sched.h 42.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3
#ifndef _LINUX_SCHED_H
#define _LINUX_SCHED_H

4
#include <uapi/linux/sched.h>
5

6
#include <linux/sched/prio.h>
7
#include <linux/nodemask.h>
8

9
#include <linux/mutex.h>
10
#include <linux/plist.h>
11
#include <linux/mm_types_task.h>
L
Linus Torvalds 已提交
12 13

#include <linux/sem.h>
14
#include <linux/shm.h>
15
#include <linux/signal_types.h>
L
Linus Torvalds 已提交
16 17
#include <linux/pid.h>
#include <linux/seccomp.h>
18
#include <linux/rcupdate.h>
L
Linus Torvalds 已提交
19

20 21
#include <linux/resource.h>
#include <linux/hrtimer.h>
D
Dmitry Vyukov 已提交
22
#include <linux/kcov.h>
23
#include <linux/task_io_accounting.h>
A
Arjan van de Ven 已提交
24
#include <linux/latencytop.h>
25

26 27
#include <asm/current.h>

28 29 30 31
/* task_struct member predeclarations: */
struct audit_context;
struct autogroup;
struct backing_dev_info;
32
struct bio_list;
33
struct blk_plug;
34
struct cfs_rq;
35
struct filename;
36 37 38 39
struct fs_struct;
struct futex_pi_state;
struct io_context;
struct mempolicy;
40
struct nameidata;
41 42 43 44 45 46 47 48 49
struct nsproxy;
struct perf_event_context;
struct pid_namespace;
struct pipe_inode_info;
struct rcu_node;
struct reclaim_state;
struct robust_list_head;
struct sched_attr;
struct sched_param;
I
Ingo Molnar 已提交
50
struct seq_file;
51 52 53
struct sighand_struct;
struct signal_struct;
struct task_delay_info;
54
struct task_group;
55 56
struct task_struct;
struct uts_namespace;
L
Linus Torvalds 已提交
57

58 59 60 61 62 63 64 65 66 67
/*
 * Task state bitmask. NOTE! These bits are also
 * encoded in fs/proc/array.c: get_task_state().
 *
 * We have two separate sets of flags: task->state
 * is about runnability, while task->exit_state are
 * about the task exiting. Confusing, but this way
 * modifying one set can't modify the other one by
 * mistake.
 */
L
Linus Torvalds 已提交
68 69 70
#define TASK_RUNNING		0
#define TASK_INTERRUPTIBLE	1
#define TASK_UNINTERRUPTIBLE	2
M
Matthew Wilcox 已提交
71 72
#define __TASK_STOPPED		4
#define __TASK_TRACED		8
73
/* in tsk->exit_state */
74 75
#define EXIT_DEAD		16
#define EXIT_ZOMBIE		32
76
#define EXIT_TRACE		(EXIT_ZOMBIE | EXIT_DEAD)
77
/* in tsk->state again */
78
#define TASK_DEAD		64
M
Matthew Wilcox 已提交
79
#define TASK_WAKEKILL		128
P
Peter Zijlstra 已提交
80
#define TASK_WAKING		256
81
#define TASK_PARKED		512
82
#define TASK_NOLOAD		1024
83 84
#define TASK_NEW		2048
#define TASK_STATE_MAX		4096
M
Matthew Wilcox 已提交
85

86
#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWPNn"
87

88
/* Convenience macros for the sake of set_current_state */
M
Matthew Wilcox 已提交
89 90 91
#define TASK_KILLABLE		(TASK_WAKEKILL | TASK_UNINTERRUPTIBLE)
#define TASK_STOPPED		(TASK_WAKEKILL | __TASK_STOPPED)
#define TASK_TRACED		(TASK_WAKEKILL | __TASK_TRACED)
L
Linus Torvalds 已提交
92

93 94
#define TASK_IDLE		(TASK_UNINTERRUPTIBLE | TASK_NOLOAD)

95 96
/* Convenience macros for the sake of wake_up */
#define TASK_NORMAL		(TASK_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE)
M
Matthew Wilcox 已提交
97
#define TASK_ALL		(TASK_NORMAL | __TASK_STOPPED | __TASK_TRACED)
98 99 100

/* get_task_state() */
#define TASK_REPORT		(TASK_RUNNING | TASK_INTERRUPTIBLE | \
M
Matthew Wilcox 已提交
101
				 TASK_UNINTERRUPTIBLE | __TASK_STOPPED | \
102
				 __TASK_TRACED | EXIT_ZOMBIE | EXIT_DEAD)
103

M
Matthew Wilcox 已提交
104 105
#define task_is_traced(task)	((task->state & __TASK_TRACED) != 0)
#define task_is_stopped(task)	((task->state & __TASK_STOPPED) != 0)
106
#define task_is_stopped_or_traced(task)	\
M
Matthew Wilcox 已提交
107
			((task->state & (__TASK_STOPPED | __TASK_TRACED)) != 0)
108
#define task_contributes_to_load(task)	\
109
				((task->state & TASK_UNINTERRUPTIBLE) != 0 && \
110 111
				 (task->flags & PF_FROZEN) == 0 && \
				 (task->state & TASK_NOLOAD) == 0)
L
Linus Torvalds 已提交
112

P
Peter Zijlstra 已提交
113 114 115 116 117 118 119 120 121 122
#ifdef CONFIG_DEBUG_ATOMIC_SLEEP

#define __set_current_state(state_value)			\
	do {							\
		current->task_state_change = _THIS_IP_;		\
		current->state = (state_value);			\
	} while (0)
#define set_current_state(state_value)				\
	do {							\
		current->task_state_change = _THIS_IP_;		\
123
		smp_store_mb(current->state, (state_value));	\
P
Peter Zijlstra 已提交
124 125 126
	} while (0)

#else
127 128 129 130 131
/*
 * set_current_state() includes a barrier so that the write of current->state
 * is correctly serialised wrt the caller's subsequent test of whether to
 * actually sleep:
 *
132
 *   for (;;) {
133
 *	set_current_state(TASK_UNINTERRUPTIBLE);
134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155
 *	if (!need_sleep)
 *		break;
 *
 *	schedule();
 *   }
 *   __set_current_state(TASK_RUNNING);
 *
 * If the caller does not need such serialisation (because, for instance, the
 * condition test and condition change and wakeup are under the same lock) then
 * use __set_current_state().
 *
 * The above is typically ordered against the wakeup, which does:
 *
 *	need_sleep = false;
 *	wake_up_state(p, TASK_UNINTERRUPTIBLE);
 *
 * Where wake_up_state() (and all other wakeup primitives) imply enough
 * barriers to order the store of the variable against wakeup.
 *
 * Wakeup will do: if (@state & p->state) p->state = TASK_RUNNING, that is,
 * once it observes the TASK_UNINTERRUPTIBLE store the waking CPU can issue a
 * TASK_RUNNING store which can collide with __set_current_state(TASK_RUNNING).
156
 *
157
 * This is obviously fine, since they both store the exact same value.
158
 *
159
 * Also see the comments of try_to_wake_up().
160
 */
P
Peter Zijlstra 已提交
161
#define __set_current_state(state_value)		\
L
Linus Torvalds 已提交
162
	do { current->state = (state_value); } while (0)
P
Peter Zijlstra 已提交
163
#define set_current_state(state_value)			\
164
	smp_store_mb(current->state, (state_value))
L
Linus Torvalds 已提交
165

P
Peter Zijlstra 已提交
166 167
#endif

L
Linus Torvalds 已提交
168 169 170
/* Task command name length */
#define TASK_COMM_LEN 16

171 172
extern cpumask_var_t cpu_isolated_map;

L
Linus Torvalds 已提交
173 174 175
extern void scheduler_tick(void);

#define	MAX_SCHEDULE_TIMEOUT	LONG_MAX
176
extern signed long schedule_timeout(signed long timeout);
177
extern signed long schedule_timeout_interruptible(signed long timeout);
M
Matthew Wilcox 已提交
178
extern signed long schedule_timeout_killable(signed long timeout);
179
extern signed long schedule_timeout_uninterruptible(signed long timeout);
180
extern signed long schedule_timeout_idle(signed long timeout);
L
Linus Torvalds 已提交
181
asmlinkage void schedule(void);
182
extern void schedule_preempt_disabled(void);
L
Linus Torvalds 已提交
183

