[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(bool preempt, struct task_struct *p)
{
#ifdef CONFIG_SCHED_DEBUG
	BUG_ON(p != current);
#endif /* CONFIG_SCHED_DEBUG */

	/*
	 * Preemption ignores task state, therefore preempted tasks are always
	 * RUNNING (we will not have dequeued if state != RUNNING).
	 */
	return preempt ? TASK_RUNNING | TASK_STATE_MAX : p->state;
}
#endif /* CREATE_TRACE_POINTS */

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

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

	TP_ARGS(preempt, 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(preempt, 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));

DECLARE_EVENT_CLASS(sched_prio_template,

	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)
);

/*
 * Tracepoint for showing priority inheritance modifying a tasks
 * priority.
 */
DEFINE_EVENT(sched_prio_template, sched_pi_setprio,
		TP_PROTO(struct task_struct *tsk, int newprio),
		TP_ARGS(tsk, newprio));

/*
 * Tracepoint for priority changes of a task.
 */
DEFINE_EVENT(sched_prio_template, sched_set_prio,
		TP_PROTO(struct task_struct *tsk, int newprio),
		TP_ARGS(tsk, 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	106	#ifdef CREATE_TRACE_POINTS
c73464b1	107	static inline long __trace_sched_switch_state(bool preempt, struct task_struct *p)
02f72694	108	{
8f9fbf09 ON	109	#ifdef CONFIG_SCHED_DEBUG
	110	BUG_ON(p != current);
	111	#endif /* CONFIG_SCHED_DEBUG */
c73464b1	112
02f72694	113	/*
c73464b1 PZ	114	* Preemption ignores task state, therefore preempted tasks are always
c73464b1 PZ	115	* RUNNING (we will not have dequeued if state != RUNNING).
02f72694	116	*/
c73464b1	117	return preempt ? TASK_RUNNING \| TASK_STATE_MAX : p->state;
02f72694	118	}
8f9fbf09	119	#endif /* CREATE_TRACE_POINTS */
02f72694	120
ea20d929 SR	121	/*
ea20d929 SR	122	* Tracepoint for task switches, performed by the scheduler:
ea20d929 SR	123	*/
	124	TRACE_EVENT(sched_switch,
	125
c73464b1 PZ	126	TP_PROTO(bool preempt,
c73464b1 PZ	127	struct task_struct *prev,
ea20d929 SR	128	struct task_struct *next),
ea20d929 SR	129
c73464b1	130	TP_ARGS(preempt, prev, next),
ea20d929 SR	131
	132	TP_STRUCT__entry(
	133	__array( char, prev_comm, TASK_COMM_LEN )
	134	__field( pid_t, prev_pid )
	135	__field( int, prev_prio )
937cdb9d	136	__field( long, prev_state )
ea20d929 SR	137	__array( char, next_comm, TASK_COMM_LEN )
	138	__field( pid_t, next_pid )
	139	__field( int, next_prio )
	140	),
	141
	142	TP_fast_assign(
	143	memcpy(__entry->next_comm, next->comm, TASK_COMM_LEN);
	144	__entry->prev_pid = prev->pid;
	145	__entry->prev_prio = prev->prio;
c73464b1	146	__entry->prev_state = __trace_sched_switch_state(preempt, prev);
ea20d929 SR	147	memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
	148	__entry->next_pid = next->pid;
	149	__entry->next_prio = next->prio;
	150	),
	151
557ab425	152	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	153	__entry->prev_comm, __entry->prev_pid, __entry->prev_prio,
557ab425 PZ	154	__entry->prev_state & (TASK_STATE_MAX-1) ?
