[net-next-2.6.git] / kernel / trace / trace.h

#ifndef _LINUX_KERNEL_TRACE_H
#define _LINUX_KERNEL_TRACE_H

#include <linux/fs.h>
#include <asm/atomic.h>
#include <linux/sched.h>
#include <linux/clocksource.h>
#include <linux/ring_buffer.h>
#include <linux/mmiotrace.h>
#include <linux/ftrace.h>
#include <trace/boot.h>

enum trace_type {
	__TRACE_FIRST_TYPE = 0,

	TRACE_FN,
	TRACE_CTX,
	TRACE_WAKE,
	TRACE_CONT,
	TRACE_STACK,
	TRACE_PRINT,
	TRACE_SPECIAL,
	TRACE_MMIO_RW,
	TRACE_MMIO_MAP,
	TRACE_BRANCH,
	TRACE_BOOT_CALL,
	TRACE_BOOT_RET,
	TRACE_FN_RET,

	__TRACE_LAST_TYPE
};

/*
 * The trace entry - the most basic unit of tracing. This is what
 * is printed in the end as a single line in the trace output, such as:
 *
 *     bash-15816 [01]   235.197585: idle_cpu <- irq_enter
 */
struct trace_entry {
	unsigned char		type;
	unsigned char		cpu;
	unsigned char		flags;
	unsigned char		preempt_count;
	int			pid;
};

/*
 * Function trace entry - function address and parent function addres:
 */
struct ftrace_entry {
	struct trace_entry	ent;
	unsigned long		ip;
	unsigned long		parent_ip;
};

/* Function return entry */
struct ftrace_ret_entry {
	struct trace_entry	ent;
	unsigned long		ip;
	unsigned long		parent_ip;
	unsigned long long	calltime;
	unsigned long long	rettime;
};
extern struct tracer boot_tracer;

/*
 * Context switch trace entry - which task (and prio) we switched from/to:
 */
struct ctx_switch_entry {
	struct trace_entry	ent;
	unsigned int		prev_pid;
	unsigned char		prev_prio;
	unsigned char		prev_state;
	unsigned int		next_pid;
	unsigned char		next_prio;
	unsigned char		next_state;
	unsigned int		next_cpu;
};

/*
 * Special (free-form) trace entry:
 */
struct special_entry {
	struct trace_entry	ent;
	unsigned long		arg1;
	unsigned long		arg2;
	unsigned long		arg3;
};

/*
 * Stack-trace entry:
 */

#define FTRACE_STACK_ENTRIES	8

struct stack_entry {
	struct trace_entry	ent;
	unsigned long		caller[FTRACE_STACK_ENTRIES];
};

/*
 * ftrace_printk entry:
 */
struct print_entry {
	struct trace_entry	ent;
	unsigned long		ip;
	char			buf[];
};

#define TRACE_OLD_SIZE		88

struct trace_field_cont {
	unsigned char		type;
	/* Temporary till we get rid of this completely */
	char			buf[TRACE_OLD_SIZE - 1];
};

struct trace_mmiotrace_rw {
	struct trace_entry	ent;
	struct mmiotrace_rw	rw;
};

struct trace_mmiotrace_map {
	struct trace_entry	ent;
	struct mmiotrace_map	map;
};

struct trace_boot_call {
	struct trace_entry	ent;
	struct boot_trace_call boot_call;
};

struct trace_boot_ret {
	struct trace_entry	ent;
	struct boot_trace_ret boot_ret;
};

#define TRACE_FUNC_SIZE 30
#define TRACE_FILE_SIZE 20
struct trace_branch {
	struct trace_entry	ent;
	unsigned	        line;
	char			func[TRACE_FUNC_SIZE+1];
	char			file[TRACE_FILE_SIZE+1];
	char			correct;
};

/*
 * trace_flag_type is an enumeration that holds different
 * states when a trace occurs. These are:
 *  IRQS_OFF		- interrupts were disabled
 *  IRQS_NOSUPPORT 	- arch does not support irqs_disabled_flags
 *  NEED_RESCED		- reschedule is requested
 *  HARDIRQ		- inside an interrupt handler
 *  SOFTIRQ		- inside a softirq handler
 *  CONT		- multiple entries hold the trace item
 */
enum trace_flag_type {
	TRACE_FLAG_IRQS_OFF		= 0x01,
	TRACE_FLAG_IRQS_NOSUPPORT	= 0x02,
	TRACE_FLAG_NEED_RESCHED		= 0x04,
	TRACE_FLAG_HARDIRQ		= 0x08,
	TRACE_FLAG_SOFTIRQ		= 0x10,
	TRACE_FLAG_CONT			= 0x20,
};

#define TRACE_BUF_SIZE		1024

/*
 * The CPU trace array - it consists of thousands of trace entries
 * plus some other descriptor data: (for example which task started
 * the trace, etc.)
 */
struct trace_array_cpu {
	atomic_t		disabled;

	/* these fields get copied into max-trace: */
	unsigned long		trace_idx;
	unsigned long		overrun;
	unsigned long		saved_latency;
	unsigned long		critical_start;
	unsigned long		critical_end;
	unsigned long		critical_sequence;
	unsigned long		nice;
	unsigned long		policy;
	unsigned long		rt_priority;
	cycle_t			preempt_timestamp;
	pid_t			pid;
	uid_t			uid;
	char			comm[TASK_COMM_LEN];
};

struct trace_iterator;

