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

/* SPDX-License-Identifier: GPL-2.0 */
#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_REPORT_MAX;

	return 1 << task_state_index(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), "|",
				{ TASK_INTERRUPTIBLE, "S" },
				{ TASK_UNINTERRUPTIBLE, "D" },
				{ __TASK_STOPPED, "T" },
				{ __TASK_TRACED, "t" },
				{ EXIT_DEAD, "X" },
				{ EXIT_ZOMBIE, "Z" },
				{ TASK_PARKED, "P" },
				{ TASK_DEAD, "I" }) :
		  "R",

		__entry->prev_state & TASK_REPORT_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 ?
				min(tsk->normal_prio, pi_task->prio) :
				tsk->normal_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
b2441318	1	/* SPDX-License-Identifier: GPL-2.0 */
d0b6e04a LZ	2	#undef TRACE_SYSTEM
	3	#define TRACE_SYSTEM sched
	4
ea20d929	5	#if !defined(_TRACE_SCHED_H) \|\| defined(TRACE_HEADER_MULTI_READ)
0a16b607 MD	6	#define _TRACE_SCHED_H
0a16b607 MD	7
6a3827d7	8	#include <linux/sched/numa_balancing.h>
0a16b607	9	#include <linux/tracepoint.h>
4ff16c25	10	#include <linux/binfmts.h>
0a16b607	11
ea20d929 SR	12	/*
	13	* Tracepoint for calling kthread_stop, performed to end a kthread:
	14	*/
	15	TRACE_EVENT(sched_kthread_stop,
	16
	17	TP_PROTO(struct task_struct *t),
	18
	19	TP_ARGS(t),
	20
	21	TP_STRUCT__entry(
	22	__array( char, comm, TASK_COMM_LEN )
	23	__field( pid_t, pid )
	24	),
	25
	26	TP_fast_assign(
	27	memcpy(__entry->comm, t->comm, TASK_COMM_LEN);
	28	__entry->pid = t->pid;
	29	),
	30
434a83c3	31	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
ea20d929 SR	32	);
	33
	34	/*
	35	* Tracepoint for the return value of the kthread stopping:
	36	*/
	37	TRACE_EVENT(sched_kthread_stop_ret,
	38
	39	TP_PROTO(int ret),
	40
	41	TP_ARGS(ret),
	42
	43	TP_STRUCT__entry(
	44	__field( int, ret )
	45	),
	46
	47	TP_fast_assign(
	48	__entry->ret = ret;
	49	),
	50
434a83c3	51	TP_printk("ret=%d", __entry->ret)
ea20d929 SR	52	);
ea20d929 SR	53
ea20d929 SR	54	/*
ea20d929 SR	55	* Tracepoint for waking up a task:
ea20d929	56	*/
091ad365	57	DECLARE_EVENT_CLASS(sched_wakeup_template,
ea20d929	58
fbd705a0	59	TP_PROTO(struct task_struct *p),
ea20d929	60
fbd705a0	61	TP_ARGS(__perf_task(p)),
ea20d929 SR	62
	63	TP_STRUCT__entry(
	64	__array( char, comm, TASK_COMM_LEN )
	65	__field( pid_t, pid )
	66	__field( int, prio )
	67	__field( int, success )
434a83c3	68	__field( int, target_cpu )
ea20d929 SR	69	),
	70
	71	TP_fast_assign(
	72	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	73	__entry->pid = p->pid;
b91473ff	74	__entry->prio = p->prio; /* XXX SCHED_DEADLINE */
fbd705a0	75	__entry->success = 1; /* rudiment, kill when possible */
434a83c3	76	__entry->target_cpu = task_cpu(p);
ea20d929 SR	77	),
ea20d929 SR	78
fbd705a0	79	TP_printk("comm=%s pid=%d prio=%d target_cpu=%03d",
ea20d929	80	__entry->comm, __entry->pid, __entry->prio,
fbd705a0	81	__entry->target_cpu)
ea20d929 SR	82	);
ea20d929 SR	83
fbd705a0 PZ	84	/*
	85	* Tracepoint called when waking a task; this tracepoint is guaranteed to be
	86	* called from the waking context.