557ab425 PZ	155	__print_flags(__entry->prev_state & (TASK_STATE_MAX-1), "\|",
937cdb9d SR	156	{ 1, "S"} , { 2, "D" }, { 4, "T" }, { 8, "t" },
937cdb9d SR	157	{ 16, "Z" }, { 32, "X" }, { 64, "x" },
80ed87c8 PZ	158	{ 128, "K" }, { 256, "W" }, { 512, "P" },
80ed87c8 PZ	159	{ 1024, "N" }) : "R",
557ab425	160	__entry->prev_state & TASK_STATE_MAX ? "+" : "",
ea20d929 SR	161	__entry->next_comm, __entry->next_pid, __entry->next_prio)
	162	);
	163
	164	/*
	165	* Tracepoint for a task being migrated:
	166	*/
	167	TRACE_EVENT(sched_migrate_task,
	168
de1d7286	169	TP_PROTO(struct task_struct *p, int dest_cpu),
ea20d929	170
de1d7286	171	TP_ARGS(p, dest_cpu),
ea20d929 SR	172
	173	TP_STRUCT__entry(
	174	__array( char, comm, TASK_COMM_LEN )
	175	__field( pid_t, pid )
	176	__field( int, prio )
	177	__field( int, orig_cpu )
	178	__field( int, dest_cpu )
	179	),
	180
	181	TP_fast_assign(
	182	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	183	__entry->pid = p->pid;
	184	__entry->prio = p->prio;
de1d7286	185	__entry->orig_cpu = task_cpu(p);
ea20d929 SR	186	__entry->dest_cpu = dest_cpu;
	187	),
	188
434a83c3	189	TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d",
ea20d929 SR	190	__entry->comm, __entry->pid, __entry->prio,
	191	__entry->orig_cpu, __entry->dest_cpu)
	192	);
	193
091ad365	194	DECLARE_EVENT_CLASS(sched_process_template,
ea20d929 SR	195
	196	TP_PROTO(struct task_struct *p),
	197
	198	TP_ARGS(p),
	199
	200	TP_STRUCT__entry(
	201	__array( char, comm, TASK_COMM_LEN )
	202	__field( pid_t, pid )
	203	__field( int, prio )
	204	),
	205
	206	TP_fast_assign(
	207	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	208	__entry->pid = p->pid;
	209	__entry->prio = p->prio;
	210	),
	211
434a83c3	212	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	213	__entry->comm, __entry->pid, __entry->prio)
	214	);
	215
	216	/*
75ec29ab	217	* Tracepoint for freeing a task:
ea20d929	218	*/
75ec29ab SR	219	DEFINE_EVENT(sched_process_template, sched_process_free,
	220	TP_PROTO(struct task_struct *p),
	221	TP_ARGS(p));
	222
ea20d929	223
75ec29ab SR	224	/*
	225	* Tracepoint for a task exiting:
	226	*/
	227	DEFINE_EVENT(sched_process_template, sched_process_exit,
	228	TP_PROTO(struct task_struct *p),
	229	TP_ARGS(p));
ea20d929	230
210f7669 LZ	231	/*
	232	* Tracepoint for waiting on task to unschedule:
	233	*/
	234	DEFINE_EVENT(sched_process_template, sched_wait_task,
	235	TP_PROTO(struct task_struct *p),
	236	TP_ARGS(p));
	237
ea20d929 SR	238	/*
	239	* Tracepoint for a waiting task:
	240	*/
	241	TRACE_EVENT(sched_process_wait,
	242
	243	TP_PROTO(struct pid *pid),
	244
	245	TP_ARGS(pid),
	246
	247	TP_STRUCT__entry(
	248	__array( char, comm, TASK_COMM_LEN )
	249	__field( pid_t, pid )
	250	__field( int, prio )
	251	),
	252
	253	TP_fast_assign(
	254	memcpy(__entry->comm, current->comm, TASK_COMM_LEN);
	255	__entry->pid = pid_nr(pid);
	256	__entry->prio = current->prio;
	257	),
	258
434a83c3	259	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	260	__entry->comm, __entry->pid, __entry->prio)
	261	);
	262
	263	/*
	264	* Tracepoint for do_fork:
	265	*/
	266	TRACE_EVENT(sched_process_fork,
	267
	268	TP_PROTO(struct task_struct parent, struct task_struct child),
