[deliverable/linux.git] / include / trace / events / sched.h

#undef TRACE_SYSTEM
#define TRACE_SYSTEM sched

#if !defined(_TRACE_SCHED_H) || defined(TRACE_HEADER_MULTI_READ)
#define _TRACE_SCHED_H

#include <linux/sched.h>
#include <linux/tracepoint.h>
#include <linux/binfmts.h>

/*
 * Tracepoint for calling kthread_stop, performed to end a kthread:
 */
TRACE_EVENT(sched_kthread_stop,

	TP_PROTO(struct task_struct *t),

	TP_ARGS(t),

	TP_STRUCT__entry(
		__array(	char,	comm,	TASK_COMM_LEN	)
		__field(	pid_t,	pid			)
	),

	TP_fast_assign(
		memcpy(__entry->comm, t->comm, TASK_COMM_LEN);
		__entry->pid	= t->pid;
	),

	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
);

/*
 * Tracepoint for the return value of the kthread stopping:
 */
TRACE_EVENT(sched_kthread_stop_ret,

	TP_PROTO(int ret),

	TP_ARGS(ret),

	TP_STRUCT__entry(
		__field(	int,	ret	)
	),

	TP_fast_assign(
		__entry->ret	= ret;
	),

	TP_printk("ret=%d", __entry->ret)
);

/*
 * Tracepoint for waking up a task:
 */
DECLARE_EVENT_CLASS(sched_wakeup_template,

	TP_PROTO(struct task_struct *p),

	TP_ARGS(__perf_task(p)),

	TP_STRUCT__entry(
		__array(	char,	comm,	TASK_COMM_LEN	)
		__field(	pid_t,	pid			)
		__field(	int,	prio			)
		__field(	int,	success			)
		__field(	int,	target_cpu		)
	),

	TP_fast_assign(
		memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
		__entry->pid		= p->pid;
		__entry->prio		= p->prio;
		__entry->success	= 1; /* rudiment, kill when possible */
		__entry->target_cpu	= task_cpu(p);
	),

	TP_printk("comm=%s pid=%d prio=%d target_cpu=%03d",
		  __entry->comm, __entry->pid, __entry->prio,
		  __entry->target_cpu)
);

/*
 * Tracepoint called when waking a task; this tracepoint is guaranteed to be
 * called from the waking context.
 */
DEFINE_EVENT(sched_wakeup_template, sched_waking,
	     TP_PROTO(struct task_struct *p),
	     TP_ARGS(p));

/*
 * Tracepoint called when the task is actually woken; p->state == TASK_RUNNNG.
 * It it not always called from the waking context.
 */
DEFINE_EVENT(sched_wakeup_template, sched_wakeup,
	     TP_PROTO(struct task_struct *p),
	     TP_ARGS(p));

/*
 * Tracepoint for waking up a new task:
 */
DEFINE_EVENT(sched_wakeup_template, sched_wakeup_new,
	     TP_PROTO(struct task_struct *p),
	     TP_ARGS(p));

#ifdef CREATE_TRACE_POINTS
static inline long __trace_sched_switch_state(struct task_struct *p)
{
	long state = p->state;

#ifdef CONFIG_PREEMPT
#ifdef CONFIG_SCHED_DEBUG
	BUG_ON(p != current);
#endif /* CONFIG_SCHED_DEBUG */
	/*
	 * For all intents and purposes a preempted task is a running task.
	 */
	if (preempt_count() & PREEMPT_ACTIVE)
		state = TASK_RUNNING | TASK_STATE_MAX;
#endif /* CONFIG_PREEMPT */

	return state;
}
#endif /* CREATE_TRACE_POINTS */

/*
 * Tracepoint for task switches, performed by the scheduler:
 */
TRACE_EVENT(sched_switch,

	TP_PROTO(struct task_struct *prev,
		 struct task_struct *next),

	TP_ARGS(prev, next),

	TP_STRUCT__entry(
		__array(	char,	prev_comm,	TASK_COMM_LEN	)
		__field(	pid_t,	prev_pid			)
		__field(	int,	prev_prio			)
		__field(	long,	prev_state			)
		__array(	char,	next_comm,	TASK_COMM_LEN	)
		__field(	pid_t,	next_pid			)
		__field(	int,	next_prio			)
	),

	TP_fast_assign(
		memcpy(__entry->next_comm, next->comm, TASK_COMM_LEN);
		__entry->prev_pid	= prev->pid;
		__entry->prev_prio	= prev->prio;
		__entry->prev_state	= __trace_sched_switch_state(prev);
		memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
		__entry->next_pid	= next->pid;
		__entry->next_prio	= next->prio;
	),

	TP_printk("prev_comm=%s prev_pid=%d prev_prio=%d prev_state=%s%s ==> next_comm=%s next_pid=%d next_prio=%d",
		__entry->prev_comm, __entry->prev_pid, __entry->prev_prio,
		__entry->prev_state & (TASK_STATE_MAX-1) ?
		  __print_flags(__entry->prev_state & (TASK_STATE_MAX-1), "|",
				{ 1, "S"} , { 2, "D" }, { 4, "T" }, { 8, "t" },
				{ 16, "Z" }, { 32, "X" }, { 64, "x" },
				{ 128, "K" }, { 256, "W" }, { 512, "P" },
				{ 1024, "N" }) : "R",
		__entry->prev_state & TASK_STATE_MAX ? "+" : "",
		__entry->next_comm, __entry->next_pid, __entry->next_prio)
);