	87	*/
	88	DEFINE_EVENT(sched_wakeup_template, sched_waking,
	89	TP_PROTO(struct task_struct *p),
	90	TP_ARGS(p));
	91
	92	/*
	93	* Tracepoint called when the task is actually woken; p->state == TASK_RUNNNG.
	94	* It it not always called from the waking context.
	95	*/
75ec29ab	96	DEFINE_EVENT(sched_wakeup_template, sched_wakeup,
fbd705a0 PZ	97	TP_PROTO(struct task_struct *p),
fbd705a0 PZ	98	TP_ARGS(p));
75ec29ab	99
ea20d929 SR	100	/*
ea20d929 SR	101	* Tracepoint for waking up a new task:
ea20d929	102	*/
75ec29ab	103	DEFINE_EVENT(sched_wakeup_template, sched_wakeup_new,
fbd705a0 PZ	104	TP_PROTO(struct task_struct *p),
fbd705a0 PZ	105	TP_ARGS(p));
ea20d929	106
02f72694	107	#ifdef CREATE_TRACE_POINTS
c73464b1	108	static inline long __trace_sched_switch_state(bool preempt, struct task_struct *p)
02f72694	109	{
8f9fbf09 ON	110	#ifdef CONFIG_SCHED_DEBUG
	111	BUG_ON(p != current);
	112	#endif /* CONFIG_SCHED_DEBUG */
c73464b1	113
02f72694	114	/*
c73464b1 PZ	115	* Preemption ignores task state, therefore preempted tasks are always
c73464b1 PZ	116	* RUNNING (we will not have dequeued if state != RUNNING).
02f72694	117	*/
efb40f58	118	if (preempt)
3f5fe9fe	119	return TASK_REPORT_MAX;
efb40f58	120
3f5fe9fe	121	return 1 << task_state_index(p);
02f72694	122	}
8f9fbf09	123	#endif /* CREATE_TRACE_POINTS */
02f72694	124
ea20d929 SR	125	/*
ea20d929 SR	126	* Tracepoint for task switches, performed by the scheduler:
ea20d929 SR	127	*/
	128	TRACE_EVENT(sched_switch,
	129
c73464b1 PZ	130	TP_PROTO(bool preempt,
c73464b1 PZ	131	struct task_struct *prev,
ea20d929 SR	132	struct task_struct *next),
ea20d929 SR	133
c73464b1	134	TP_ARGS(preempt, 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;
c73464b1	150	__entry->prev_state = __trace_sched_switch_state(preempt, 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;
b91473ff	154	/* XXX SCHED_DEADLINE */
ea20d929 SR	155	),
ea20d929 SR	156
557ab425	157	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	158	__entry->prev_comm, __entry->prev_pid, __entry->prev_prio,
efb40f58	159
06eb6184 PZ	160	(__entry->prev_state & (TASK_REPORT_MAX - 1)) ?