184 185
extern int __must_check io_schedule_prepare(void);
extern void io_schedule_finish(int token);
186
extern long io_schedule_timeout(long timeout);
187
extern void io_schedule(void);
188

189
/**
190
 * struct prev_cputime - snaphsot of system and user cputime
191 192
 * @utime: time spent in user mode
 * @stime: time spent in system mode
193
 * @lock: protects the above two fields
194
 *
195 196
 * Stores previous user/system time values such that we can guarantee
 * monotonicity.
197
 */
198 199
struct prev_cputime {
#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
200 201
	u64 utime;
	u64 stime;
202 203
	raw_spinlock_t lock;
#endif
204 205
};

206 207
/**
 * struct task_cputime - collected CPU time counts
208 209
 * @utime:		time spent in user mode, in nanoseconds
 * @stime:		time spent in kernel mode, in nanoseconds
210
 * @sum_exec_runtime:	total time spent on the CPU, in nanoseconds
211
 *
212 213 214
 * This structure groups together three kinds of CPU time that are tracked for
 * threads and thread groups.  Most things considering CPU time want to group
 * these counts together and treat all three of them in parallel.
215 216
 */
struct task_cputime {
217 218
	u64 utime;
	u64 stime;
219 220
	unsigned long long sum_exec_runtime;
};
221

222 223
/* Alternate field names when used to cache expirations. */
#define virt_exp	utime
224
#define prof_exp	stime
225 226
#define sched_exp	sum_exec_runtime

L
Linus Torvalds 已提交
227
struct sched_info {
228
#ifdef CONFIG_SCHED_INFO
L
Linus Torvalds 已提交
229
	/* cumulative counters */
230
	unsigned long pcount;	      /* # of times run on this cpu */
231
	unsigned long long run_delay; /* time spent waiting on a runqueue */
L
Linus Torvalds 已提交
232 233

	/* timestamps */
234 235
	unsigned long long last_arrival,/* when we last ran on a cpu */
			   last_queued;	/* when we were last queued to run */
236
#endif /* CONFIG_SCHED_INFO */
237
};
L
Linus Torvalds 已提交
238

239 240 241 242 243 244 245 246 247 248
/*
 * Integer metrics need fixed point arithmetic, e.g., sched/fair
 * has a few: load, load_avg, util_avg, freq, and capacity.
 *
 * We define a basic fixed point arithmetic range, and then formalize
 * all these metrics based on that basic range.
 */
# define SCHED_FIXEDPOINT_SHIFT	10
# define SCHED_FIXEDPOINT_SCALE	(1L << SCHED_FIXEDPOINT_SHIFT)

I
Ingo Molnar 已提交
249
struct load_weight {
250 251
	unsigned long weight;
	u32 inv_weight;
I
Ingo Molnar 已提交
252 253
};

254
/*
255 256 257 258 259 260 261 262 263
 * The load_avg/util_avg accumulates an infinite geometric series
 * (see __update_load_avg() in kernel/sched/fair.c).
 *
 * [load_avg definition]
 *
 *   load_avg = runnable% * scale_load_down(load)
 *
 * where runnable% is the time ratio that a sched_entity is runnable.
 * For cfs_rq, it is the aggregated load_avg of all runnable and
264
 * blocked sched_entities.
265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304
 *
 * load_avg may also take frequency scaling into account:
 *
 *   load_avg = runnable% * scale_load_down(load) * freq%
 *
 * where freq% is the CPU frequency normalized to the highest frequency.
 *
 * [util_avg definition]
 *
 *   util_avg = running% * SCHED_CAPACITY_SCALE
 *
 * where running% is the time ratio that a sched_entity is running on
 * a CPU. For cfs_rq, it is the aggregated util_avg of all runnable
 * and blocked sched_entities.
 *
 * util_avg may also factor frequency scaling and CPU capacity scaling:
 *
 *   util_avg = running% * SCHED_CAPACITY_SCALE * freq% * capacity%
 *
 * where freq% is the same as above, and capacity% is the CPU capacity
 * normalized to the greatest capacity (due to uarch differences, etc).
 *
 * N.B., the above ratios (runnable%, running%, freq%, and capacity%)
 * themselves are in the range of [0, 1]. To do fixed point arithmetics,
 * we therefore scale them to as large a range as necessary. This is for
 * example reflected by util_avg's SCHED_CAPACITY_SCALE.
 *
 * [Overflow issue]
 *
 * The 64-bit load_sum can have 4353082796 (=2^64/47742/88761) entities
 * with the highest load (=88761), always runnable on a single cfs_rq,
 * and should not overflow as the number already hits PID_MAX_LIMIT.
 *
 * For all other cases (including 32-bit kernels), struct load_weight's
 * weight will overflow first before we do, because:
 *
 *    Max(load_avg) <= Max(load.weight)
 *
 * Then it is the load_weight's responsibility to consider overflow
 * issues.
305
 */
306
struct sched_avg {
307 308 309
	u64 last_update_time, load_sum;
	u32 util_sum, period_contrib;
	unsigned long load_avg, util_avg;
310 311
};

312
struct sched_statistics {
313
#ifdef CONFIG_SCHEDSTATS
I
Ingo Molnar 已提交
314
	u64			wait_start;
315
	u64			wait_max;
316 317
	u64			wait_count;
	u64			wait_sum;
318 319
	u64			iowait_count;
	u64			iowait_sum;
320

I
Ingo Molnar 已提交
321 322
	u64			sleep_start;
	u64			sleep_max;
323 324 325
	s64			sum_sleep_runtime;

	u64			block_start;
I
Ingo Molnar 已提交
326 327
	u64			block_max;
	u64			exec_max;
I
Ingo Molnar 已提交
328
	u64			slice_max;
329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344

	u64			nr_migrations_cold;
	u64			nr_failed_migrations_affine;
	u64			nr_failed_migrations_running;
	u64			nr_failed_migrations_hot;
	u64			nr_forced_migrations;

	u64			nr_wakeups;
	u64			nr_wakeups_sync;
	u64			nr_wakeups_migrate;
	u64			nr_wakeups_local;
	u64			nr_wakeups_remote;
	u64			nr_wakeups_affine;
	u64			nr_wakeups_affine_attempts;
	u64			nr_wakeups_passive;
	u64			nr_wakeups_idle;
345
#endif
346
};
347 348 349 350 351 352 353 354 355 356 357 358 359 360 361

struct sched_entity {
	struct load_weight	load;		/* for load-balancing */
	struct rb_node		run_node;
	struct list_head	group_node;
	unsigned int		on_rq;

	u64			exec_start;
	u64			sum_exec_runtime;
	u64			vruntime;
	u64			prev_sum_exec_runtime;

	u64			nr_migrations;

	struct sched_statistics statistics;
362

I
Ingo Molnar 已提交
363
#ifdef CONFIG_FAIR_GROUP_SCHED
P
Peter Zijlstra 已提交
364
	int			depth;
I
Ingo Molnar 已提交
365 366 367 368 369 370
	struct sched_entity	*parent;
	/* rq on which this entity is (to be) queued: */
	struct cfs_rq		*cfs_rq;
	/* rq "owned" by this entity/group: */
	struct cfs_rq		*my_q;
#endif
371

372
#ifdef CONFIG_SMP
373 374 375 376 377 378 379
	/*
	 * Per entity load average tracking.
	 *
	 * Put into separate cache line so it does not
	 * collide with read-mostly values above.
	 */
	struct sched_avg	avg ____cacheline_aligned_in_smp;
380
#endif
I
Ingo Molnar 已提交
381
};
382

P
Peter Zijlstra 已提交
383 384
struct sched_rt_entity {
	struct list_head run_list;
385
	unsigned long timeout;
386
	unsigned long watchdog_stamp;
387
	unsigned int time_slice;
388 389
	unsigned short on_rq;
	unsigned short on_list;
P
Peter Zijlstra 已提交
390