/*
 * Tracepoint for a task being migrated:
 */
TRACE_EVENT(sched_migrate_task,

	TP_PROTO(struct task_struct *p, int dest_cpu),

	TP_ARGS(p, dest_cpu),

	TP_STRUCT__entry(
		__array(	char,	comm,	TASK_COMM_LEN	)
		__field(	pid_t,	pid			)
		__field(	int,	prio			)
		__field(	int,	orig_cpu		)
		__field(	int,	dest_cpu		)
	),

	TP_fast_assign(
		memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
		__entry->pid		= p->pid;
		__entry->prio		= p->prio;
		__entry->orig_cpu	= task_cpu(p);
		__entry->dest_cpu	= dest_cpu;
	),

	TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d",
		  __entry->comm, __entry->pid, __entry->prio,
		  __entry->orig_cpu, __entry->dest_cpu)
);

DECLARE_EVENT_CLASS(sched_process_template,

	TP_PROTO(struct task_struct *p),

	TP_ARGS(p),

	TP_STRUCT__entry(
		__array(	char,	comm,	TASK_COMM_LEN	)
		__field(	pid_t,	pid			)
		__field(	int,	prio			)
	),

	TP_fast_assign(
		memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
		__entry->pid		= p->pid;
		__entry->prio		= p->prio;
	),

	TP_printk("comm=%s pid=%d prio=%d",
		  __entry->comm, __entry->pid, __entry->prio)
);

/*
 * Tracepoint for freeing a task:
 */
DEFINE_EVENT(sched_process_template, sched_process_free,
	     TP_PROTO(struct task_struct *p),
	     TP_ARGS(p));
	     

/*
 * Tracepoint for a task exiting:
 */
DEFINE_EVENT(sched_process_template, sched_process_exit,
	     TP_PROTO(struct task_struct *p),
	     TP_ARGS(p));

/*
 * Tracepoint for waiting on task to unschedule:
 */
DEFINE_EVENT(sched_process_template, sched_wait_task,
	TP_PROTO(struct task_struct *p),
	TP_ARGS(p));

/*
 * Tracepoint for a waiting task:
 */
TRACE_EVENT(sched_process_wait,

	TP_PROTO(struct pid *pid),

	TP_ARGS(pid),

	TP_STRUCT__entry(
		__array(	char,	comm,	TASK_COMM_LEN	)
		__field(	pid_t,	pid			)
		__field(	int,	prio			)
	),

	TP_fast_assign(
		memcpy(__entry->comm, current->comm, TASK_COMM_LEN);
		__entry->pid		= pid_nr(pid);
		__entry->prio		= current->prio;
	),

	TP_printk("comm=%s pid=%d prio=%d",
		  __entry->comm, __entry->pid, __entry->prio)
);

/*
 * Tracepoint for do_fork:
 */
TRACE_EVENT(sched_process_fork,

	TP_PROTO(struct task_struct *parent, struct task_struct *child),

	TP_ARGS(parent, child),

	TP_STRUCT__entry(
		__array(	char,	parent_comm,	TASK_COMM_LEN	)
		__field(	pid_t,	parent_pid			)
		__array(	char,	child_comm,	TASK_COMM_LEN	)
		__field(	pid_t,	child_pid			)
	),

	TP_fast_assign(
		memcpy(__entry->parent_comm, parent->comm, TASK_COMM_LEN);
		__entry->parent_pid	= parent->pid;
		memcpy(__entry->child_comm, child->comm, TASK_COMM_LEN);
		__entry->child_pid	= child->pid;
	),

	TP_printk("comm=%s pid=%d child_comm=%s child_pid=%d",
		__entry->parent_comm, __entry->parent_pid,
		__entry->child_comm, __entry->child_pid)
);

/*
 * Tracepoint for exec:
 */
TRACE_EVENT(sched_process_exec,

	TP_PROTO(struct task_struct *p, pid_t old_pid,
		 struct linux_binprm *bprm),

	TP_ARGS(p, old_pid, bprm),

	TP_STRUCT__entry(
		__string(	filename,	bprm->filename	)
		__field(	pid_t,		pid		)
		__field(	pid_t,		old_pid		)
	),

	TP_fast_assign(
		__assign_str(filename, bprm->filename);
		__entry->pid		= p->pid;
		__entry->old_pid	= old_pid;
	),

	TP_printk("filename=%s pid=%d old_pid=%d", __get_str(filename),
		  __entry->pid, __entry->old_pid)
);

/*
 * XXX the below sched_stat tracepoints only apply to SCHED_OTHER/BATCH/IDLE
 *     adding sched_stat support to SCHED_FIFO/RR would be welcome.
 */
DECLARE_EVENT_CLASS(sched_stat_template,

	TP_PROTO(struct task_struct *tsk, u64 delay),

	TP_ARGS(__perf_task(tsk), __perf_count(delay)),

	TP_STRUCT__entry(
		__array( char,	comm,	TASK_COMM_LEN	)
		__field( pid_t,	pid			)
		__field( u64,	delay			)
	),