06eb6184 PZ	161	__print_flags(__entry->prev_state & (TASK_REPORT_MAX - 1), "\|",
ff28915f UKK	162	{ TASK_INTERRUPTIBLE, "S" },
	163	{ TASK_UNINTERRUPTIBLE, "D" },
	164	{ __TASK_STOPPED, "T" },
	165	{ __TASK_TRACED, "t" },
	166	{ EXIT_DEAD, "X" },
	167	{ EXIT_ZOMBIE, "Z" },
	168	{ TASK_PARKED, "P" },
	169	{ TASK_DEAD, "I" }) :
efb40f58 PZ	170	"R",
efb40f58 PZ	171
3f5fe9fe	172	__entry->prev_state & TASK_REPORT_MAX ? "+" : "",
ea20d929 SR	173	__entry->next_comm, __entry->next_pid, __entry->next_prio)
	174	);
	175
	176	/*
	177	* Tracepoint for a task being migrated:
	178	*/
	179	TRACE_EVENT(sched_migrate_task,
	180
de1d7286	181	TP_PROTO(struct task_struct *p, int dest_cpu),
ea20d929	182
de1d7286	183	TP_ARGS(p, dest_cpu),
ea20d929 SR	184
	185	TP_STRUCT__entry(
	186	__array( char, comm, TASK_COMM_LEN )
	187	__field( pid_t, pid )
	188	__field( int, prio )
	189	__field( int, orig_cpu )
	190	__field( int, dest_cpu )
	191	),
	192
	193	TP_fast_assign(
	194	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	195	__entry->pid = p->pid;
b91473ff	196	__entry->prio = p->prio; /* XXX SCHED_DEADLINE */
de1d7286	197	__entry->orig_cpu = task_cpu(p);
ea20d929 SR	198	__entry->dest_cpu = dest_cpu;
	199	),
	200
434a83c3	201	TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d",
ea20d929 SR	202	__entry->comm, __entry->pid, __entry->prio,
	203	__entry->orig_cpu, __entry->dest_cpu)
	204	);
	205
091ad365	206	DECLARE_EVENT_CLASS(sched_process_template,
ea20d929 SR	207
	208	TP_PROTO(struct task_struct *p),
	209
	210	TP_ARGS(p),
	211
	212	TP_STRUCT__entry(
	213	__array( char, comm, TASK_COMM_LEN )
	214	__field( pid_t, pid )
	215	__field( int, prio )
	216	),
	217
	218	TP_fast_assign(
	219	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	220	__entry->pid = p->pid;
b91473ff	221	__entry->prio = p->prio; /* XXX SCHED_DEADLINE */
ea20d929 SR	222	),
ea20d929 SR	223
434a83c3	224	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	225	__entry->comm, __entry->pid, __entry->prio)
	226	);
	227
	228	/*
75ec29ab	229	* Tracepoint for freeing a task:
ea20d929	230	*/
75ec29ab SR	231	DEFINE_EVENT(sched_process_template, sched_process_free,
	232	TP_PROTO(struct task_struct *p),
	233	TP_ARGS(p));
	234
ea20d929	235
75ec29ab SR	236	/*
	237	* Tracepoint for a task exiting:
	238	*/
	239	DEFINE_EVENT(sched_process_template, sched_process_exit,
	240	TP_PROTO(struct task_struct *p),
	241	TP_ARGS(p));
ea20d929	242
210f7669 LZ	243	/*
	244	* Tracepoint for waiting on task to unschedule:
	245	*/
	246	DEFINE_EVENT(sched_process_template, sched_wait_task,
	247	TP_PROTO(struct task_struct *p),
	248	TP_ARGS(p));
	249
ea20d929 SR	250	/*
	251	* Tracepoint for a waiting task:
	252	*/
	253	TRACE_EVENT(sched_process_wait,
	254
	255	TP_PROTO(struct pid *pid),
	256
	257	TP_ARGS(pid),
	258
	259	TP_STRUCT__entry(
	260	__array( char, comm, TASK_COMM_LEN )
	261	__field( pid_t, pid )
	262	__field( int, prio )
	263	),
	264
	265	TP_fast_assign(
	266	memcpy(__entry->comm, current->comm, TASK_COMM_LEN);
	267	__entry->pid = pid_nr(pid);
b91473ff	268	__entry->prio = current->prio; /* XXX SCHED_DEADLINE */
ea20d929 SR	269	),
ea20d929 SR	270
434a83c3	271	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	272	__entry->comm, __entry->pid, __entry->prio)
	273	);
	274
	275	/*