391
	struct sched_rt_entity *back;
392
#ifdef CONFIG_RT_GROUP_SCHED
P
Peter Zijlstra 已提交
393 394 395 396 397 398
	struct sched_rt_entity	*parent;
	/* rq on which this entity is (to be) queued: */
	struct rt_rq		*rt_rq;
	/* rq "owned" by this entity/group: */
	struct rt_rq		*my_q;
#endif
P
Peter Zijlstra 已提交
399 400
};

401 402 403 404 405
struct sched_dl_entity {
	struct rb_node	rb_node;

	/*
	 * Original scheduling parameters. Copied here from sched_attr
406 407
	 * during sched_setattr(), they will remain the same until
	 * the next sched_setattr().
408 409 410
	 */
	u64 dl_runtime;		/* maximum runtime for each instance	*/
	u64 dl_deadline;	/* relative deadline of each instance	*/
411
	u64 dl_period;		/* separation of two instances (period) */
412
	u64 dl_bw;		/* dl_runtime / dl_deadline		*/
413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429

	/*
	 * Actual scheduling parameters. Initialized with the values above,
	 * they are continously updated during task execution. Note that
	 * the remaining runtime could be < 0 in case we are in overrun.
	 */
	s64 runtime;		/* remaining runtime for this instance	*/
	u64 deadline;		/* absolute deadline for this instance	*/
	unsigned int flags;	/* specifying the scheduler behaviour	*/

	/*
	 * Some bool flags:
	 *
	 * @dl_throttled tells if we exhausted the runtime. If so, the
	 * task has to wait for a replenishment to be performed at the
	 * next firing of dl_timer.
	 *
430 431
	 * @dl_boosted tells if we are boosted due to DI. If so we are
	 * outside bandwidth enforcement mechanism (but only until we
432 433 434 435
	 * exit the critical section);
	 *
	 * @dl_yielded tells if task gave up the cpu before consuming
	 * all its available runtime during the last job.
436
	 */
437
	int dl_throttled, dl_boosted, dl_yielded;
438 439 440 441 442 443 444

	/*
	 * Bandwidth enforcement timer. Each -deadline task has its
	 * own bandwidth to be enforced, thus we need one timer per task.
	 */
	struct hrtimer dl_timer;
};
445

446 447
union rcu_special {
	struct {
448 449 450 451 452 453
		u8 blocked;
		u8 need_qs;
		u8 exp_need_qs;
		u8 pad;	/* Otherwise the compiler can store garbage here. */
	} b; /* Bits. */
	u32 s; /* Set of bits. */
454
};
455

P
Peter Zijlstra 已提交
456 457 458
enum perf_event_task_context {
	perf_invalid_context = -1,
	perf_hw_context = 0,
459
	perf_sw_context,
P
Peter Zijlstra 已提交
460 461 462
	perf_nr_task_contexts,
};

463 464 465 466
struct wake_q_node {
	struct wake_q_node *next;
};

L
Linus Torvalds 已提交
467
struct task_struct {
468 469 470 471 472 473 474
#ifdef CONFIG_THREAD_INFO_IN_TASK
	/*
	 * For reasons of header soup (see current_thread_info()), this
	 * must be the first element of task_struct.
	 */
	struct thread_info thread_info;
#endif
L
Linus Torvalds 已提交
475
	volatile long state;	/* -1 unrunnable, 0 runnable, >0 stopped */
R
Roman Zippel 已提交
476
	void *stack;
L
Linus Torvalds 已提交
477
	atomic_t usage;
478 479
	unsigned int flags;	/* per process flags, defined below */
	unsigned int ptrace;
L
Linus Torvalds 已提交
480

481
#ifdef CONFIG_SMP
P
Peter Zijlstra 已提交
482
	struct llist_node wake_entry;
P
Peter Zijlstra 已提交
483
	int on_cpu;
484 485 486
#ifdef CONFIG_THREAD_INFO_IN_TASK
	unsigned int cpu;	/* current CPU */
#endif
M
Mike Galbraith 已提交
487
	unsigned int wakee_flips;
488
	unsigned long wakee_flip_decay_ts;
M
Mike Galbraith 已提交
489
	struct task_struct *last_wakee;
490 491

	int wake_cpu;
492
#endif
P
Peter Zijlstra 已提交
493
	int on_rq;
494

495
	int prio, static_prio, normal_prio;
496
	unsigned int rt_priority;
497
	const struct sched_class *sched_class;
I
Ingo Molnar 已提交
498
	struct sched_entity se;
P
Peter Zijlstra 已提交
499
	struct sched_rt_entity rt;
P
Peter Zijlstra 已提交
500 501 502
#ifdef CONFIG_CGROUP_SCHED
	struct task_group *sched_task_group;
#endif
503
	struct sched_dl_entity dl;
L
Linus Torvalds 已提交
504

505 506 507 508 509
#ifdef CONFIG_PREEMPT_NOTIFIERS
	/* list of struct preempt_notifier: */
	struct hlist_head preempt_notifiers;
#endif

510
#ifdef CONFIG_BLK_DEV_IO_TRACE
511
	unsigned int btrace_seq;
512
#endif
L
Linus Torvalds 已提交
513

514
	unsigned int policy;
515
	int nr_cpus_allowed;
L
Linus Torvalds 已提交
516 517
	cpumask_t cpus_allowed;

P
Paul E. McKenney 已提交
518
#ifdef CONFIG_PREEMPT_RCU
P
Paul E. McKenney 已提交
519
	int rcu_read_lock_nesting;
520
	union rcu_special rcu_read_unlock_special;
521
	struct list_head rcu_node_entry;
P
Paul E. McKenney 已提交
522
	struct rcu_node *rcu_blocked_node;
523
#endif /* #ifdef CONFIG_PREEMPT_RCU */
P
Paul E. McKenney 已提交
524 525 526 527
#ifdef CONFIG_TASKS_RCU
	unsigned long rcu_tasks_nvcsw;
	bool rcu_tasks_holdout;
	struct list_head rcu_tasks_holdout_list;
528
	int rcu_tasks_idle_cpu;
P
Paul E. McKenney 已提交
529
#endif /* #ifdef CONFIG_TASKS_RCU */
P
Paul E. McKenney 已提交
530

L
Linus Torvalds 已提交
531 532 533
	struct sched_info sched_info;

	struct list_head tasks;
534
#ifdef CONFIG_SMP
535
	struct plist_node pushable_tasks;
536
	struct rb_node pushable_dl_tasks;
537
#endif
L
Linus Torvalds 已提交
538 539

	struct mm_struct *mm, *active_mm;
540 541 542 543

	/* Per-thread vma caching: */
	struct vmacache vmacache;

544 545 546
#if defined(SPLIT_RSS_COUNTING)
	struct task_rss_stat	rss_stat;
#endif
L
Linus Torvalds 已提交
547
/* task state */
548
	int exit_state;
L
Linus Torvalds 已提交
549 550
	int exit_code, exit_signal;
	int pdeath_signal;  /*  The signal sent when the parent dies  */
551
	unsigned long jobctl;	/* JOBCTL_*, siglock protected */
552 553

	/* Used for emulating ABI behavior of previous Linux versions */
554
	unsigned int personality;
555

556
	/* scheduler bits, serialized by scheduler locks */
557
	unsigned sched_reset_on_fork:1;
558
	unsigned sched_contributes_to_load:1;
559
	unsigned sched_migrated:1;
P
Peter Zijlstra 已提交
560
	unsigned sched_remote_wakeup:1;
561 562 563 564 565
	unsigned :0; /* force alignment to the next boundary */

	/* unserialized, strictly 'current' */
	unsigned in_execve:1; /* bit to tell LSMs we're in execve */
	unsigned in_iowait:1;
566 567 568
#if !defined(TIF_RESTORE_SIGMASK)
	unsigned restore_sigmask:1;
#endif
T
Tejun Heo 已提交
569 570
#ifdef CONFIG_MEMCG
	unsigned memcg_may_oom:1;
571
#ifndef CONFIG_SLOB
572 573
	unsigned memcg_kmem_skip_account:1;
#endif
574
#endif
575 576 577
#ifdef CONFIG_COMPAT_BRK
	unsigned brk_randomized:1;
#endif
578