/*
 * The trace array - an array of per-CPU trace arrays. This is the
 * highest level data structure that individual tracers deal with.
 * They have on/off state as well:
 */
struct trace_array {
	struct ring_buffer	*buffer;
	unsigned long		entries;
	int			cpu;
	cycle_t			time_start;
	struct task_struct	*waiter;
	struct trace_array_cpu	*data[NR_CPUS];
};

#define FTRACE_CMP_TYPE(var, type) \
	__builtin_types_compatible_p(typeof(var), type *)

#undef IF_ASSIGN
#define IF_ASSIGN(var, entry, etype, id)		\
	if (FTRACE_CMP_TYPE(var, etype)) {		\
		var = (typeof(var))(entry);		\
		WARN_ON(id && (entry)->type != id);	\
		break;					\
	}

/* Will cause compile errors if type is not found. */
extern void __ftrace_bad_type(void);

/*
 * The trace_assign_type is a verifier that the entry type is
 * the same as the type being assigned. To add new types simply
 * add a line with the following format:
 *
 * IF_ASSIGN(var, ent, type, id);
 *
 *  Where "type" is the trace type that includes the trace_entry
 *  as the "ent" item. And "id" is the trace identifier that is
 *  used in the trace_type enum.
 *
 *  If the type can have more than one id, then use zero.
 */
#define trace_assign_type(var, ent)					\
	do {								\
		IF_ASSIGN(var, ent, struct ftrace_entry, TRACE_FN);	\
		IF_ASSIGN(var, ent, struct ctx_switch_entry, 0);	\
		IF_ASSIGN(var, ent, struct trace_field_cont, TRACE_CONT); \
		IF_ASSIGN(var, ent, struct stack_entry, TRACE_STACK);	\
		IF_ASSIGN(var, ent, struct print_entry, TRACE_PRINT);	\
		IF_ASSIGN(var, ent, struct special_entry, 0);		\
		IF_ASSIGN(var, ent, struct trace_mmiotrace_rw,		\
			  TRACE_MMIO_RW);				\
		IF_ASSIGN(var, ent, struct trace_mmiotrace_map,		\
			  TRACE_MMIO_MAP);				\
		IF_ASSIGN(var, ent, struct trace_boot_call, TRACE_BOOT_CALL);\
		IF_ASSIGN(var, ent, struct trace_boot_ret, TRACE_BOOT_RET);\
		IF_ASSIGN(var, ent, struct trace_branch, TRACE_BRANCH); \
		IF_ASSIGN(var, ent, struct ftrace_ret_entry, TRACE_FN_RET);\
		__ftrace_bad_type();					\
	} while (0)

/* Return values for print_line callback */
enum print_line_t {
	TRACE_TYPE_PARTIAL_LINE	= 0,	/* Retry after flushing the seq */
	TRACE_TYPE_HANDLED	= 1,
	TRACE_TYPE_UNHANDLED	= 2	/* Relay to other output functions */
};

/*
 * A specific tracer, represented by methods that operate on a trace array:
 */
struct tracer {
	const char		*name;
	/* Your tracer should raise a warning if init fails */
	int			(*init)(struct trace_array *tr);
	void			(*reset)(struct trace_array *tr);
	void			(*start)(struct trace_array *tr);
	void			(*stop)(struct trace_array *tr);
	void			(*open)(struct trace_iterator *iter);
	void			(*pipe_open)(struct trace_iterator *iter);
	void			(*close)(struct trace_iterator *iter);
	ssize_t			(*read)(struct trace_iterator *iter,
					struct file *filp, char __user *ubuf,
					size_t cnt, loff_t *ppos);
#ifdef CONFIG_FTRACE_STARTUP_TEST
	int			(*selftest)(struct tracer *trace,
					    struct trace_array *tr);
#endif
	enum print_line_t	(*print_line)(struct trace_iterator *iter);
	struct tracer		*next;
	int			print_max;
};

struct trace_seq {
	unsigned char		buffer[PAGE_SIZE];
	unsigned int		len;
	unsigned int		readpos;
};

/*
 * Trace iterator - used by printout routines who present trace
 * results to users and which routines might sleep, etc:
 */
struct trace_iterator {
	struct trace_array	*tr;
	struct tracer		*trace;
	void			*private;
	struct ring_buffer_iter	*buffer_iter[NR_CPUS];

	/* The below is zeroed out in pipe_read */
	struct trace_seq	seq;
	struct trace_entry	*ent;
	int			cpu;
	u64			ts;

	unsigned long		iter_flags;
	loff_t			pos;
	long			idx;

	cpumask_t		started;
};

int tracing_is_enabled(void);
void trace_wake_up(void);
void tracing_reset(struct trace_array *tr, int cpu);
int tracing_open_generic(struct inode *inode, struct file *filp);
struct dentry *tracing_init_dentry(void);
void init_tracer_sysprof_debugfs(struct dentry *d_tracer);

struct trace_entry *tracing_get_trace_entry(struct trace_array *tr,
						struct trace_array_cpu *data);
void tracing_generic_entry_update(struct trace_entry *entry,
				  unsigned long flags,
				  int pc);

void ftrace(struct trace_array *tr,
			    struct trace_array_cpu *data,
			    unsigned long ip,
			    unsigned long parent_ip,
			    unsigned long flags, int pc);
void tracing_sched_switch_trace(struct trace_array *tr,
				struct trace_array_cpu *data,
				struct task_struct *prev,
				struct task_struct *next,
				unsigned long flags, int pc);
void tracing_record_cmdline(struct task_struct *tsk);

