sched.h 87.4 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 7
#include <linux/sched/prio.h>

8 9 10 11 12

struct sched_param {
	int sched_priority;
};

L
Linus Torvalds 已提交
13 14 15 16 17 18 19 20
#include <asm/param.h>	/* for HZ */

#include <linux/capability.h>
#include <linux/threads.h>
#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/timex.h>
#include <linux/jiffies.h>
21
#include <linux/plist.h>
L
Linus Torvalds 已提交
22 23 24 25 26
#include <linux/rbtree.h>
#include <linux/thread_info.h>
#include <linux/cpumask.h>
#include <linux/errno.h>
#include <linux/nodemask.h>
27
#include <linux/mm_types.h>
28
#include <linux/preempt_mask.h>
L
Linus Torvalds 已提交
29 30 31

#include <asm/page.h>
#include <asm/ptrace.h>
32
#include <linux/cputime.h>
L
Linus Torvalds 已提交
33 34 35

#include <linux/smp.h>
#include <linux/sem.h>
36
#include <linux/shm.h>
L
Linus Torvalds 已提交
37 38 39 40 41 42
#include <linux/signal.h>
#include <linux/compiler.h>
#include <linux/completion.h>
#include <linux/pid.h>
#include <linux/percpu.h>
#include <linux/topology.h>
P
Peter Zijlstra 已提交
43
#include <linux/proportions.h>
L
Linus Torvalds 已提交
44
#include <linux/seccomp.h>
I
Ingo Molnar 已提交
45
#include <linux/rcupdate.h>
46
#include <linux/rculist.h>
I
Ingo Molnar 已提交
47
#include <linux/rtmutex.h>
L
Linus Torvalds 已提交
48

49 50 51 52 53
#include <linux/time.h>
#include <linux/param.h>
#include <linux/resource.h>
#include <linux/timer.h>
#include <linux/hrtimer.h>
54
#include <linux/task_io_accounting.h>
A
Arjan van de Ven 已提交
55
#include <linux/latencytop.h>
56
#include <linux/cred.h>
P
Peter Zijlstra 已提交
57
#include <linux/llist.h>
58
#include <linux/uidgid.h>
59
#include <linux/gfp.h>
60
#include <linux/magic.h>
61 62

#include <asm/processor.h>
H
H. J. Lu 已提交
63

64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104
#define SCHED_ATTR_SIZE_VER0	48	/* sizeof first published struct */

/*
 * Extended scheduling parameters data structure.
 *
 * This is needed because the original struct sched_param can not be
 * altered without introducing ABI issues with legacy applications
 * (e.g., in sched_getparam()).
 *
 * However, the possibility of specifying more than just a priority for
 * the tasks may be useful for a wide variety of application fields, e.g.,
 * multimedia, streaming, automation and control, and many others.
 *
 * This variant (sched_attr) is meant at describing a so-called
 * sporadic time-constrained task. In such model a task is specified by:
 *  - the activation period or minimum instance inter-arrival time;
 *  - the maximum (or average, depending on the actual scheduling
 *    discipline) computation time of all instances, a.k.a. runtime;
 *  - the deadline (relative to the actual activation time) of each
 *    instance.
 * Very briefly, a periodic (sporadic) task asks for the execution of
 * some specific computation --which is typically called an instance--
 * (at most) every period. Moreover, each instance typically lasts no more
 * than the runtime and must be completed by time instant t equal to
 * the instance activation time + the deadline.
 *
 * This is reflected by the actual fields of the sched_attr structure:
 *
 *  @size		size of the structure, for fwd/bwd compat.
 *
 *  @sched_policy	task's scheduling policy
 *  @sched_flags	for customizing the scheduler behaviour
 *  @sched_nice		task's nice value      (SCHED_NORMAL/BATCH)
 *  @sched_priority	task's static priority (SCHED_FIFO/RR)
 *  @sched_deadline	representative of the task's deadline
 *  @sched_runtime	representative of the task's runtime
 *  @sched_period	representative of the task's period
 *
 * Given this task model, there are a multiplicity of scheduling algorithms
 * and policies, that can be used to ensure all the tasks will make their
 * timing constraints.
105 106 107 108
 *
 * As of now, the SCHED_DEADLINE policy (sched_dl scheduling class) is the
 * only user of this new interface. More information about the algorithm
 * available in the scheduling class file or in Documentation/.
109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
 */
struct sched_attr {
	u32 size;

	u32 sched_policy;
	u64 sched_flags;

	/* SCHED_NORMAL, SCHED_BATCH */
	s32 sched_nice;

	/* SCHED_FIFO, SCHED_RR */
	u32 sched_priority;

	/* SCHED_DEADLINE */
	u64 sched_runtime;
	u64 sched_deadline;
	u64 sched_period;
};

L
Linus Torvalds 已提交
128
struct exec_domain;
129
struct futex_pi_state;
130
struct robust_list_head;
131
struct bio_list;
132
struct fs_struct;
133
struct perf_event_context;
134
struct blk_plug;
135
struct filename;
L
Linus Torvalds 已提交
136

D
Davidlohr Bueso 已提交
137 138 139 140
#define VMACACHE_BITS 2
#define VMACACHE_SIZE (1U << VMACACHE_BITS)
#define VMACACHE_MASK (VMACACHE_SIZE - 1)

L
Linus Torvalds 已提交
141 142 143 144 145 146 147 148 149 150 151
/*
 * These are the constant used to fake the fixed-point load-average
 * counting. Some notes:
 *  - 11 bit fractions expand to 22 bits by the multiplies: this gives
 *    a load-average precision of 10 bits integer + 11 bits fractional
 *  - if you want to count load-averages more often, you need more
 *    precision, or rounding will get you. With 2-second counting freq,
 *    the EXP_n values would be 1981, 2034 and 2043 if still using only
 *    11 bit fractions.
 */
extern unsigned long avenrun[];		/* Load averages */
152
extern void get_avenrun(unsigned long *loads, unsigned long offset, int shift);
L
Linus Torvalds 已提交
153 154 155

#define FSHIFT		11		/* nr of bits of precision */
#define FIXED_1		(1<<FSHIFT)	/* 1.0 as fixed-point */
156
#define LOAD_FREQ	(5*HZ+1)	/* 5 sec intervals */
L
Linus Torvalds 已提交
157 158 159 160 161 162 163 164 165 166 167 168 169 170
#define EXP_1		1884		/* 1/exp(5sec/1min) as fixed-point */
#define EXP_5		2014		/* 1/exp(5sec/5min) */
#define EXP_15		2037		/* 1/exp(5sec/15min) */

#define CALC_LOAD(load,exp,n) \
	load *= exp; \
	load += n*(FIXED_1-exp); \
	load >>= FSHIFT;

extern unsigned long total_forks;
extern int nr_threads;
DECLARE_PER_CPU(unsigned long, process_counts);
extern int nr_processes(void);
extern unsigned long nr_running(void);
171
extern bool single_task_running(void);
L
Linus Torvalds 已提交
172
extern unsigned long nr_iowait(void);
173
extern unsigned long nr_iowait_cpu(int cpu);
174
extern void get_iowait_load(unsigned long *nr_waiters, unsigned long *load);
175

176
extern void calc_global_load(unsigned long ticks);
177
extern void update_cpu_load_nohz(void);
L
Linus Torvalds 已提交
178

179 180
extern unsigned long get_parent_ip(unsigned long addr);

181 182
extern void dump_cpu_task(int cpu);

I
Ingo Molnar 已提交
183 184
struct seq_file;
struct cfs_rq;
185
struct task_group;
I
Ingo Molnar 已提交
186 187 188 189
#ifdef CONFIG_SCHED_DEBUG
extern void proc_sched_show_task(struct task_struct *p, struct seq_file *m);
extern void proc_sched_set_task(struct task_struct *p);
extern void
190
print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq);
I
Ingo Molnar 已提交
191
#endif
L
Linus Torvalds 已提交
192

193 194 195 196 197 198 199 200 201 202
/*
 * 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
203 204 205
#define TASK_RUNNING		0
#define TASK_INTERRUPTIBLE	1
#define TASK_UNINTERRUPTIBLE	2
M
Matthew Wilcox 已提交
206 207
#define __TASK_STOPPED		4
#define __TASK_TRACED		8
208
/* in tsk->exit_state */
209 210
#define EXIT_DEAD		16
#define EXIT_ZOMBIE		32
211
#define EXIT_TRACE		(EXIT_ZOMBIE | EXIT_DEAD)
212
/* in tsk->state again */
213
#define TASK_DEAD		64
M
Matthew Wilcox 已提交
214
#define TASK_WAKEKILL		128
P
Peter Zijlstra 已提交
215
#define TASK_WAKING		256
216 217
#define TASK_PARKED		512
#define TASK_STATE_MAX		1024
M
Matthew Wilcox 已提交
218

219
#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWP"
220

221 222
extern char ___assert_task_state[1 - 2*!!(
		sizeof(TASK_STATE_TO_CHAR_STR)-1 != ilog2(TASK_STATE_MAX)+1)];
M
Matthew Wilcox 已提交
223 224 225 226 227

/* Convenience macros for the sake of set_task_state */
#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
228

229 230
/* Convenience macros for the sake of wake_up */
#define TASK_NORMAL		(TASK_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE)
M
Matthew Wilcox 已提交
231
#define TASK_ALL		(TASK_NORMAL | __TASK_STOPPED | __TASK_TRACED)
232 233 234

/* get_task_state() */
#define TASK_REPORT		(TASK_RUNNING | TASK_INTERRUPTIBLE | \
M
Matthew Wilcox 已提交
235
				 TASK_UNINTERRUPTIBLE | __TASK_STOPPED | \
236
				 __TASK_TRACED | EXIT_ZOMBIE | EXIT_DEAD)
237

M
Matthew Wilcox 已提交
238 239
#define task_is_traced(task)	((task->state & __TASK_TRACED) != 0)
#define task_is_stopped(task)	((task->state & __TASK_STOPPED) != 0)
240
#define task_is_stopped_or_traced(task)	\
M
Matthew Wilcox 已提交
241
			((task->state & (__TASK_STOPPED | __TASK_TRACED)) != 0)
242
#define task_contributes_to_load(task)	\
243
				((task->state & TASK_UNINTERRUPTIBLE) != 0 && \
T
Tejun Heo 已提交
244
				 (task->flags & PF_FROZEN) == 0)
L
Linus Torvalds 已提交
245

P
Peter Zijlstra 已提交
246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
#ifdef CONFIG_DEBUG_ATOMIC_SLEEP

#define __set_task_state(tsk, state_value)			\
	do {							\
		(tsk)->task_state_change = _THIS_IP_;		\
		(tsk)->state = (state_value);			\
	} while (0)
#define set_task_state(tsk, state_value)			\
	do {							\
		(tsk)->task_state_change = _THIS_IP_;		\
		set_mb((tsk)->state, (state_value));		\
	} while (0)

/*
 * 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:
 *
 *	set_current_state(TASK_UNINTERRUPTIBLE);
 *	if (do_i_need_to_sleep())
 *		schedule();
 *
 * If the caller does not need such serialisation then use __set_current_state()
 */
#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_;		\
		set_mb(current->state, (state_value));		\
	} while (0)

#else

L
Linus Torvalds 已提交
283 284 285 286 287
#define __set_task_state(tsk, state_value)		\
	do { (tsk)->state = (state_value); } while (0)
#define set_task_state(tsk, state_value)		\
	set_mb((tsk)->state, (state_value))

288 289 290 291 292 293 294 295 296 297 298
/*
 * 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:
 *
 *	set_current_state(TASK_UNINTERRUPTIBLE);
 *	if (do_i_need_to_sleep())
 *		schedule();
 *
 * If the caller does not need such serialisation then use __set_current_state()
 */
P
Peter Zijlstra 已提交
299
#define __set_current_state(state_value)		\
L
Linus Torvalds 已提交
300
	do { current->state = (state_value); } while (0)
P
Peter Zijlstra 已提交
301
#define set_current_state(state_value)			\
L
Linus Torvalds 已提交
302 303
	set_mb(current->state, (state_value))

P
Peter Zijlstra 已提交
304 305
#endif

L
Linus Torvalds 已提交
306 307 308 309 310 311 312 313 314 315 316 317 318 319
/* Task command name length */
#define TASK_COMM_LEN 16

#include <linux/spinlock.h>

/*
 * This serializes "schedule()" and also protects
 * the run-queue from deletions/modifications (but
 * _adding_ to the beginning of the run-queue has
 * a separate lock).
 */
extern rwlock_t tasklist_lock;
extern spinlock_t mmlist_lock;

320
struct task_struct;
L
Linus Torvalds 已提交
321

322 323 324 325
#ifdef CONFIG_PROVE_RCU
extern int lockdep_tasklist_lock_is_held(void);
#endif /* #ifdef CONFIG_PROVE_RCU */

L
Linus Torvalds 已提交
326 327
extern void sched_init(void);
extern void sched_init_smp(void);
328
extern asmlinkage void schedule_tail(struct task_struct *prev);
329
extern void init_idle(struct task_struct *idle, int cpu);
I
Ingo Molnar 已提交
330
extern void init_idle_bootup_task(struct task_struct *idle);
L
Linus Torvalds 已提交
331

332
extern int runqueue_is_locked(int cpu);
I
Ingo Molnar 已提交
333

334
#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON)
335
extern void nohz_balance_enter_idle(int cpu);
336
extern void set_cpu_sd_state_idle(void);
337
extern int get_nohz_timer_target(int pinned);
338
#else
339
static inline void nohz_balance_enter_idle(int cpu) { }
340
static inline void set_cpu_sd_state_idle(void) { }
341 342 343 344
static inline int get_nohz_timer_target(int pinned)
{
	return smp_processor_id();
}
345
#endif
L
Linus Torvalds 已提交
346