579 580
	unsigned long atomic_flags; /* Flags needing atomic access. */

581 582
	struct restart_block restart_block;

L
Linus Torvalds 已提交
583 584
	pid_t pid;
	pid_t tgid;
585

586
#ifdef CONFIG_CC_STACKPROTECTOR
587 588
	/* Canary value for the -fstack-protector gcc feature */
	unsigned long stack_canary;
589
#endif
590
	/*
L
Linus Torvalds 已提交
591
	 * pointers to (original) parent process, youngest child, younger sibling,
592
	 * older sibling, respectively.  (p->father can be replaced with
R
Roland McGrath 已提交
593
	 * p->real_parent->pid)
L
Linus Torvalds 已提交
594
	 */
595 596
	struct task_struct __rcu *real_parent; /* real parent process */
	struct task_struct __rcu *parent; /* recipient of SIGCHLD, wait4() reports */
L
Linus Torvalds 已提交
597
	/*
R
Roland McGrath 已提交
598
	 * children/sibling forms the list of my natural children
L
Linus Torvalds 已提交
599 600 601 602 603
	 */
	struct list_head children;	/* list of my children */
	struct list_head sibling;	/* linkage in my parent's children list */
	struct task_struct *group_leader;	/* threadgroup leader */

R
Roland McGrath 已提交
604 605 606 607 608 609 610 611
	/*
	 * ptraced is the list of tasks this task is using ptrace on.
	 * This includes both natural children and PTRACE_ATTACH targets.
	 * p->ptrace_entry is p's link on the p->parent->ptraced list.
	 */
	struct list_head ptraced;
	struct list_head ptrace_entry;

L
Linus Torvalds 已提交
612
	/* PID/PID hash table linkage. */
613
	struct pid_link pids[PIDTYPE_MAX];
O
Oleg Nesterov 已提交
614
	struct list_head thread_group;
615
	struct list_head thread_node;
L
Linus Torvalds 已提交
616 617 618 619 620

	struct completion *vfork_done;		/* for vfork() */
	int __user *set_child_tid;		/* CLONE_CHILD_SETTID */
	int __user *clear_child_tid;		/* CLONE_CHILD_CLEARTID */

621
	u64 utime, stime;
622
#ifdef CONFIG_ARCH_HAS_SCALED_CPUTIME
623
	u64 utimescaled, stimescaled;
624
#endif
625
	u64 gtime;
626
	struct prev_cputime prev_cputime;
627
#ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN
628
	seqcount_t vtime_seqcount;
629 630
	unsigned long long vtime_snap;
	enum {
631 632 633
		/* Task is sleeping or running in a CPU with VTIME inactive */
		VTIME_INACTIVE = 0,
		/* Task runs in userspace in a CPU with VTIME active */
634
		VTIME_USER,
635
		/* Task runs in kernelspace in a CPU with VTIME active */
636 637
		VTIME_SYS,
	} vtime_snap_whence;
638
#endif
639 640

#ifdef CONFIG_NO_HZ_FULL
641
	atomic_t tick_dep_mask;
642
#endif
L
Linus Torvalds 已提交
643
	unsigned long nvcsw, nivcsw; /* context switch counts */
644
	u64 start_time;		/* monotonic time in nsec */
645
	u64 real_start_time;	/* boot based time in nsec */
L
Linus Torvalds 已提交
646 647 648
/* mm fault and swap info: this can arguably be seen as either mm-specific or thread-specific */
	unsigned long min_flt, maj_flt;

649
#ifdef CONFIG_POSIX_TIMERS
650
	struct task_cputime cputime_expires;
L
Linus Torvalds 已提交
651
	struct list_head cpu_timers[3];
652
#endif
L
Linus Torvalds 已提交
653 654

/* process credentials */
655
	const struct cred __rcu *ptracer_cred; /* Tracer's credentials at attach */
A
Arnd Bergmann 已提交
656
	const struct cred __rcu *real_cred; /* objective and real subjective task
657
					 * credentials (COW) */
A
Arnd Bergmann 已提交
658
	const struct cred __rcu *cred;	/* effective (overridable) subjective task
659
					 * credentials (COW) */
660 661 662
	char comm[TASK_COMM_LEN]; /* executable name excluding path
				     - access with [gs]et_task_comm (which lock
				       it with task_lock())
663
				     - initialized normally by setup_new_exec */
L
Linus Torvalds 已提交
664
/* file system info */
665
	struct nameidata *nameidata;
666
#ifdef CONFIG_SYSVIPC
L
Linus Torvalds 已提交
667 668
/* ipc stuff */
	struct sysv_sem sysvsem;
669
	struct sysv_shm sysvshm;
670
#endif
671
#ifdef CONFIG_DETECT_HUNG_TASK
672 673 674
/* hung task detection */
	unsigned long last_switch_count;
#endif
L
Linus Torvalds 已提交
675 676 677 678
/* filesystem information */
	struct fs_struct *fs;
/* open file information */
	struct files_struct *files;
679
/* namespaces */
S
Serge E. Hallyn 已提交
680
	struct nsproxy *nsproxy;
L
Linus Torvalds 已提交
681 682 683 684 685
/* signal handlers */
	struct signal_struct *signal;
	struct sighand_struct *sighand;

	sigset_t blocked, real_blocked;
686
	sigset_t saved_sigmask;	/* restored if set_restore_sigmask() was used */
L
Linus Torvalds 已提交
687 688 689 690
	struct sigpending pending;

	unsigned long sas_ss_sp;
	size_t sas_ss_size;
691
	unsigned sas_ss_flags;
692

693
	struct callback_head *task_works;
694

L
Linus Torvalds 已提交
695
	struct audit_context *audit_context;
A
Al Viro 已提交
696
#ifdef CONFIG_AUDITSYSCALL
697
	kuid_t loginuid;
698
	unsigned int sessionid;
A
Al Viro 已提交
699
#endif
700
	struct seccomp seccomp;
L
Linus Torvalds 已提交
701 702 703 704

/* Thread group tracking */
   	u32 parent_exec_id;
   	u32 self_exec_id;
705 706
/* Protection of (de-)allocation: mm, files, fs, tty, keyrings, mems_allowed,
 * mempolicy */
L
Linus Torvalds 已提交
707 708
	spinlock_t alloc_lock;

709
	/* Protection of the PI data structures: */
710
	raw_spinlock_t pi_lock;
711

712 713
	struct wake_q_node wake_q;

I
Ingo Molnar 已提交
714 715
#ifdef CONFIG_RT_MUTEXES
	/* PI waiters blocked on a rt_mutex held by this task */
716 717
	struct rb_root pi_waiters;
	struct rb_node *pi_waiters_leftmost;
I
Ingo Molnar 已提交
718 719 720 721
	/* Deadlock detection and priority inheritance handling */
	struct rt_mutex_waiter *pi_blocked_on;
#endif

722 723 724 725
#ifdef CONFIG_DEBUG_MUTEXES
	/* mutex deadlock detection */
	struct mutex_waiter *blocked_on;
#endif
726 727 728 729
#ifdef CONFIG_TRACE_IRQFLAGS
	unsigned int irq_events;
	unsigned long hardirq_enable_ip;
	unsigned long hardirq_disable_ip;
730
	unsigned int hardirq_enable_event;
731
	unsigned int hardirq_disable_event;
732 733
	int hardirqs_enabled;
	int hardirq_context;
734 735
	unsigned long softirq_disable_ip;
	unsigned long softirq_enable_ip;
736
	unsigned int softirq_disable_event;
737
	unsigned int softirq_enable_event;
738
	int softirqs_enabled;
739 740
	int softirq_context;
#endif
I
Ingo Molnar 已提交
741
#ifdef CONFIG_LOCKDEP
742
# define MAX_LOCK_DEPTH 48UL
I
Ingo Molnar 已提交
743 744 745
	u64 curr_chain_key;
	int lockdep_depth;
	unsigned int lockdep_recursion;
746
	struct held_lock held_locks[MAX_LOCK_DEPTH];
747
	gfp_t lockdep_reclaim_gfp;
I
Ingo Molnar 已提交
748
#endif
749 750 751
#ifdef CONFIG_UBSAN
	unsigned int in_ubsan;
#endif
752