void tracing_sched_wakeup_trace(struct trace_array *tr,
				struct trace_array_cpu *data,
				struct task_struct *wakee,
				struct task_struct *cur,
				unsigned long flags, int pc);
void trace_special(struct trace_array *tr,
		   struct trace_array_cpu *data,
		   unsigned long arg1,
		   unsigned long arg2,
		   unsigned long arg3, int pc);
void trace_function(struct trace_array *tr,
		    struct trace_array_cpu *data,
		    unsigned long ip,
		    unsigned long parent_ip,
		    unsigned long flags, int pc);
void
trace_function_return(struct ftrace_retfunc *trace);

void tracing_start_cmdline_record(void);
void tracing_stop_cmdline_record(void);
void tracing_sched_switch_assign_trace(struct trace_array *tr);
void tracing_stop_sched_switch_record(void);
void tracing_start_sched_switch_record(void);
int register_tracer(struct tracer *type);
void unregister_tracer(struct tracer *type);

extern unsigned long nsecs_to_usecs(unsigned long nsecs);

extern unsigned long tracing_max_latency;
extern unsigned long tracing_thresh;

void update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu);
void update_max_tr_single(struct trace_array *tr,
			  struct task_struct *tsk, int cpu);

extern cycle_t ftrace_now(int cpu);

#ifdef CONFIG_FUNCTION_TRACER
void tracing_start_function_trace(void);
void tracing_stop_function_trace(void);
#else
# define tracing_start_function_trace()		do { } while (0)
# define tracing_stop_function_trace()		do { } while (0)
#endif

#ifdef CONFIG_CONTEXT_SWITCH_TRACER
typedef void
(*tracer_switch_func_t)(void *private,
			void *__rq,
			struct task_struct *prev,
			struct task_struct *next);

struct tracer_switch_ops {
	tracer_switch_func_t		func;
	void				*private;
	struct tracer_switch_ops	*next;
};

#endif /* CONFIG_CONTEXT_SWITCH_TRACER */

#ifdef CONFIG_DYNAMIC_FTRACE
extern unsigned long ftrace_update_tot_cnt;
#define DYN_FTRACE_TEST_NAME trace_selftest_dynamic_test_func
extern int DYN_FTRACE_TEST_NAME(void);
#endif

#ifdef CONFIG_FTRACE_STARTUP_TEST
extern int trace_selftest_startup_function(struct tracer *trace,
					   struct trace_array *tr);
extern int trace_selftest_startup_irqsoff(struct tracer *trace,
					  struct trace_array *tr);
extern int trace_selftest_startup_preemptoff(struct tracer *trace,
					     struct trace_array *tr);
extern int trace_selftest_startup_preemptirqsoff(struct tracer *trace,
						 struct trace_array *tr);
extern int trace_selftest_startup_wakeup(struct tracer *trace,
					 struct trace_array *tr);
extern int trace_selftest_startup_nop(struct tracer *trace,
					 struct trace_array *tr);
extern int trace_selftest_startup_sched_switch(struct tracer *trace,
					       struct trace_array *tr);
extern int trace_selftest_startup_sysprof(struct tracer *trace,
					       struct trace_array *tr);
extern int trace_selftest_startup_branch(struct tracer *trace,
					 struct trace_array *tr);
#endif /* CONFIG_FTRACE_STARTUP_TEST */

extern void *head_page(struct trace_array_cpu *data);
extern int trace_seq_printf(struct trace_seq *s, const char *fmt, ...);
extern void trace_seq_print_cont(struct trace_seq *s,
				 struct trace_iterator *iter);

extern int
seq_print_ip_sym(struct trace_seq *s, unsigned long ip,
		unsigned long sym_flags);
extern ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf,
				 size_t cnt);
extern long ns2usecs(cycle_t nsec);
extern int trace_vprintk(unsigned long ip, const char *fmt, va_list args);

extern unsigned long trace_flags;

/* Standard output formatting function used for function return traces */
#ifdef CONFIG_FUNCTION_RET_TRACER
extern enum print_line_t print_return_function(struct trace_iterator *iter);
#else
static inline enum print_line_t
print_return_function(struct trace_iterator *iter)
{
	return TRACE_TYPE_UNHANDLED;
}
#endif

/*
 * trace_iterator_flags is an enumeration that defines bit
 * positions into trace_flags that controls the output.
 *
 * NOTE: These bits must match the trace_options array in
 *       trace.c.
 */
enum trace_iterator_flags {
	TRACE_ITER_PRINT_PARENT		= 0x01,
	TRACE_ITER_SYM_OFFSET		= 0x02,
	TRACE_ITER_SYM_ADDR		= 0x04,
	TRACE_ITER_VERBOSE		= 0x08,
	TRACE_ITER_RAW			= 0x10,
	TRACE_ITER_HEX			= 0x20,
	TRACE_ITER_BIN			= 0x40,
	TRACE_ITER_BLOCK		= 0x80,
	TRACE_ITER_STACKTRACE		= 0x100,
	TRACE_ITER_SCHED_TREE		= 0x200,
	TRACE_ITER_PRINTK		= 0x400,
	TRACE_ITER_PREEMPTONLY		= 0x800,
	TRACE_ITER_BRANCH		= 0x1000,
	TRACE_ITER_ANNOTATE		= 0x2000,
};

/*
 * TRACE_ITER_SYM_MASK masks the options in trace_flags that
 * control the output of kernel symbols.
 */
#define TRACE_ITER_SYM_MASK \
	(TRACE_ITER_PRINT_PARENT|TRACE_ITER_SYM_OFFSET|TRACE_ITER_SYM_ADDR)

extern struct tracer nop_trace;