I
Ingo Molnar 已提交
347
/*
I
Ingo Molnar 已提交
348
 * Only dump TASK_* tasks. (0 for all tasks)
I
Ingo Molnar 已提交
349 350 351 352 353
 */
extern void show_state_filter(unsigned long state_filter);

static inline void show_state(void)
{
I
Ingo Molnar 已提交
354
	show_state_filter(0);
I
Ingo Molnar 已提交
355 356
}

L
Linus Torvalds 已提交
357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373
extern void show_regs(struct pt_regs *);

/*
 * TASK is a pointer to the task whose backtrace we want to see (or NULL for current
 * task), SP is the stack pointer of the first frame that should be shown in the back
 * trace (or NULL if the entire call-chain of the task should be shown).
 */
extern void show_stack(struct task_struct *task, unsigned long *sp);

void io_schedule(void);
long io_schedule_timeout(long timeout);

extern void cpu_init (void);
extern void trap_init(void);
extern void update_process_times(int user);
extern void scheduler_tick(void);

374 375
extern void sched_show_task(struct task_struct *p);

376
#ifdef CONFIG_LOCKUP_DETECTOR
I
Ingo Molnar 已提交
377
extern void touch_softlockup_watchdog(void);
378
extern void touch_softlockup_watchdog_sync(void);
379
extern void touch_all_softlockup_watchdogs(void);
380 381 382
extern int proc_dowatchdog_thresh(struct ctl_table *table, int write,
				  void __user *buffer,
				  size_t *lenp, loff_t *ppos);
I
Ingo Molnar 已提交
383
extern unsigned int  softlockup_panic;
384
void lockup_detector_init(void);
I
Ingo Molnar 已提交
385 386 387 388
#else
static inline void touch_softlockup_watchdog(void)
{
}
389 390 391
static inline void touch_softlockup_watchdog_sync(void)
{
}
392 393 394
static inline void touch_all_softlockup_watchdogs(void)
{
}
395 396 397
static inline void lockup_detector_init(void)
{
}
I
Ingo Molnar 已提交
398 399
#endif

400 401 402 403 404 405 406 407
#ifdef CONFIG_DETECT_HUNG_TASK
void reset_hung_task_detector(void);
#else
static inline void reset_hung_task_detector(void)
{
}
#endif

L
Linus Torvalds 已提交
408 409
/* Attach to any functions which should be ignored in wchan output. */
#define __sched		__attribute__((__section__(".sched.text")))
410 411 412 413

/* Linker adds these: start and end of __sched functions */
extern char __sched_text_start[], __sched_text_end[];

L
Linus Torvalds 已提交
414 415 416 417
/* Is this address in the __sched functions? */
extern int in_sched_functions(unsigned long addr);

#define	MAX_SCHEDULE_TIMEOUT	LONG_MAX
418
extern signed long schedule_timeout(signed long timeout);
419
extern signed long schedule_timeout_interruptible(signed long timeout);
M
Matthew Wilcox 已提交
420
extern signed long schedule_timeout_killable(signed long timeout);
421
extern signed long schedule_timeout_uninterruptible(signed long timeout);
L
Linus Torvalds 已提交
422
asmlinkage void schedule(void);
423
extern void schedule_preempt_disabled(void);
L
Linus Torvalds 已提交
424

S
Serge E. Hallyn 已提交
425
struct nsproxy;
426
struct user_namespace;
L
Linus Torvalds 已提交
427

428 429
#ifdef CONFIG_MMU
extern void arch_pick_mmap_layout(struct mm_struct *mm);
L
Linus Torvalds 已提交
430 431 432 433 434 435 436
extern unsigned long
arch_get_unmapped_area(struct file *, unsigned long, unsigned long,
		       unsigned long, unsigned long);
extern unsigned long
arch_get_unmapped_area_topdown(struct file *filp, unsigned long addr,
			  unsigned long len, unsigned long pgoff,
			  unsigned long flags);
437 438 439
#else
static inline void arch_pick_mmap_layout(struct mm_struct *mm) {}
#endif
L
Linus Torvalds 已提交
440

441 442 443 444
#define SUID_DUMP_DISABLE	0	/* No setuid dumping */
#define SUID_DUMP_USER		1	/* Dump as user of process */
#define SUID_DUMP_ROOT		2	/* Dump as root */

445
/* mm flags */
H
Hugh Dickins 已提交
446

447
/* for SUID_DUMP_* above */
448
#define MMF_DUMPABLE_BITS 2
H
Hugh Dickins 已提交
449
#define MMF_DUMPABLE_MASK ((1 << MMF_DUMPABLE_BITS) - 1)
450

451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467
extern void set_dumpable(struct mm_struct *mm, int value);
/*
 * This returns the actual value of the suid_dumpable flag. For things
 * that are using this for checking for privilege transitions, it must
 * test against SUID_DUMP_USER rather than treating it as a boolean
 * value.
 */
static inline int __get_dumpable(unsigned long mm_flags)
{
	return mm_flags & MMF_DUMPABLE_MASK;
}

static inline int get_dumpable(struct mm_struct *mm)
{
	return __get_dumpable(mm->flags);
}

468 469 470 471 472
/* coredump filter bits */
#define MMF_DUMP_ANON_PRIVATE	2
#define MMF_DUMP_ANON_SHARED	3
#define MMF_DUMP_MAPPED_PRIVATE	4
#define MMF_DUMP_MAPPED_SHARED	5
R
Roland McGrath 已提交
473
#define MMF_DUMP_ELF_HEADERS	6
474 475
#define MMF_DUMP_HUGETLB_PRIVATE 7
#define MMF_DUMP_HUGETLB_SHARED  8
H
Hugh Dickins 已提交
476

477
#define MMF_DUMP_FILTER_SHIFT	MMF_DUMPABLE_BITS
478
#define MMF_DUMP_FILTER_BITS	7
479 480 481
#define MMF_DUMP_FILTER_MASK \
	(((1 << MMF_DUMP_FILTER_BITS) - 1) << MMF_DUMP_FILTER_SHIFT)
#define MMF_DUMP_FILTER_DEFAULT \
482
	((1 << MMF_DUMP_ANON_PRIVATE) |	(1 << MMF_DUMP_ANON_SHARED) |\
483 484 485 486 487 488 489
	 (1 << MMF_DUMP_HUGETLB_PRIVATE) | MMF_DUMP_MASK_DEFAULT_ELF)

#ifdef CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS
# define MMF_DUMP_MASK_DEFAULT_ELF	(1 << MMF_DUMP_ELF_HEADERS)
#else
# define MMF_DUMP_MASK_DEFAULT_ELF	0
#endif
H
Hugh Dickins 已提交
490 491
					/* leave room for more dump flags */
#define MMF_VM_MERGEABLE	16	/* KSM may merge identical pages */
A
Andrea Arcangeli 已提交
492
#define MMF_VM_HUGEPAGE		17	/* set when VM_HUGEPAGE is set on vma */
493
#define MMF_EXE_FILE_CHANGED	18	/* see prctl_set_mm_exe_file() */
H
Hugh Dickins 已提交
494

495 496
#define MMF_HAS_UPROBES		19	/* has uprobes */
#define MMF_RECALC_UPROBES	20	/* MMF_HAS_UPROBES can be wrong */
497

H
Hugh Dickins 已提交
498
#define MMF_INIT_MASK		(MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK)
499

L
Linus Torvalds 已提交
500 501 502 503
struct sighand_struct {
	atomic_t		count;
	struct k_sigaction	action[_NSIG];
	spinlock_t		siglock;
D
Davide Libenzi 已提交
504
	wait_queue_head_t	signalfd_wqh;
L
Linus Torvalds 已提交
505 506
};

507
struct pacct_struct {
508 509
	int			ac_flag;
	long			ac_exitcode;
510
	unsigned long		ac_mem;
511 512
	cputime_t		ac_utime, ac_stime;
	unsigned long		ac_minflt, ac_majflt;
513 514
};

515 516 517
struct cpu_itimer {
	cputime_t expires;
	cputime_t incr;
518 519
	u32 error;
	u32 incr_error;
520 521
};

522 523 524 525 526 527 528 529 530 531 532 533
/**
 * struct cputime - snaphsot of system and user cputime
 * @utime: time spent in user mode
 * @stime: time spent in system mode
 *
 * Gathers a generic snapshot of user and system time.
 */
struct cputime {
	cputime_t utime;
	cputime_t stime;
};

534 535 536 537 538
/**
 * struct task_cputime - collected CPU time counts
 * @utime:		time spent in user mode, in &cputime_t units
 * @stime:		time spent in kernel mode, in &cputime_t units
 * @sum_exec_runtime:	total time spent on the CPU, in nanoseconds
539
 *
540 541 542 543 544
 * This is an extension of struct cputime that includes the total runtime
 * spent by the task from the scheduler point of view.
 *
 * As a result, this structure groups together three kinds of CPU time
 * that are tracked for threads and thread groups.  Most things considering
545 546 547 548 549 550 551 552 553 554 555 556 557
 * CPU time want to group these counts together and treat all three
 * of them in parallel.
 */
struct task_cputime {
	cputime_t utime;
	cputime_t stime;
	unsigned long long sum_exec_runtime;
};
/* Alternate field names when used to cache expirations. */
#define prof_exp	stime
#define virt_exp	utime
#define sched_exp	sum_exec_runtime

558 559
#define INIT_CPUTIME	\
	(struct task_cputime) {					\
560 561
		.utime = 0,					\
		.stime = 0,					\
562 563 564
		.sum_exec_runtime = 0,				\
	}

565 566 567 568 569 570
#ifdef CONFIG_PREEMPT_COUNT
#define PREEMPT_DISABLED	(1 + PREEMPT_ENABLED)
#else
#define PREEMPT_DISABLED	PREEMPT_ENABLED
#endif

P
Peter Zijlstra 已提交
571 572 573
/*
 * Disable preemption until the scheduler is running.
 * Reset by start_kernel()->sched_init()->init_idle().
P
Peter Zijlstra 已提交
574 575 576
 *
 * We include PREEMPT_ACTIVE to avoid cond_resched() from working
 * before the scheduler is active -- see should_resched().
P
Peter Zijlstra 已提交
577
 */
578
#define INIT_PREEMPT_COUNT	(PREEMPT_DISABLED + PREEMPT_ACTIVE)
P
Peter Zijlstra 已提交
579

580
/**
581 582 583 584 585
 * struct thread_group_cputimer - thread group interval timer counts
 * @cputime:		thread group interval timers.
 * @running:		non-zero when there are timers running and
 * 			@cputime receives updates.
 * @lock:		lock for fields in this struct.
586 587
 *
 * This structure contains the version of task_cputime, above, that is
588
 * used for thread group CPU timer calculations.
589
 */
590 591 592
struct thread_group_cputimer {
	struct task_cputime cputime;
	int running;
593
	raw_spinlock_t lock;
594 595
};

596
#include <linux/rwsem.h>
597 598
struct autogroup;

L
Linus Torvalds 已提交
599
/*
600
 * NOTE! "signal_struct" does not have its own
L
Linus Torvalds 已提交
601 602 603 604 605 606
 * locking, because a shared signal_struct always
 * implies a shared sighand_struct, so locking
 * sighand_struct is always a proper superset of
 * the locking of signal_struct.
 */
struct signal_struct {
607
	atomic_t		sigcnt;
L
Linus Torvalds 已提交
608
	atomic_t		live;
609
	int			nr_threads;
610
	struct list_head	thread_head;
L
Linus Torvalds 已提交
611 612 613 614

	wait_queue_head_t	wait_chldexit;	/* for wait4() */

	/* current thread group signal load-balancing target: */
615
	struct task_struct	*curr_target;
L
Linus Torvalds 已提交
616 617 618 619 620 621 622 623 624 625 626 627

	/* shared signal handling: */
	struct sigpending	shared_pending;

	/* thread group exit support */
	int			group_exit_code;
	/* overloaded:
	 * - notify group_exit_task when ->count is equal to notify_count
	 * - everyone except group_exit_task is stopped during signal delivery
	 *   of fatal signals, group_exit_task processes the signal.
	 */
	int			notify_count;
628
	struct task_struct	*group_exit_task;
L
Linus Torvalds 已提交
629 630 631 632 633

	/* thread group stop support, overloads group_exit_code too */
	int			group_stop_count;
	unsigned int		flags; /* see SIGNAL_* flags below */

634 635 636 637 638 639 640 641 642 643 644 645
	/*
	 * PR_SET_CHILD_SUBREAPER marks a process, like a service
	 * manager, to re-parent orphan (double-forking) child processes
	 * to this process instead of 'init'. The service manager is
	 * able to receive SIGCHLD signals and is able to investigate
	 * the process until it calls wait(). All children of this
	 * process will inherit a flag if they should look for a
	 * child_subreaper process at exit.
	 */
	unsigned int		is_child_subreaper:1;
	unsigned int		has_child_subreaper:1;

L
Linus Torvalds 已提交
646
	/* POSIX.1b Interval Timers */
647 648
	int			posix_timer_id;
	struct list_head	posix_timers;
L
Linus Torvalds 已提交
649 650

	/* ITIMER_REAL timer for the process */
651
	struct hrtimer real_timer;
652
	struct pid *leader_pid;
653
	ktime_t it_real_incr;
L
Linus Torvalds 已提交
654

655 656 657 658 659 660
	/*
	 * ITIMER_PROF and ITIMER_VIRTUAL timers for the process, we use
	 * CPUCLOCK_PROF and CPUCLOCK_VIRT for indexing array as these
	 * values are defined to 0 and 1 respectively
	 */
	struct cpu_itimer it[2];
L
Linus Torvalds 已提交
661

662
	/*
663 664
	 * Thread group totals for process CPU timers.
	 * See thread_group_cputimer(), et al, for details.
665
	 */
666
	struct thread_group_cputimer cputimer;
667 668 669 670 671 672

	/* Earliest-expiration cache. */
	struct task_cputime cputime_expires;

	struct list_head cpu_timers[3];