L
Linus Torvalds 已提交
753 754 755
/* journalling filesystem info */
	void *journal_info;

756
/* stacked block device info */
757
	struct bio_list *bio_list;
758

759 760 761 762 763
#ifdef CONFIG_BLOCK
/* stack plugging */
	struct blk_plug *plug;
#endif

L
Linus Torvalds 已提交
764 765 766 767 768 769 770 771 772
/* VM state */
	struct reclaim_state *reclaim_state;

	struct backing_dev_info *backing_dev_info;

	struct io_context *io_context;

	unsigned long ptrace_message;
	siginfo_t *last_siginfo; /* For ptrace use.  */
773
	struct task_io_accounting ioac;
774
#if defined(CONFIG_TASK_XACCT)
L
Linus Torvalds 已提交
775 776
	u64 acct_rss_mem1;	/* accumulated rss usage */
	u64 acct_vm_mem1;	/* accumulated virtual memory usage */
777
	u64 acct_timexpd;	/* stime + utime since last update */
L
Linus Torvalds 已提交
778 779
#endif
#ifdef CONFIG_CPUSETS
780
	nodemask_t mems_allowed;	/* Protected by alloc_lock */
781
	seqcount_t mems_allowed_seq;	/* Seqence no to catch updates */
782
	int cpuset_mem_spread_rotor;
783
	int cpuset_slab_spread_rotor;
L
Linus Torvalds 已提交
784
#endif
785
#ifdef CONFIG_CGROUPS
786
	/* Control Group info protected by css_set_lock */
A
Arnd Bergmann 已提交
787
	struct css_set __rcu *cgroups;
788 789
	/* cg_list protected by css_set_lock and tsk->alloc_lock */
	struct list_head cg_list;
790
#endif
F
Fenghua Yu 已提交
791 792 793
#ifdef CONFIG_INTEL_RDT_A
	int closid;
#endif
794
#ifdef CONFIG_FUTEX
795
	struct robust_list_head __user *robust_list;
796 797 798
#ifdef CONFIG_COMPAT
	struct compat_robust_list_head __user *compat_robust_list;
#endif
799 800
	struct list_head pi_state_list;
	struct futex_pi_state *pi_state_cache;
801
#endif
802
#ifdef CONFIG_PERF_EVENTS
P
Peter Zijlstra 已提交
803
	struct perf_event_context *perf_event_ctxp[perf_nr_task_contexts];
804 805
	struct mutex perf_event_mutex;
	struct list_head perf_event_list;
806
#endif
807 808 809
#ifdef CONFIG_DEBUG_PREEMPT
	unsigned long preempt_disable_ip;
#endif
810
#ifdef CONFIG_NUMA
811
	struct mempolicy *mempolicy;	/* Protected by alloc_lock */
812
	short il_next;
813
	short pref_node_fork;
814
#endif
815 816 817
#ifdef CONFIG_NUMA_BALANCING
	int numa_scan_seq;
	unsigned int numa_scan_period;
818
	unsigned int numa_scan_period_max;
819
	int numa_preferred_nid;
820
	unsigned long numa_migrate_retry;
821
	u64 node_stamp;			/* migration stamp  */
822 823
	u64 last_task_numa_placement;
	u64 last_sum_exec_runtime;
824
	struct callback_head numa_work;
825

826 827 828
	struct list_head numa_entry;
	struct numa_group *numa_group;

829
	/*
830 831 832 833 834 835 836 837 838 839 840 841
	 * numa_faults is an array split into four regions:
	 * faults_memory, faults_cpu, faults_memory_buffer, faults_cpu_buffer
	 * in this precise order.
	 *
	 * faults_memory: Exponential decaying average of faults on a per-node
	 * basis. Scheduling placement decisions are made based on these
	 * counts. The values remain static for the duration of a PTE scan.
	 * faults_cpu: Track the nodes the process was running on when a NUMA
	 * hinting fault was incurred.
	 * faults_memory_buffer and faults_cpu_buffer: Record faults per node
	 * during the current scan window. When the scan completes, the counts
	 * in faults_memory and faults_cpu decay and these values are copied.
842
	 */
843
	unsigned long *numa_faults;
844
	unsigned long total_numa_faults;
845

846 847
	/*
	 * numa_faults_locality tracks if faults recorded during the last
848 849 850
	 * scan window were remote/local or failed to migrate. The task scan
	 * period is adapted based on the locality of the faults with different
	 * weights depending on whether they were shared or private faults
851
	 */
852
	unsigned long numa_faults_locality[3];
853

I
Ingo Molnar 已提交
854
	unsigned long numa_pages_migrated;
855 856
#endif /* CONFIG_NUMA_BALANCING */

857 858
	struct tlbflush_unmap_batch tlb_ubc;

I
Ingo Molnar 已提交
859
	struct rcu_head rcu;
860 861 862 863 864

	/*
	 * cache last used pipe for splice
	 */
	struct pipe_inode_info *splice_pipe;
865 866 867

	struct page_frag task_frag;

868 869
#ifdef CONFIG_TASK_DELAY_ACCT
	struct task_delay_info		*delays;
870
#endif
871

872 873
#ifdef CONFIG_FAULT_INJECTION
	int make_it_fail;
874
#endif
875 876 877 878 879 880
	/*
	 * when (nr_dirtied >= nr_dirtied_pause), it's time to call
	 * balance_dirty_pages() for some dirty throttling pause
	 */
	int nr_dirtied;
	int nr_dirtied_pause;
881
	unsigned long dirty_paused_when; /* start of a write-and-pause period */
882

A
Arjan van de Ven 已提交
883 884 885 886
#ifdef CONFIG_LATENCYTOP
	int latency_record_count;
	struct latency_record latency_record[LT_SAVECOUNT];
#endif
887 888 889 890
	/*
	 * time slack values; these are used to round up poll() and
	 * select() etc timeout values. These are in nanoseconds.
	 */
891 892
	u64 timer_slack_ns;
	u64 default_timer_slack_ns;
893

894 895 896
#ifdef CONFIG_KASAN
	unsigned int kasan_depth;
#endif
897
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
D
Daniel Mack 已提交
898
	/* Index of current stored address in ret_stack */
899 900 901
	int curr_ret_stack;
	/* Stack of return addresses for return function tracing */
	struct ftrace_ret_stack	*ret_stack;
902 903
	/* time stamp for last schedule */
	unsigned long long ftrace_timestamp;
904 905 906 907 908
	/*
	 * Number of functions that haven't been traced
	 * because of depth overrun.
	 */
	atomic_t trace_overrun;
909 910
	/* Pause for the tracing */
	atomic_t tracing_graph_pause;
911
#endif
912 913 914
#ifdef CONFIG_TRACING
	/* state flags for use by tracers */
	unsigned long trace;
915
	/* bitmask and counter of trace recursion */
916 917
	unsigned long trace_recursion;
#endif /* CONFIG_TRACING */
D
Dmitry Vyukov 已提交
918 919 920 921 922 923 924 925 926 927
#ifdef CONFIG_KCOV
	/* Coverage collection mode enabled for this task (0 if disabled). */
	enum kcov_mode kcov_mode;
	/* Size of the kcov_area. */
	unsigned	kcov_size;
	/* Buffer for coverage collection. */
	void		*kcov_area;
	/* kcov desciptor wired with this task or NULL. */
	struct kcov	*kcov;
#endif
928
#ifdef CONFIG_MEMCG
T
Tejun Heo 已提交
929 930 931
	struct mem_cgroup *memcg_in_oom;
	gfp_t memcg_oom_gfp_mask;
	int memcg_oom_order;
932 933 934