/**
 * ftrace_preempt_disable - disable preemption scheduler safe
 *
 * When tracing can happen inside the scheduler, there exists
 * cases that the tracing might happen before the need_resched
 * flag is checked. If this happens and the tracer calls
 * preempt_enable (after a disable), a schedule might take place
 * causing an infinite recursion.
 *
 * To prevent this, we read the need_recshed flag before
 * disabling preemption. When we want to enable preemption we
 * check the flag, if it is set, then we call preempt_enable_no_resched.
 * Otherwise, we call preempt_enable.
 *
 * The rational for doing the above is that if need resched is set
 * and we have yet to reschedule, we are either in an atomic location
 * (where we do not need to check for scheduling) or we are inside
 * the scheduler and do not want to resched.
 */
static inline int ftrace_preempt_disable(void)
{
	int resched;

	resched = need_resched();
	preempt_disable_notrace();

	return resched;
}

/**
 * ftrace_preempt_enable - enable preemption scheduler safe
 * @resched: the return value from ftrace_preempt_disable
 *
 * This is a scheduler safe way to enable preemption and not miss
 * any preemption checks. The disabled saved the state of preemption.
 * If resched is set, then we were either inside an atomic or
 * are inside the scheduler (we would have already scheduled
 * otherwise). In this case, we do not want to call normal
 * preempt_enable, but preempt_enable_no_resched instead.
 */
static inline void ftrace_preempt_enable(int resched)
{
	if (resched)
		preempt_enable_no_resched_notrace();
	else
		preempt_enable_notrace();
}

#ifdef CONFIG_BRANCH_TRACER
extern int enable_branch_tracing(struct trace_array *tr);
extern void disable_branch_tracing(void);
static inline int trace_branch_enable(struct trace_array *tr)
{
	if (trace_flags & TRACE_ITER_BRANCH)
		return enable_branch_tracing(tr);
	return 0;
}
static inline void trace_branch_disable(void)
{
	/* due to races, always disable */
	disable_branch_tracing();
}
#else
static inline int trace_branch_enable(struct trace_array *tr)
{
	return 0;
}
static inline void trace_branch_disable(void)
{
}
#endif /* CONFIG_BRANCH_TRACER */

