[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/numa_balancing.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; /* XXX SCHED_DEADLINE */
		__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).
	 */
	if (preempt)
		return TASK_STATE_MAX;

	return __get_task_state(p);
}
#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;
		/* XXX SCHED_DEADLINE */
	),

	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_REPORT_MAX - 1)) ?
		  __print_flags(__entry->prev_state & (TASK_REPORT_MAX - 1), "|",
				{ 0x01, "S" }, { 0x02, "D" }, { 0x04, "T" },
				{ 0x08, "t" }, { 0x10, "X" }, { 0x20, "Z" },
				{ 0x40, "P" }, { 0x80, "I" }) :
		  "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; /* XXX SCHED_DEADLINE */
		__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; /* XXX SCHED_DEADLINE */
	),

	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; /* XXX SCHED_DEADLINE */
	),

	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, struct task_struct *pi_task),

	TP_ARGS(tsk, pi_task),

	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	= pi_task ? pi_task->prio : tsk->prio;
		/* XXX SCHED_DEADLINE bits missing */
	),

	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
0a16b607 MD	6
6a3827d7	7	#include <linux/sched/numa_balancing.h>
0a16b607	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;
b91473ff	73	__entry->prio = p->prio; /* XXX SCHED_DEADLINE */
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	*/
efb40f58 PZ	117	if (preempt)
	118	return TASK_STATE_MAX;
	119
	120	return __get_task_state(p);
02f72694	121	}
8f9fbf09	122	#endif /* CREATE_TRACE_POINTS */
02f72694	123
ea20d929 SR	124	/*
ea20d929 SR	125	* Tracepoint for task switches, performed by the scheduler:
ea20d929 SR	126	*/
	127	TRACE_EVENT(sched_switch,
	128
c73464b1 PZ	129	TP_PROTO(bool preempt,
c73464b1 PZ	130	struct task_struct *prev,
ea20d929 SR	131	struct task_struct *next),
ea20d929 SR	132
c73464b1	133	TP_ARGS(preempt, prev, next),
ea20d929 SR	134
	135	TP_STRUCT__entry(
	136	__array( char, prev_comm, TASK_COMM_LEN )
	137	__field( pid_t, prev_pid )
	138	__field( int, prev_prio )
937cdb9d	139	__field( long, prev_state )
ea20d929 SR	140	__array( char, next_comm, TASK_COMM_LEN )
	141	__field( pid_t, next_pid )
	142	__field( int, next_prio )
	143	),
	144
	145	TP_fast_assign(
	146	memcpy(__entry->next_comm, next->comm, TASK_COMM_LEN);
	147	__entry->prev_pid = prev->pid;
	148	__entry->prev_prio = prev->prio;
c73464b1	149	__entry->prev_state = __trace_sched_switch_state(preempt, prev);
ea20d929 SR	150	memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
	151	__entry->next_pid = next->pid;
	152	__entry->next_prio = next->prio;
b91473ff	153	/* XXX SCHED_DEADLINE */
ea20d929 SR	154	),
ea20d929 SR	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,
efb40f58	158
06eb6184 PZ	159	(__entry->prev_state & (TASK_REPORT_MAX - 1)) ?