	TP_fast_assign(
		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
		__entry->pid	= tsk->pid;
		__entry->delay	= delay;
	),

	TP_printk("comm=%s pid=%d delay=%Lu [ns]",
			__entry->comm, __entry->pid,
			(unsigned long long)__entry->delay)
);


/*
 * Tracepoint for accounting wait time (time the task is runnable
 * but not actually running due to scheduler contention).
 */
DEFINE_EVENT(sched_stat_template, sched_stat_wait,
	     TP_PROTO(struct task_struct *tsk, u64 delay),
	     TP_ARGS(tsk, delay));

/*
 * Tracepoint for accounting sleep time (time the task is not runnable,
 * including iowait, see below).
 */
DEFINE_EVENT(sched_stat_template, sched_stat_sleep,
	     TP_PROTO(struct task_struct *tsk, u64 delay),
	     TP_ARGS(tsk, delay));

/*
 * Tracepoint for accounting iowait time (time the task is not runnable
 * due to waiting on IO to complete).
 */
DEFINE_EVENT(sched_stat_template, sched_stat_iowait,
	     TP_PROTO(struct task_struct *tsk, u64 delay),
	     TP_ARGS(tsk, delay));

/*
 * Tracepoint for accounting blocked time (time the task is in uninterruptible).
 */
DEFINE_EVENT(sched_stat_template, sched_stat_blocked,
	     TP_PROTO(struct task_struct *tsk, u64 delay),
	     TP_ARGS(tsk, delay));

/*
 * Tracepoint for accounting runtime (time the task is executing
 * on a CPU).
 */
DECLARE_EVENT_CLASS(sched_stat_runtime,

	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),

	TP_ARGS(tsk, __perf_count(runtime), vruntime),

	TP_STRUCT__entry(
		__array( char,	comm,	TASK_COMM_LEN	)
		__field( pid_t,	pid			)
		__field( u64,	runtime			)
		__field( u64,	vruntime			)
	),

	TP_fast_assign(
		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
		__entry->pid		= tsk->pid;
		__entry->runtime	= runtime;
		__entry->vruntime	= vruntime;
	),

	TP_printk("comm=%s pid=%d runtime=%Lu [ns] vruntime=%Lu [ns]",
			__entry->comm, __entry->pid,
			(unsigned long long)__entry->runtime,
			(unsigned long long)__entry->vruntime)
);

DEFINE_EVENT(sched_stat_runtime, sched_stat_runtime,
	     TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	     TP_ARGS(tsk, runtime, vruntime));

/*
 * Tracepoint for showing priority inheritance modifying a tasks
 * priority.
 */
TRACE_EVENT(sched_pi_setprio,

	TP_PROTO(struct task_struct *tsk, int newprio),

	TP_ARGS(tsk, newprio),

	TP_STRUCT__entry(
		__array( char,	comm,	TASK_COMM_LEN	)
		__field( pid_t,	pid			)
		__field( int,	oldprio			)
		__field( int,	newprio			)
	),

	TP_fast_assign(
		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
		__entry->pid		= tsk->pid;
		__entry->oldprio	= tsk->prio;
		__entry->newprio	= newprio;
	),

	TP_printk("comm=%s pid=%d oldprio=%d newprio=%d",
			__entry->comm, __entry->pid,
			__entry->oldprio, __entry->newprio)
);

#ifdef CONFIG_DETECT_HUNG_TASK
TRACE_EVENT(sched_process_hang,
	TP_PROTO(struct task_struct *tsk),
	TP_ARGS(tsk),

	TP_STRUCT__entry(
		__array( char,	comm,	TASK_COMM_LEN	)
		__field( pid_t,	pid			)
	),

	TP_fast_assign(
		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
		__entry->pid = tsk->pid;
	),

	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
);
#endif /* CONFIG_DETECT_HUNG_TASK */

DECLARE_EVENT_CLASS(sched_move_task_template,

	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),

	TP_ARGS(tsk, src_cpu, dst_cpu),

	TP_STRUCT__entry(
		__field( pid_t,	pid			)
		__field( pid_t,	tgid			)
		__field( pid_t,	ngid			)
		__field( int,	src_cpu			)
		__field( int,	src_nid			)
		__field( int,	dst_cpu			)
		__field( int,	dst_nid			)
	),

	TP_fast_assign(
		__entry->pid		= task_pid_nr(tsk);
		__entry->tgid		= task_tgid_nr(tsk);
		__entry->ngid		= task_numa_group_id(tsk);
		__entry->src_cpu	= src_cpu;
		__entry->src_nid	= cpu_to_node(src_cpu);
		__entry->dst_cpu	= dst_cpu;
		__entry->dst_nid	= cpu_to_node(dst_cpu);
	),

	TP_printk("pid=%d tgid=%d ngid=%d src_cpu=%d src_nid=%d dst_cpu=%d dst_nid=%d",
			__entry->pid, __entry->tgid, __entry->ngid,
			__entry->src_cpu, __entry->src_nid,
			__entry->dst_cpu, __entry->dst_nid)
);

/*
 * Tracks migration of tasks from one runqueue to another. Can be used to
 * detect if automatic NUMA balancing is bouncing between nodes
 */
DEFINE_EVENT(sched_move_task_template, sched_move_numa,
	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),