#endif /* _LINUX_KERNEL_TRACE_H */
Commit	Line	Data
bc0c38d1 SR	1	#ifndef _LINUX_KERNEL_TRACE_H
	2	#define _LINUX_KERNEL_TRACE_H
	3
	4	#include <linux/fs.h>
	5	#include <asm/atomic.h>
	6	#include <linux/sched.h>
	7	#include <linux/clocksource.h>
3928a8a2	8	#include <linux/ring_buffer.h>
bd8ac686	9	#include <linux/mmiotrace.h>
d13744cd	10	#include <linux/ftrace.h>
3f5ec136	11	#include <trace/boot.h>
bc0c38d1	12
72829bc3 TG	13	enum trace_type {
	14	__TRACE_FIRST_TYPE = 0,
	15
	16	TRACE_FN,
	17	TRACE_CTX,
	18	TRACE_WAKE,
dd0e545f	19	TRACE_CONT,
72829bc3	20	TRACE_STACK,
dd0e545f	21	TRACE_PRINT,
72829bc3	22	TRACE_SPECIAL,
bd8ac686 PP	23	TRACE_MMIO_RW,
bd8ac686 PP	24	TRACE_MMIO_MAP,
9f029e83	25	TRACE_BRANCH,
74239072 FW	26	TRACE_BOOT_CALL,
74239072 FW	27	TRACE_BOOT_RET,
15e6cb36	28	TRACE_FN_RET,
72829bc3 TG	29
	30	__TRACE_LAST_TYPE
	31	};
	32
777e208d SR	33	/*
	34	* The trace entry - the most basic unit of tracing. This is what
	35	* is printed in the end as a single line in the trace output, such as:
	36	*
	37	* bash-15816 [01] 235.197585: idle_cpu <- irq_enter
	38	*/
	39	struct trace_entry {
	40	unsigned char type;
	41	unsigned char cpu;
	42	unsigned char flags;
	43	unsigned char preempt_count;
	44	int pid;
	45	};
	46
bc0c38d1 SR	47	/*
	48	* Function trace entry - function address and parent function addres:
	49	*/
	50	struct ftrace_entry {
777e208d	51	struct trace_entry ent;
bc0c38d1 SR	52	unsigned long ip;
	53	unsigned long parent_ip;
	54	};
15e6cb36 FW	55
	56	/* Function return entry */
	57	struct ftrace_ret_entry {
	58	struct trace_entry ent;
	59	unsigned long ip;
	60	unsigned long parent_ip;
	61	unsigned long long calltime;
	62	unsigned long long rettime;
	63	};
d13744cd	64	extern struct tracer boot_tracer;
bc0c38d1 SR	65
	66	/*
	67	* Context switch trace entry - which task (and prio) we switched from/to:
	68	*/
	69	struct ctx_switch_entry {
777e208d	70	struct trace_entry ent;
bc0c38d1 SR	71	unsigned int prev_pid;
	72	unsigned char prev_prio;
	73	unsigned char prev_state;
	74	unsigned int next_pid;
	75	unsigned char next_prio;
bac524d3	76	unsigned char next_state;
80b5e940	77	unsigned int next_cpu;
bc0c38d1 SR	78	};
bc0c38d1 SR	79
f0a920d5 IM	80	/*
	81	* Special (free-form) trace entry:
	82	*/
	83	struct special_entry {
777e208d	84	struct trace_entry ent;
f0a920d5 IM	85	unsigned long arg1;
	86	unsigned long arg2;
	87	unsigned long arg3;
	88	};
	89
86387f7e IM	90	/*
	91	* Stack-trace entry:
	92	*/
	93
74f4e369	94	#define FTRACE_STACK_ENTRIES 8
86387f7e IM	95
86387f7e IM	96	struct stack_entry {
777e208d	97	struct trace_entry ent;
86387f7e IM	98	unsigned long caller[FTRACE_STACK_ENTRIES];
	99	};
	100
dd0e545f SR	101	/*
	102	* ftrace_printk entry:
	103	*/
	104	struct print_entry {
777e208d	105	struct trace_entry ent;
dd0e545f SR	106	unsigned long ip;
	107	char buf[];
	108	};
	109
777e208d SR	110	#define TRACE_OLD_SIZE 88
	111
	112	struct trace_field_cont {
	113	unsigned char type;
	114	/* Temporary till we get rid of this completely */
	115	char buf[TRACE_OLD_SIZE - 1];
	116	};
	117
	118	struct trace_mmiotrace_rw {
	119	struct trace_entry ent;
	120	struct mmiotrace_rw rw;
	121	};
	122
	123	struct trace_mmiotrace_map {
	124	struct trace_entry ent;
	125	struct mmiotrace_map map;
	126	};
	127
74239072	128	struct trace_boot_call {
777e208d	129	struct trace_entry ent;
74239072 FW	130	struct boot_trace_call boot_call;
	131	};
	132
	133	struct trace_boot_ret {
	134	struct trace_entry ent;
	135	struct boot_trace_ret boot_ret;
777e208d SR	136	};
777e208d SR	137
52f232cb SR	138	#define TRACE_FUNC_SIZE 30
52f232cb SR	139	#define TRACE_FILE_SIZE 20
9f029e83	140	struct trace_branch {
52f232cb SR	141	struct trace_entry ent;
	142	unsigned line;
	143	char func[TRACE_FUNC_SIZE+1];
	144	char file[TRACE_FILE_SIZE+1];
	145	char correct;
	146	};
	147
fc5e27ae PP	148	/*
	149	* trace_flag_type is an enumeration that holds different
	150	* states when a trace occurs. These are:
9244489a SR	151	* IRQS_OFF - interrupts were disabled
	152	* IRQS_NOSUPPORT - arch does not support irqs_disabled_flags
	153	* NEED_RESCED - reschedule is requested
	154	* HARDIRQ - inside an interrupt handler
	155	* SOFTIRQ - inside a softirq handler
	156	* CONT - multiple entries hold the trace item
fc5e27ae PP	157	*/
	158	enum trace_flag_type {
	159	TRACE_FLAG_IRQS_OFF = 0x01,
9244489a SR	160	TRACE_FLAG_IRQS_NOSUPPORT = 0x02,
	161	TRACE_FLAG_NEED_RESCHED = 0x04,
	162	TRACE_FLAG_HARDIRQ = 0x08,
	163	TRACE_FLAG_SOFTIRQ = 0x10,
	164	TRACE_FLAG_CONT = 0x20,
fc5e27ae PP	165	};
fc5e27ae PP	166
5bf9a1ee	167	#define TRACE_BUF_SIZE 1024
bc0c38d1 SR	168
	169	/*
	170	* The CPU trace array - it consists of thousands of trace entries
	171	* plus some other descriptor data: (for example which task started
	172	* the trace, etc.)
	173	*/
	174	struct trace_array_cpu {
bc0c38d1	175	atomic_t disabled;
4e3c3333	176
c7aafc54	177	/* these fields get copied into max-trace: */