06eb6184 PZ	160	__print_flags(__entry->prev_state & (TASK_REPORT_MAX - 1), "\|",
efb40f58	161	{ 0x01, "S" }, { 0x02, "D" }, { 0x04, "T" },
06eb6184	162	{ 0x08, "t" }, { 0x10, "X" }, { 0x20, "Z" },
8ef9925b	163	{ 0x40, "P" }, { 0x80, "I" }) :
efb40f58 PZ	164	"R",
efb40f58 PZ	165
557ab425	166	__entry->prev_state & TASK_STATE_MAX ? "+" : "",
ea20d929 SR	167	__entry->next_comm, __entry->next_pid, __entry->next_prio)
	168	);
	169
	170	/*
	171	* Tracepoint for a task being migrated:
	172	*/
	173	TRACE_EVENT(sched_migrate_task,
	174
de1d7286	175	TP_PROTO(struct task_struct *p, int dest_cpu),
ea20d929	176
de1d7286	177	TP_ARGS(p, dest_cpu),
ea20d929 SR	178
	179	TP_STRUCT__entry(
	180	__array( char, comm, TASK_COMM_LEN )
	181	__field( pid_t, pid )
	182	__field( int, prio )
	183	__field( int, orig_cpu )
	184	__field( int, dest_cpu )
	185	),
	186
	187	TP_fast_assign(
	188	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	189	__entry->pid = p->pid;
b91473ff	190	__entry->prio = p->prio; /* XXX SCHED_DEADLINE */
de1d7286	191	__entry->orig_cpu = task_cpu(p);
ea20d929 SR	192	__entry->dest_cpu = dest_cpu;
	193	),
	194
434a83c3	195	TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d",
ea20d929 SR	196	__entry->comm, __entry->pid, __entry->prio,
	197	__entry->orig_cpu, __entry->dest_cpu)
	198	);
	199
091ad365	200	DECLARE_EVENT_CLASS(sched_process_template,
ea20d929 SR	201
	202	TP_PROTO(struct task_struct *p),
	203
	204	TP_ARGS(p),
	205
	206	TP_STRUCT__entry(
	207	__array( char, comm, TASK_COMM_LEN )
	208	__field( pid_t, pid )
	209	__field( int, prio )
	210	),
	211
	212	TP_fast_assign(
	213	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	214	__entry->pid = p->pid;
b91473ff	215	__entry->prio = p->prio; /* XXX SCHED_DEADLINE */
ea20d929 SR	216	),
ea20d929 SR	217
434a83c3	218	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	219	__entry->comm, __entry->pid, __entry->prio)
	220	);
	221
	222	/*
75ec29ab	223	* Tracepoint for freeing a task:
ea20d929	224	*/
75ec29ab SR	225	DEFINE_EVENT(sched_process_template, sched_process_free,
	226	TP_PROTO(struct task_struct *p),
	227	TP_ARGS(p));
	228
ea20d929	229
75ec29ab SR	230	/*
	231	* Tracepoint for a task exiting:
	232	*/
	233	DEFINE_EVENT(sched_process_template, sched_process_exit,
	234	TP_PROTO(struct task_struct *p),
	235	TP_ARGS(p));
ea20d929	236
210f7669 LZ	237	/*
	238	* Tracepoint for waiting on task to unschedule:
	239	*/
	240	DEFINE_EVENT(sched_process_template, sched_wait_task,
	241	TP_PROTO(struct task_struct *p),
	242	TP_ARGS(p));
	243
ea20d929 SR	244	/*
	245	* Tracepoint for a waiting task:
	246	*/
	247	TRACE_EVENT(sched_process_wait,
	248
	249	TP_PROTO(struct pid *pid),
	250
	251	TP_ARGS(pid),
	252
	253	TP_STRUCT__entry(
	254	__array( char, comm, TASK_COMM_LEN )
	255	__field( pid_t, pid )
	256	__field( int, prio )
	257	),
	258
	259	TP_fast_assign(
	260	memcpy(__entry->comm, current->comm, TASK_COMM_LEN);
	261	__entry->pid = pid_nr(pid);
b91473ff	262	__entry->prio = current->prio; /* XXX SCHED_DEADLINE */
ea20d929 SR	263	),
ea20d929 SR	264
434a83c3	265	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	266	__entry->comm, __entry->pid, __entry->prio)
	267	);
	268
	269	/*
	270	* Tracepoint for do_fork:
	271	*/