	/* number of pages to reclaim on returning to userland */
	unsigned int memcg_nr_pages_over_high;
935
#endif
936 937 938
#ifdef CONFIG_UPROBES
	struct uprobe_task *utask;
#endif
K
Kent Overstreet 已提交
939 940 941 942
#if defined(CONFIG_BCACHE) || defined(CONFIG_BCACHE_MODULE)
	unsigned int	sequential_io;
	unsigned int	sequential_io_avg;
#endif
P
Peter Zijlstra 已提交
943 944 945
#ifdef CONFIG_DEBUG_ATOMIC_SLEEP
	unsigned long	task_state_change;
#endif
946
	int pagefault_disabled;
947
#ifdef CONFIG_MMU
948
	struct task_struct *oom_reaper_list;
949
#endif
950 951 952
#ifdef CONFIG_VMAP_STACK
	struct vm_struct *stack_vm_area;
#endif
953 954 955 956
#ifdef CONFIG_THREAD_INFO_IN_TASK
	/* A live task holds one reference. */
	atomic_t stack_refcount;
#endif
957 958 959 960 961 962 963 964
/* CPU-specific state of this task */
	struct thread_struct thread;
/*
 * WARNING: on x86, 'thread_struct' contains a variable-sized
 * structure.  It *MUST* be at the end of 'task_struct'.
 *
 * Do not put anything below here!
 */
L
Linus Torvalds 已提交
965 966
};

A
Alexey Dobriyan 已提交
967
static inline struct pid *task_pid(struct task_struct *task)
968 969 970 971
{
	return task->pids[PIDTYPE_PID].pid;
}

A
Alexey Dobriyan 已提交
972
static inline struct pid *task_tgid(struct task_struct *task)
973 974 975 976
{
	return task->group_leader->pids[PIDTYPE_PID].pid;
}

977 978 979 980 981
/*
 * Without tasklist or rcu lock it is not safe to dereference
 * the result of task_pgrp/task_session even if task == current,
 * we can race with another thread doing sys_setsid/sys_setpgid.
 */
A
Alexey Dobriyan 已提交
982
static inline struct pid *task_pgrp(struct task_struct *task)
983 984 985 986
{
	return task->group_leader->pids[PIDTYPE_PGID].pid;
}

A
Alexey Dobriyan 已提交
987
static inline struct pid *task_session(struct task_struct *task)
988 989 990 991
{
	return task->group_leader->pids[PIDTYPE_SID].pid;
}

992 993 994 995 996
/*
 * the helpers to get the task's different pids as they are seen
 * from various namespaces
 *
 * task_xid_nr()     : global id, i.e. the id seen from the init namespace;
E
Eric W. Biederman 已提交
997 998
 * task_xid_vnr()    : virtual id, i.e. the id seen from the pid namespace of
 *                     current.
999 1000 1001 1002 1003 1004
 * task_xid_nr_ns()  : id seen from the ns specified;
 *
 * set_task_vxid()   : assigns a virtual id to a task;
 *
 * see also pid_nr() etc in include/linux/pid.h
 */
1005 1006
pid_t __task_pid_nr_ns(struct task_struct *task, enum pid_type type,
			struct pid_namespace *ns);
1007

A
Alexey Dobriyan 已提交
1008
static inline pid_t task_pid_nr(struct task_struct *tsk)
1009 1010 1011 1012
{
	return tsk->pid;
}

1013 1014 1015 1016 1017
static inline pid_t task_pid_nr_ns(struct task_struct *tsk,
					struct pid_namespace *ns)
{
	return __task_pid_nr_ns(tsk, PIDTYPE_PID, ns);
}
1018 1019 1020

static inline pid_t task_pid_vnr(struct task_struct *tsk)
{
1021
	return __task_pid_nr_ns(tsk, PIDTYPE_PID, NULL);
1022 1023 1024
}


A
Alexey Dobriyan 已提交
1025
static inline pid_t task_tgid_nr(struct task_struct *tsk)
1026 1027 1028 1029
{
	return tsk->tgid;
}

1030
pid_t task_tgid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns);
1031 1032 1033 1034 1035 1036 1037

static inline pid_t task_tgid_vnr(struct task_struct *tsk)
{
	return pid_vnr(task_tgid(tsk));
}


1038
static inline int pid_alive(const struct task_struct *p);
1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055
static inline pid_t task_ppid_nr_ns(const struct task_struct *tsk, struct pid_namespace *ns)
{
	pid_t pid = 0;

	rcu_read_lock();
	if (pid_alive(tsk))
		pid = task_tgid_nr_ns(rcu_dereference(tsk->real_parent), ns);
	rcu_read_unlock();

	return pid;
}

static inline pid_t task_ppid_nr(const struct task_struct *tsk)
{
	return task_ppid_nr_ns(tsk, &init_pid_ns);
}

1056 1057
static inline pid_t task_pgrp_nr_ns(struct task_struct *tsk,
					struct pid_namespace *ns)
1058
{
1059
	return __task_pid_nr_ns(tsk, PIDTYPE_PGID, ns);
1060 1061 1062 1063
}

static inline pid_t task_pgrp_vnr(struct task_struct *tsk)
{
1064
	return __task_pid_nr_ns(tsk, PIDTYPE_PGID, NULL);
1065 1066 1067
}


1068 1069
static inline pid_t task_session_nr_ns(struct task_struct *tsk,
					struct pid_namespace *ns)
1070
{
1071
	return __task_pid_nr_ns(tsk, PIDTYPE_SID, ns);
1072 1073 1074 1075
}

static inline pid_t task_session_vnr(struct task_struct *tsk)
{
1076
	return __task_pid_nr_ns(tsk, PIDTYPE_SID, NULL);
1077 1078
}

1079 1080 1081 1082 1083
/* obsolete, do not use */
static inline pid_t task_pgrp_nr(struct task_struct *tsk)
{
	return task_pgrp_nr_ns(tsk, &init_pid_ns);
}
1084

L
Linus Torvalds 已提交
1085 1086 1087 1088 1089 1090 1091
/**
 * pid_alive - check that a task structure is not stale
 * @p: Task structure to be checked.
 *
 * Test if a process is not yet dead (at most zombie state)
 * If pid_alive fails, then pointers within the task structure
 * can be stale and must not be dereferenced.
1092 1093
 *
 * Return: 1 if the process is alive. 0 otherwise.
L
Linus Torvalds 已提交
1094
 */
1095
static inline int pid_alive(const struct task_struct *p)
L
Linus Torvalds 已提交
1096
{
1097
	return p->pids[PIDTYPE_PID].pid != NULL;
L
Linus Torvalds 已提交
1098 1099
}

1100
/**
1101 1102
 * is_global_init - check if a task structure is init. Since init
 * is free to have sub-threads we need to check tgid.
1103 1104 1105
 * @tsk: Task structure to be checked.
 *
 * Check if a task structure is the first user space task the kernel created.
1106 1107
 *
 * Return: 1 if the task structure is init. 0 otherwise.
1108
 */
A
Alexey Dobriyan 已提交
1109
static inline int is_global_init(struct task_struct *tsk)
1110
{
1111
	return task_tgid_nr(tsk) == 1;
1112
}
1113

1114 1115
extern struct pid *cad_pid;