c7aafc54	178	unsigned long trace_idx;
53d0aa77	179	unsigned long overrun;
bc0c38d1 SR	180	unsigned long saved_latency;
	181	unsigned long critical_start;
	182	unsigned long critical_end;
	183	unsigned long critical_sequence;
	184	unsigned long nice;
	185	unsigned long policy;
	186	unsigned long rt_priority;
	187	cycle_t preempt_timestamp;
	188	pid_t pid;
	189	uid_t uid;
	190	char comm[TASK_COMM_LEN];
	191	};
	192
	193	struct trace_iterator;
	194
	195	/*
	196	* The trace array - an array of per-CPU trace arrays. This is the
	197	* highest level data structure that individual tracers deal with.
	198	* They have on/off state as well:
	199	*/
	200	struct trace_array {
3928a8a2	201	struct ring_buffer *buffer;
bc0c38d1	202	unsigned long entries;
bc0c38d1 SR	203	int cpu;
bc0c38d1 SR	204	cycle_t time_start;
b3806b43	205	struct task_struct *waiter;
bc0c38d1 SR	206	struct trace_array_cpu *data[NR_CPUS];
	207	};
	208
7104f300 SR	209	#define FTRACE_CMP_TYPE(var, type) \
	210	__builtin_types_compatible_p(typeof(var), type *)
	211
	212	#undef IF_ASSIGN
	213	#define IF_ASSIGN(var, entry, etype, id) \
	214	if (FTRACE_CMP_TYPE(var, etype)) { \
	215	var = (typeof(var))(entry); \
	216	WARN_ON(id && (entry)->type != id); \
	217	break; \
	218	}
	219
	220	/* Will cause compile errors if type is not found. */
	221	extern void __ftrace_bad_type(void);
	222
	223	/*
	224	* The trace_assign_type is a verifier that the entry type is
	225	* the same as the type being assigned. To add new types simply
	226	* add a line with the following format:
	227	*
	228	* IF_ASSIGN(var, ent, type, id);
	229	*
	230	* Where "type" is the trace type that includes the trace_entry
	231	* as the "ent" item. And "id" is the trace identifier that is
	232	* used in the trace_type enum.
	233	*
	234	* If the type can have more than one id, then use zero.
	235	*/
	236	#define trace_assign_type(var, ent) \
	237	do { \
	238	IF_ASSIGN(var, ent, struct ftrace_entry, TRACE_FN); \
	239	IF_ASSIGN(var, ent, struct ctx_switch_entry, 0); \
	240	IF_ASSIGN(var, ent, struct trace_field_cont, TRACE_CONT); \
	241	IF_ASSIGN(var, ent, struct stack_entry, TRACE_STACK); \
	242	IF_ASSIGN(var, ent, struct print_entry, TRACE_PRINT); \
	243	IF_ASSIGN(var, ent, struct special_entry, 0); \
	244	IF_ASSIGN(var, ent, struct trace_mmiotrace_rw, \
	245	TRACE_MMIO_RW); \
	246	IF_ASSIGN(var, ent, struct trace_mmiotrace_map, \
	247	TRACE_MMIO_MAP); \
74239072 FW	248	IF_ASSIGN(var, ent, struct trace_boot_call, TRACE_BOOT_CALL);\
74239072 FW	249	IF_ASSIGN(var, ent, struct trace_boot_ret, TRACE_BOOT_RET);\
9f029e83	250	IF_ASSIGN(var, ent, struct trace_branch, TRACE_BRANCH); \
74239072	251	IF_ASSIGN(var, ent, struct ftrace_ret_entry, TRACE_FN_RET);\
7104f300 SR	252	__ftrace_bad_type(); \
7104f300 SR	253	} while (0)
2c4f035f FW	254
	255	/* Return values for print_line callback */
	256	enum print_line_t {
	257	TRACE_TYPE_PARTIAL_LINE = 0, /* Retry after flushing the seq */
	258	TRACE_TYPE_HANDLED = 1,
	259	TRACE_TYPE_UNHANDLED = 2 /* Relay to other output functions */
	260	};
	261
bc0c38d1 SR	262	/*
	263	* A specific tracer, represented by methods that operate on a trace array:
	264	*/
	265	struct tracer {
	266	const char *name;
1c80025a FW	267	/* Your tracer should raise a warning if init fails */
1c80025a FW	268	int (init)(struct trace_array tr);
bc0c38d1	269	void (reset)(struct trace_array tr);
9036990d SR	270	void (start)(struct trace_array tr);
9036990d SR	271	void (stop)(struct trace_array tr);
bc0c38d1	272	void (open)(struct trace_iterator iter);
107bad8b	273	void (pipe_open)(struct trace_iterator iter);
bc0c38d1	274	void (close)(struct trace_iterator iter);
107bad8b SR	275	ssize_t (read)(struct trace_iterator iter,
	276	struct file filp, char __user ubuf,
	277	size_t cnt, loff_t *ppos);
60a11774 SR	278	#ifdef CONFIG_FTRACE_STARTUP_TEST
	279	int (selftest)(struct tracer trace,
	280	struct trace_array *tr);
	281	#endif
2c4f035f	282	enum print_line_t (print_line)(struct trace_iterator iter);
bc0c38d1 SR	283	struct tracer *next;
	284	int print_max;
	285	};
	286
214023c3 SR	287	struct trace_seq {
	288	unsigned char buffer[PAGE_SIZE];
	289	unsigned int len;
6c6c2796	290	unsigned int readpos;
214023c3 SR	291	};
214023c3 SR	292
bc0c38d1 SR	293	/*
	294	* Trace iterator - used by printout routines who present trace
	295	* results to users and which routines might sleep, etc:
	296	*/
	297	struct trace_iterator {
	298	struct trace_array *tr;
	299	struct tracer *trace;
107bad8b	300	void *private;
3928a8a2	301	struct ring_buffer_iter *buffer_iter[NR_CPUS];
4e3c3333	302
53d0aa77 SR	303	/* The below is zeroed out in pipe_read */
53d0aa77 SR	304	struct trace_seq seq;
bc0c38d1	305	struct trace_entry *ent;
4e3c3333	306	int cpu;
3928a8a2	307	u64 ts;
4e3c3333	308
bc0c38d1 SR	309	unsigned long iter_flags;
bc0c38d1 SR	310	loff_t pos;
4c11d7ae	311	long idx;
a309720c SR	312
a309720c SR	313	cpumask_t started;
bc0c38d1 SR	314	};