	269
	270	TP_ARGS(parent, child),
	271
	272	TP_STRUCT__entry(
	273	__array( char, parent_comm, TASK_COMM_LEN )
	274	__field( pid_t, parent_pid )
	275	__array( char, child_comm, TASK_COMM_LEN )
	276	__field( pid_t, child_pid )
	277	),
	278
	279	TP_fast_assign(
	280	memcpy(__entry->parent_comm, parent->comm, TASK_COMM_LEN);
	281	__entry->parent_pid = parent->pid;
	282	memcpy(__entry->child_comm, child->comm, TASK_COMM_LEN);
	283	__entry->child_pid = child->pid;
	284	),
	285
434a83c3	286	TP_printk("comm=%s pid=%d child_comm=%s child_pid=%d",
ea20d929 SR	287	__entry->parent_comm, __entry->parent_pid,
	288	__entry->child_comm, __entry->child_pid)
	289	);
	290
4ff16c25 DS	291	/*
	292	* Tracepoint for exec:
	293	*/
	294	TRACE_EVENT(sched_process_exec,
	295
	296	TP_PROTO(struct task_struct *p, pid_t old_pid,
	297	struct linux_binprm *bprm),
	298
	299	TP_ARGS(p, old_pid, bprm),
	300
	301	TP_STRUCT__entry(
	302	__string( filename, bprm->filename )
	303	__field( pid_t, pid )
	304	__field( pid_t, old_pid )
	305	),
	306
	307	TP_fast_assign(
	308	__assign_str(filename, bprm->filename);
	309	__entry->pid = p->pid;
6308191f	310	__entry->old_pid = old_pid;
4ff16c25 DS	311	),
	312
	313	TP_printk("filename=%s pid=%d old_pid=%d", __get_str(filename),
	314	__entry->pid, __entry->old_pid)
	315	);
	316
768d0c27 PZ	317	/*
	318	* XXX the below sched_stat tracepoints only apply to SCHED_OTHER/BATCH/IDLE
	319	* adding sched_stat support to SCHED_FIFO/RR would be welcome.
	320	*/
091ad365	321	DECLARE_EVENT_CLASS(sched_stat_template,
768d0c27 PZ	322
	323	TP_PROTO(struct task_struct *tsk, u64 delay),
	324
12473965	325	TP_ARGS(__perf_task(tsk), __perf_count(delay)),
768d0c27 PZ	326
	327	TP_STRUCT__entry(
	328	__array( char, comm, TASK_COMM_LEN )
	329	__field( pid_t, pid )
	330	__field( u64, delay )
	331	),
	332
	333	TP_fast_assign(
	334	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	335	__entry->pid = tsk->pid;
	336	__entry->delay = delay;
768d0c27 PZ	337	),
768d0c27 PZ	338
434a83c3	339	TP_printk("comm=%s pid=%d delay=%Lu [ns]",
768d0c27 PZ	340	__entry->comm, __entry->pid,
	341	(unsigned long long)__entry->delay)
	342	);
	343
75ec29ab SR	344
	345	/*
	346	* Tracepoint for accounting wait time (time the task is runnable
	347	* but not actually running due to scheduler contention).
	348	*/
	349	DEFINE_EVENT(sched_stat_template, sched_stat_wait,
	350	TP_PROTO(struct task_struct *tsk, u64 delay),
	351	TP_ARGS(tsk, delay));
	352
	353	/*
	354	* Tracepoint for accounting sleep time (time the task is not runnable,
	355	* including iowait, see below).
	356	*/
470dda74 LZ	357	DEFINE_EVENT(sched_stat_template, sched_stat_sleep,
	358	TP_PROTO(struct task_struct *tsk, u64 delay),
	359	TP_ARGS(tsk, delay));
75ec29ab SR	360
	361	/*
	362	* Tracepoint for accounting iowait time (time the task is not runnable
	363	* due to waiting on IO to complete).
	364	*/
470dda74 LZ	365	DEFINE_EVENT(sched_stat_template, sched_stat_iowait,
	366	TP_PROTO(struct task_struct *tsk, u64 delay),
	367	TP_ARGS(tsk, delay));
75ec29ab	368
b781a602 AV	369	/*
	370	* Tracepoint for accounting blocked time (time the task is in uninterruptible).