673
	struct pid *tty_old_pgrp;
674

L
Linus Torvalds 已提交
675 676 677 678 679
	/* boolean value for session group leader */
	int leader;

	struct tty_struct *tty; /* NULL if no tty */

680 681 682
#ifdef CONFIG_SCHED_AUTOGROUP
	struct autogroup *autogroup;
#endif
L
Linus Torvalds 已提交
683 684 685 686 687 688
	/*
	 * Cumulative resource counters for dead threads in the group,
	 * and for reaped dead child processes forked by this group.
	 * Live threads maintain their own counters and add to these
	 * in __exit_signal, except for the group leader.
	 */
689
	seqlock_t stats_lock;
690
	cputime_t utime, stime, cutime, cstime;
691 692
	cputime_t gtime;
	cputime_t cgtime;
693
#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
694
	struct cputime prev_cputime;
695
#endif
L
Linus Torvalds 已提交
696 697
	unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw;
	unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt;
698
	unsigned long inblock, oublock, cinblock, coublock;
J
Jiri Pirko 已提交
699
	unsigned long maxrss, cmaxrss;
700
	struct task_io_accounting ioac;
L
Linus Torvalds 已提交
701

702 703 704 705 706 707 708 709
	/*
	 * Cumulative ns of schedule CPU time fo dead threads in the
	 * group, not including a zombie group leader, (This only differs
	 * from jiffies_to_ns(utime + stime) if sched_clock uses something
	 * other than jiffies.)
	 */
	unsigned long long sum_sched_runtime;

L
Linus Torvalds 已提交
710 711 712 713 714 715 716 717 718 719 720
	/*
	 * We don't bother to synchronize most readers of this at all,
	 * because there is no reader checking a limit that actually needs
	 * to get both rlim_cur and rlim_max atomically, and either one
	 * alone is a single word that can safely be read normally.
	 * getrlimit/setrlimit use task_lock(current->group_leader) to
	 * protect this instead of the siglock, because they really
	 * have no need to disable irqs.
	 */
	struct rlimit rlim[RLIM_NLIMITS];

721 722 723
#ifdef CONFIG_BSD_PROCESS_ACCT
	struct pacct_struct pacct;	/* per-process accounting information */
#endif
724 725 726
#ifdef CONFIG_TASKSTATS
	struct taskstats *stats;
#endif
M
Miloslav Trmac 已提交
727 728
#ifdef CONFIG_AUDIT
	unsigned audit_tty;
729
	unsigned audit_tty_log_passwd;
M
Miloslav Trmac 已提交
730 731
	struct tty_audit_buf *tty_audit_buf;
#endif
732 733
#ifdef CONFIG_CGROUPS
	/*
734 735 736 737 738 739 740
	 * group_rwsem prevents new tasks from entering the threadgroup and
	 * member tasks from exiting,a more specifically, setting of
	 * PF_EXITING.  fork and exit paths are protected with this rwsem
	 * using threadgroup_change_begin/end().  Users which require
	 * threadgroup to remain stable should use threadgroup_[un]lock()
	 * which also takes care of exec path.  Currently, cgroup is the
	 * only user.
741
	 */
742
	struct rw_semaphore group_rwsem;
743
#endif
744

745
	oom_flags_t oom_flags;
746 747 748
	short oom_score_adj;		/* OOM kill score adjustment */
	short oom_score_adj_min;	/* OOM kill score adjustment min value.
					 * Only settable by CAP_SYS_RESOURCE. */
749 750 751 752

	struct mutex cred_guard_mutex;	/* guard against foreign influences on
					 * credential calculations
					 * (notably. ptrace) */
L
Linus Torvalds 已提交
753 754 755 756 757 758
};

/*
 * Bits in flags field of signal_struct.
 */
#define SIGNAL_STOP_STOPPED	0x00000001 /* job control stop in effect */
759 760
#define SIGNAL_STOP_CONTINUED	0x00000002 /* SIGCONT since WCONTINUED reap */
#define SIGNAL_GROUP_EXIT	0x00000004 /* group exit in progress */
761
#define SIGNAL_GROUP_COREDUMP	0x00000008 /* coredump in progress */
762 763 764 765 766 767
/*
 * Pending notifications to parent.
 */
#define SIGNAL_CLD_STOPPED	0x00000010
#define SIGNAL_CLD_CONTINUED	0x00000020
#define SIGNAL_CLD_MASK		(SIGNAL_CLD_STOPPED|SIGNAL_CLD_CONTINUED)
L
Linus Torvalds 已提交
768

769 770
#define SIGNAL_UNKILLABLE	0x00000040 /* for init: ignore fatal signals */

771 772 773 774 775 776 777
/* If true, all threads except ->group_exit_task have pending SIGKILL */
static inline int signal_group_exit(const struct signal_struct *sig)
{
	return	(sig->flags & SIGNAL_GROUP_EXIT) ||
		(sig->group_exit_task != NULL);
}

L
Linus Torvalds 已提交
778 779 780 781 782 783 784
/*
 * Some day this will be a full-fledged user tracking system..
 */
struct user_struct {
	atomic_t __count;	/* reference count */
	atomic_t processes;	/* How many processes does this user have? */
	atomic_t sigpending;	/* How many pending signals does this user have? */
785
#ifdef CONFIG_INOTIFY_USER
R
Robert Love 已提交
786 787 788
	atomic_t inotify_watches; /* How many inotify watches does this user have? */
	atomic_t inotify_devs;	/* How many inotify devs does this user have opened? */
#endif
789 790 791
#ifdef CONFIG_FANOTIFY
	atomic_t fanotify_listeners;
#endif
792
#ifdef CONFIG_EPOLL
793
	atomic_long_t epoll_watches; /* The number of file descriptors currently watched */
794
#endif
A
Alexey Dobriyan 已提交
795
#ifdef CONFIG_POSIX_MQUEUE
L
Linus Torvalds 已提交
796 797
	/* protected by mq_lock	*/
	unsigned long mq_bytes;	/* How many bytes can be allocated to mqueue? */
A
Alexey Dobriyan 已提交
798
#endif
L
Linus Torvalds 已提交
799 800 801 802 803 804 805 806
	unsigned long locked_shm; /* How many pages of mlocked shm ? */

#ifdef CONFIG_KEYS
	struct key *uid_keyring;	/* UID specific keyring */
	struct key *session_keyring;	/* UID's default session keyring */
#endif

	/* Hash table maintenance information */
P
Pavel Emelyanov 已提交
807
	struct hlist_node uidhash_node;
808
	kuid_t uid;
809

810
#ifdef CONFIG_PERF_EVENTS
811 812
	atomic_long_t locked_vm;
#endif
L
Linus Torvalds 已提交
813 814
};

815
extern int uids_sysfs_init(void);
816

817
extern struct user_struct *find_user(kuid_t);
L
Linus Torvalds 已提交
818 819 820 821

extern struct user_struct root_user;
#define INIT_USER (&root_user)

822

L
Linus Torvalds 已提交
823 824 825
struct backing_dev_info;
struct reclaim_state;

826
#if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT)
L
Linus Torvalds 已提交
827 828
struct sched_info {
	/* cumulative counters */
829
	unsigned long pcount;	      /* # of times run on this cpu */
830
	unsigned long long run_delay; /* time spent waiting on a runqueue */
L
Linus Torvalds 已提交
831 832

	/* timestamps */
833 834
	unsigned long long last_arrival,/* when we last ran on a cpu */
			   last_queued;	/* when we were last queued to run */
L
Linus Torvalds 已提交
835
};
836
#endif /* defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) */
L
Linus Torvalds 已提交
837

838 839 840 841 842 843 844 845 846 847 848 849 850 851
#ifdef CONFIG_TASK_DELAY_ACCT
struct task_delay_info {
	spinlock_t	lock;
	unsigned int	flags;	/* Private per-task flags */

	/* For each stat XXX, add following, aligned appropriately
	 *
	 * struct timespec XXX_start, XXX_end;
	 * u64 XXX_delay;
	 * u32 XXX_count;
	 *
	 * Atomicity of updates to XXX_delay, XXX_count protected by
	 * single lock above (split into XXX_lock if contention is an issue).
	 */
852 853 854 855 856 857

	/*
	 * XXX_count is incremented on every XXX operation, the delay
	 * associated with the operation is added to XXX_delay.
	 * XXX_delay contains the accumulated delay time in nanoseconds.
	 */
858
	u64 blkio_start;	/* Shared by blkio, swapin */
859 860 861 862 863 864
	u64 blkio_delay;	/* wait for sync block io completion */
	u64 swapin_delay;	/* wait for swapin block io completion */
	u32 blkio_count;	/* total count of the number of sync block */
				/* io operations performed */
	u32 swapin_count;	/* total count of the number of swapin block */
				/* io operations performed */
865

866
	u64 freepages_start;
867 868
	u64 freepages_delay;	/* wait for memory reclaim */
	u32 freepages_count;	/* total count of memory reclaim */
869
};
870 871 872 873 874 875 876 877 878 879 880
#endif	/* CONFIG_TASK_DELAY_ACCT */

static inline int sched_info_on(void)
{
#ifdef CONFIG_SCHEDSTATS
	return 1;
#elif defined(CONFIG_TASK_DELAY_ACCT)
	extern int delayacct_on;
	return delayacct_on;
#else
	return 0;
881
#endif
882
}
883

I
Ingo Molnar 已提交
884 885 886 887 888
enum cpu_idle_type {
	CPU_IDLE,
	CPU_NOT_IDLE,
	CPU_NEWLY_IDLE,
	CPU_MAX_IDLE_TYPES
L
Linus Torvalds 已提交
889 890
};

891
/*
892
 * Increase resolution of cpu_capacity calculations
893
 */
894 895
#define SCHED_CAPACITY_SHIFT	10
#define SCHED_CAPACITY_SCALE	(1L << SCHED_CAPACITY_SHIFT)
L
Linus Torvalds 已提交
896

897 898 899
/*
 * sched-domains (multiprocessor balancing) declarations:
 */
900
#ifdef CONFIG_SMP
P
Peter Zijlstra 已提交
901 902 903 904
#define SD_LOAD_BALANCE		0x0001	/* Do load balancing on this domain. */
#define SD_BALANCE_NEWIDLE	0x0002	/* Balance when about to become idle */
#define SD_BALANCE_EXEC		0x0004	/* Balance on exec */
#define SD_BALANCE_FORK		0x0008	/* Balance on fork, clone */
905
#define SD_BALANCE_WAKE		0x0010  /* Balance on wakeup */
P
Peter Zijlstra 已提交
906
#define SD_WAKE_AFFINE		0x0020	/* Wake task to waking CPU */
907
#define SD_SHARE_CPUCAPACITY	0x0080	/* Domain members share cpu power */
908
#define SD_SHARE_POWERDOMAIN	0x0100	/* Domain members share power domain */
P
Peter Zijlstra 已提交
909 910
#define SD_SHARE_PKG_RESOURCES	0x0200	/* Domain members share cpu pkg resources */
#define SD_SERIALIZE		0x0400	/* Only a single load balancing instance */
911
#define SD_ASYM_PACKING		0x0800  /* Place busy groups earlier in the domain */
P
Peter Zijlstra 已提交
912
#define SD_PREFER_SIBLING	0x1000	/* Prefer to place tasks in a sibling domain */
913
#define SD_OVERLAP		0x2000	/* sched_domains of this level overlap */
914
#define SD_NUMA			0x4000	/* cross-node balancing */
915

916
#ifdef CONFIG_SCHED_SMT
G
Guenter Roeck 已提交
917
static inline int cpu_smt_flags(void)
918
{
919
	return SD_SHARE_CPUCAPACITY | SD_SHARE_PKG_RESOURCES;
920 921 922 923
}
#endif

#ifdef CONFIG_SCHED_MC
G
Guenter Roeck 已提交
924
static inline int cpu_core_flags(void)
925 926 927 928 929 930
{
	return SD_SHARE_PKG_RESOURCES;
}
#endif

#ifdef CONFIG_NUMA
G
Guenter Roeck 已提交
931
static inline int cpu_numa_flags(void)
932 933 934 935
{
	return SD_NUMA;
}
#endif
936

937 938 939 940 941 942 943 944
struct sched_domain_attr {
	int relax_domain_level;
};

#define SD_ATTR_INIT	(struct sched_domain_attr) {	\
	.relax_domain_level = -1,			\
}

945 946
extern int sched_domain_level_max;

947 948
struct sched_group;

L
Linus Torvalds 已提交
949 950 951
struct sched_domain {
	/* These fields must be setup */
	struct sched_domain *parent;	/* top domain must be null terminated */
952
	struct sched_domain *child;	/* bottom domain must be null terminated */
L
Linus Torvalds 已提交
953 954 955 956 957 958
	struct sched_group *groups;	/* the balancing groups of the domain */
	unsigned long min_interval;	/* Minimum balance interval ms */
	unsigned long max_interval;	/* Maximum balance interval ms */
	unsigned int busy_factor;	/* less balancing by factor if busy */
	unsigned int imbalance_pct;	/* No balance until over watermark */
	unsigned int cache_nice_tries;	/* Leave cache hot tasks for # tries */
N
Nick Piggin 已提交
959 960 961 962
	unsigned int busy_idx;
	unsigned int idle_idx;
	unsigned int newidle_idx;
	unsigned int wake_idx;
N
Nick Piggin 已提交
963
	unsigned int forkexec_idx;
P
Peter Zijlstra 已提交
964
	unsigned int smt_gain;
V
Vincent Guittot 已提交
965 966

	int nohz_idle;			/* NOHZ IDLE status */
L
Linus Torvalds 已提交
967
	int flags;			/* See SD_* */
968
	int level;
L
Linus Torvalds 已提交
969 970 971 972 973 974

	/* Runtime fields. */
	unsigned long last_balance;	/* init to jiffies. units in jiffies */
	unsigned int balance_interval;	/* initialise to 1. units in ms. */
	unsigned int nr_balance_failed; /* initialise to 0 */

975
	/* idle_balance() stats */
976
	u64 max_newidle_lb_cost;
977
	unsigned long next_decay_max_lb_cost;
P
Peter Zijlstra 已提交
978