	TP_ARGS(tsk, src_cpu, dst_cpu)
);

DEFINE_EVENT(sched_move_task_template, sched_stick_numa,
	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),

	TP_ARGS(tsk, src_cpu, dst_cpu)
);

TRACE_EVENT(sched_swap_numa,

	TP_PROTO(struct task_struct *src_tsk, int src_cpu,
		 struct task_struct *dst_tsk, int dst_cpu),

	TP_ARGS(src_tsk, src_cpu, dst_tsk, dst_cpu),

	TP_STRUCT__entry(
		__field( pid_t,	src_pid			)
		__field( pid_t,	src_tgid		)
		__field( pid_t,	src_ngid		)
		__field( int,	src_cpu			)
		__field( int,	src_nid			)
		__field( pid_t,	dst_pid			)
		__field( pid_t,	dst_tgid		)
		__field( pid_t,	dst_ngid		)
		__field( int,	dst_cpu			)
		__field( int,	dst_nid			)
	),

	TP_fast_assign(
		__entry->src_pid	= task_pid_nr(src_tsk);
		__entry->src_tgid	= task_tgid_nr(src_tsk);
		__entry->src_ngid	= task_numa_group_id(src_tsk);
		__entry->src_cpu	= src_cpu;
		__entry->src_nid	= cpu_to_node(src_cpu);
		__entry->dst_pid	= task_pid_nr(dst_tsk);
		__entry->dst_tgid	= task_tgid_nr(dst_tsk);
		__entry->dst_ngid	= task_numa_group_id(dst_tsk);
		__entry->dst_cpu	= dst_cpu;
		__entry->dst_nid	= cpu_to_node(dst_cpu);
	),

	TP_printk("src_pid=%d src_tgid=%d src_ngid=%d src_cpu=%d src_nid=%d dst_pid=%d dst_tgid=%d dst_ngid=%d dst_cpu=%d dst_nid=%d",
			__entry->src_pid, __entry->src_tgid, __entry->src_ngid,
			__entry->src_cpu, __entry->src_nid,
			__entry->dst_pid, __entry->dst_tgid, __entry->dst_ngid,
			__entry->dst_cpu, __entry->dst_nid)
);

/*
 * Tracepoint for waking a polling cpu without an IPI.
 */
TRACE_EVENT(sched_wake_idle_without_ipi,

