[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), "|",
				{ 0x01, "S" }, { 0x02, "D" }, { 0x04, "T" },
				{ 0x08, "t" }, { 0x10, "X" }, { 0x20, "Z" },
				{ 0x40, "P" }, { 0x80, "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 ? 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
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), "\|",
efb40f58	162	{ 0x01, "S" }, { 0x02, "D" }, { 0x04, "T" },
06eb6184	163	{ 0x08, "t" }, { 0x10, "X" }, { 0x20, "Z" },
8ef9925b	164	{ 0x40, "P" }, { 0x80, "I" }) :
efb40f58 PZ	165	"R",
efb40f58 PZ	166
3f5fe9fe	167	__entry->prev_state & TASK_REPORT_MAX ? "+" : "",
ea20d929 SR	168	__entry->next_comm, __entry->next_pid, __entry->next_prio)
	169	);
	170
	171	/*
	172	* Tracepoint for a task being migrated:
	173	*/
	174	TRACE_EVENT(sched_migrate_task,
	175
de1d7286	176	TP_PROTO(struct task_struct *p, int dest_cpu),
ea20d929	177
de1d7286	178	TP_ARGS(p, dest_cpu),
ea20d929 SR	179
	180	TP_STRUCT__entry(
	181	__array( char, comm, TASK_COMM_LEN )
	182	__field( pid_t, pid )
	183	__field( int, prio )
	184	__field( int, orig_cpu )
	185	__field( int, dest_cpu )
	186	),
	187
	188	TP_fast_assign(
	189	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	190	__entry->pid = p->pid;
b91473ff	191	__entry->prio = p->prio; /* XXX SCHED_DEADLINE */
de1d7286	192	__entry->orig_cpu = task_cpu(p);
ea20d929 SR	193	__entry->dest_cpu = dest_cpu;
	194	),
	195
434a83c3	196	TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d",
ea20d929 SR	197	__entry->comm, __entry->pid, __entry->prio,
	198	__entry->orig_cpu, __entry->dest_cpu)
	199	);
	200
091ad365	201	DECLARE_EVENT_CLASS(sched_process_template,
ea20d929 SR	202
	203	TP_PROTO(struct task_struct *p),
	204
	205	TP_ARGS(p),
	206
	207	TP_STRUCT__entry(
	208	__array( char, comm, TASK_COMM_LEN )
	209	__field( pid_t, pid )
	210	__field( int, prio )
	211	),
	212
	213	TP_fast_assign(
	214	memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
	215	__entry->pid = p->pid;
b91473ff	216	__entry->prio = p->prio; /* XXX SCHED_DEADLINE */
ea20d929 SR	217	),
ea20d929 SR	218
434a83c3	219	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	220	__entry->comm, __entry->pid, __entry->prio)
	221	);
	222
	223	/*
75ec29ab	224	* Tracepoint for freeing a task:
ea20d929	225	*/
75ec29ab SR	226	DEFINE_EVENT(sched_process_template, sched_process_free,
	227	TP_PROTO(struct task_struct *p),
	228	TP_ARGS(p));
	229
ea20d929	230
75ec29ab SR	231	/*
	232	* Tracepoint for a task exiting:
	233	*/
	234	DEFINE_EVENT(sched_process_template, sched_process_exit,
	235	TP_PROTO(struct task_struct *p),
	236	TP_ARGS(p));
ea20d929	237
210f7669 LZ	238	/*
	239	* Tracepoint for waiting on task to unschedule:
	240	*/
	241	DEFINE_EVENT(sched_process_template, sched_wait_task,
	242	TP_PROTO(struct task_struct *p),
	243	TP_ARGS(p));
	244
ea20d929 SR	245	/*
	246	* Tracepoint for a waiting task:
	247	*/
	248	TRACE_EVENT(sched_process_wait,
	249
	250	TP_PROTO(struct pid *pid),
	251
	252	TP_ARGS(pid),
	253
	254	TP_STRUCT__entry(
	255	__array( char, comm, TASK_COMM_LEN )
	256	__field( pid_t, pid )
	257	__field( int, prio )
	258	),
	259
	260	TP_fast_assign(
	261	memcpy(__entry->comm, current->comm, TASK_COMM_LEN);
	262	__entry->pid = pid_nr(pid);