L
Linus Torvalds 已提交
1116 1117 1118
/*
 * Per process flags
 */
1119
#define PF_IDLE		0x00000002	/* I am an IDLE thread */
L
Linus Torvalds 已提交
1120
#define PF_EXITING	0x00000004	/* getting shut down */
1121
#define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
1122
#define PF_VCPU		0x00000010	/* I'm a virtual CPU */
T
Tejun Heo 已提交
1123
#define PF_WQ_WORKER	0x00000020	/* I'm a workqueue worker */
L
Linus Torvalds 已提交
1124
#define PF_FORKNOEXEC	0x00000040	/* forked but didn't exec */
1125
#define PF_MCE_PROCESS  0x00000080      /* process policy on mce errors */
L
Linus Torvalds 已提交
1126 1127 1128 1129
#define PF_SUPERPRIV	0x00000100	/* used super-user privileges */
#define PF_DUMPCORE	0x00000200	/* dumped core */
#define PF_SIGNALED	0x00000400	/* killed by a signal */
#define PF_MEMALLOC	0x00000800	/* Allocating memory */
1130
#define PF_NPROC_EXCEEDED 0x00001000	/* set_user noticed that RLIMIT_NPROC was exceeded */
L
Linus Torvalds 已提交
1131
#define PF_USED_MATH	0x00002000	/* if unset the fpu must be initialized before use */
1132
#define PF_USED_ASYNC	0x00004000	/* used async_schedule*(), used by module init */
L
Linus Torvalds 已提交
1133 1134 1135 1136
#define PF_NOFREEZE	0x00008000	/* this thread should not be frozen */
#define PF_FROZEN	0x00010000	/* frozen for system suspend */
#define PF_FSTRANS	0x00020000	/* inside a filesystem transaction */
#define PF_KSWAPD	0x00040000	/* I am kswapd */
1137
#define PF_MEMALLOC_NOIO 0x00080000	/* Allocating memory without IO involved */
L
Linus Torvalds 已提交
1138
#define PF_LESS_THROTTLE 0x00100000	/* Throttle me less: I clean memory */
1139
#define PF_KTHREAD	0x00200000	/* I am a kernel thread */
J
Jens Axboe 已提交
1140 1141
#define PF_RANDOMIZE	0x00400000	/* randomize virtual address space */
#define PF_SWAPWRITE	0x00800000	/* Allowed to write to swap */
1142
#define PF_NO_SETAFFINITY 0x04000000	/* Userland is not allowed to meddle with cpus_allowed */
1143
#define PF_MCE_EARLY    0x08000000      /* Early kill for mce process policy */
1144
#define PF_MUTEX_TESTER	0x20000000	/* Thread belongs to the rt mutex tester */
1145
#define PF_FREEZER_SKIP	0x40000000	/* Freezer should not count it as freezable */
1146
#define PF_SUSPEND_TASK 0x80000000      /* this thread called freeze_processes and should not be frozen */
L
Linus Torvalds 已提交
1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172

/*
 * Only the _current_ task can read/write to tsk->flags, but other
 * tasks can access tsk->flags in readonly mode for example
 * with tsk_used_math (like during threaded core dumping).
 * There is however an exception to this rule during ptrace
 * or during fork: the ptracer task is allowed to write to the
 * child->flags of its traced child (same goes for fork, the parent
 * can write to the child->flags), because we're guaranteed the
 * child is not running and in turn not changing child->flags
 * at the same time the parent does it.
 */
#define clear_stopped_child_used_math(child) do { (child)->flags &= ~PF_USED_MATH; } while (0)
#define set_stopped_child_used_math(child) do { (child)->flags |= PF_USED_MATH; } while (0)
#define clear_used_math() clear_stopped_child_used_math(current)
#define set_used_math() set_stopped_child_used_math(current)
#define conditional_stopped_child_used_math(condition, child) \
	do { (child)->flags &= ~PF_USED_MATH, (child)->flags |= (condition) ? PF_USED_MATH : 0; } while (0)
#define conditional_used_math(condition) \
	conditional_stopped_child_used_math(condition, current)
#define copy_to_stopped_child_used_math(child) \
	do { (child)->flags &= ~PF_USED_MATH, (child)->flags |= current->flags & PF_USED_MATH; } while (0)
/* NOTE: this will return 0 or PF_USED_MATH, it will never return 1 */
#define tsk_used_math(p) ((p)->flags & PF_USED_MATH)
#define used_math() tsk_used_math(current)

1173
/* Per-process atomic flags. */
1174
#define PFA_NO_NEW_PRIVS 0	/* May not gain new privileges. */
1175 1176
#define PFA_SPREAD_PAGE  1      /* Spread page cache over cpuset */
#define PFA_SPREAD_SLAB  2      /* Spread some slab caches over cpuset */
1177
#define PFA_LMK_WAITING  3      /* Lowmemorykiller is waiting */
1178

1179

1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191
#define TASK_PFA_TEST(name, func)					\
	static inline bool task_##func(struct task_struct *p)		\
	{ return test_bit(PFA_##name, &p->atomic_flags); }
#define TASK_PFA_SET(name, func)					\
	static inline void task_set_##func(struct task_struct *p)	\
	{ set_bit(PFA_##name, &p->atomic_flags); }
#define TASK_PFA_CLEAR(name, func)					\
	static inline void task_clear_##func(struct task_struct *p)	\
	{ clear_bit(PFA_##name, &p->atomic_flags); }

TASK_PFA_TEST(NO_NEW_PRIVS, no_new_privs)
TASK_PFA_SET(NO_NEW_PRIVS, no_new_privs)
1192

1193 1194 1195 1196 1197 1198 1199
TASK_PFA_TEST(SPREAD_PAGE, spread_page)
TASK_PFA_SET(SPREAD_PAGE, spread_page)
TASK_PFA_CLEAR(SPREAD_PAGE, spread_page)

TASK_PFA_TEST(SPREAD_SLAB, spread_slab)
TASK_PFA_SET(SPREAD_SLAB, spread_slab)
TASK_PFA_CLEAR(SPREAD_SLAB, spread_slab)
1200

1201 1202 1203
TASK_PFA_TEST(LMK_WAITING, lmk_waiting)
TASK_PFA_SET(LMK_WAITING, lmk_waiting)

1204 1205 1206 1207 1208 1209 1210
static inline void tsk_restore_flags(struct task_struct *task,
				unsigned long orig_flags, unsigned long flags)
{
	task->flags &= ~flags;
	task->flags |= orig_flags & flags;
}

1211 1212
extern int cpuset_cpumask_can_shrink(const struct cpumask *cur,
				     const struct cpumask *trial);
1213 1214
extern int task_can_attach(struct task_struct *p,
			   const struct cpumask *cs_cpus_allowed);
L
Linus Torvalds 已提交
1215
#ifdef CONFIG_SMP
1216 1217 1218
extern void do_set_cpus_allowed(struct task_struct *p,
			       const struct cpumask *new_mask);

1219
extern int set_cpus_allowed_ptr(struct task_struct *p,
1220
				const struct cpumask *new_mask);
L
Linus Torvalds 已提交
1221
#else
1222 1223 1224 1225
static inline void do_set_cpus_allowed(struct task_struct *p,
				      const struct cpumask *new_mask)
{
}
1226
static inline int set_cpus_allowed_ptr(struct task_struct *p,
1227
				       const struct cpumask *new_mask)
L
Linus Torvalds 已提交
1228
{
1229
	if (!cpumask_test_cpu(0, new_mask))
L
Linus Torvalds 已提交
1230 1231 1232 1233
		return -EINVAL;
	return 0;
}
#endif
1234

1235 1236 1237 1238
#ifndef cpu_relax_yield
#define cpu_relax_yield() cpu_relax()
#endif

1239
extern int yield_to(struct task_struct *p, bool preempt);
1240 1241
extern void set_user_nice(struct task_struct *p, long nice);
extern int task_prio(const struct task_struct *p);
1242 1243 1244 1245 1246 1247 1248 1249 1250 1251
/**
 * task_nice - return the nice value of a given task.
 * @p: the task in question.
 *
 * Return: The nice value [ -20 ... 0 ... 19 ].
 */
static inline int task_nice(const struct task_struct *p)
{
	return PRIO_TO_NICE((p)->static_prio);
}
1252 1253
extern int can_nice(const struct task_struct *p, const int nice);
extern int task_curr(const struct task_struct *p);
L
Linus Torvalds 已提交
1254
extern int idle_cpu(int cpu);
1255 1256
extern int sched_setscheduler(struct task_struct *, int,
			      const struct sched_param *);
1257
extern int sched_setscheduler_nocheck(struct task_struct *, int,
1258
				      const struct sched_param *);
1259 1260
extern int sched_setattr(struct task_struct *,
			 const struct sched_attr *);
1261
extern struct task_struct *idle_task(int cpu);
1262 1263
/**
 * is_idle_task - is the specified task an idle task?
1264
 * @p: the task in question.
1265 1266
 *
 * Return: 1 if @p is an idle task. 0 otherwise.
1267
 */
1268
static inline bool is_idle_task(const struct task_struct *p)
1269
{
1270
	return !!(p->flags & PF_IDLE);
1271
}
1272
extern struct task_struct *curr_task(int cpu);
1273
extern void ia64_set_curr_task(int cpu, struct task_struct *p);
L
Linus Torvalds 已提交
1274 1275 1276 1277

void yield(void);

union thread_union {
1278
#ifndef CONFIG_THREAD_INFO_IN_TASK
L
Linus Torvalds 已提交
1279
	struct thread_info thread_info;
1280
#endif
L
Linus Torvalds 已提交
1281 1282 1283
	unsigned long stack[THREAD_SIZE/sizeof(long)];
};