	TP_PROTO(int cpu),

	TP_ARGS(cpu),

	TP_STRUCT__entry(
		__field(	int,	cpu	)
	),

	TP_fast_assign(
		__entry->cpu	= cpu;
	),

	TP_printk("cpu=%d", __entry->cpu)
);
#endif /* _TRACE_SCHED_H */

/* This part must be outside protection */
#include <trace/define_trace.h>
Commit	Line	Data
d0b6e04a LZ	1	#undef TRACE_SYSTEM
	2	#define TRACE_SYSTEM sched
	3
ea20d929	4	#if !defined(_TRACE_SCHED_H) \|\| defined(TRACE_HEADER_MULTI_READ)
0a16b607 MD	5	#define _TRACE_SCHED_H
	6
	7	#include <linux/sched.h>
	8	#include <linux/tracepoint.h>
4ff16c25	9	#include <linux/binfmts.h>
0a16b607	10
ea20d929 SR	11	/*
	12	* Tracepoint for calling kthread_stop, performed to end a kthread:
	13	*/
	14	TRACE_EVENT(sched_kthread_stop,
	15
	16	TP_PROTO(struct task_struct *t),
	17
	18	TP_ARGS(t),
	19
	20	TP_STRUCT__entry(
	21	__array( char, comm, TASK_COMM_LEN )
	22	__field( pid_t, pid )
	23	),
	24
	25	TP_fast_assign(
	26	memcpy(__entry->comm, t->comm, TASK_COMM_LEN);
	27	__entry->pid = t->pid;
	28	),
	29
434a83c3	30	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
ea20d929 SR	31	);
	32
	33	/*
	34	* Tracepoint for the return value of the kthread stopping:
	35	*/
	36	TRACE_EVENT(sched_kthread_stop_ret,
	37
	38	TP_PROTO(int ret),
	39
	40	TP_ARGS(ret),
	41
	42	TP_STRUCT__entry(
	43	__field( int, ret )
	44	),
	45
	46	TP_fast_assign(
	47	__entry->ret = ret;
	48	),
	49
434a83c3	50	TP_printk("ret=%d", __entry->ret)
ea20d929 SR	51	);
ea20d929 SR	52
ea20d929 SR	53	/*
ea20d929 SR	54	* Tracepoint for waking up a task:
ea20d929	55	*/
091ad365	56	DECLARE_EVENT_CLASS(sched_wakeup_template,
ea20d929	57
fbd705a0	58	TP_PROTO(struct task_struct *p),
ea20d929	59
fbd705a0	60	TP_ARGS(__perf_task(p)),
ea20d929 SR	61
	62	TP_STRUCT__entry(
	63	__array( char, comm, TASK_COMM_LEN )
	64	__field( pid_t, pid )
	65	__field( int, prio )
	66	__field( int, success )
434a83c3	67	__field( int, target_cpu )
ea20d929 SR	68	),
	69
	70	TP_fast_assign(
	71	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	72	__entry->pid = p->pid;
	73	__entry->prio = p->prio;
fbd705a0	74	__entry->success = 1; /* rudiment, kill when possible */
434a83c3	75	__entry->target_cpu = task_cpu(p);
ea20d929 SR	76	),
ea20d929 SR	77
fbd705a0	78	TP_printk("comm=%s pid=%d prio=%d target_cpu=%03d",
ea20d929	79	__entry->comm, __entry->pid, __entry->prio,
fbd705a0	80	__entry->target_cpu)
ea20d929 SR	81	);
ea20d929 SR	82
fbd705a0 PZ	83	/*
	84	* Tracepoint called when waking a task; this tracepoint is guaranteed to be
	85	* called from the waking context.
	86	*/
	87	DEFINE_EVENT(sched_wakeup_template, sched_waking,
	88	TP_PROTO(struct task_struct *p),
	89	TP_ARGS(p));
	90
	91	/*
	92	* Tracepoint called when the task is actually woken; p->state == TASK_RUNNNG.
	93	* It it not always called from the waking context.
	94	*/
75ec29ab	95	DEFINE_EVENT(sched_wakeup_template, sched_wakeup,
fbd705a0 PZ	96	TP_PROTO(struct task_struct *p),
fbd705a0 PZ	97	TP_ARGS(p));
75ec29ab	98
ea20d929 SR	99	/*
ea20d929 SR	100	* Tracepoint for waking up a new task:
ea20d929	101	*/
75ec29ab	102	DEFINE_EVENT(sched_wakeup_template, sched_wakeup_new,
fbd705a0 PZ	103	TP_PROTO(struct task_struct *p),
fbd705a0 PZ	104	TP_ARGS(p));
ea20d929	105
02f72694 PZ	106	#ifdef CREATE_TRACE_POINTS
	107	static inline long __trace_sched_switch_state(struct task_struct *p)
	108	{
	109	long state = p->state;
	110
	111	#ifdef CONFIG_PREEMPT
8f9fbf09 ON	112	#ifdef CONFIG_SCHED_DEBUG
	113	BUG_ON(p != current);
	114	#endif /* CONFIG_SCHED_DEBUG */
02f72694 PZ	115	/*
	116	* For all intents and purposes a preempted task is a running task.
	117	*/
8f9fbf09	118	if (preempt_count() & PREEMPT_ACTIVE)
557ab425	119	state = TASK_RUNNING \| TASK_STATE_MAX;
8f9fbf09	120	#endif /* CONFIG_PREEMPT */
02f72694 PZ	121
	122	return state;
	123	}
8f9fbf09	124	#endif /* CREATE_TRACE_POINTS */
02f72694	125
ea20d929 SR	126	/*
ea20d929 SR	127	* Tracepoint for task switches, performed by the scheduler:
ea20d929 SR	128	*/
	129	TRACE_EVENT(sched_switch,
	130
27a9da65	131	TP_PROTO(struct task_struct *prev,
ea20d929 SR	132	struct task_struct *next),
ea20d929 SR	133
27a9da65	134	TP_ARGS(prev, next),
ea20d929 SR	135
	136	TP_STRUCT__entry(
	137	__array( char, prev_comm, TASK_COMM_LEN )
	138	__field( pid_t, prev_pid )
	139	__field( int, prev_prio )
937cdb9d	140	__field( long, prev_state )