b91473ff	263	__entry->prio = current->prio; /* XXX SCHED_DEADLINE */
ea20d929 SR	264	),
ea20d929 SR	265
434a83c3	266	TP_printk("comm=%s pid=%d prio=%d",
ea20d929 SR	267	__entry->comm, __entry->pid, __entry->prio)
	268	);
	269
	270	/*
	271	* Tracepoint for do_fork:
	272	*/
	273	TRACE_EVENT(sched_process_fork,
	274
	275	TP_PROTO(struct task_struct parent, struct task_struct child),
	276
	277	TP_ARGS(parent, child),
	278
	279	TP_STRUCT__entry(
	280	__array( char, parent_comm, TASK_COMM_LEN )
	281	__field( pid_t, parent_pid )
	282	__array( char, child_comm, TASK_COMM_LEN )
	283	__field( pid_t, child_pid )
	284	),
	285
	286	TP_fast_assign(
	287	memcpy(__entry->parent_comm, parent->comm, TASK_COMM_LEN);
	288	__entry->parent_pid = parent->pid;
	289	memcpy(__entry->child_comm, child->comm, TASK_COMM_LEN);
	290	__entry->child_pid = child->pid;
	291	),
	292
434a83c3	293	TP_printk("comm=%s pid=%d child_comm=%s child_pid=%d",
ea20d929 SR	294	__entry->parent_comm, __entry->parent_pid,
	295	__entry->child_comm, __entry->child_pid)
	296	);
	297
4ff16c25 DS	298	/*
	299	* Tracepoint for exec:
	300	*/
	301	TRACE_EVENT(sched_process_exec,
	302
	303	TP_PROTO(struct task_struct *p, pid_t old_pid,
	304	struct linux_binprm *bprm),
	305
	306	TP_ARGS(p, old_pid, bprm),
	307
	308	TP_STRUCT__entry(
	309	__string( filename, bprm->filename )
	310	__field( pid_t, pid )
	311	__field( pid_t, old_pid )
	312	),
	313
	314	TP_fast_assign(
	315	__assign_str(filename, bprm->filename);
	316	__entry->pid = p->pid;
6308191f	317	__entry->old_pid = old_pid;
4ff16c25 DS	318	),
	319
	320	TP_printk("filename=%s pid=%d old_pid=%d", __get_str(filename),
	321	__entry->pid, __entry->old_pid)
	322	);
	323
768d0c27 PZ	324	/*
	325	* XXX the below sched_stat tracepoints only apply to SCHED_OTHER/BATCH/IDLE
	326	* adding sched_stat support to SCHED_FIFO/RR would be welcome.
	327	*/
091ad365	328	DECLARE_EVENT_CLASS(sched_stat_template,
768d0c27 PZ	329
	330	TP_PROTO(struct task_struct *tsk, u64 delay),
	331
12473965	332	TP_ARGS(__perf_task(tsk), __perf_count(delay)),
768d0c27 PZ	333
	334	TP_STRUCT__entry(
	335	__array( char, comm, TASK_COMM_LEN )
	336	__field( pid_t, pid )
	337	__field( u64, delay )
	338	),
	339
	340	TP_fast_assign(
	341	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	342	__entry->pid = tsk->pid;
	343	__entry->delay = delay;
768d0c27 PZ	344	),
768d0c27 PZ	345
434a83c3	346	TP_printk("comm=%s pid=%d delay=%Lu [ns]",
768d0c27 PZ	347	__entry->comm, __entry->pid,
	348	(unsigned long long)__entry->delay)
	349	);
	350
75ec29ab SR	351
	352	/*
	353	* Tracepoint for accounting wait time (time the task is runnable
	354	* but not actually running due to scheduler contention).
	355	*/
	356	DEFINE_EVENT(sched_stat_template, sched_stat_wait,
	357	TP_PROTO(struct task_struct *tsk, u64 delay),
	358	TP_ARGS(tsk, delay));
	359
	360	/*
	361	* Tracepoint for accounting sleep time (time the task is not runnable,
	362	* including iowait, see below).
	363	*/
470dda74 LZ	364	DEFINE_EVENT(sched_stat_template, sched_stat_sleep,
	365	TP_PROTO(struct task_struct *tsk, u64 delay),
	366	TP_ARGS(tsk, delay));
75ec29ab SR	367
	368	/*
	369	* Tracepoint for accounting iowait time (time the task is not runnable
	370	* due to waiting on IO to complete).