	276	* Tracepoint for do_fork:
	277	*/
	278	TRACE_EVENT(sched_process_fork,
	279
	280	TP_PROTO(struct task_struct parent, struct task_struct child),
	281
	282	TP_ARGS(parent, child),
	283
	284	TP_STRUCT__entry(
	285	__array( char, parent_comm, TASK_COMM_LEN )
	286	__field( pid_t, parent_pid )
	287	__array( char, child_comm, TASK_COMM_LEN )
	288	__field( pid_t, child_pid )
	289	),
	290
	291	TP_fast_assign(
	292	memcpy(__entry->parent_comm, parent->comm, TASK_COMM_LEN);
	293	__entry->parent_pid = parent->pid;
	294	memcpy(__entry->child_comm, child->comm, TASK_COMM_LEN);
	295	__entry->child_pid = child->pid;
	296	),
	297
434a83c3	298	TP_printk("comm=%s pid=%d child_comm=%s child_pid=%d",
ea20d929 SR	299	__entry->parent_comm, __entry->parent_pid,
	300	__entry->child_comm, __entry->child_pid)
	301	);
	302
4ff16c25 DS	303	/*
	304	* Tracepoint for exec:
	305	*/
	306	TRACE_EVENT(sched_process_exec,
	307
	308	TP_PROTO(struct task_struct *p, pid_t old_pid,
	309	struct linux_binprm *bprm),
	310
	311	TP_ARGS(p, old_pid, bprm),
	312
	313	TP_STRUCT__entry(
	314	__string( filename, bprm->filename )
	315	__field( pid_t, pid )
	316	__field( pid_t, old_pid )
	317	),
	318
	319	TP_fast_assign(
	320	__assign_str(filename, bprm->filename);
	321	__entry->pid = p->pid;
6308191f	322	__entry->old_pid = old_pid;
4ff16c25 DS	323	),
	324
	325	TP_printk("filename=%s pid=%d old_pid=%d", __get_str(filename),
	326	__entry->pid, __entry->old_pid)
	327	);
	328
768d0c27 PZ	329	/*
	330	* XXX the below sched_stat tracepoints only apply to SCHED_OTHER/BATCH/IDLE
	331	* adding sched_stat support to SCHED_FIFO/RR would be welcome.
	332	*/
091ad365	333	DECLARE_EVENT_CLASS(sched_stat_template,
768d0c27 PZ	334
	335	TP_PROTO(struct task_struct *tsk, u64 delay),
	336
12473965	337	TP_ARGS(__perf_task(tsk), __perf_count(delay)),
768d0c27 PZ	338
	339	TP_STRUCT__entry(
	340	__array( char, comm, TASK_COMM_LEN )
	341	__field( pid_t, pid )
	342	__field( u64, delay )
	343	),
	344
	345	TP_fast_assign(
	346	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	347	__entry->pid = tsk->pid;
	348	__entry->delay = delay;
768d0c27 PZ	349	),
768d0c27 PZ	350
434a83c3	351	TP_printk("comm=%s pid=%d delay=%Lu [ns]",
768d0c27 PZ	352	__entry->comm, __entry->pid,
	353	(unsigned long long)__entry->delay)
	354	);
	355
75ec29ab SR	356
	357	/*
	358	* Tracepoint for accounting wait time (time the task is runnable
	359	* but not actually running due to scheduler contention).
	360	*/
	361	DEFINE_EVENT(sched_stat_template, sched_stat_wait,
	362	TP_PROTO(struct task_struct *tsk, u64 delay),
	363	TP_ARGS(tsk, delay));
	364
	365	/*
	366	* Tracepoint for accounting sleep time (time the task is not runnable,
	367	* including iowait, see below).
	368	*/
470dda74 LZ	369	DEFINE_EVENT(sched_stat_template, sched_stat_sleep,
	370	TP_PROTO(struct task_struct *tsk, u64 delay),
	371	TP_ARGS(tsk, delay));
75ec29ab SR	372
	373	/*
	374	* Tracepoint for accounting iowait time (time the task is not runnable
	375	* due to waiting on IO to complete).
	376	*/
470dda74 LZ	377	DEFINE_EVENT(sched_stat_template, sched_stat_iowait,
	378	TP_PROTO(struct task_struct *tsk, u64 delay),
	379	TP_ARGS(tsk, delay));
75ec29ab	380
b781a602 AV	381	/*
	382	* Tracepoint for accounting blocked time (time the task is in uninterruptible).