	272	TRACE_EVENT(sched_process_fork,
	273
	274	TP_PROTO(struct task_struct parent, struct task_struct child),
	275
	276	TP_ARGS(parent, child),
	277
	278	TP_STRUCT__entry(
	279	__array( char, parent_comm, TASK_COMM_LEN )
	280	__field( pid_t, parent_pid )
	281	__array( char, child_comm, TASK_COMM_LEN )
	282	__field( pid_t, child_pid )
	283	),
	284
	285	TP_fast_assign(
	286	memcpy(__entry->parent_comm, parent->comm, TASK_COMM_LEN);
	287	__entry->parent_pid = parent->pid;
	288	memcpy(__entry->child_comm, child->comm, TASK_COMM_LEN);
	289	__entry->child_pid = child->pid;
	290	),
	291
434a83c3	292	TP_printk("comm=%s pid=%d child_comm=%s child_pid=%d",
ea20d929 SR	293	__entry->parent_comm, __entry->parent_pid,
	294	__entry->child_comm, __entry->child_pid)
	295	);
	296
4ff16c25 DS	297	/*
	298	* Tracepoint for exec:
	299	*/
	300	TRACE_EVENT(sched_process_exec,
	301
	302	TP_PROTO(struct task_struct *p, pid_t old_pid,
	303	struct linux_binprm *bprm),
	304
	305	TP_ARGS(p, old_pid, bprm),
	306
	307	TP_STRUCT__entry(
	308	__string( filename, bprm->filename )
	309	__field( pid_t, pid )
	310	__field( pid_t, old_pid )
	311	),
	312
	313	TP_fast_assign(
	314	__assign_str(filename, bprm->filename);
	315	__entry->pid = p->pid;
6308191f	316	__entry->old_pid = old_pid;
4ff16c25 DS	317	),
	318
	319	TP_printk("filename=%s pid=%d old_pid=%d", __get_str(filename),
	320	__entry->pid, __entry->old_pid)
	321	);
	322
768d0c27 PZ	323	/*
	324	* XXX the below sched_stat tracepoints only apply to SCHED_OTHER/BATCH/IDLE
	325	* adding sched_stat support to SCHED_FIFO/RR would be welcome.
	326	*/
091ad365	327	DECLARE_EVENT_CLASS(sched_stat_template,
768d0c27 PZ	328
	329	TP_PROTO(struct task_struct *tsk, u64 delay),
	330
12473965	331	TP_ARGS(__perf_task(tsk), __perf_count(delay)),
768d0c27 PZ	332
	333	TP_STRUCT__entry(
	334	__array( char, comm, TASK_COMM_LEN )
	335	__field( pid_t, pid )
	336	__field( u64, delay )
	337	),
	338
	339	TP_fast_assign(
	340	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	341	__entry->pid = tsk->pid;
	342	__entry->delay = delay;
768d0c27 PZ	343	),
768d0c27 PZ	344
434a83c3	345	TP_printk("comm=%s pid=%d delay=%Lu [ns]",
768d0c27 PZ	346	__entry->comm, __entry->pid,
	347	(unsigned long long)__entry->delay)
	348	);
	349
75ec29ab SR	350
	351	/*
	352	* Tracepoint for accounting wait time (time the task is runnable
	353	* but not actually running due to scheduler contention).
	354	*/
	355	DEFINE_EVENT(sched_stat_template, sched_stat_wait,
	356	TP_PROTO(struct task_struct *tsk, u64 delay),
	357	TP_ARGS(tsk, delay));
	358
	359	/*
	360	* Tracepoint for accounting sleep time (time the task is not runnable,
	361	* including iowait, see below).
	362	*/
470dda74 LZ	363	DEFINE_EVENT(sched_stat_template, sched_stat_sleep,
	364	TP_PROTO(struct task_struct *tsk, u64 delay),
	365	TP_ARGS(tsk, delay));
75ec29ab SR	366
	367	/*
	368	* Tracepoint for accounting iowait time (time the task is not runnable
	369	* due to waiting on IO to complete).
	370	*/
470dda74 LZ	371	DEFINE_EVENT(sched_stat_template, sched_stat_iowait,
	372	TP_PROTO(struct task_struct *tsk, u64 delay),
	373	TP_ARGS(tsk, delay));
75ec29ab	374
b781a602 AV	375	/*
	376	* Tracepoint for accounting blocked time (time the task is in uninterruptible).