	371	*/
470dda74 LZ	372	DEFINE_EVENT(sched_stat_template, sched_stat_iowait,
	373	TP_PROTO(struct task_struct *tsk, u64 delay),
	374	TP_ARGS(tsk, delay));
75ec29ab	375
b781a602 AV	376	/*
	377	* Tracepoint for accounting blocked time (time the task is in uninterruptible).
	378	*/
	379	DEFINE_EVENT(sched_stat_template, sched_stat_blocked,
	380	TP_PROTO(struct task_struct *tsk, u64 delay),
	381	TP_ARGS(tsk, delay));
	382
f977bb49 IM	383	/*
	384	* Tracepoint for accounting runtime (time the task is executing
	385	* on a CPU).
	386	*/
36009d07	387	DECLARE_EVENT_CLASS(sched_stat_runtime,
f977bb49 IM	388
	389	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	390
12473965	391	TP_ARGS(tsk, __perf_count(runtime), vruntime),
f977bb49 IM	392
	393	TP_STRUCT__entry(
	394	__array( char, comm, TASK_COMM_LEN )
	395	__field( pid_t, pid )
	396	__field( u64, runtime )
	397	__field( u64, vruntime )
	398	),
	399
	400	TP_fast_assign(
	401	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	402	__entry->pid = tsk->pid;
	403	__entry->runtime = runtime;
	404	__entry->vruntime = vruntime;
f977bb49 IM	405	),
f977bb49 IM	406
434a83c3	407	TP_printk("comm=%s pid=%d runtime=%Lu [ns] vruntime=%Lu [ns]",
f977bb49 IM	408	__entry->comm, __entry->pid,
	409	(unsigned long long)__entry->runtime,
	410	(unsigned long long)__entry->vruntime)
	411	);
	412
36009d07 ON	413	DEFINE_EVENT(sched_stat_runtime, sched_stat_runtime,
	414	TP_PROTO(struct task_struct *tsk, u64 runtime, u64 vruntime),
	415	TP_ARGS(tsk, runtime, vruntime));
	416
a8027073 SR	417	/*
	418	* Tracepoint for showing priority inheritance modifying a tasks
	419	* priority.
	420	*/
	421	TRACE_EVENT(sched_pi_setprio,
	422
b91473ff	423	TP_PROTO(struct task_struct tsk, struct task_struct pi_task),
a8027073	424
b91473ff	425	TP_ARGS(tsk, pi_task),
a8027073 SR	426
	427	TP_STRUCT__entry(
	428	__array( char, comm, TASK_COMM_LEN )
	429	__field( pid_t, pid )
	430	__field( int, oldprio )
	431	__field( int, newprio )
	432	),
	433
	434	TP_fast_assign(
	435	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	436	__entry->pid = tsk->pid;
	437	__entry->oldprio = tsk->prio;
b91473ff PZ	438	__entry->newprio = pi_task ? pi_task->prio : tsk->prio;
b91473ff PZ	439	/* XXX SCHED_DEADLINE bits missing */
a8027073 SR	440	),
	441
	442	TP_printk("comm=%s pid=%d oldprio=%d newprio=%d",
	443	__entry->comm, __entry->pid,
	444	__entry->oldprio, __entry->newprio)
	445	);
	446
6a716c90 ON	447	#ifdef CONFIG_DETECT_HUNG_TASK
	448	TRACE_EVENT(sched_process_hang,
	449	TP_PROTO(struct task_struct *tsk),
	450	TP_ARGS(tsk),
	451
	452	TP_STRUCT__entry(
	453	__array( char, comm, TASK_COMM_LEN )
	454	__field( pid_t, pid )
	455	),
	456
	457	TP_fast_assign(
	458	memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
	459	__entry->pid = tsk->pid;
	460	),
	461
	462	TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
	463	);
	464	#endif /* CONFIG_DETECT_HUNG_TASK */
	465
286549dc MG	466	DECLARE_EVENT_CLASS(sched_move_task_template,
	467
	468	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	469
	470	TP_ARGS(tsk, src_cpu, dst_cpu),
	471