	383	*/
	384	DEFINE_EVENT(sched_stat_template, sched_stat_blocked,
	385	TP_PROTO(struct task_struct *tsk, u64 delay),
	386	TP_ARGS(tsk, delay));
	387
f977bb49 IM	388	/*
	389	* Tracepoint for accounting runtime (time the task is executing
	390	* on a CPU).
	391	*/
36009d07	392	DECLARE_EVENT_CLASS(sched_stat_runtime,
f977bb49 IM	393
	394	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	395
12473965	396	TP_ARGS(tsk, __perf_count(runtime), vruntime),
f977bb49 IM	397
	398	TP_STRUCT__entry(
	399	__array( char, comm, TASK_COMM_LEN )
	400	__field( pid_t, pid )
	401	__field( u64, runtime )
	402	__field( u64, vruntime )
	403	),
	404
	405	TP_fast_assign(
	406	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	407	__entry->pid = tsk->pid;
	408	__entry->runtime = runtime;
	409	__entry->vruntime = vruntime;
f977bb49 IM	410	),
f977bb49 IM	411
434a83c3	412	TP_printk("comm=%s pid=%d runtime=%Lu [ns] vruntime=%Lu [ns]",
f977bb49 IM	413	__entry->comm, __entry->pid,
	414	(unsigned long long)__entry->runtime,
	415	(unsigned long long)__entry->vruntime)
	416	);
	417
36009d07 ON	418	DEFINE_EVENT(sched_stat_runtime, sched_stat_runtime,
	419	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	420	TP_ARGS(tsk, runtime, vruntime));
	421
a8027073 SR	422	/*
	423	* Tracepoint for showing priority inheritance modifying a tasks
	424	* priority.
	425	*/
	426	TRACE_EVENT(sched_pi_setprio,
	427
b91473ff	428	TP_PROTO(struct task_struct tsk, struct task_struct pi_task),
a8027073	429
b91473ff	430	TP_ARGS(tsk, pi_task),
a8027073 SR	431
	432	TP_STRUCT__entry(
	433	__array( char, comm, TASK_COMM_LEN )
	434	__field( pid_t, pid )
	435	__field( int, oldprio )
	436	__field( int, newprio )
	437	),
	438
	439	TP_fast_assign(
	440	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	441	__entry->pid = tsk->pid;
	442	__entry->oldprio = tsk->prio;
4ff648de SAS	443	__entry->newprio = pi_task ?
	444	min(tsk->normal_prio, pi_task->prio) :
	445	tsk->normal_prio;
b91473ff	446	/* XXX SCHED_DEADLINE bits missing */
a8027073 SR	447	),
	448
	449	TP_printk("comm=%s pid=%d oldprio=%d newprio=%d",
	450	__entry->comm, __entry->pid,
	451	__entry->oldprio, __entry->newprio)
	452	);
	453
6a716c90 ON	454	#ifdef CONFIG_DETECT_HUNG_TASK
	455	TRACE_EVENT(sched_process_hang,
	456	TP_PROTO(struct task_struct *tsk),
	457	TP_ARGS(tsk),
	458
	459	TP_STRUCT__entry(
	460	__array( char, comm, TASK_COMM_LEN )
	461	__field( pid_t, pid )
	462	),
	463
	464	TP_fast_assign(
	465	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	466	__entry->pid = tsk->pid;
	467	),
	468
	469	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
	470	);
	471	#endif /* CONFIG_DETECT_HUNG_TASK */
	472
286549dc MG	473	DECLARE_EVENT_CLASS(sched_move_task_template,
	474
	475	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	476
	477	TP_ARGS(tsk, src_cpu, dst_cpu),
	478
	479	TP_STRUCT__entry(
	480	__field( pid_t, pid )
	481	__field( pid_t, tgid )
	482	__field( pid_t, ngid )
	483	__field( int, src_cpu )