1284 1285 1286 1287 1288 1289 1290 1291 1292
#ifdef CONFIG_THREAD_INFO_IN_TASK
static inline struct thread_info *task_thread_info(struct task_struct *task)
{
	return &task->thread_info;
}
#elif !defined(__HAVE_THREAD_FUNCTIONS)
# define task_thread_info(task)	((struct thread_info *)(task)->stack)
#endif

1293 1294 1295 1296 1297
/*
 * find a task by one of its numerical ids
 *
 * find_task_by_pid_ns():
 *      finds a task by its pid in the specified namespace
1298 1299
 * find_task_by_vpid():
 *      finds a task by its virtual pid
1300
 *
1301
 * see also find_vpid() etc in include/linux/pid.h
1302 1303
 */

1304 1305 1306
extern struct task_struct *find_task_by_vpid(pid_t nr);
extern struct task_struct *find_task_by_pid_ns(pid_t nr,
		struct pid_namespace *ns);
1307

1308 1309
extern int wake_up_state(struct task_struct *tsk, unsigned int state);
extern int wake_up_process(struct task_struct *tsk);
1310
extern void wake_up_new_task(struct task_struct *tsk);
L
Linus Torvalds 已提交
1311 1312 1313 1314 1315 1316
#ifdef CONFIG_SMP
 extern void kick_process(struct task_struct *tsk);
#else
 static inline void kick_process(struct task_struct *tsk) { }
#endif

1317 1318 1319 1320 1321
extern void __set_task_comm(struct task_struct *tsk, const char *from, bool exec);
static inline void set_task_comm(struct task_struct *tsk, const char *from)
{
	__set_task_comm(tsk, from, false);
}
1322
extern char *get_task_comm(char *to, struct task_struct *tsk);
L
Linus Torvalds 已提交
1323 1324

#ifdef CONFIG_SMP
1325
void scheduler_ipi(void);
R
Roland McGrath 已提交
1326
extern unsigned long wait_task_inactive(struct task_struct *, long match_state);
L
Linus Torvalds 已提交
1327
#else
1328
static inline void scheduler_ipi(void) { }
R
Roland McGrath 已提交
1329 1330 1331 1332 1333
static inline unsigned long wait_task_inactive(struct task_struct *p,
					       long match_state)
{
	return 1;
}
L
Linus Torvalds 已提交
1334 1335 1336 1337 1338 1339 1340
#endif

/* set thread flags in other task's structures
 * - see asm/thread_info.h for TIF_xxxx flags available
 */
static inline void set_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
1341
	set_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
1342 1343 1344 1345
}

static inline void clear_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
1346
	clear_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
1347 1348 1349 1350
}

static inline int test_and_set_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
1351
	return test_and_set_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
1352 1353 1354 1355
}

static inline int test_and_clear_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
1356
	return test_and_clear_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
1357 1358 1359 1360
}

static inline int test_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
1361
	return test_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373
}

static inline void set_tsk_need_resched(struct task_struct *tsk)
{
	set_tsk_thread_flag(tsk,TIF_NEED_RESCHED);
}

static inline void clear_tsk_need_resched(struct task_struct *tsk)
{
	clear_tsk_thread_flag(tsk,TIF_NEED_RESCHED);
}

1374 1375 1376 1377 1378
static inline int test_tsk_need_resched(struct task_struct *tsk)
{
	return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED));
}

L
Linus Torvalds 已提交
1379 1380 1381 1382 1383 1384 1385
/*
 * cond_resched() and cond_resched_lock(): latency reduction via
 * explicit rescheduling in places that are safe. The return
 * value indicates whether a reschedule was done in fact.
 * cond_resched_lock() will drop the spinlock before scheduling,
 * cond_resched_softirq() will enable bhs before scheduling.
 */
1386
#ifndef CONFIG_PREEMPT
1387
extern int _cond_resched(void);
1388 1389 1390
#else
static inline int _cond_resched(void) { return 0; }
#endif
1391

1392
#define cond_resched() ({			\
1393
	___might_sleep(__FILE__, __LINE__, 0);	\
1394 1395
	_cond_resched();			\
})
1396

1397 1398 1399
extern int __cond_resched_lock(spinlock_t *lock);

#define cond_resched_lock(lock) ({				\
1400
	___might_sleep(__FILE__, __LINE__, PREEMPT_LOCK_OFFSET);\
1401 1402 1403 1404 1405
	__cond_resched_lock(lock);				\
})

extern int __cond_resched_softirq(void);

1406
#define cond_resched_softirq() ({					\
1407
	___might_sleep(__FILE__, __LINE__, SOFTIRQ_DISABLE_OFFSET);	\
1408
	__cond_resched_softirq();					\
1409
})
L
Linus Torvalds 已提交
1410

1411 1412 1413 1414 1415 1416 1417 1418 1419
static inline void cond_resched_rcu(void)
{
#if defined(CONFIG_DEBUG_ATOMIC_SLEEP) || !defined(CONFIG_PREEMPT_RCU)
	rcu_read_unlock();
	cond_resched();
	rcu_read_lock();
#endif
}

L
Linus Torvalds 已提交
1420 1421
/*
 * Does a critical section need to be broken due to another
N
Nick Piggin 已提交
1422 1423
 * task waiting?: (technically does not depend on CONFIG_PREEMPT,
 * but a general need for low latency)
L
Linus Torvalds 已提交
1424
 */
N
Nick Piggin 已提交
1425
static inline int spin_needbreak(spinlock_t *lock)
L
Linus Torvalds 已提交
1426
{
N
Nick Piggin 已提交
1427 1428 1429
#ifdef CONFIG_PREEMPT
	return spin_is_contended(lock);
#else
L
Linus Torvalds 已提交
1430
	return 0;
N
Nick Piggin 已提交
1431
#endif
L
Linus Torvalds 已提交
1432 1433
}

1434 1435 1436 1437 1438
static __always_inline bool need_resched(void)
{
	return unlikely(tif_need_resched());
}

L
Linus Torvalds 已提交
1439 1440 1441 1442 1443 1444 1445
/*
 * Wrappers for p->thread_info->cpu access. No-op on UP.
 */
#ifdef CONFIG_SMP

static inline unsigned int task_cpu(const struct task_struct *p)
{
1446 1447 1448
#ifdef CONFIG_THREAD_INFO_IN_TASK
	return p->cpu;
#else
A
Al Viro 已提交
1449
	return task_thread_info(p)->cpu;
1450
#endif
L
Linus Torvalds 已提交
1451 1452
}

I
Ingo Molnar 已提交
1453
extern void set_task_cpu(struct task_struct *p, unsigned int cpu);
L
Linus Torvalds 已提交
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467

#else

static inline unsigned int task_cpu(const struct task_struct *p)
{
	return 0;
}

static inline void set_task_cpu(struct task_struct *p, unsigned int cpu)
{
}

#endif /* CONFIG_SMP */

1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479
/*
 * In order to reduce various lock holder preemption latencies provide an
 * interface to see if a vCPU is currently running or not.
 *
 * This allows us to terminate optimistic spin loops and block, analogous to
 * the native optimistic spin heuristic of testing if the lock owner task is
 * running or not.
 */
#ifndef vcpu_is_preempted
# define vcpu_is_preempted(cpu)	false
#endif

1480 1481
extern long sched_setaffinity(pid_t pid, const struct cpumask *new_mask);
extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
1482

D
Dave Hansen 已提交
1483 1484 1485 1486
#ifndef TASK_SIZE_OF
#define TASK_SIZE_OF(tsk)	TASK_SIZE
#endif

L
Linus Torvalds 已提交
1487
#endif