L
Linus Torvalds 已提交
979 980
#ifdef CONFIG_SCHEDSTATS
	/* load_balance() stats */
981 982 983 984 985 986 987 988
	unsigned int lb_count[CPU_MAX_IDLE_TYPES];
	unsigned int lb_failed[CPU_MAX_IDLE_TYPES];
	unsigned int lb_balanced[CPU_MAX_IDLE_TYPES];
	unsigned int lb_imbalance[CPU_MAX_IDLE_TYPES];
	unsigned int lb_gained[CPU_MAX_IDLE_TYPES];
	unsigned int lb_hot_gained[CPU_MAX_IDLE_TYPES];
	unsigned int lb_nobusyg[CPU_MAX_IDLE_TYPES];
	unsigned int lb_nobusyq[CPU_MAX_IDLE_TYPES];
L
Linus Torvalds 已提交
989 990

	/* Active load balancing */
991 992 993
	unsigned int alb_count;
	unsigned int alb_failed;
	unsigned int alb_pushed;
L
Linus Torvalds 已提交
994

995
	/* SD_BALANCE_EXEC stats */
996 997 998
	unsigned int sbe_count;
	unsigned int sbe_balanced;
	unsigned int sbe_pushed;
L
Linus Torvalds 已提交
999

1000
	/* SD_BALANCE_FORK stats */
1001 1002 1003
	unsigned int sbf_count;
	unsigned int sbf_balanced;
	unsigned int sbf_pushed;
1004

L
Linus Torvalds 已提交
1005
	/* try_to_wake_up() stats */
1006 1007 1008
	unsigned int ttwu_wake_remote;
	unsigned int ttwu_move_affine;
	unsigned int ttwu_move_balance;
L
Linus Torvalds 已提交
1009
#endif
1010 1011 1012
#ifdef CONFIG_SCHED_DEBUG
	char *name;
#endif
1013 1014 1015 1016
	union {
		void *private;		/* used during construction */
		struct rcu_head rcu;	/* used during destruction */
	};
1017

1018
	unsigned int span_weight;
1019 1020 1021 1022 1023 1024 1025 1026
	/*
	 * Span of all CPUs in this domain.
	 *
	 * NOTE: this field is variable length. (Allocated dynamically
	 * by attaching extra space to the end of the structure,
	 * depending on how many CPUs the kernel has booted up with)
	 */
	unsigned long span[0];
L
Linus Torvalds 已提交
1027 1028
};

1029 1030
static inline struct cpumask *sched_domain_span(struct sched_domain *sd)
{
1031
	return to_cpumask(sd->span);
1032 1033
}

1034
extern void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[],
1035
				    struct sched_domain_attr *dattr_new);
P
Paul Jackson 已提交
1036

1037 1038 1039 1040
/* Allocate an array of sched domains, for partition_sched_domains(). */
cpumask_var_t *alloc_sched_domains(unsigned int ndoms);
void free_sched_domains(cpumask_var_t doms[], unsigned int ndoms);

1041 1042
bool cpus_share_cache(int this_cpu, int that_cpu);

1043
typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
G
Guenter Roeck 已提交
1044
typedef int (*sched_domain_flags_f)(void);
1045 1046 1047 1048 1049 1050

#define SDTL_OVERLAP	0x01

struct sd_data {
	struct sched_domain **__percpu sd;
	struct sched_group **__percpu sg;
1051
	struct sched_group_capacity **__percpu sgc;
1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067
};

struct sched_domain_topology_level {
	sched_domain_mask_f mask;
	sched_domain_flags_f sd_flags;
	int		    flags;
	int		    numa_level;
	struct sd_data      data;
#ifdef CONFIG_SCHED_DEBUG
	char                *name;
#endif
};

extern struct sched_domain_topology_level *sched_domain_topology;

extern void set_sched_topology(struct sched_domain_topology_level *tl);
1068
extern void wake_up_if_idle(int cpu);
1069 1070 1071 1072 1073 1074 1075

#ifdef CONFIG_SCHED_DEBUG
# define SD_INIT_NAME(type)		.name = #type
#else
# define SD_INIT_NAME(type)
#endif

1076
#else /* CONFIG_SMP */
L
Linus Torvalds 已提交
1077

1078
struct sched_domain_attr;
1079

1080
static inline void
1081
partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[],
1082 1083
			struct sched_domain_attr *dattr_new)
{
1084
}
1085 1086 1087 1088 1089 1090

static inline bool cpus_share_cache(int this_cpu, int that_cpu)
{
	return true;
}

1091
#endif	/* !CONFIG_SMP */
L
Linus Torvalds 已提交
1092

1093

L
Linus Torvalds 已提交
1094 1095 1096
struct io_context;			/* See blkdev.h */


1097
#ifdef ARCH_HAS_PREFETCH_SWITCH_STACK
1098
extern void prefetch_stack(struct task_struct *t);
1099 1100 1101
#else
static inline void prefetch_stack(struct task_struct *t) { }
#endif
L
Linus Torvalds 已提交
1102 1103 1104

struct audit_context;		/* See audit.c */
struct mempolicy;
1105
struct pipe_inode_info;
1106
struct uts_namespace;
L
Linus Torvalds 已提交
1107

I
Ingo Molnar 已提交
1108
struct load_weight {
1109 1110
	unsigned long weight;
	u32 inv_weight;
I
Ingo Molnar 已提交
1111 1112
};

1113 1114 1115
struct sched_avg {
	/*
	 * These sums represent an infinite geometric series and so are bound
1116
	 * above by 1024/(1-y).  Thus we only need a u32 to store them for all
1117 1118 1119 1120
	 * choices of y < 1-2^(-32)*1024.
	 */
	u32 runnable_avg_sum, runnable_avg_period;
	u64 last_runnable_update;
1121
	s64 decay_count;
1122
	unsigned long load_avg_contrib;
1123 1124
};

1125
#ifdef CONFIG_SCHEDSTATS
1126
struct sched_statistics {
I
Ingo Molnar 已提交
1127
	u64			wait_start;
1128
	u64			wait_max;
1129 1130
	u64			wait_count;
	u64			wait_sum;
1131 1132
	u64			iowait_count;
	u64			iowait_sum;
1133

I
Ingo Molnar 已提交
1134 1135
	u64			sleep_start;
	u64			sleep_max;
1136 1137 1138
	s64			sum_sleep_runtime;

	u64			block_start;
I
Ingo Molnar 已提交
1139 1140
	u64			block_max;
	u64			exec_max;
I
Ingo Molnar 已提交
1141
	u64			slice_max;
1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157

	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;
1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175
};
#endif

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;

#ifdef CONFIG_SCHEDSTATS
	struct sched_statistics statistics;
1176 1177
#endif

I
Ingo Molnar 已提交
1178
#ifdef CONFIG_FAIR_GROUP_SCHED
P
Peter Zijlstra 已提交
1179
	int			depth;
I
Ingo Molnar 已提交
1180 1181 1182 1183 1184 1185
	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
1186

1187
#ifdef CONFIG_SMP
1188
	/* Per-entity load-tracking */
1189 1190
	struct sched_avg	avg;
#endif
I
Ingo Molnar 已提交
1191
};
1192

P
Peter Zijlstra 已提交
1193 1194
struct sched_rt_entity {
	struct list_head run_list;
1195
	unsigned long timeout;
1196
	unsigned long watchdog_stamp;
1197
	unsigned int time_slice;
P
Peter Zijlstra 已提交
1198

1199
	struct sched_rt_entity *back;
1200
#ifdef CONFIG_RT_GROUP_SCHED
P
Peter Zijlstra 已提交
1201 1202 1203 1204 1205 1206
	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
1207 1208
};

1209 1210 1211 1212 1213
struct sched_dl_entity {
	struct rb_node	rb_node;

	/*
	 * Original scheduling parameters. Copied here from sched_attr
1214 1215
	 * during sched_setattr(), they will remain the same until
	 * the next sched_setattr().
1216 1217 1218
	 */
	u64 dl_runtime;		/* maximum runtime for each instance	*/
	u64 dl_deadline;	/* relative deadline of each instance	*/
1219
	u64 dl_period;		/* separation of two instances (period) */
1220
	u64 dl_bw;		/* dl_runtime / dl_deadline		*/
1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240

	/*
	 * 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.
	 *
	 * @dl_new tells if a new instance arrived. If so we must
	 * start executing it with full runtime and reset its absolute
	 * deadline;
1241 1242 1243
	 *
	 * @dl_boosted tells if we are boosted due to DI. If so we are
	 * outside bandwidth enforcement mechanism (but only until we
1244 1245 1246 1247
	 * exit the critical section);
	 *
	 * @dl_yielded tells if task gave up the cpu before consuming
	 * all its available runtime during the last job.
1248
	 */
1249
	int dl_throttled, dl_new, dl_boosted, dl_yielded;
1250 1251 1252 1253 1254 1255 1256

	/*
	 * Bandwidth enforcement timer. Each -deadline task has its
	 * own bandwidth to be enforced, thus we need one timer per task.
	 */
	struct hrtimer dl_timer;
};
1257

1258 1259 1260 1261 1262 1263 1264
union rcu_special {
	struct {
		bool blocked;
		bool need_qs;
	} b;
	short s;
};
1265 1266
struct rcu_node;

P
Peter Zijlstra 已提交
1267 1268 1269
enum perf_event_task_context {
	perf_invalid_context = -1,
	perf_hw_context = 0,
1270
	perf_sw_context,
P
Peter Zijlstra 已提交
1271 1272 1273
	perf_nr_task_contexts,
};

L
Linus Torvalds 已提交
1274 1275
struct task_struct {
	volatile long state;	/* -1 unrunnable, 0 runnable, >0 stopped */
R
Roman Zippel 已提交
1276
	void *stack;
L
Linus Torvalds 已提交
1277
	atomic_t usage;
1278 1279
	unsigned int flags;	/* per process flags, defined below */
	unsigned int ptrace;
L
Linus Torvalds 已提交
1280

1281
#ifdef CONFIG_SMP
P
Peter Zijlstra 已提交
1282
	struct llist_node wake_entry;
P
Peter Zijlstra 已提交
1283
	int on_cpu;
1284 1285 1286
	struct task_struct *last_wakee;
	unsigned long wakee_flips;
	unsigned long wakee_flip_decay_ts;
1287 1288

	int wake_cpu;
1289
#endif
P
Peter Zijlstra 已提交
1290
	int on_rq;
1291

1292
	int prio, static_prio, normal_prio;
1293
	unsigned int rt_priority;
1294
	const struct sched_class *sched_class;
I
Ingo Molnar 已提交
1295
	struct sched_entity se;
P
Peter Zijlstra 已提交
1296
	struct sched_rt_entity rt;
P
Peter Zijlstra 已提交
1297 1298 1299
#ifdef CONFIG_CGROUP_SCHED
	struct task_group *sched_task_group;
#endif
1300
	struct sched_dl_entity dl;
L
Linus Torvalds 已提交
1301

1302 1303 1304 1305 1306
#ifdef CONFIG_PREEMPT_NOTIFIERS
	/* list of struct preempt_notifier: */
	struct hlist_head preempt_notifiers;
#endif

1307
#ifdef CONFIG_BLK_DEV_IO_TRACE
1308
	unsigned int btrace_seq;
1309
#endif
L
Linus Torvalds 已提交
1310

1311
	unsigned int policy;
1312
	int nr_cpus_allowed;
L
Linus Torvalds 已提交
1313 1314
	cpumask_t cpus_allowed;

P
Paul E. McKenney 已提交
1315
#ifdef CONFIG_PREEMPT_RCU
P
Paul E. McKenney 已提交
1316
	int rcu_read_lock_nesting;
1317
	union rcu_special rcu_read_unlock_special;
1318
	struct list_head rcu_node_entry;
P
Paul E. McKenney 已提交
1319
#endif /* #ifdef CONFIG_PREEMPT_RCU */
1320
#ifdef CONFIG_PREEMPT_RCU
P
Paul E. McKenney 已提交
1321
	struct rcu_node *rcu_blocked_node;
1322
#endif /* #ifdef CONFIG_PREEMPT_RCU */
P
Paul E. McKenney 已提交
1323 1324 1325 1326
#ifdef CONFIG_TASKS_RCU
	unsigned long rcu_tasks_nvcsw;
	bool rcu_tasks_holdout;
	struct list_head rcu_tasks_holdout_list;
1327
	int rcu_tasks_idle_cpu;
P
Paul E. McKenney 已提交
1328
#endif /* #ifdef CONFIG_TASKS_RCU */
P
Paul E. McKenney 已提交
1329

1330
#if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT)
L
Linus Torvalds 已提交
1331 1332 1333 1334
	struct sched_info sched_info;
#endif

	struct list_head tasks;
1335
#ifdef CONFIG_SMP
1336
	struct plist_node pushable_tasks;
1337
	struct rb_node pushable_dl_tasks;
1338
#endif
L
Linus Torvalds 已提交
1339 1340

	struct mm_struct *mm, *active_mm;
1341 1342 1343
#ifdef CONFIG_COMPAT_BRK
	unsigned brk_randomized:1;
#endif
D
Davidlohr Bueso 已提交
1344 1345 1346
	/* per-thread vma caching */
	u32 vmacache_seqnum;
	struct vm_area_struct *vmacache[VMACACHE_SIZE];
1347 1348 1349
#if defined(SPLIT_RSS_COUNTING)
	struct task_rss_stat	rss_stat;
#endif
L
Linus Torvalds 已提交
1350
/* task state */
1351
	int exit_state;
L
Linus Torvalds 已提交
1352 1353
	int exit_code, exit_signal;
	int pdeath_signal;  /*  The signal sent when the parent dies  */
1354
	unsigned int jobctl;	/* JOBCTL_*, siglock protected */
1355 1356

	/* Used for emulating ABI behavior of previous Linux versions */
1357
	unsigned int personality;
1358