	371	*/
	372	DEFINE_EVENT(sched_stat_template, sched_stat_blocked,
	373	TP_PROTO(struct task_struct *tsk, u64 delay),
	374	TP_ARGS(tsk, delay));
	375
f977bb49 IM	376	/*
	377	* Tracepoint for accounting runtime (time the task is executing
	378	* on a CPU).
	379	*/
36009d07	380	DECLARE_EVENT_CLASS(sched_stat_runtime,
f977bb49 IM	381
	382	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	383
12473965	384	TP_ARGS(tsk, __perf_count(runtime), vruntime),
f977bb49 IM	385
	386	TP_STRUCT__entry(
	387	__array( char, comm, TASK_COMM_LEN )
	388	__field( pid_t, pid )
	389	__field( u64, runtime )
	390	__field( u64, vruntime )
	391	),
	392
	393	TP_fast_assign(
	394	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	395	__entry->pid = tsk->pid;
	396	__entry->runtime = runtime;
	397	__entry->vruntime = vruntime;
f977bb49 IM	398	),
f977bb49 IM	399
434a83c3	400	TP_printk("comm=%s pid=%d runtime=%Lu [ns] vruntime=%Lu [ns]",
f977bb49 IM	401	__entry->comm, __entry->pid,
	402	(unsigned long long)__entry->runtime,
	403	(unsigned long long)__entry->vruntime)
	404	);
	405
36009d07 ON	406	DEFINE_EVENT(sched_stat_runtime, sched_stat_runtime,
	407	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	408	TP_ARGS(tsk, runtime, vruntime));
	409
3b866741	410	DECLARE_EVENT_CLASS(sched_prio_template,
a8027073 SR	411
	412	TP_PROTO(struct task_struct *tsk, int newprio),
	413
	414	TP_ARGS(tsk, newprio),
	415
	416	TP_STRUCT__entry(
	417	__array( char, comm, TASK_COMM_LEN )
	418	__field( pid_t, pid )
	419	__field( int, oldprio )
	420	__field( int, newprio )
	421	),
	422
	423	TP_fast_assign(
	424	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	425	__entry->pid = tsk->pid;
	426	__entry->oldprio = tsk->prio;
	427	__entry->newprio = newprio;
	428	),
	429
	430	TP_printk("comm=%s pid=%d oldprio=%d newprio=%d",
	431	__entry->comm, __entry->pid,
	432	__entry->oldprio, __entry->newprio)
	433	);
	434
3b866741 JD	435	/*
	436	* Tracepoint for showing priority inheritance modifying a tasks
	437	* priority.
	438	*/
	439	DEFINE_EVENT(sched_prio_template, sched_pi_setprio,
	440	TP_PROTO(struct task_struct *tsk, int newprio),
	441	TP_ARGS(tsk, newprio));
	442
	443	/*
	444	* Tracepoint for priority changes of a task.
	445	*/
	446	DEFINE_EVENT(sched_prio_template, sched_set_prio,
	447	TP_PROTO(struct task_struct *tsk, int newprio),
	448	TP_ARGS(tsk, newprio));
	449
6a716c90 ON	450	#ifdef CONFIG_DETECT_HUNG_TASK
	451	TRACE_EVENT(sched_process_hang,
	452	TP_PROTO(struct task_struct *tsk),
	453	TP_ARGS(tsk),
	454
	455	TP_STRUCT__entry(
	456	__array( char, comm, TASK_COMM_LEN )
	457	__field( pid_t, pid )
	458	),
	459
	460	TP_fast_assign(
	461	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	462	__entry->pid = tsk->pid;
	463	),
	464
	465	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
	466	);
	467	#endif /* CONFIG_DETECT_HUNG_TASK */
	468
286549dc MG	469	DECLARE_EVENT_CLASS(sched_move_task_template,
	470
	471	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	472
	473	TP_ARGS(tsk, src_cpu, dst_cpu),
	474
	475	TP_STRUCT__entry(