	472	TP_STRUCT__entry(
	473	__field( pid_t, pid )
	474	__field( pid_t, tgid )
	475	__field( pid_t, ngid )
	476	__field( int, src_cpu )
	477	__field( int, src_nid )
	478	__field( int, dst_cpu )
	479	__field( int, dst_nid )
	480	),
	481
	482	TP_fast_assign(
	483	__entry->pid = task_pid_nr(tsk);
	484	__entry->tgid = task_tgid_nr(tsk);
	485	__entry->ngid = task_numa_group_id(tsk);
	486	__entry->src_cpu = src_cpu;
	487	__entry->src_nid = cpu_to_node(src_cpu);
	488	__entry->dst_cpu = dst_cpu;
	489	__entry->dst_nid = cpu_to_node(dst_cpu);
	490	),
	491
	492	TP_printk("pid=%d tgid=%d ngid=%d src_cpu=%d src_nid=%d dst_cpu=%d dst_nid=%d",
	493	__entry->pid, __entry->tgid, __entry->ngid,
	494	__entry->src_cpu, __entry->src_nid,
	495	__entry->dst_cpu, __entry->dst_nid)
	496	);
	497
	498	/*
	499	* Tracks migration of tasks from one runqueue to another. Can be used to
	500	* detect if automatic NUMA balancing is bouncing between nodes
	501	*/
	502	DEFINE_EVENT(sched_move_task_template, sched_move_numa,
	503	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	504
	505	TP_ARGS(tsk, src_cpu, dst_cpu)
	506	);
	507
	508	DEFINE_EVENT(sched_move_task_template, sched_stick_numa,
	509	TP_PROTO(struct task_struct *tsk, int src_cpu, int dst_cpu),
	510
	511	TP_ARGS(tsk, src_cpu, dst_cpu)
	512	);
	513
	514	TRACE_EVENT(sched_swap_numa,
	515
	516	TP_PROTO(struct task_struct *src_tsk, int src_cpu,
	517	struct task_struct *dst_tsk, int dst_cpu),
	518
	519	TP_ARGS(src_tsk, src_cpu, dst_tsk, dst_cpu),
	520
	521	TP_STRUCT__entry(
	522	__field( pid_t, src_pid )
	523	__field( pid_t, src_tgid )
	524	__field( pid_t, src_ngid )
	525	__field( int, src_cpu )
	526	__field( int, src_nid )
	527	__field( pid_t, dst_pid )
	528	__field( pid_t, dst_tgid )
	529	__field( pid_t, dst_ngid )
530	__field( int, dst_cpu )
531	__field( int, dst_nid )
532	),
533
534	TP_fast_assign(
535	__entry->src_pid = task_pid_nr(src_tsk);
536	__entry->src_tgid = task_tgid_nr(src_tsk);
537	__entry->src_ngid = task_numa_group_id(src_tsk);
538	__entry->src_cpu = src_cpu;
539	__entry->src_nid = cpu_to_node(src_cpu);
540	__entry->dst_pid = task_pid_nr(dst_tsk);
541	__entry->dst_tgid = task_tgid_nr(dst_tsk);
542	__entry->dst_ngid = task_numa_group_id(dst_tsk);
543	__entry->dst_cpu = dst_cpu;
544	__entry->dst_nid = cpu_to_node(dst_cpu);
545	),
546
547	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",
548	__entry->src_pid, __entry->src_tgid, __entry->src_ngid,
549	__entry->src_cpu, __entry->src_nid,
550	__entry->dst_pid, __entry->dst_tgid, __entry->dst_ngid,
551	__entry->dst_cpu, __entry->dst_nid)
552	);
dfc68f29 AL	553
	554	/*
	555	* Tracepoint for waking a polling cpu without an IPI.
	556	*/
	557	TRACE_EVENT(sched_wake_idle_without_ipi,
	558
	559	TP_PROTO(int cpu),
	560
	561	TP_ARGS(cpu),
	562
	563	TP_STRUCT__entry(
	564	__field( int, cpu )
	565	),
	566
	567	TP_fast_assign(
	568	__entry->cpu = cpu;
	569	),
	570
	571	TP_printk("cpu=%d", __entry->cpu)
	572	);
ea20d929	573	#endif /* _TRACE_SCHED_H */
a8d154b0 SR	574
	575	/* This part must be outside protection */
	576	#include <trace/define_trace.h>