1359 1360
	unsigned in_execve:1;	/* Tell the LSMs that the process is doing an
				 * execve */
1361 1362
	unsigned in_iowait:1;

1363 1364
	/* Revert to default priority/policy when forking */
	unsigned sched_reset_on_fork:1;
1365
	unsigned sched_contributes_to_load:1;
1366

1367 1368 1369 1370
#ifdef CONFIG_MEMCG_KMEM
	unsigned memcg_kmem_skip_account:1;
#endif

1371 1372
	unsigned long atomic_flags; /* Flags needing atomic access. */

L
Linus Torvalds 已提交
1373 1374
	pid_t pid;
	pid_t tgid;
1375

1376
#ifdef CONFIG_CC_STACKPROTECTOR
1377 1378
	/* Canary value for the -fstack-protector gcc feature */
	unsigned long stack_canary;
1379
#endif
1380
	/*
L
Linus Torvalds 已提交
1381
	 * pointers to (original) parent process, youngest child, younger sibling,
1382
	 * older sibling, respectively.  (p->father can be replaced with
R
Roland McGrath 已提交
1383
	 * p->real_parent->pid)
L
Linus Torvalds 已提交
1384
	 */
1385 1386
	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
1387
	/*
R
Roland McGrath 已提交
1388
	 * children/sibling forms the list of my natural children
L
Linus Torvalds 已提交
1389 1390 1391 1392 1393
	 */
	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
1394 1395 1396 1397 1398 1399 1400 1401
	/*
	 * 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
1402
	/* PID/PID hash table linkage. */
1403
	struct pid_link pids[PIDTYPE_MAX];
O
Oleg Nesterov 已提交
1404
	struct list_head thread_group;
1405
	struct list_head thread_node;
L
Linus Torvalds 已提交
1406 1407 1408 1409 1410

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

1411
	cputime_t utime, stime, utimescaled, stimescaled;
1412
	cputime_t gtime;
1413
#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
1414
	struct cputime prev_cputime;
1415 1416 1417 1418 1419 1420 1421 1422 1423
#endif
#ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN
	seqlock_t vtime_seqlock;
	unsigned long long vtime_snap;
	enum {
		VTIME_SLEEPING = 0,
		VTIME_USER,
		VTIME_SYS,
	} vtime_snap_whence;
1424
#endif
L
Linus Torvalds 已提交
1425
	unsigned long nvcsw, nivcsw; /* context switch counts */
1426
	u64 start_time;		/* monotonic time in nsec */
1427
	u64 real_start_time;	/* boot based time in nsec */
L
Linus Torvalds 已提交
1428 1429 1430
/* mm fault and swap info: this can arguably be seen as either mm-specific or thread-specific */
	unsigned long min_flt, maj_flt;

1431
	struct task_cputime cputime_expires;
L
Linus Torvalds 已提交
1432 1433 1434
	struct list_head cpu_timers[3];

/* process credentials */
A
Arnd Bergmann 已提交
1435
	const struct cred __rcu *real_cred; /* objective and real subjective task
1436
					 * credentials (COW) */
A
Arnd Bergmann 已提交
1437
	const struct cred __rcu *cred;	/* effective (overridable) subjective task
1438
					 * credentials (COW) */
1439 1440 1441
	char comm[TASK_COMM_LEN]; /* executable name excluding path
				     - access with [gs]et_task_comm (which lock
				       it with task_lock())
1442
				     - initialized normally by setup_new_exec */
L
Linus Torvalds 已提交
1443 1444
/* file system info */
	int link_count, total_link_count;
1445
#ifdef CONFIG_SYSVIPC
L
Linus Torvalds 已提交
1446 1447
/* ipc stuff */
	struct sysv_sem sysvsem;
1448
	struct sysv_shm sysvshm;
1449
#endif
1450
#ifdef CONFIG_DETECT_HUNG_TASK
1451 1452 1453
/* hung task detection */
	unsigned long last_switch_count;
#endif
L
Linus Torvalds 已提交
1454 1455 1456 1457 1458 1459
/* CPU-specific state of this task */
	struct thread_struct thread;
/* filesystem information */
	struct fs_struct *fs;
/* open file information */
	struct files_struct *files;
1460
/* namespaces */
S
Serge E. Hallyn 已提交
1461
	struct nsproxy *nsproxy;
L
Linus Torvalds 已提交
1462 1463 1464 1465 1466
/* signal handlers */
	struct signal_struct *signal;
	struct sighand_struct *sighand;

	sigset_t blocked, real_blocked;
1467
	sigset_t saved_sigmask;	/* restored if set_restore_sigmask() was used */
L
Linus Torvalds 已提交
1468 1469 1470 1471 1472 1473 1474
	struct sigpending pending;

	unsigned long sas_ss_sp;
	size_t sas_ss_size;
	int (*notifier)(void *priv);
	void *notifier_data;
	sigset_t *notifier_mask;
1475
	struct callback_head *task_works;
1476

L
Linus Torvalds 已提交
1477
	struct audit_context *audit_context;
A
Al Viro 已提交
1478
#ifdef CONFIG_AUDITSYSCALL
1479
	kuid_t loginuid;
1480
	unsigned int sessionid;
A
Al Viro 已提交
1481
#endif
1482
	struct seccomp seccomp;
L
Linus Torvalds 已提交
1483 1484 1485 1486

/* Thread group tracking */
   	u32 parent_exec_id;
   	u32 self_exec_id;
1487 1488
/* Protection of (de-)allocation: mm, files, fs, tty, keyrings, mems_allowed,
 * mempolicy */
L
Linus Torvalds 已提交
1489 1490
	spinlock_t alloc_lock;

1491
	/* Protection of the PI data structures: */
1492
	raw_spinlock_t pi_lock;
1493

I
Ingo Molnar 已提交
1494 1495
#ifdef CONFIG_RT_MUTEXES
	/* PI waiters blocked on a rt_mutex held by this task */
1496 1497
	struct rb_root pi_waiters;
	struct rb_node *pi_waiters_leftmost;
I
Ingo Molnar 已提交
1498 1499 1500 1501
	/* Deadlock detection and priority inheritance handling */
	struct rt_mutex_waiter *pi_blocked_on;
#endif

1502 1503 1504 1505
#ifdef CONFIG_DEBUG_MUTEXES
	/* mutex deadlock detection */
	struct mutex_waiter *blocked_on;
#endif
1506 1507 1508 1509
#ifdef CONFIG_TRACE_IRQFLAGS
	unsigned int irq_events;
	unsigned long hardirq_enable_ip;
	unsigned long hardirq_disable_ip;
1510
	unsigned int hardirq_enable_event;
1511
	unsigned int hardirq_disable_event;
1512 1513
	int hardirqs_enabled;
	int hardirq_context;
1514 1515
	unsigned long softirq_disable_ip;
	unsigned long softirq_enable_ip;
1516
	unsigned int softirq_disable_event;
1517
	unsigned int softirq_enable_event;
1518
	int softirqs_enabled;
1519 1520
	int softirq_context;
#endif
I
Ingo Molnar 已提交
1521
#ifdef CONFIG_LOCKDEP
1522
# define MAX_LOCK_DEPTH 48UL
I
Ingo Molnar 已提交
1523 1524 1525
	u64 curr_chain_key;
	int lockdep_depth;
	unsigned int lockdep_recursion;
1526
	struct held_lock held_locks[MAX_LOCK_DEPTH];
1527
	gfp_t lockdep_reclaim_gfp;
I
Ingo Molnar 已提交
1528
#endif
1529

L
Linus Torvalds 已提交
1530 1531 1532
/* journalling filesystem info */
	void *journal_info;

1533
/* stacked block device info */
1534
	struct bio_list *bio_list;
1535

1536 1537 1538 1539 1540
#ifdef CONFIG_BLOCK
/* stack plugging */
	struct blk_plug *plug;
#endif

L
Linus Torvalds 已提交
1541 1542 1543 1544 1545 1546 1547 1548 1549
/* 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.  */
1550
	struct task_io_accounting ioac;
1551
#if defined(CONFIG_TASK_XACCT)
L
Linus Torvalds 已提交
1552 1553
	u64 acct_rss_mem1;	/* accumulated rss usage */
	u64 acct_vm_mem1;	/* accumulated virtual memory usage */
1554
	cputime_t acct_timexpd;	/* stime + utime since last update */
L
Linus Torvalds 已提交
1555 1556
#endif
#ifdef CONFIG_CPUSETS
1557
	nodemask_t mems_allowed;	/* Protected by alloc_lock */
1558
	seqcount_t mems_allowed_seq;	/* Seqence no to catch updates */
1559
	int cpuset_mem_spread_rotor;
1560
	int cpuset_slab_spread_rotor;
L
Linus Torvalds 已提交
1561
#endif
1562
#ifdef CONFIG_CGROUPS
1563
	/* Control Group info protected by css_set_lock */
A
Arnd Bergmann 已提交
1564
	struct css_set __rcu *cgroups;
1565 1566
	/* cg_list protected by css_set_lock and tsk->alloc_lock */
	struct list_head cg_list;
1567
#endif
1568
#ifdef CONFIG_FUTEX
1569
	struct robust_list_head __user *robust_list;
1570 1571 1572
#ifdef CONFIG_COMPAT
	struct compat_robust_list_head __user *compat_robust_list;
#endif
1573 1574
	struct list_head pi_state_list;
	struct futex_pi_state *pi_state_cache;
1575
#endif
1576
#ifdef CONFIG_PERF_EVENTS
P
Peter Zijlstra 已提交
1577
	struct perf_event_context *perf_event_ctxp[perf_nr_task_contexts];
1578 1579
	struct mutex perf_event_mutex;
	struct list_head perf_event_list;
1580
#endif
1581 1582 1583
#ifdef CONFIG_DEBUG_PREEMPT
	unsigned long preempt_disable_ip;
#endif
1584
#ifdef CONFIG_NUMA
1585
	struct mempolicy *mempolicy;	/* Protected by alloc_lock */
1586
	short il_next;
1587
	short pref_node_fork;
1588
#endif
1589 1590 1591
#ifdef CONFIG_NUMA_BALANCING
	int numa_scan_seq;
	unsigned int numa_scan_period;
1592
	unsigned int numa_scan_period_max;
1593
	int numa_preferred_nid;
1594
	unsigned long numa_migrate_retry;
1595
	u64 node_stamp;			/* migration stamp  */
1596 1597
	u64 last_task_numa_placement;
	u64 last_sum_exec_runtime;
1598
	struct callback_head numa_work;
1599

1600 1601 1602
	struct list_head numa_entry;
	struct numa_group *numa_group;

1603
	/*
1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615
	 * 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.
1616
	 */
1617
	unsigned long *numa_faults;
1618
	unsigned long total_numa_faults;
1619

1620 1621 1622 1623 1624 1625 1626 1627
	/*
	 * numa_faults_locality tracks if faults recorded during the last
	 * scan window were remote/local. 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
	 */
	unsigned long numa_faults_locality[2];

I
Ingo Molnar 已提交
1628
	unsigned long numa_pages_migrated;
1629 1630
#endif /* CONFIG_NUMA_BALANCING */

I
Ingo Molnar 已提交
1631
	struct rcu_head rcu;
1632 1633 1634 1635 1636

	/*
	 * cache last used pipe for splice
	 */
	struct pipe_inode_info *splice_pipe;
1637 1638 1639

	struct page_frag task_frag;

1640 1641
#ifdef	CONFIG_TASK_DELAY_ACCT
	struct task_delay_info *delays;
1642 1643 1644
#endif
#ifdef CONFIG_FAULT_INJECTION
	int make_it_fail;
1645
#endif
1646 1647 1648 1649 1650 1651
	/*
	 * 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;
1652
	unsigned long dirty_paused_when; /* start of a write-and-pause period */
1653

A
Arjan van de Ven 已提交
1654 1655 1656 1657
#ifdef CONFIG_LATENCYTOP
	int latency_record_count;
	struct latency_record latency_record[LT_SAVECOUNT];
#endif
1658 1659 1660 1661 1662 1663
	/*
	 * time slack values; these are used to round up poll() and
	 * select() etc timeout values. These are in nanoseconds.
	 */
	unsigned long timer_slack_ns;
	unsigned long default_timer_slack_ns;
1664

1665
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
D
Daniel Mack 已提交
1666
	/* Index of current stored address in ret_stack */
1667 1668 1669
	int curr_ret_stack;
	/* Stack of return addresses for return function tracing */
	struct ftrace_ret_stack	*ret_stack;
1670 1671
	/* time stamp for last schedule */
	unsigned long long ftrace_timestamp;
1672 1673 1674 1675 1676
	/*
	 * Number of functions that haven't been traced
	 * because of depth overrun.
	 */
	atomic_t trace_overrun;
1677 1678
	/* Pause for the tracing */
	atomic_t tracing_graph_pause;
1679
#endif
1680 1681 1682
#ifdef CONFIG_TRACING
	/* state flags for use by tracers */
	unsigned long trace;
1683
	/* bitmask and counter of trace recursion */
1684 1685
	unsigned long trace_recursion;
#endif /* CONFIG_TRACING */
1686
#ifdef CONFIG_MEMCG
1687
	struct memcg_oom_info {
1688 1689 1690
		struct mem_cgroup *memcg;
		gfp_t gfp_mask;
		int order;
1691 1692
		unsigned int may_oom:1;
	} memcg_oom;
1693
#endif
1694 1695 1696
#ifdef CONFIG_UPROBES
	struct uprobe_task *utask;
#endif
K
Kent Overstreet 已提交
1697 1698 1699 1700
#if defined(CONFIG_BCACHE) || defined(CONFIG_BCACHE_MODULE)
	unsigned int	sequential_io;
	unsigned int	sequential_io_avg;
#endif
P
Peter Zijlstra 已提交
1701 1702 1703
#ifdef CONFIG_DEBUG_ATOMIC_SLEEP
	unsigned long	task_state_change;
#endif
L
Linus Torvalds 已提交
1704 1705
};