	476	__field( pid_t, pid )
	477	__field( pid_t, tgid )
	478	__field( pid_t, ngid )
	479	__field( int, src_cpu )
	480	__field( int, src_nid )
	481	__field( int, dst_cpu )
	482	__field( int, dst_nid )
	483	),
	484
	485	TP_fast_assign(
	486	__entry->pid = task_pid_nr(tsk);
	487	__entry->tgid = task_tgid_nr(tsk);
	488	__entry->ngid = task_numa_group_id(tsk);
	489	__entry->src_cpu = src_cpu;
	490	__entry->src_nid = cpu_to_node(src_cpu);
	491	__entry->dst_cpu = dst_cpu;
	492	__entry->dst_nid = cpu_to_node(dst_cpu);
	493	),
	494
	495	TP_printk("pid=%d tgid=%d ngid=%d src_cpu=%d src_nid=%d dst_cpu=%d dst_nid=%d",
	496	__entry->pid, __entry->tgid, __entry->ngid,
	497	__entry->src_cpu, __entry->src_nid,
	498	__entry->dst_cpu, __entry->dst_nid)
	499	);
	500
	501	/*
	502	* Tracks migration of tasks from one runqueue to another. Can be used to
	503	* detect if automatic NUMA balancing is bouncing between nodes
	504	*/
	505	DEFINE_EVENT(sched_move_task_template, sched_move_numa,
	506	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	507
	508	TP_ARGS(tsk, src_cpu, dst_cpu)
	509	);
	510
	511	DEFINE_EVENT(sched_move_task_template, sched_stick_numa,
	512	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	513
	514	TP_ARGS(tsk, src_cpu, dst_cpu)
	515	);
	516
	517	TRACE_EVENT(sched_swap_numa,
	518
	519	TP_PROTO(struct task_struct *src_tsk, int src_cpu,
	520	struct task_struct *dst_tsk, int dst_cpu),
	521
	522	TP_ARGS(src_tsk, src_cpu, dst_tsk, dst_cpu),
	523
	524	TP_STRUCT__entry(
	525	__field( pid_t, src_pid )
	526	__field( pid_t, src_tgid )
	527	__field( pid_t, src_ngid )
	528	__field( int, src_cpu )
	529	__field( int, src_nid )
	530	__field( pid_t, dst_pid )
	531	__field( pid_t, dst_tgid )
	532	__field( pid_t, dst_ngid )
533	__field( int, dst_cpu )
534	__field( int, dst_nid )
535	),
536
537	TP_fast_assign(
538	__entry->src_pid = task_pid_nr(src_tsk);
539	__entry->src_tgid = task_tgid_nr(src_tsk);
540	__entry->src_ngid = task_numa_group_id(src_tsk);
541	__entry->src_cpu = src_cpu;
542	__entry->src_nid = cpu_to_node(src_cpu);
543	__entry->dst_pid = task_pid_nr(dst_tsk);
544	__entry->dst_tgid = task_tgid_nr(dst_tsk);
545	__entry->dst_ngid = task_numa_group_id(dst_tsk);
546	__entry->dst_cpu = dst_cpu;
547	__entry->dst_nid = cpu_to_node(dst_cpu);
548	),
549
550	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",
551	__entry->src_pid, __entry->src_tgid, __entry->src_ngid,
552	__entry->src_cpu, __entry->src_nid,
553	__entry->dst_pid, __entry->dst_tgid, __entry->dst_ngid,
554	__entry->dst_cpu, __entry->dst_nid)
555	);
dfc68f29 AL	556
	557	/*
	558	* Tracepoint for waking a polling cpu without an IPI.
	559	*/
	560	TRACE_EVENT(sched_wake_idle_without_ipi,
	561
	562	TP_PROTO(int cpu),
	563
	564	TP_ARGS(cpu),
	565
	566	TP_STRUCT__entry(
	567	__field( int, cpu )
	568	),
	569
	570	TP_fast_assign(
	571	__entry->cpu = cpu;
	572	),
	573
	574	TP_printk("cpu=%d", __entry->cpu)
	575	);
ea20d929	576	#endif /* _TRACE_SCHED_H */
a8d154b0 SR	577
	578	/* This part must be outside protection */
	579	#include <trace/define_trace.h>