	377	*/
	378	DEFINE_EVENT(sched_stat_template, sched_stat_blocked,
	379	TP_PROTO(struct task_struct *tsk, u64 delay),
	380	TP_ARGS(tsk, delay));
	381
f977bb49 IM	382	/*
	383	* Tracepoint for accounting runtime (time the task is executing
	384	* on a CPU).
	385	*/
36009d07	386	DECLARE_EVENT_CLASS(sched_stat_runtime,
f977bb49 IM	387
	388	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	389
12473965	390	TP_ARGS(tsk, __perf_count(runtime), vruntime),
f977bb49 IM	391
	392	TP_STRUCT__entry(
	393	__array( char, comm, TASK_COMM_LEN )
	394	__field( pid_t, pid )
	395	__field( u64, runtime )
	396	__field( u64, vruntime )
	397	),
	398
	399	TP_fast_assign(
	400	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	401	__entry->pid = tsk->pid;
	402	__entry->runtime = runtime;
	403	__entry->vruntime = vruntime;
f977bb49 IM	404	),
f977bb49 IM	405
434a83c3	406	TP_printk("comm=%s pid=%d runtime=%Lu [ns] vruntime=%Lu [ns]",
f977bb49 IM	407	__entry->comm, __entry->pid,
	408	(unsigned long long)__entry->runtime,
	409	(unsigned long long)__entry->vruntime)
	410	);
	411
36009d07 ON	412	DEFINE_EVENT(sched_stat_runtime, sched_stat_runtime,
	413	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	414	TP_ARGS(tsk, runtime, vruntime));
	415
a8027073 SR	416	/*
	417	* Tracepoint for showing priority inheritance modifying a tasks
	418	* priority.
	419	*/
	420	TRACE_EVENT(sched_pi_setprio,
	421
b91473ff	422	TP_PROTO(struct task_struct tsk, struct task_struct pi_task),
a8027073	423
b91473ff	424	TP_ARGS(tsk, pi_task),
a8027073 SR	425
	426	TP_STRUCT__entry(
	427	__array( char, comm, TASK_COMM_LEN )
	428	__field( pid_t, pid )
	429	__field( int, oldprio )
	430	__field( int, newprio )
	431	),
	432
	433	TP_fast_assign(
	434	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	435	__entry->pid = tsk->pid;
	436	__entry->oldprio = tsk->prio;
b91473ff PZ	437	__entry->newprio = pi_task ? pi_task->prio : tsk->prio;
b91473ff PZ	438	/* XXX SCHED_DEADLINE bits missing */
a8027073 SR	439	),
	440
	441	TP_printk("comm=%s pid=%d oldprio=%d newprio=%d",
	442	__entry->comm, __entry->pid,
	443	__entry->oldprio, __entry->newprio)
	444	);
	445
6a716c90 ON	446	#ifdef CONFIG_DETECT_HUNG_TASK
	447	TRACE_EVENT(sched_process_hang,
	448	TP_PROTO(struct task_struct *tsk),
	449	TP_ARGS(tsk),
	450
	451	TP_STRUCT__entry(
	452	__array( char, comm, TASK_COMM_LEN )
	453	__field( pid_t, pid )
	454	),
	455
	456	TP_fast_assign(
	457	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	458	__entry->pid = tsk->pid;
	459	),
	460
	461	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
	462	);
	463	#endif /* CONFIG_DETECT_HUNG_TASK */
	464
286549dc MG	465	DECLARE_EVENT_CLASS(sched_move_task_template,
	466
	467	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	468
	469	TP_ARGS(tsk, src_cpu, dst_cpu),
	470
	471	TP_STRUCT__entry(
	472	__field( pid_t, pid )
	473	__field( pid_t, tgid )
	474	__field( pid_t, ngid )
	475	__field( int, src_cpu )