1706
/* Future-safe accessor for struct task_struct's cpus_allowed. */
1707
#define tsk_cpus_allowed(tsk) (&(tsk)->cpus_allowed)
1708

1709 1710
#define TNF_MIGRATED	0x01
#define TNF_NO_GROUP	0x02
1711
#define TNF_SHARED	0x04
1712
#define TNF_FAULT_LOCAL	0x08
1713

1714
#ifdef CONFIG_NUMA_BALANCING
1715
extern void task_numa_fault(int last_node, int node, int pages, int flags);
1716
extern pid_t task_numa_group_id(struct task_struct *p);
1717
extern void set_numabalancing_state(bool enabled);
1718
extern void task_numa_free(struct task_struct *p);
1719 1720
extern bool should_numa_migrate_memory(struct task_struct *p, struct page *page,
					int src_nid, int dst_cpu);
1721
#else
1722
static inline void task_numa_fault(int last_node, int node, int pages,
1723
				   int flags)
1724 1725
{
}
1726 1727 1728 1729
static inline pid_t task_numa_group_id(struct task_struct *p)
{
	return 0;
}
1730 1731 1732
static inline void set_numabalancing_state(bool enabled)
{
}
1733 1734 1735
static inline void task_numa_free(struct task_struct *p)
{
}
1736 1737 1738 1739 1740
static inline bool should_numa_migrate_memory(struct task_struct *p,
				struct page *page, int src_nid, int dst_cpu)
{
	return true;
}
1741 1742
#endif

A
Alexey Dobriyan 已提交
1743
static inline struct pid *task_pid(struct task_struct *task)
1744 1745 1746 1747
{
	return task->pids[PIDTYPE_PID].pid;
}

A
Alexey Dobriyan 已提交
1748
static inline struct pid *task_tgid(struct task_struct *task)
1749 1750 1751 1752
{
	return task->group_leader->pids[PIDTYPE_PID].pid;
}

1753 1754 1755 1756 1757
/*
 * 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
1758
static inline struct pid *task_pgrp(struct task_struct *task)
1759 1760 1761 1762
{
	return task->group_leader->pids[PIDTYPE_PGID].pid;
}

A
Alexey Dobriyan 已提交
1763
static inline struct pid *task_session(struct task_struct *task)
1764 1765 1766 1767
{
	return task->group_leader->pids[PIDTYPE_SID].pid;
}

1768 1769 1770 1771 1772 1773 1774
struct pid_namespace;

/*
 * 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
1775 1776
 * task_xid_vnr()    : virtual id, i.e. the id seen from the pid namespace of
 *                     current.
1777 1778 1779 1780 1781 1782
 * 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
 */
1783 1784
pid_t __task_pid_nr_ns(struct task_struct *task, enum pid_type type,
			struct pid_namespace *ns);
1785

A
Alexey Dobriyan 已提交
1786
static inline pid_t task_pid_nr(struct task_struct *tsk)
1787 1788 1789 1790
{
	return tsk->pid;
}

1791 1792 1793 1794 1795
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);
}
1796 1797 1798

static inline pid_t task_pid_vnr(struct task_struct *tsk)
{
1799
	return __task_pid_nr_ns(tsk, PIDTYPE_PID, NULL);
1800 1801 1802
}


A
Alexey Dobriyan 已提交
1803
static inline pid_t task_tgid_nr(struct task_struct *tsk)
1804 1805 1806 1807
{
	return tsk->tgid;
}

1808
pid_t task_tgid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns);
1809 1810 1811 1812 1813 1814 1815

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


1816
static inline int pid_alive(const struct task_struct *p);
1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833
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);
}

1834 1835
static inline pid_t task_pgrp_nr_ns(struct task_struct *tsk,
					struct pid_namespace *ns)
1836
{
1837
	return __task_pid_nr_ns(tsk, PIDTYPE_PGID, ns);
1838 1839 1840 1841
}

static inline pid_t task_pgrp_vnr(struct task_struct *tsk)
{
1842
	return __task_pid_nr_ns(tsk, PIDTYPE_PGID, NULL);
1843 1844 1845
}


1846 1847
static inline pid_t task_session_nr_ns(struct task_struct *tsk,
					struct pid_namespace *ns)
1848
{
1849
	return __task_pid_nr_ns(tsk, PIDTYPE_SID, ns);
1850 1851 1852 1853
}

static inline pid_t task_session_vnr(struct task_struct *tsk)
{
1854
	return __task_pid_nr_ns(tsk, PIDTYPE_SID, NULL);
1855 1856
}

1857 1858 1859 1860 1861
/* obsolete, do not use */
static inline pid_t task_pgrp_nr(struct task_struct *tsk)
{
	return task_pgrp_nr_ns(tsk, &init_pid_ns);
}
1862

L
Linus Torvalds 已提交
1863 1864 1865 1866 1867 1868 1869
/**
 * 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.
1870 1871
 *
 * Return: 1 if the process is alive. 0 otherwise.
L
Linus Torvalds 已提交
1872
 */
1873
static inline int pid_alive(const struct task_struct *p)
L
Linus Torvalds 已提交
1874
{
1875
	return p->pids[PIDTYPE_PID].pid != NULL;
L
Linus Torvalds 已提交
1876 1877
}

1878
/**
1879
 * is_global_init - check if a task structure is init
1880 1881 1882
 * @tsk: Task structure to be checked.
 *
 * Check if a task structure is the first user space task the kernel created.
1883 1884
 *
 * Return: 1 if the task structure is init. 0 otherwise.
1885
 */
A
Alexey Dobriyan 已提交
1886
static inline int is_global_init(struct task_struct *tsk)
1887 1888 1889
{
	return tsk->pid == 1;
}
1890

1891 1892
extern struct pid *cad_pid;

L
Linus Torvalds 已提交
1893 1894
extern void free_task(struct task_struct *tsk);
#define get_task_struct(tsk) do { atomic_inc(&(tsk)->usage); } while(0)
I
Ingo Molnar 已提交
1895

1896
extern void __put_task_struct(struct task_struct *t);
I
Ingo Molnar 已提交
1897 1898 1899 1900

static inline void put_task_struct(struct task_struct *t)
{
	if (atomic_dec_and_test(&t->usage))
1901
		__put_task_struct(t);
I
Ingo Molnar 已提交
1902
}
L
Linus Torvalds 已提交
1903

1904 1905 1906 1907 1908 1909 1910
#ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN
extern void task_cputime(struct task_struct *t,
			 cputime_t *utime, cputime_t *stime);
extern void task_cputime_scaled(struct task_struct *t,
				cputime_t *utimescaled, cputime_t *stimescaled);
extern cputime_t task_gtime(struct task_struct *t);
#else
1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928
static inline void task_cputime(struct task_struct *t,
				cputime_t *utime, cputime_t *stime)
{
	if (utime)
		*utime = t->utime;
	if (stime)
		*stime = t->stime;
}

static inline void task_cputime_scaled(struct task_struct *t,
				       cputime_t *utimescaled,
				       cputime_t *stimescaled)
{
	if (utimescaled)
		*utimescaled = t->utimescaled;
	if (stimescaled)
		*stimescaled = t->stimescaled;
}
1929 1930 1931 1932 1933 1934

static inline cputime_t task_gtime(struct task_struct *t)
{
	return t->gtime;
}
#endif
1935 1936
extern void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st);
extern void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st);
1937

L
Linus Torvalds 已提交
1938 1939 1940 1941
/*
 * Per process flags
 */
#define PF_EXITING	0x00000004	/* getting shut down */
1942
#define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
1943
#define PF_VCPU		0x00000010	/* I'm a virtual CPU */
T
Tejun Heo 已提交
1944
#define PF_WQ_WORKER	0x00000020	/* I'm a workqueue worker */
L
Linus Torvalds 已提交
1945
#define PF_FORKNOEXEC	0x00000040	/* forked but didn't exec */
1946
#define PF_MCE_PROCESS  0x00000080      /* process policy on mce errors */
L
Linus Torvalds 已提交
1947 1948 1949 1950
#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 */
1951
#define PF_NPROC_EXCEEDED 0x00001000	/* set_user noticed that RLIMIT_NPROC was exceeded */
L
Linus Torvalds 已提交
1952
#define PF_USED_MATH	0x00002000	/* if unset the fpu must be initialized before use */
1953
#define PF_USED_ASYNC	0x00004000	/* used async_schedule*(), used by module init */
L
Linus Torvalds 已提交
1954 1955 1956 1957
#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 */
1958
#define PF_MEMALLOC_NOIO 0x00080000	/* Allocating memory without IO involved */
L
Linus Torvalds 已提交
1959
#define PF_LESS_THROTTLE 0x00100000	/* Throttle me less: I clean memory */
1960
#define PF_KTHREAD	0x00200000	/* I am a kernel thread */
J
Jens Axboe 已提交
1961 1962
#define PF_RANDOMIZE	0x00400000	/* randomize virtual address space */
#define PF_SWAPWRITE	0x00800000	/* Allowed to write to swap */
1963
#define PF_NO_SETAFFINITY 0x04000000	/* Userland is not allowed to meddle with cpus_allowed */
1964
#define PF_MCE_EARLY    0x08000000      /* Early kill for mce process policy */
1965
#define PF_MUTEX_TESTER	0x20000000	/* Thread belongs to the rt mutex tester */
1966
#define PF_FREEZER_SKIP	0x40000000	/* Freezer should not count it as freezable */
1967
#define PF_SUSPEND_TASK 0x80000000      /* this thread called freeze_processes and should not be frozen */
L
Linus Torvalds 已提交
1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993

/*
 * 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)

1994 1995 1996
/* __GFP_IO isn't allowed if PF_MEMALLOC_NOIO is set in current->flags
 * __GFP_FS is also cleared as it implies __GFP_IO.
 */
1997 1998 1999
static inline gfp_t memalloc_noio_flags(gfp_t flags)
{
	if (unlikely(current->flags & PF_MEMALLOC_NOIO))
2000
		flags &= ~(__GFP_IO | __GFP_FS);
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
	return flags;
}

static inline unsigned int memalloc_noio_save(void)
{
	unsigned int flags = current->flags & PF_MEMALLOC_NOIO;
	current->flags |= PF_MEMALLOC_NOIO;
	return flags;
}

static inline void memalloc_noio_restore(unsigned int flags)
{
	current->flags = (current->flags & ~PF_MEMALLOC_NOIO) | flags;
}

2016
/* Per-process atomic flags. */
2017
#define PFA_NO_NEW_PRIVS 0	/* May not gain new privileges. */
2018 2019 2020
#define PFA_SPREAD_PAGE  1      /* Spread page cache over cpuset */
#define PFA_SPREAD_SLAB  2      /* Spread some slab caches over cpuset */

2021

2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033
#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)
2034

2035 2036 2037 2038 2039 2040 2041
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)
2042

2043
/*
2044
 * task->jobctl flags
2045
 */
2046
#define JOBCTL_STOP_SIGMASK	0xffff	/* signr of the last group stop */
2047

2048 2049 2050
#define JOBCTL_STOP_DEQUEUED_BIT 16	/* stop signal dequeued */
#define JOBCTL_STOP_PENDING_BIT	17	/* task should stop for group stop */
#define JOBCTL_STOP_CONSUME_BIT	18	/* consume group stop count */
2051
#define JOBCTL_TRAP_STOP_BIT	19	/* trap for STOP */
2052
#define JOBCTL_TRAP_NOTIFY_BIT	20	/* trap for NOTIFY */
2053
#define JOBCTL_TRAPPING_BIT	21	/* switching to TRACED */
T
Tejun Heo 已提交
2054
#define JOBCTL_LISTENING_BIT	22	/* ptracer is listening for events */
2055 2056 2057 2058

#define JOBCTL_STOP_DEQUEUED	(1 << JOBCTL_STOP_DEQUEUED_BIT)
#define JOBCTL_STOP_PENDING	(1 << JOBCTL_STOP_PENDING_BIT)
#define JOBCTL_STOP_CONSUME	(1 << JOBCTL_STOP_CONSUME_BIT)
2059
#define JOBCTL_TRAP_STOP	(1 << JOBCTL_TRAP_STOP_BIT)
2060
#define JOBCTL_TRAP_NOTIFY	(1 << JOBCTL_TRAP_NOTIFY_BIT)
2061
#define JOBCTL_TRAPPING		(1 << JOBCTL_TRAPPING_BIT)
T
Tejun Heo 已提交
2062
#define JOBCTL_LISTENING	(1 << JOBCTL_LISTENING_BIT)
2063

2064
#define JOBCTL_TRAP_MASK	(JOBCTL_TRAP_STOP | JOBCTL_TRAP_NOTIFY)
2065
#define JOBCTL_PENDING_MASK	(JOBCTL_STOP_PENDING | JOBCTL_TRAP_MASK)
2066

2067 2068
extern bool task_set_jobctl_pending(struct task_struct *task,
				    unsigned int mask);
2069
extern void task_clear_jobctl_trapping(struct task_struct *task);
2070 2071
extern void task_clear_jobctl_pending(struct task_struct *task,
				      unsigned int mask);
2072

2073 2074
static inline void rcu_copy_process(struct task_struct *p)
{
P
Paul E. McKenney 已提交
2075
#ifdef CONFIG_PREEMPT_RCU
2076
	p->rcu_read_lock_nesting = 0;
2077
	p->rcu_read_unlock_special.s = 0;
2078
	p->rcu_blocked_node = NULL;
2079
	INIT_LIST_HEAD(&p->rcu_node_entry);
P
Paul E. McKenney 已提交
2080 2081 2082 2083
#endif /* #ifdef CONFIG_PREEMPT_RCU */
#ifdef CONFIG_TASKS_RCU
	p->rcu_tasks_holdout = false;
	INIT_LIST_HEAD(&p->rcu_tasks_holdout_list);
2084
	p->rcu_tasks_idle_cpu = -1;
P
Paul E. McKenney 已提交
2085
#endif /* #ifdef CONFIG_TASKS_RCU */
2086 2087
}