bc0c38d1 SR	315
9036990d	316	int tracing_is_enabled(void);
45dcd8b8	317	void trace_wake_up(void);
3928a8a2	318	void tracing_reset(struct trace_array *tr, int cpu);
bc0c38d1 SR	319	int tracing_open_generic(struct inode inode, struct file filp);
bc0c38d1 SR	320	struct dentry *tracing_init_dentry(void);
d618b3e6 IM	321	void init_tracer_sysprof_debugfs(struct dentry *d_tracer);
d618b3e6 IM	322
45dcd8b8 PP	323	struct trace_entry tracing_get_trace_entry(struct trace_array tr,
	324	struct trace_array_cpu *data);
	325	void tracing_generic_entry_update(struct trace_entry *entry,
38697053 SR	326	unsigned long flags,
38697053 SR	327	int pc);
45dcd8b8	328
bc0c38d1 SR	329	void ftrace(struct trace_array *tr,
	330	struct trace_array_cpu *data,
	331	unsigned long ip,
	332	unsigned long parent_ip,
38697053	333	unsigned long flags, int pc);
bc0c38d1 SR	334	void tracing_sched_switch_trace(struct trace_array *tr,
	335	struct trace_array_cpu *data,
	336	struct task_struct *prev,
	337	struct task_struct *next,
38697053	338	unsigned long flags, int pc);
bc0c38d1	339	void tracing_record_cmdline(struct task_struct *tsk);
57422797 IM	340
	341	void tracing_sched_wakeup_trace(struct trace_array *tr,
	342	struct trace_array_cpu *data,
	343	struct task_struct *wakee,
	344	struct task_struct *cur,
38697053	345	unsigned long flags, int pc);
f0a920d5 IM	346	void trace_special(struct trace_array *tr,
	347	struct trace_array_cpu *data,
	348	unsigned long arg1,
	349	unsigned long arg2,
38697053	350	unsigned long arg3, int pc);
6fb44b71 SR	351	void trace_function(struct trace_array *tr,
	352	struct trace_array_cpu *data,
	353	unsigned long ip,
	354	unsigned long parent_ip,
38697053	355	unsigned long flags, int pc);
15e6cb36 FW	356	void
15e6cb36 FW	357	trace_function_return(struct ftrace_retfunc *trace);
bc0c38d1	358
41bc8144 SR	359	void tracing_start_cmdline_record(void);
41bc8144 SR	360	void tracing_stop_cmdline_record(void);
e168e051 SR	361	void tracing_sched_switch_assign_trace(struct trace_array *tr);
	362	void tracing_stop_sched_switch_record(void);
	363	void tracing_start_sched_switch_record(void);
bc0c38d1 SR	364	int register_tracer(struct tracer *type);
	365	void unregister_tracer(struct tracer *type);
	366
	367	extern unsigned long nsecs_to_usecs(unsigned long nsecs);
	368
	369	extern unsigned long tracing_max_latency;
	370	extern unsigned long tracing_thresh;
	371
	372	void update_max_tr(struct trace_array tr, struct task_struct tsk, int cpu);
	373	void update_max_tr_single(struct trace_array *tr,
	374	struct task_struct *tsk, int cpu);
	375
e309b41d	376	extern cycle_t ftrace_now(int cpu);
bc0c38d1	377
606576ce	378	#ifdef CONFIG_FUNCTION_TRACER
001b6767 SR	379	void tracing_start_function_trace(void);
	380	void tracing_stop_function_trace(void);
	381	#else
	382	# define tracing_start_function_trace() do { } while (0)
	383	# define tracing_stop_function_trace() do { } while (0)
	384	#endif
	385
bc0c38d1 SR	386	#ifdef CONFIG_CONTEXT_SWITCH_TRACER
	387	typedef void
	388	(tracer_switch_func_t)(void private,
5b82a1b0	389	void *__rq,
bc0c38d1 SR	390	struct task_struct *prev,
	391	struct task_struct *next);
	392
	393	struct tracer_switch_ops {
	394	tracer_switch_func_t func;
	395	void *private;
	396	struct tracer_switch_ops *next;
	397	};
	398
bc0c38d1 SR	399	#endif /* CONFIG_CONTEXT_SWITCH_TRACER */
	400
	401	#ifdef CONFIG_DYNAMIC_FTRACE
	402	extern unsigned long ftrace_update_tot_cnt;
d05cdb25 SR	403	#define DYN_FTRACE_TEST_NAME trace_selftest_dynamic_test_func
d05cdb25 SR	404	extern int DYN_FTRACE_TEST_NAME(void);
bc0c38d1 SR	405	#endif
bc0c38d1 SR	406
60a11774	407	#ifdef CONFIG_FTRACE_STARTUP_TEST
60a11774 SR	408	extern int trace_selftest_startup_function(struct tracer *trace,
60a11774 SR	409	struct trace_array *tr);
60a11774 SR	410	extern int trace_selftest_startup_irqsoff(struct tracer *trace,
60a11774 SR	411	struct trace_array *tr);
60a11774 SR	412	extern int trace_selftest_startup_preemptoff(struct tracer *trace,
60a11774 SR	413	struct trace_array *tr);
60a11774 SR	414	extern int trace_selftest_startup_preemptirqsoff(struct tracer *trace,
60a11774 SR	415	struct trace_array *tr);
60a11774 SR	416	extern int trace_selftest_startup_wakeup(struct tracer *trace,
60a11774 SR	417	struct trace_array *tr);
fb1b6d8b SN	418	extern int trace_selftest_startup_nop(struct tracer *trace,
fb1b6d8b SN	419	struct trace_array *tr);
60a11774 SR	420	extern int trace_selftest_startup_sched_switch(struct tracer *trace,
60a11774 SR	421	struct trace_array *tr);
a6dd24f8 IM	422	extern int trace_selftest_startup_sysprof(struct tracer *trace,
a6dd24f8 IM	423	struct trace_array *tr);
80e5ea45 SR	424	extern int trace_selftest_startup_branch(struct tracer *trace,
80e5ea45 SR	425	struct trace_array *tr);
60a11774 SR	426	#endif /* CONFIG_FTRACE_STARTUP_TEST */
60a11774 SR	427
c7aafc54	428	extern void head_page(struct trace_array_cpu data);
