole construct as the
	404	* 'trace_sched_switch() function' from now on.
	405	*
	406	*
	407	* TRACE_EVENT(sched_switch,
	408	*
	409	* *
	410	* * A function has a regular function arguments
	411	* * prototype, declare it via TP_PROTO():
	412	* *
	413	*
ef18012b SR	414	* TP_PROTO(struct rq rq, struct task_struct prev,
ef18012b SR	415	* struct task_struct *next),
823f9124 SR	416	*
	417	* *
	418	* * Define the call signature of the 'function'.
	419	* * (Design sidenote: we use this instead of a
	420	* * TP_PROTO1/TP_PROTO2/TP_PROTO3 ugliness.)
	421	* *
	422	*
ef18012b	423	* TP_ARGS(rq, prev, next),
823f9124 SR	424	*
	425	* *
	426	* * Fast binary tracing: define the trace record via
	427	* * TP_STRUCT__entry(). You can think about it like a
	428	* * regular C structure local variable definition.
	429	* *
	430	* * This is how the trace record is structured and will
	431	* * be saved into the ring buffer. These are the fields
	432	* * that will be exposed to user-space in
156f5a78	433	* * /sys/kernel/debug/tracing/events/<*>/format.
823f9124 SR	434	* *
	435	* * The declared 'local variable' is called '__entry'
	436	* *
	437	* * __field(pid_t, prev_prid) is equivalent to a standard declariton:
	438	* *
	439	* * pid_t prev_pid;
	440	* *
	441	* * __array(char, prev_comm, TASK_COMM_LEN) is equivalent to:
	442	* *
	443	* * char prev_comm[TASK_COMM_LEN];
	444	* *
	445	*
	446	* TP_STRUCT__entry(
	447	* __array( char, prev_comm, TASK_COMM_LEN )
	448	* __field( pid_t, prev_pid )
	449	* __field( int, prev_prio )
	450	* __array( char, next_comm, TASK_COMM_LEN )
	451	* __field( pid_t, next_pid )
	452	* __field( int, next_prio )
	453	* ),
	454	*
	455	* *
	456	* * Assign the entry into the trace record, by embedding
	457	* * a full C statement block into TP_fast_assign(). You
	458	* * can refer to the trace record as '__entry' -
	459	* * otherwise you can put arbitrary C code in here.
	460	* *
	461	* * Note: this C code will execute every time a trace event
	462	* * happens, on an active tracepoint.
	463	* *
	464	*
ef18012b SR	465	* TP_fast_assign(
	466	* memcpy(__entry->next_comm, next->comm, TASK_COMM_LEN);
	467	* __entry->prev_pid = prev->pid;
	468	* __entry->prev_prio = prev->prio;
823f9124 SR	469	* memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
823f9124 SR	470	* __entry->next_pid = next->pid;
ef18012b	471	* __entry->next_prio = next->prio;
ec6e7c3a	472	* ),
823f9124 SR	473	*
	474	* *
	475	* * Formatted output of a trace record via TP_printk().
	476	* * This is how the tracepoint will appear under ftrace
	477	* * plugins that make use of this tracepoint.
	478	* *
	479	* * (raw-binary tracing wont actually perform this step.)
	480	* *
	481	*
	482	* TP_printk("task %s:%d [%d] ==> %s:%d [%d]",
	483	* __entry->prev_comm, __entry->prev_pid, __entry->prev_prio,
	484	* __entry->next_comm, __entry->next_pid, __entry->next_prio),
	485	*
	486	* );
	487	*
	488	* This macro construct is thus used for the regular printk format
	489	* tracing setup, it is used to construct a function pointer based
	490	* tracepoint callback (this is used by programmatic plugins and
	491	* can also by used by generic instrumentation like SystemTap), and
	492	* it is also used to expose a structured trace record in
156f5a78	493	* /sys/kernel/debug/tracing/events/.
97419875 JS	494	*
	495	* A set of (un)registration functions can be passed to the variant
	496	* TRACE_EVENT_FN to perform any (un)registration work.
823f9124 SR	497	*/
823f9124 SR	498
091ad365	499	#define DECLARE_EVENT_CLASS(name, proto, args, tstruct, assign, print)
ff038f5c SR	500	#define DEFINE_EVENT(template, name, proto, args) \
ff038f5c SR	501	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
f5abaa1b SR	502	#define DEFINE_EVENT_FN(template, name, proto, args, reg, unreg)\
f5abaa1b SR	503	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
e5bc9721 SR	504	#define DEFINE_EVENT_PRINT(template, name, proto, args, print) \
e5bc9721 SR	505	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
287050d3 SR	506	#define DEFINE_EVENT_CONDITION(template, name, proto, \
	507	args, cond) \
	508	DECLARE_TRACE_CONDITION(name, PARAMS(proto), \
	509	PARAMS(args), PARAMS(cond))
ff038f5c	510
30a8fecc	511	#define TRACE_EVENT(name, proto, args, struct, assign, print) \
da4d0302	512	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
97419875 JS	513	#define TRACE_EVENT_FN(name, proto, args, struct, \
	514	assign, print, reg, unreg) \
	515	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
2701121b DK	516	#define TRACE_EVENT_FN_COND(name, proto, args, cond, struct, \
	517	assign, print, reg, unreg) \
	518	DECLARE_TRACE_CONDITION(name, PARAMS(proto), \
	519	PARAMS(args), PARAMS(cond))
287050d3 SR	520	#define TRACE_EVENT_CONDITION(name, proto, args, cond, \
	521	struct, assign, print) \
	522	DECLARE_TRACE_CONDITION(name, PARAMS(proto), \
	523	PARAMS(args), PARAMS(cond))
7cb2e3ee	524
1ed0c597 FW	525	#define TRACE_EVENT_FLAGS(event, flag)
1ed0c597 FW	526
d5b5f391 PZ	527	#define TRACE_EVENT_PERF_PERM(event, expr...)
d5b5f391 PZ	528
16336345 YS	529	#define DECLARE_EVENT_NOP(name, proto, args) \
	530	static inline void trace_##name(proto) \
	531	{ } \
	532	static inline bool trace_##name##_enabled(void) \
	533	{ \
	534	return false; \
	535	}
	536
	537	#define TRACE_EVENT_NOP(name, proto, args, struct, assign, print) \
	538	DECLARE_EVENT_NOP(name, PARAMS(proto), PARAMS(args))
	539
	540	#define DECLARE_EVENT_CLASS_NOP(name, proto, args, tstruct, assign, print)
	541	#define DEFINE_EVENT_NOP(template, name, proto, args) \
	542	DECLARE_EVENT_NOP(name, PARAMS(proto), PARAMS(args))
	543
7cb2e3ee	544	#endif /* ifdef TRACE_EVENT (see note above) */