2088 2089 2090 2091 2092 2093 2094
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;
}

2095 2096
extern int cpuset_cpumask_can_shrink(const struct cpumask *cur,
				     const struct cpumask *trial);
2097 2098
extern int task_can_attach(struct task_struct *p,
			   const struct cpumask *cs_cpus_allowed);
L
Linus Torvalds 已提交
2099
#ifdef CONFIG_SMP
2100 2101 2102
extern void do_set_cpus_allowed(struct task_struct *p,
			       const struct cpumask *new_mask);

2103
extern int set_cpus_allowed_ptr(struct task_struct *p,
2104
				const struct cpumask *new_mask);
L
Linus Torvalds 已提交
2105
#else
2106 2107 2108 2109
static inline void do_set_cpus_allowed(struct task_struct *p,
				      const struct cpumask *new_mask)
{
}
2110
static inline int set_cpus_allowed_ptr(struct task_struct *p,
2111
				       const struct cpumask *new_mask)
L
Linus Torvalds 已提交
2112
{
2113
	if (!cpumask_test_cpu(0, new_mask))
L
Linus Torvalds 已提交
2114 2115 2116 2117
		return -EINVAL;
	return 0;
}
#endif
2118

2119
#ifdef CONFIG_NO_HZ_COMMON
2120 2121 2122 2123 2124
void calc_load_enter_idle(void);
void calc_load_exit_idle(void);
#else
static inline void calc_load_enter_idle(void) { }
static inline void calc_load_exit_idle(void) { }
2125
#endif /* CONFIG_NO_HZ_COMMON */
2126

2127
#ifndef CONFIG_CPUMASK_OFFSTACK
2128 2129 2130 2131
static inline int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask)
{
	return set_cpus_allowed_ptr(p, &new_mask);
}
2132
#endif
L
Linus Torvalds 已提交
2133

2134
/*
2135 2136 2137 2138 2139 2140
 * Do not use outside of architecture code which knows its limitations.
 *
 * sched_clock() has no promise of monotonicity or bounded drift between
 * CPUs, use (which you should not) requires disabling IRQs.
 *
 * Please use one of the three interfaces below.
2141
 */
2142
extern unsigned long long notrace sched_clock(void);
2143
/*
2144
 * See the comment in kernel/sched/clock.c
2145 2146 2147 2148 2149
 */
extern u64 cpu_clock(int cpu);
extern u64 local_clock(void);
extern u64 sched_clock_cpu(int cpu);

2150

2151
extern void sched_clock_init(void);
2152

2153
#ifndef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165
static inline void sched_clock_tick(void)
{
}

static inline void sched_clock_idle_sleep_event(void)
{
}

static inline void sched_clock_idle_wakeup_event(u64 delta_ns)
{
}
#else
2166 2167 2168 2169 2170 2171
/*
 * Architectures can set this to 1 if they have specified
 * CONFIG_HAVE_UNSTABLE_SCHED_CLOCK in their arch Kconfig,
 * but then during bootup it turns out that sched_clock()
 * is reliable after all:
 */
2172 2173 2174
extern int sched_clock_stable(void);
extern void set_sched_clock_stable(void);
extern void clear_sched_clock_stable(void);
2175

2176 2177 2178 2179 2180
extern void sched_clock_tick(void);
extern void sched_clock_idle_sleep_event(void);
extern void sched_clock_idle_wakeup_event(u64 delta_ns);
#endif

2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193
#ifdef CONFIG_IRQ_TIME_ACCOUNTING
/*
 * An i/f to runtime opt-in for irq time accounting based off of sched_clock.
 * The reason for this explicit opt-in is not to have perf penalty with
 * slow sched_clocks.
 */
extern void enable_sched_clock_irqtime(void);
extern void disable_sched_clock_irqtime(void);
#else
static inline void enable_sched_clock_irqtime(void) {}
static inline void disable_sched_clock_irqtime(void) {}
#endif

2194
extern unsigned long long
2195
task_sched_runtime(struct task_struct *task);
L
Linus Torvalds 已提交
2196 2197 2198 2199 2200 2201 2202 2203

/* sched_exec is called by processes performing an exec */
#ifdef CONFIG_SMP
extern void sched_exec(void);
#else
#define sched_exec()   {}
#endif

2204 2205
extern void sched_clock_idle_sleep_event(void);
extern void sched_clock_idle_wakeup_event(u64 delta_ns);
2206

L
Linus Torvalds 已提交
2207 2208 2209 2210 2211 2212
#ifdef CONFIG_HOTPLUG_CPU
extern void idle_task_exit(void);
#else
static inline void idle_task_exit(void) {}
#endif

2213
#if defined(CONFIG_NO_HZ_COMMON) && defined(CONFIG_SMP)
2214
extern void wake_up_nohz_cpu(int cpu);
2215
#else
2216
static inline void wake_up_nohz_cpu(int cpu) { }
2217 2218
#endif

2219 2220
#ifdef CONFIG_NO_HZ_FULL
extern bool sched_can_stop_tick(void);
2221
extern u64 scheduler_tick_max_deferment(void);
2222 2223
#else
static inline bool sched_can_stop_tick(void) { return false; }
2224 2225
#endif

2226 2227 2228 2229 2230 2231 2232
#ifdef CONFIG_SCHED_AUTOGROUP
extern void sched_autogroup_create_attach(struct task_struct *p);
extern void sched_autogroup_detach(struct task_struct *p);
extern void sched_autogroup_fork(struct signal_struct *sig);
extern void sched_autogroup_exit(struct signal_struct *sig);
#ifdef CONFIG_PROC_FS
extern void proc_sched_autogroup_show_task(struct task_struct *p, struct seq_file *m);
2233
extern int proc_sched_autogroup_set_nice(struct task_struct *p, int nice);
2234 2235 2236 2237 2238 2239 2240 2241
#endif
#else
static inline void sched_autogroup_create_attach(struct task_struct *p) { }
static inline void sched_autogroup_detach(struct task_struct *p) { }
static inline void sched_autogroup_fork(struct signal_struct *sig) { }
static inline void sched_autogroup_exit(struct signal_struct *sig) { }
#endif

2242
extern int yield_to(struct task_struct *p, bool preempt);
2243 2244
extern void set_user_nice(struct task_struct *p, long nice);
extern int task_prio(const struct task_struct *p);
2245 2246 2247 2248 2249 2250 2251 2252 2253 2254
/**
 * 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);
}
2255 2256
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
2257
extern int idle_cpu(int cpu);
2258 2259
extern int sched_setscheduler(struct task_struct *, int,
			      const struct sched_param *);
2260
extern int sched_setscheduler_nocheck(struct task_struct *, int,
2261
				      const struct sched_param *);
2262 2263
extern int sched_setattr(struct task_struct *,
			 const struct sched_attr *);
2264
extern struct task_struct *idle_task(int cpu);
2265 2266
/**
 * is_idle_task - is the specified task an idle task?
2267
 * @p: the task in question.
2268 2269
 *
 * Return: 1 if @p is an idle task. 0 otherwise.
2270
 */
2271
static inline bool is_idle_task(const struct task_struct *p)
2272 2273 2274
{
	return p->pid == 0;
}
2275 2276
extern struct task_struct *curr_task(int cpu);
extern void set_curr_task(int cpu, struct task_struct *p);
L
Linus Torvalds 已提交
2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304

void yield(void);

/*
 * The default (Linux) execution domain.
 */
extern struct exec_domain	default_exec_domain;

union thread_union {
	struct thread_info thread_info;
	unsigned long stack[THREAD_SIZE/sizeof(long)];
};

#ifndef __HAVE_ARCH_KSTACK_END
static inline int kstack_end(void *addr)
{
	/* Reliable end of stack detection:
	 * Some APM bios versions misalign the stack
	 */
	return !(((unsigned long)addr+sizeof(void*)-1) & (THREAD_SIZE-sizeof(void*)));
}
#endif

extern union thread_union init_thread_union;
extern struct task_struct init_task;

extern struct   mm_struct init_mm;

2305 2306 2307 2308 2309 2310 2311
extern struct pid_namespace init_pid_ns;

/*
 * find a task by one of its numerical ids
 *
 * find_task_by_pid_ns():
 *      finds a task by its pid in the specified namespace
2312 2313
 * find_task_by_vpid():
 *      finds a task by its virtual pid
2314
 *
2315
 * see also find_vpid() etc in include/linux/pid.h
2316 2317
 */

2318 2319 2320
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);
2321

L
Linus Torvalds 已提交
2322
/* per-UID process charging. */
2323
extern struct user_struct * alloc_uid(kuid_t);
L
Linus Torvalds 已提交
2324 2325 2326 2327 2328 2329 2330 2331 2332
static inline struct user_struct *get_uid(struct user_struct *u)
{
	atomic_inc(&u->__count);
	return u;
}
extern void free_uid(struct user_struct *);

#include <asm/current.h>

T
Torben Hohn 已提交
2333
extern void xtime_update(unsigned long ticks);
L
Linus Torvalds 已提交
2334

2335 2336
extern int wake_up_state(struct task_struct *tsk, unsigned int state);
extern int wake_up_process(struct task_struct *tsk);
2337
extern void wake_up_new_task(struct task_struct *tsk);
L
Linus Torvalds 已提交
2338 2339 2340 2341 2342
#ifdef CONFIG_SMP
 extern void kick_process(struct task_struct *tsk);
#else
 static inline void kick_process(struct task_struct *tsk) { }
#endif
2343
extern int sched_fork(unsigned long clone_flags, struct task_struct *p);
2344
extern void sched_dead(struct task_struct *p);
L
Linus Torvalds 已提交
2345 2346 2347

extern void proc_caches_init(void);
extern void flush_signals(struct task_struct *);
2348
extern void __flush_signals(struct task_struct *);
2349
extern void ignore_signals(struct task_struct *);
L
Linus Torvalds 已提交
2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362
extern void flush_signal_handlers(struct task_struct *, int force_default);
extern int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info);

static inline int dequeue_signal_lock(struct task_struct *tsk, sigset_t *mask, siginfo_t *info)
{
	unsigned long flags;
	int ret;

	spin_lock_irqsave(&tsk->sighand->siglock, flags);
	ret = dequeue_signal(tsk, mask, info);
	spin_unlock_irqrestore(&tsk->sighand->siglock, flags);

	return ret;
2363
}
L
Linus Torvalds 已提交
2364 2365 2366 2367 2368 2369 2370 2371

extern void block_all_signals(int (*notifier)(void *priv), void *priv,
			      sigset_t *mask);
extern void unblock_all_signals(void);
extern void release_task(struct task_struct * p);
extern int send_sig_info(int, struct siginfo *, struct task_struct *);
extern int force_sigsegv(int, struct task_struct *);
extern int force_sig_info(int, struct siginfo *, struct task_struct *);
2372 2373
extern int __kill_pgrp_info(int sig, struct siginfo *info, struct pid *pgrp);
extern int kill_pid_info(int sig, struct siginfo *info, struct pid *pid);
2374 2375
extern int kill_pid_info_as_cred(int, struct siginfo *, struct pid *,
				const struct cred *, u32);
2376 2377
extern int kill_pgrp(struct pid *pid, int sig, int priv);
extern int kill_pid(struct pid *pid, int sig, int priv);
2378
extern int kill_proc_info(int, struct siginfo *, pid_t);
2379
extern __must_check bool do_notify_parent(struct task_struct *, int);
2380
extern void __wake_up_parent(struct task_struct *p, struct task_struct *parent);
L
Linus Torvalds 已提交
2381 2382
extern void force_sig(int, struct task_struct *);
extern int send_sig(int, struct task_struct *, int);
2383
extern int zap_other_threads(struct task_struct *p);
L
Linus Torvalds 已提交
2384 2385
extern struct sigqueue *sigqueue_alloc(void);
extern void sigqueue_free(struct sigqueue *);
2386
extern int send_sigqueue(struct sigqueue *,  struct task_struct *, int group);
2387
extern int do_sigaction(int, struct k_sigaction *, struct k_sigaction *);
L
Linus Torvalds 已提交
2388

A
Al Viro 已提交
2389 2390 2391
static inline void restore_saved_sigmask(void)
{
	if (test_and_clear_restore_sigmask())
2392
		__set_current_blocked(&current->saved_sigmask);
A
Al Viro 已提交
2393 2394
}

A
Al Viro 已提交
2395 2396 2397 2398 2399 2400 2401 2402
static inline sigset_t *sigmask_to_save(void)
{
	sigset_t *res = &current->blocked;
	if (unlikely(test_restore_sigmask()))
		res = &current->saved_sigmask;
	return res;
}

2403 2404 2405 2406 2407
static inline int kill_cad_pid(int sig, int priv)
{
	return kill_pid(cad_pid, sig, priv);
}

L
Linus Torvalds 已提交
2408 2409 2410 2411 2412
/* These can be the second arg to send_sig_info/send_group_sig_info.  */
#define SEND_SIG_NOINFO ((struct siginfo *) 0)
#define SEND_SIG_PRIV	((struct siginfo *) 1)
#define SEND_SIG_FORCED	((struct siginfo *) 2)

2413 2414 2415
/*
 * True if we are on the alternate signal stack.
 */
L
Linus Torvalds 已提交
2416 2417
static inline int on_sig_stack(unsigned long sp)
{
2418 2419 2420 2421 2422 2423 2424
#ifdef CONFIG_STACK_GROWSUP
	return sp >= current->sas_ss_sp &&
		sp - current->sas_ss_sp < current->sas_ss_size;
#else
	return sp > current->sas_ss_sp &&
		sp - current->sas_ss_sp <= current->sas_ss_size;
#endif
L
Linus Torvalds 已提交
2425 2426 2427 2428
}

static inline int sas_ss_flags(unsigned long sp)
{
2429 2430 2431 2432
	if (!current->sas_ss_size)
		return SS_DISABLE;

	return on_sig_stack(sp) ? SS_ONSTACK : 0;
L
Linus Torvalds 已提交
2433 2434
}

A
Al Viro 已提交
2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445
static inline unsigned long sigsp(unsigned long sp, struct ksignal *ksig)
{
	if (unlikely((ksig->ka.sa.sa_flags & SA_ONSTACK)) && ! sas_ss_flags(sp))
#ifdef CONFIG_STACK_GROWSUP
		return current->sas_ss_sp;
#else
		return current->sas_ss_sp + current->sas_ss_size;
#endif
	return sp;
}