72829bc3	429	extern int trace_seq_printf(struct trace_seq s, const char fmt, ...);
fc5e27ae PP	430	extern void trace_seq_print_cont(struct trace_seq *s,
fc5e27ae PP	431	struct trace_iterator *iter);
15e6cb36 FW	432
	433	extern int
	434	seq_print_ip_sym(struct trace_seq *s, unsigned long ip,
	435	unsigned long sym_flags);
6c6c2796 PP	436	extern ssize_t trace_seq_to_user(struct trace_seq s, char __user ubuf,
6c6c2796 PP	437	size_t cnt);
72829bc3	438	extern long ns2usecs(cycle_t nsec);
801fe400	439	extern int trace_vprintk(unsigned long ip, const char *fmt, va_list args);
c7aafc54	440
4e655519 IM	441	extern unsigned long trace_flags;
4e655519 IM	442
15e6cb36 FW	443	/* Standard output formatting function used for function return traces */
	444	#ifdef CONFIG_FUNCTION_RET_TRACER
	445	extern enum print_line_t print_return_function(struct trace_iterator *iter);
	446	#else
	447	static inline enum print_line_t
	448	print_return_function(struct trace_iterator *iter)
	449	{
	450	return TRACE_TYPE_UNHANDLED;
	451	}
	452	#endif
	453
4fcdae83 SR	454	/*
	455	* trace_iterator_flags is an enumeration that defines bit
	456	* positions into trace_flags that controls the output.
	457	*
	458	* NOTE: These bits must match the trace_options array in
	459	* trace.c.
	460	*/
4e655519 IM	461	enum trace_iterator_flags {
	462	TRACE_ITER_PRINT_PARENT = 0x01,
	463	TRACE_ITER_SYM_OFFSET = 0x02,
	464	TRACE_ITER_SYM_ADDR = 0x04,
	465	TRACE_ITER_VERBOSE = 0x08,
	466	TRACE_ITER_RAW = 0x10,
	467	TRACE_ITER_HEX = 0x20,
	468	TRACE_ITER_BIN = 0x40,
	469	TRACE_ITER_BLOCK = 0x80,
	470	TRACE_ITER_STACKTRACE = 0x100,
4ac3ba41	471	TRACE_ITER_SCHED_TREE = 0x200,
f09ce573	472	TRACE_ITER_PRINTK = 0x400,
b2a866f9	473	TRACE_ITER_PREEMPTONLY = 0x800,
9f029e83	474	TRACE_ITER_BRANCH = 0x1000,
12ef7d44	475	TRACE_ITER_ANNOTATE = 0x2000,
4e655519 IM	476	};
4e655519 IM	477
15e6cb36 FW	478	/*
	479	* TRACE_ITER_SYM_MASK masks the options in trace_flags that
	480	* control the output of kernel symbols.
	481	*/
	482	#define TRACE_ITER_SYM_MASK \
	483	(TRACE_ITER_PRINT_PARENT\|TRACE_ITER_SYM_OFFSET\|TRACE_ITER_SYM_ADDR)
	484
43a15386 FW	485	extern struct tracer nop_trace;
43a15386 FW	486
8f0a056f SR	487	/**
	488	* ftrace_preempt_disable - disable preemption scheduler safe
	489	*
	490	* When tracing can happen inside the scheduler, there exists
	491	* cases that the tracing might happen before the need_resched
	492	* flag is checked. If this happens and the tracer calls
	493	* preempt_enable (after a disable), a schedule might take place
	494	* causing an infinite recursion.
	495	*
	496	* To prevent this, we read the need_recshed flag before
	497	* disabling preemption. When we want to enable preemption we
	498	* check the flag, if it is set, then we call preempt_enable_no_resched.
	499	* Otherwise, we call preempt_enable.
	500	*
	501	* The rational for doing the above is that if need resched is set
	502	* and we have yet to reschedule, we are either in an atomic location
	503	* (where we do not need to check for scheduling) or we are inside
	504	* the scheduler and do not want to resched.
	505	*/
	506	static inline int ftrace_preempt_disable(void)
	507	{
	508	int resched;
	509
	510	resched = need_resched();
	511	preempt_disable_notrace();
	512
	513	return resched;
	514	}
	515
	516	/**
	517	* ftrace_preempt_enable - enable preemption scheduler safe
	518	* @resched: the return value from ftrace_preempt_disable
	519	*
	520	* This is a scheduler safe way to enable preemption and not miss
	521	* any preemption checks. The disabled saved the state of preemption.
	522	* If resched is set, then we were either inside an atomic or
	523	* are inside the scheduler (we would have already scheduled
	524	* otherwise). In this case, we do not want to call normal
	525	* preempt_enable, but preempt_enable_no_resched instead.
	526	*/
	527	static inline void ftrace_preempt_enable(int resched)
	528	{
	529	if (resched)
	530	preempt_enable_no_resched_notrace();
	531	else
	532	preempt_enable_notrace();
	533	}
	534
2ed84eeb	535	#ifdef CONFIG_BRANCH_TRACER
9f029e83 SR	536	extern int enable_branch_tracing(struct trace_array *tr);
	537	extern void disable_branch_tracing(void);
	538	static inline int trace_branch_enable(struct trace_array *tr)
52f232cb	539	{
9f029e83 SR	540	if (trace_flags & TRACE_ITER_BRANCH)
9f029e83 SR	541	return enable_branch_tracing(tr);
52f232cb SR	542	return 0;
52f232cb SR	543	}
9f029e83	544	static inline void trace_branch_disable(void)
52f232cb SR	545	{
52f232cb SR	546	/* due to races, always disable */
9f029e83	547	disable_branch_tracing();
52f232cb SR	548	}
52f232cb SR	549	#else
9f029e83	550	static inline int trace_branch_enable(struct trace_array *tr)
52f232cb SR	551	{
	552	return 0;
	553	}
9f029e83	554	static inline void trace_branch_disable(void)
52f232cb SR	555	{
52f232cb SR	556	}
2ed84eeb	557	#endif /* CONFIG_BRANCH_TRACER */
52f232cb	558
bc0c38d1	559	#endif /* _LINUX_KERNEL_TRACE_H */