ea20d929 SR	141	__array( char, next_comm, TASK_COMM_LEN )
	142	__field( pid_t, next_pid )
	143	__field( int, next_prio )
	144	),
	145
	146	TP_fast_assign(
	147	memcpy(__entry->next_comm, next->comm, TASK_COMM_LEN);
	148	__entry->prev_pid = prev->pid;
	149	__entry->prev_prio = prev->prio;
02f72694	150	__entry->prev_state = __trace_sched_switch_state(prev);
ea20d929 SR	151	memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
	152	__entry->next_pid = next->pid;
	153	__entry->next_prio = next->prio;
	154	),
	155
557ab425	156	TP_printk("prev_comm=%s prev_pid=%d prev_prio=%d prev_state=%s%s ==> next_comm=%s next_pid=%d next_prio=%d",
ea20d929	157	__entry->prev_comm, __entry->prev_pid, __entry->prev_prio,
557ab425 PZ	158	__entry->prev_state & (TASK_STATE_MAX-1) ?
557ab425 PZ	159	__print_flags(__entry->prev_state & (TASK_STATE_MAX-1), "\|",
937cdb9d SR	160	{ 1, "S"} , { 2, "D" }, { 4, "T" }, { 8, "t" },
937cdb9d SR	161	{ 16, "Z" }, { 32, "X" }, { 64, "x" },
80ed87c8 PZ	162	{ 128, "K" }, { 256, "W" }, { 512, "P" },
80ed87c8 PZ	163	{ 1024, "N" }) : "R",
557ab425	164	__entry->prev_state & TASK_STATE_MAX ? "+" : "",
ea20d929 SR	165	__entry->next_comm, __entry->next_pid, __entry->next_prio)
	166	);
	167
	168	/*
	169	* Tracepoint for a task being migrated:
	170	*/
	171	TRACE_EVENT(sched_migrate_task,
	172
de1d7286	173	TP_PROTO(struct task_struct *p, int dest_cpu),
ea20d929	174
de1d7286	175	TP_ARGS(p, dest_cpu),
ea20d929 SR	176
	177	TP_STRUCT__entry(
	178	__array( char, comm, TASK_COMM_LEN )
	179	__field( pid_t, pid )
	180	__field( int, prio )
	181	__field( int, orig_cpu )
	182	__field( int, dest_cpu )
	183	),
	184
	185	TP_fast_assign(
	186	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	187	__entry->pid = p->pid;
	188	__entry->prio = p->prio;
de1d7286	189	__entry->orig_cpu = task_cpu(p);
ea20d929 SR	190	__entry->dest_cpu = dest_cpu;
	191	),
	192
434a83c3	193	TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d",
ea20d929 SR	194	__entry->comm, __entry->pid, __entry->prio,
	195	__entry->orig_cpu, __entry->dest_cpu)
	196	);
	197
091ad365	198	DECLARE_EVENT_CLASS(sched_process_template,
ea20d929 SR	199
	200	TP_PROTO(struct task_struct *p),
	201
	202	TP_ARGS(p),
	203
	204	TP_STRUCT__entry(
	205	__array( char, comm, TASK_COMM_LEN )
	206	__field( pid_t, pid )
	207	__field( int, prio )
	208	),
	209
	210	TP_fast_assign(
	211	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	212	__entry->pid = p->pid;
	213	__entry->prio = p->prio;
	214	),
	215
434a83c3	216	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	217	__entry->comm, __entry->pid, __entry->prio)
	218	);
	219
	220	/*
75ec29ab	221	* Tracepoint for freeing a task:
ea20d929	222	*/
75ec29ab SR	223	DEFINE_EVENT(sched_process_template, sched_process_free,
	224	TP_PROTO(struct task_struct *p),
	225	TP_ARGS(p));
	226
ea20d929	227
75ec29ab SR	228	/*
	229	* Tracepoint for a task exiting:
	230	*/
	231	DEFINE_EVENT(sched_process_template, sched_process_exit,
	232	TP_PROTO(struct task_struct *p),
	233	TP_ARGS(p));
ea20d929	234
210f7669 LZ	235	/*
	236	* Tracepoint for waiting on task to unschedule:
	237	*/
	238	DEFINE_EVENT(sched_process_template, sched_wait_task,
	239	TP_PROTO(struct task_struct *p),
	240	TP_ARGS(p));
	241
ea20d929 SR	242	/*
	243	* Tracepoint for a waiting task:
	244	*/
	245	TRACE_EVENT(sched_process_wait,
	246
	247	TP_PROTO(struct pid *pid),
	248
	249	TP_ARGS(pid),
	250
	251	TP_STRUCT__entry(
	252	__array( char, comm, TASK_COMM_LEN )
	253	__field( pid_t, pid )
	254	__field( int, prio )
	255	),
	256
	257	TP_fast_assign(
	258	memcpy(__entry->comm, current->comm, TASK_COMM_LEN);
	259	__entry->pid = pid_nr(pid);
	260	__entry->prio = current->prio;
	261	),
	262
434a83c3	263	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	264	__entry->comm, __entry->pid, __entry->prio)
	265	);
	266
	267	/*
	268	* Tracepoint for do_fork:
	269	*/
	270	TRACE_EVENT(sched_process_fork,
	271
	272	TP_PROTO(struct task_struct parent, struct task_struct child),
	273
	274	TP_ARGS(parent, child),
	275
	276	TP_STRUCT__entry(
	277	__array( char, parent_comm, TASK_COMM_LEN )
	278	__field( pid_t, parent_pid )
	279	__array( char, child_comm, TASK_COMM_LEN )
	280	__field( pid_t, child_pid )
	281	),
	282
	283	TP_fast_assign(
	284	memcpy(__entry->parent_comm, parent->comm, TASK_COMM_LEN);