	484	__field( int, src_nid )
	485	__field( int, dst_cpu )
	486	__field( int, dst_nid )
	487	),
	488
	489	TP_fast_assign(
	490	__entry->pid = task_pid_nr(tsk);
	491	__entry->tgid = task_tgid_nr(tsk);
	492	__entry->ngid = task_numa_group_id(tsk);
	493	__entry->src_cpu = src_cpu;
	494	__entry->src_nid = cpu_to_node(src_cpu);
	495	__entry->dst_cpu = dst_cpu;
	496	__entry->dst_nid = cpu_to_node(dst_cpu);
	497	),
	498
	499	TP_printk("pid=%d tgid=%d ngid=%d src_cpu=%d src_nid=%d dst_cpu=%d dst_nid=%d",
	500	__entry->pid, __entry->tgid, __entry->ngid,
	501	__entry->src_cpu, __entry->src_nid,
	502	__entry->dst_cpu, __entry->dst_nid)
	503	);
	504
	505	/*
	506	* Tracks migration of tasks from one runqueue to another. Can be used to
	507	* detect if automatic NUMA balancing is bouncing between nodes
	508	*/
	509	DEFINE_EVENT(sched_move_task_template, sched_move_numa,
	510	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	511
	512	TP_ARGS(tsk, src_cpu, dst_cpu)
	513	);
	514
	515	DEFINE_EVENT(sched_move_task_template, sched_stick_numa,
	516	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	517
	518	TP_ARGS(tsk, src_cpu, dst_cpu)
	519	);
	520
	521	TRACE_EVENT(sched_swap_numa,
	522
	523	TP_PROTO(struct task_struct *src_tsk, int src_cpu,
	524	struct task_struct *dst_tsk, int dst_cpu),
	525
	526	TP_ARGS(src_tsk, src_cpu, dst_tsk, dst_cpu),
	527
	528	TP_STRUCT__entry(
	529	__field( pid_t, src_pid )
	530	__field( pid_t, src_tgid )
	531	__field( pid_t, src_ngid )
	532	__field( int, src_cpu )
	533	__field( int, src_nid )
	534	__field( pid_t, dst_pid )
	535	__field( pid_t, dst_tgid )
	536	__field( pid_t, dst_ngid )
537	__field( int, dst_cpu )
538	__field( int, dst_nid )
539	),
540
541	TP_fast_assign(
542	__entry->src_pid = task_pid_nr(src_tsk);
543	__entry->src_tgid = task_tgid_nr(src_tsk);
544	__entry->src_ngid = task_numa_group_id(src_tsk);
545	__entry->src_cpu = src_cpu;
546	__entry->src_nid = cpu_to_node(src_cpu);
547	__entry->dst_pid = task_pid_nr(dst_tsk);
548	__entry->dst_tgid = task_tgid_nr(dst_tsk);
549	__entry->dst_ngid = task_numa_group_id(dst_tsk);
550	__entry->dst_cpu = dst_cpu;
551	__entry->dst_nid = cpu_to_node(dst_cpu);
552	),
553
554	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",
555	__entry->src_pid, __entry->src_tgid, __entry->src_ngid,
556	__entry->src_cpu, __entry->src_nid,
557	__entry->dst_pid, __entry->dst_tgid, __entry->dst_ngid,
558	__entry->dst_cpu, __entry->dst_nid)
559	);
dfc68f29 AL	560
	561	/*
	562	* Tracepoint for waking a polling cpu without an IPI.
	563	*/
	564	TRACE_EVENT(sched_wake_idle_without_ipi,
	565
	566	TP_PROTO(int cpu),
	567
	568	TP_ARGS(cpu),
	569
	570	TP_STRUCT__entry(
	571	__field( int, cpu )
	572	),
	573
	574	TP_fast_assign(
	575	__entry->cpu = cpu;
	576	),
	577
	578	TP_printk("cpu=%d", __entry->cpu)
	579	);
ea20d929	580	#endif /* _TRACE_SCHED_H */
a8d154b0 SR	581
	582	/* This part must be outside protection */
	583	#include <trace/define_trace.h>