	476	__field( int, src_nid )
	477	__field( int, dst_cpu )
	478	__field( int, dst_nid )
	479	),
	480
	481	TP_fast_assign(
	482	__entry->pid = task_pid_nr(tsk);
	483	__entry->tgid = task_tgid_nr(tsk);
	484	__entry->ngid = task_numa_group_id(tsk);
	485	__entry->src_cpu = src_cpu;
	486	__entry->src_nid = cpu_to_node(src_cpu);
	487	__entry->dst_cpu = dst_cpu;
	488	__entry->dst_nid = cpu_to_node(dst_cpu);
	489	),
	490
	491	TP_printk("pid=%d tgid=%d ngid=%d src_cpu=%d src_nid=%d dst_cpu=%d dst_nid=%d",
	492	__entry->pid, __entry->tgid, __entry->ngid,
	493	__entry->src_cpu, __entry->src_nid,
	494	__entry->dst_cpu, __entry->dst_nid)
	495	);
	496
	497	/*
	498	* Tracks migration of tasks from one runqueue to another. Can be used to
	499	* detect if automatic NUMA balancing is bouncing between nodes
	500	*/
	501	DEFINE_EVENT(sched_move_task_template, sched_move_numa,
	502	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	503
	504	TP_ARGS(tsk, src_cpu, dst_cpu)
	505	);
	506
	507	DEFINE_EVENT(sched_move_task_template, sched_stick_numa,
	508	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	509
	510	TP_ARGS(tsk, src_cpu, dst_cpu)
	511	);
	512
	513	TRACE_EVENT(sched_swap_numa,
	514
	515	TP_PROTO(struct task_struct *src_tsk, int src_cpu,
	516	struct task_struct *dst_tsk, int dst_cpu),
	517
	518	TP_ARGS(src_tsk, src_cpu, dst_tsk, dst_cpu),
	519
	520	TP_STRUCT__entry(
	521	__field( pid_t, src_pid )
	522	__field( pid_t, src_tgid )
	523	__field( pid_t, src_ngid )
	524	__field( int, src_cpu )
	525	__field( int, src_nid )
	526	__field( pid_t, dst_pid )
	527	__field( pid_t, dst_tgid )
	528	__field( pid_t, dst_ngid )
529	__field( int, dst_cpu )
530	__field( int, dst_nid )
531	),
532
533	TP_fast_assign(
534	__entry->src_pid = task_pid_nr(src_tsk);
535	__entry->src_tgid = task_tgid_nr(src_tsk);
536	__entry->src_ngid = task_numa_group_id(src_tsk);
537	__entry->src_cpu = src_cpu;
538	__entry->src_nid = cpu_to_node(src_cpu);
539	__entry->dst_pid = task_pid_nr(dst_tsk);
540	__entry->dst_tgid = task_tgid_nr(dst_tsk);
541	__entry->dst_ngid = task_numa_group_id(dst_tsk);
542	__entry->dst_cpu = dst_cpu;
543	__entry->dst_nid = cpu_to_node(dst_cpu);
544	),
545
546	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",
547	__entry->src_pid, __entry->src_tgid, __entry->src_ngid,
548	__entry->src_cpu, __entry->src_nid,
549	__entry->dst_pid, __entry->dst_tgid, __entry->dst_ngid,
550	__entry->dst_cpu, __entry->dst_nid)
551	);
dfc68f29 AL	552
	553	/*
	554	* Tracepoint for waking a polling cpu without an IPI.
	555	*/
	556	TRACE_EVENT(sched_wake_idle_without_ipi,
	557
	558	TP_PROTO(int cpu),
	559
	560	TP_ARGS(cpu),
	561
	562	TP_STRUCT__entry(
	563	__field( int, cpu )
	564	),
	565
	566	TP_fast_assign(
	567	__entry->cpu = cpu;
	568	),
	569
	570	TP_printk("cpu=%d", __entry->cpu)
	571	);
ea20d929	572	#endif /* _TRACE_SCHED_H */
a8d154b0 SR	573
	574	/* This part must be outside protection */
	575	#include <trace/define_trace.h>