	285	__entry->parent_pid = parent->pid;
	286	memcpy(__entry->child_comm, child->comm, TASK_COMM_LEN);
	287	__entry->child_pid = child->pid;
	288	),
	289
434a83c3	290	TP_printk("comm=%s pid=%d child_comm=%s child_pid=%d",
ea20d929 SR	291	__entry->parent_comm, __entry->parent_pid,
	292	__entry->child_comm, __entry->child_pid)
	293	);
	294
4ff16c25 DS	295	/*
	296	* Tracepoint for exec:
	297	*/
	298	TRACE_EVENT(sched_process_exec,
	299
	300	TP_PROTO(struct task_struct *p, pid_t old_pid,
	301	struct linux_binprm *bprm),
	302
	303	TP_ARGS(p, old_pid, bprm),
	304
	305	TP_STRUCT__entry(
	306	__string( filename, bprm->filename )
	307	__field( pid_t, pid )
	308	__field( pid_t, old_pid )
	309	),
	310
	311	TP_fast_assign(
	312	__assign_str(filename, bprm->filename);
	313	__entry->pid = p->pid;
6308191f	314	__entry->old_pid = old_pid;
4ff16c25 DS	315	),
	316
	317	TP_printk("filename=%s pid=%d old_pid=%d", __get_str(filename),
	318	__entry->pid, __entry->old_pid)
	319	);
	320
768d0c27 PZ	321	/*
	322	* XXX the below sched_stat tracepoints only apply to SCHED_OTHER/BATCH/IDLE
	323	* adding sched_stat support to SCHED_FIFO/RR would be welcome.
	324	*/
091ad365	325	DECLARE_EVENT_CLASS(sched_stat_template,
768d0c27 PZ	326
	327	TP_PROTO(struct task_struct *tsk, u64 delay),
	328
12473965	329	TP_ARGS(__perf_task(tsk), __perf_count(delay)),
768d0c27 PZ	330
	331	TP_STRUCT__entry(
	332	__array( char, comm, TASK_COMM_LEN )
	333	__field( pid_t, pid )
	334	__field( u64, delay )
	335	),
	336
	337	TP_fast_assign(
	338	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	339	__entry->pid = tsk->pid;
	340	__entry->delay = delay;
768d0c27 PZ	341	),
768d0c27 PZ	342
434a83c3	343	TP_printk("comm=%s pid=%d delay=%Lu [ns]",
768d0c27 PZ	344	__entry->comm, __entry->pid,
	345	(unsigned long long)__entry->delay)
	346	);
	347
75ec29ab SR	348
	349	/*
	350	* Tracepoint for accounting wait time (time the task is runnable
	351	* but not actually running due to scheduler contention).
	352	*/
	353	DEFINE_EVENT(sched_stat_template, sched_stat_wait,
	354	TP_PROTO(struct task_struct *tsk, u64 delay),
	355	TP_ARGS(tsk, delay));
	356
	357	/*
	358	* Tracepoint for accounting sleep time (time the task is not runnable,
	359	* including iowait, see below).
	360	*/
470dda74 LZ	361	DEFINE_EVENT(sched_stat_template, sched_stat_sleep,
	362	TP_PROTO(struct task_struct *tsk, u64 delay),
	363	TP_ARGS(tsk, delay));
75ec29ab SR	364
	365	/*
	366	* Tracepoint for accounting iowait time (time the task is not runnable
	367	* due to waiting on IO to complete).
	368	*/
470dda74 LZ	369	DEFINE_EVENT(sched_stat_template, sched_stat_iowait,
	370	TP_PROTO(struct task_struct *tsk, u64 delay),
	371	TP_ARGS(tsk, delay));
75ec29ab	372
b781a602 AV	373	/*
	374	* Tracepoint for accounting blocked time (time the task is in uninterruptible).
	375	*/
	376	DEFINE_EVENT(sched_stat_template, sched_stat_blocked,
	377	TP_PROTO(struct task_struct *tsk, u64 delay),
	378	TP_ARGS(tsk, delay));
	379
f977bb49 IM	380	/*
	381	* Tracepoint for accounting runtime (time the task is executing
	382	* on a CPU).
	383	*/
36009d07	384	DECLARE_EVENT_CLASS(sched_stat_runtime,
f977bb49 IM	385
	386	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	387
12473965	388	TP_ARGS(tsk, __perf_count(runtime), vruntime),
f977bb49 IM	389
	390	TP_STRUCT__entry(
	391	__array( char, comm, TASK_COMM_LEN )
	392	__field( pid_t, pid )
	393	__field( u64, runtime )
	394	__field( u64, vruntime )
	395	),
	396
	397	TP_fast_assign(
	398	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	399	__entry->pid = tsk->pid;
	400	__entry->runtime = runtime;
	401	__entry->vruntime = vruntime;
f977bb49 IM	402	),
f977bb49 IM	403
434a83c3	404	TP_printk("comm=%s pid=%d runtime=%Lu [ns] vruntime=%Lu [ns]",
f977bb49 IM	405	__entry->comm, __entry->pid,
	406	(unsigned long long)__entry->runtime,
	407	(unsigned long long)__entry->vruntime)
	408	);
	409
36009d07 ON	410	DEFINE_EVENT(sched_stat_runtime, sched_stat_runtime,
	411	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	412	TP_ARGS(tsk, runtime, vruntime));
	413
a8027073 SR	414	/*
	415	* Tracepoint for showing priority inheritance modifying a tasks
	416	* priority.
	417	*/
	418	TRACE_EVENT(sched_pi_setprio,
	419
	420	TP_PROTO(struct task_struct *tsk, int newprio),
	421
	422	TP_ARGS(tsk, newprio),
	423
	424	TP_STRUCT__entry(
	425	__array( char, comm, TASK_COMM_LEN )
	426	__field( pid_t, pid )
	427	__field( int, oldprio )
	428	__field( int, newprio )
	429	),
	430
	431	TP_fast_assign(
	432	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	433	__entry->pid = tsk->pid;
	434	__entry->oldprio = tsk->prio;
	435	__entry->newprio = newprio;
	436	),
	437
	438	TP_printk("comm=%s pid=%d oldprio=%d newprio=%d",
	439	__entry->comm, __entry->pid,
	440	__entry->oldprio, __entry->newprio)
	441	);
	442
6a716c90 ON	443	#ifdef CONFIG_DETECT_HUNG_TASK
	444	TRACE_EVENT(sched_process_hang,
	445	TP_PROTO(struct task_struct *tsk),
	446	TP_ARGS(tsk),
	447
	448	TP_STRUCT__entry(
	449	__array( char, comm, TASK_COMM_LEN )
	450	__field( pid_t, pid )
	451	),
	452
	453	TP_fast_assign(
	454	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	455	__entry->pid = tsk->pid;
	456	),
	457
	458	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
	459	);
	460	#endif /* CONFIG_DETECT_HUNG_TASK */
	461
286549dc MG	462	DECLARE_EVENT_CLASS(sched_move_task_template,
	463
	464	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	465
	466	TP_ARGS(tsk, src_cpu, dst_cpu),
	467
	468	TP_STRUCT__entry(
	469	__field( pid_t, pid )
	470	__field( pid_t, tgid )
	471	__field( pid_t, ngid )
	472	__field( int, src_cpu )
	473	__field( int, src_nid )
	474	__field( int, dst_cpu )
	475	__field( int, dst_nid )
	476	),
	477
	478	TP_fast_assign(
	479	__entry->pid = task_pid_nr(tsk);
	480	__entry->tgid = task_tgid_nr(tsk);
	481	__entry->ngid = task_numa_group_id(tsk);
	482	__entry->src_cpu = src_cpu;
	483	__entry->src_nid = cpu_to_node(src_cpu);
	484	__entry->dst_cpu = dst_cpu;
	485	__entry->dst_nid = cpu_to_node(dst_cpu);
	486	),
	487
	488	TP_printk("pid=%d tgid=%d ngid=%d src_cpu=%d src_nid=%d dst_cpu=%d dst_nid=%d",
	489	__entry->pid, __entry->tgid, __entry->ngid,
	490	__entry->src_cpu, __entry->src_nid,
	491	__entry->dst_cpu, __entry->dst_nid)
	492	);
	493
	494	/*
	495	* Tracks migration of tasks from one runqueue to another. Can be used to
	496	* detect if automatic NUMA balancing is bouncing between nodes
	497	*/
	498	DEFINE_EVENT(sched_move_task_template, sched_move_numa,
	499	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	500
	501	TP_ARGS(tsk, src_cpu, dst_cpu)
	502	);
	503
	504	DEFINE_EVENT(sched_move_task_template, sched_stick_numa,
	505	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	506
	507	TP_ARGS(tsk, src_cpu, dst_cpu)
	508	);
	509
	510	TRACE_EVENT(sched_swap_numa,
	511
	512	TP_PROTO(struct task_struct *src_tsk, int src_cpu,
	513	struct task_struct *dst_tsk, int dst_cpu),
	514
	515	TP_ARGS(src_tsk, src_cpu, dst_tsk, dst_cpu),
	516
	517	TP_STRUCT__entry(
	518	__field( pid_t, src_pid )
	519	__field( pid_t, src_tgid )
	520	__field( pid_t, src_ngid )
	521	__field( int, src_cpu )
	522	__field( int, src_nid )
	523	__field( pid_t, dst_pid )
	524	__field( pid_t, dst_tgid )
	525	__field( pid_t, dst_ngid )
526	__field( int, dst_cpu )
527	__field( int, dst_nid )
528	),
529
530	TP_fast_assign(
531	__entry->src_pid = task_pid_nr(src_tsk);
532	__entry->src_tgid = task_tgid_nr(src_tsk);
533	__entry->src_ngid = task_numa_group_id(src_tsk);
534	__entry->src_cpu = src_cpu;
535	__entry->src_nid = cpu_to_node(src_cpu);
536	__entry->dst_pid = task_pid_nr(dst_tsk);
537	__entry->dst_tgid = task_tgid_nr(dst_tsk);
538	__entry->dst_ngid = task_numa_group_id(dst_tsk);
539	__entry->dst_cpu = dst_cpu;
540	__entry->dst_nid = cpu_to_node(dst_cpu);
541	),
542
543	TP_printk("src_pid=%d src_tgid=%d src_ngid=%d src_cpu=%d src_nid=%d dst_pid=%d dst_tgid=%d dst_ngid=%d dst_cpu=%d dst_nid=%d",
544	__entry->src_pid, __entry->src_tgid, __entry->src_ngid,
545	__entry->src_cpu, __entry->src_nid,
546	__entry->dst_pid, __entry->dst_tgid, __entry->dst_ngid,
547	__entry->dst_cpu, __entry->dst_nid)
548	);
dfc68f29 AL	549
	550	/*
	551	* Tracepoint for waking a polling cpu without an IPI.
	552	*/
	553	TRACE_EVENT(sched_wake_idle_without_ipi,
	554
	555	TP_PROTO(int cpu),
	556
	557	TP_ARGS(cpu),
	558
	559	TP_STRUCT__entry(
	560	__field( int, cpu )
	561	),
	562
	563	TP_fast_assign(
	564	__entry->cpu = cpu;
	565	),
	566
	567	TP_printk("cpu=%d", __entry->cpu)
	568	);
ea20d929	569	#endif /* _TRACE_SCHED_H */
a8d154b0 SR	570
	571	/* This part must be outside protection */
	572	#include <trace/define_trace.h>