L
Linus Torvalds 已提交
2446 2447 2448 2449 2450 2451
/*
 * Routines for handling mm_structs
 */
extern struct mm_struct * mm_alloc(void);

/* mmdrop drops the mm and the page tables */
2452
extern void __mmdrop(struct mm_struct *);
L
Linus Torvalds 已提交
2453 2454
static inline void mmdrop(struct mm_struct * mm)
{
I
Ingo Molnar 已提交
2455
	if (unlikely(atomic_dec_and_test(&mm->mm_count)))
L
Linus Torvalds 已提交
2456 2457 2458 2459 2460 2461 2462
		__mmdrop(mm);
}

/* mmput gets rid of the mappings and all user-space */
extern void mmput(struct mm_struct *);
/* Grab a reference to a task's mm, if it is not already going away */
extern struct mm_struct *get_task_mm(struct task_struct *task);
2463 2464 2465 2466 2467 2468
/*
 * Grab a reference to a task's mm, if it is not already going away
 * and ptrace_may_access with the mode parameter passed to it
 * succeeds.
 */
extern struct mm_struct *mm_access(struct task_struct *task, unsigned int mode);
L
Linus Torvalds 已提交
2469 2470 2471
/* Remove the current tasks stale references to the old mm_struct */
extern void mm_release(struct task_struct *, struct mm_struct *);

A
Alexey Dobriyan 已提交
2472
extern int copy_thread(unsigned long, unsigned long, unsigned long,
2473
			struct task_struct *);
L
Linus Torvalds 已提交
2474 2475 2476 2477
extern void flush_thread(void);
extern void exit_thread(void);

extern void exit_files(struct task_struct *);
2478
extern void __cleanup_sighand(struct sighand_struct *);
2479

L
Linus Torvalds 已提交
2480
extern void exit_itimers(struct signal_struct *);
2481
extern void flush_itimer_signals(void);
L
Linus Torvalds 已提交
2482

2483
extern void do_group_exit(int);
L
Linus Torvalds 已提交
2484

2485
extern int do_execve(struct filename *,
2486
		     const char __user * const __user *,
2487
		     const char __user * const __user *);
2488
extern long do_fork(unsigned long, unsigned long, unsigned long, int __user *, int __user *);
2489
struct task_struct *fork_idle(int);
2490
extern pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
L
Linus Torvalds 已提交
2491

2492 2493 2494 2495 2496
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);
}
2497
extern char *get_task_comm(char *to, struct task_struct *tsk);
L
Linus Torvalds 已提交
2498 2499

#ifdef CONFIG_SMP
2500
void scheduler_ipi(void);
R
Roland McGrath 已提交
2501
extern unsigned long wait_task_inactive(struct task_struct *, long match_state);
L
Linus Torvalds 已提交
2502
#else
2503
static inline void scheduler_ipi(void) { }
R
Roland McGrath 已提交
2504 2505 2506 2507 2508
static inline unsigned long wait_task_inactive(struct task_struct *p,
					       long match_state)
{
	return 1;
}
L
Linus Torvalds 已提交
2509 2510
#endif

2511 2512
#define next_task(p) \
	list_entry_rcu((p)->tasks.next, struct task_struct, tasks)
L
Linus Torvalds 已提交
2513 2514 2515 2516

#define for_each_process(p) \
	for (p = &init_task ; (p = next_task(p)) != &init_task ; )

2517
extern bool current_is_single_threaded(void);
D
David Howells 已提交
2518

L
Linus Torvalds 已提交
2519 2520 2521 2522 2523 2524 2525 2526 2527 2528
/*
 * Careful: do_each_thread/while_each_thread is a double loop so
 *          'break' will not work as expected - use goto instead.
 */
#define do_each_thread(g, t) \
	for (g = t = &init_task ; (g = t = next_task(g)) != &init_task ; ) do

#define while_each_thread(g, t) \
	while ((t = next_thread(t)) != g)

2529 2530 2531 2532 2533 2534 2535 2536 2537 2538
#define __for_each_thread(signal, t)	\
	list_for_each_entry_rcu(t, &(signal)->thread_head, thread_node)

#define for_each_thread(p, t)		\
	__for_each_thread((p)->signal, t)

/* Careful: this is a double loop, 'break' won't work as expected. */
#define for_each_process_thread(p, t)	\
	for_each_process(p) for_each_thread(p, t)

2539 2540
static inline int get_nr_threads(struct task_struct *tsk)
{
2541
	return tsk->signal->nr_threads;
2542 2543
}

2544 2545 2546 2547
static inline bool thread_group_leader(struct task_struct *p)
{
	return p->exit_signal >= 0;
}
L
Linus Torvalds 已提交
2548

2549 2550 2551 2552 2553 2554
/* Do to the insanities of de_thread it is possible for a process
 * to have the pid of the thread group leader without actually being
 * the thread group leader.  For iteration through the pids in proc
 * all we care about is that we have a task with the appropriate
 * pid, we don't actually care if we have the right task.
 */
2555
static inline bool has_group_leader_pid(struct task_struct *p)
2556
{
2557
	return task_pid(p) == p->signal->leader_pid;
2558 2559
}

2560
static inline
2561
bool same_thread_group(struct task_struct *p1, struct task_struct *p2)
2562
{
2563
	return p1->signal == p2->signal;
2564 2565
}

2566
static inline struct task_struct *next_thread(const struct task_struct *p)
O
Oleg Nesterov 已提交
2567
{
2568 2569
	return list_entry_rcu(p->thread_group.next,
			      struct task_struct, thread_group);
O
Oleg Nesterov 已提交
2570 2571
}

A
Alexey Dobriyan 已提交
2572
static inline int thread_group_empty(struct task_struct *p)
L
Linus Torvalds 已提交
2573
{
O
Oleg Nesterov 已提交
2574
	return list_empty(&p->thread_group);
L
Linus Torvalds 已提交
2575 2576 2577 2578 2579 2580
}

#define delay_group_leader(p) \
		(thread_group_leader(p) && !thread_group_empty(p))

/*
2581
 * Protects ->fs, ->files, ->mm, ->group_info, ->comm, keyring
2582
 * subscriptions and synchronises with wait4().  Also used in procfs.  Also
2583
 * pins the final release of task.io_context.  Also protects ->cpuset and
O
Oleg Nesterov 已提交
2584
 * ->cgroup.subsys[]. And ->vfork_done.
L
Linus Torvalds 已提交
2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599
 *
 * Nests both inside and outside of read_lock(&tasklist_lock).
 * It must not be nested with write_lock_irq(&tasklist_lock),
 * neither inside nor outside.
 */
static inline void task_lock(struct task_struct *p)
{
	spin_lock(&p->alloc_lock);
}

static inline void task_unlock(struct task_struct *p)
{
	spin_unlock(&p->alloc_lock);
}

2600
extern struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
2601 2602
							unsigned long *flags);

2603 2604 2605 2606 2607 2608 2609 2610 2611
static inline struct sighand_struct *lock_task_sighand(struct task_struct *tsk,
						       unsigned long *flags)
{
	struct sighand_struct *ret;

	ret = __lock_task_sighand(tsk, flags);
	(void)__cond_lock(&tsk->sighand->siglock, ret);
	return ret;
}
2612

2613 2614 2615 2616 2617 2618
static inline void unlock_task_sighand(struct task_struct *tsk,
						unsigned long *flags)
{
	spin_unlock_irqrestore(&tsk->sighand->siglock, *flags);
}

2619
#ifdef CONFIG_CGROUPS
2620
static inline void threadgroup_change_begin(struct task_struct *tsk)
2621
{
2622
	down_read(&tsk->signal->group_rwsem);
2623
}
2624
static inline void threadgroup_change_end(struct task_struct *tsk)
2625
{
2626
	up_read(&tsk->signal->group_rwsem);
2627
}
2628 2629 2630 2631 2632 2633 2634

/**
 * threadgroup_lock - lock threadgroup
 * @tsk: member task of the threadgroup to lock
 *
 * Lock the threadgroup @tsk belongs to.  No new task is allowed to enter
 * and member tasks aren't allowed to exit (as indicated by PF_EXITING) or
2635 2636
 * change ->group_leader/pid.  This is useful for cases where the threadgroup
 * needs to stay stable across blockable operations.
2637 2638 2639 2640 2641
 *
 * fork and exit paths explicitly call threadgroup_change_{begin|end}() for
 * synchronization.  While held, no new task will be added to threadgroup
 * and no existing live task will have its PF_EXITING set.
 *
2642 2643
 * de_thread() does threadgroup_change_{begin|end}() when a non-leader
 * sub-thread becomes a new leader.
2644
 */
2645
static inline void threadgroup_lock(struct task_struct *tsk)
2646
{
2647
	down_write(&tsk->signal->group_rwsem);
2648
}
2649 2650 2651 2652 2653 2654 2655

/**
 * threadgroup_unlock - unlock threadgroup
 * @tsk: member task of the threadgroup to unlock
 *
 * Reverse threadgroup_lock().
 */
2656
static inline void threadgroup_unlock(struct task_struct *tsk)
2657
{
2658
	up_write(&tsk->signal->group_rwsem);
2659 2660
}
#else
2661 2662 2663 2664
static inline void threadgroup_change_begin(struct task_struct *tsk) {}
static inline void threadgroup_change_end(struct task_struct *tsk) {}
static inline void threadgroup_lock(struct task_struct *tsk) {}
static inline void threadgroup_unlock(struct task_struct *tsk) {}
2665 2666
#endif

A
Al Viro 已提交
2667 2668
#ifndef __HAVE_THREAD_FUNCTIONS

R
Roman Zippel 已提交
2669 2670
#define task_thread_info(task)	((struct thread_info *)(task)->stack)
#define task_stack_page(task)	((task)->stack)
A
Al Viro 已提交
2671

2672 2673 2674 2675 2676 2677
static inline void setup_thread_stack(struct task_struct *p, struct task_struct *org)
{
	*task_thread_info(p) = *task_thread_info(org);
	task_thread_info(p)->task = p;
}

2678 2679 2680 2681 2682 2683 2684 2685 2686
/*
 * Return the address of the last usable long on the stack.
 *
 * When the stack grows down, this is just above the thread
 * info struct. Going any lower will corrupt the threadinfo.
 *
 * When the stack grows up, this is the highest address.
 * Beyond that position, we corrupt data on the next page.
 */
2687 2688
static inline unsigned long *end_of_stack(struct task_struct *p)
{
2689 2690 2691
#ifdef CONFIG_STACK_GROWSUP
	return (unsigned long *)((unsigned long)task_thread_info(p) + THREAD_SIZE) - 1;
#else
R
Roman Zippel 已提交
2692
	return (unsigned long *)(task_thread_info(p) + 1);
2693
#endif
2694 2695
}

A
Al Viro 已提交
2696
#endif
2697 2698
#define task_stack_end_corrupted(task) \
		(*(end_of_stack(task)) != STACK_END_MAGIC)
A
Al Viro 已提交
2699

2700 2701 2702 2703 2704 2705 2706
static inline int object_is_on_stack(void *obj)
{
	void *stack = task_stack_page(current);

	return (obj >= stack) && (obj < (stack + THREAD_SIZE));
}

2707 2708
extern void thread_info_cache_init(void);

2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720
#ifdef CONFIG_DEBUG_STACK_USAGE
static inline unsigned long stack_not_used(struct task_struct *p)
{
	unsigned long *n = end_of_stack(p);

	do { 	/* Skip over canary */
		n++;
	} while (!*n);

	return (unsigned long)n - (unsigned long)end_of_stack(p);
}
#endif
2721
extern void set_task_stack_end_magic(struct task_struct *tsk);
2722

L
Linus Torvalds 已提交
2723 2724 2725 2726 2727
/* 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
2728
	set_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2729 2730 2731 2732
}

static inline void clear_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
2733
	clear_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2734 2735 2736 2737
}

static inline int test_and_set_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
2738
	return test_and_set_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2739 2740 2741 2742
}

static inline int test_and_clear_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
2743
	return test_and_clear_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2744 2745 2746 2747
}

static inline int test_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
2748
	return test_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760
}

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);
}

2761 2762 2763 2764 2765
static inline int test_tsk_need_resched(struct task_struct *tsk)
{
	return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED));
}

2766 2767 2768 2769 2770 2771
static inline int restart_syscall(void)
{
	set_tsk_thread_flag(current, TIF_SIGPENDING);
	return -ERESTARTNOINTR;
}

L
Linus Torvalds 已提交
2772 2773 2774 2775
static inline int signal_pending(struct task_struct *p)
{
	return unlikely(test_tsk_thread_flag(p,TIF_SIGPENDING));
}
M
Matthew Wilcox 已提交
2776

2777 2778 2779 2780
static inline int __fatal_signal_pending(struct task_struct *p)
{
	return unlikely(sigismember(&p->pending.signal, SIGKILL));
}
M
Matthew Wilcox 已提交
2781 2782 2783 2784 2785 2786

static inline int fatal_signal_pending(struct task_struct *p)
{
	return signal_pending(p) && __fatal_signal_pending(p);
}

2787 2788 2789 2790 2791 2792 2793 2794 2795 2796
static inline int signal_pending_state(long state, struct task_struct *p)
{
	if (!(state & (TASK_INTERRUPTIBLE | TASK_WAKEKILL)))
		return 0;
	if (!signal_pending(p))
		return 0;

	return (state & TASK_INTERRUPTIBLE) || __fatal_signal_pending(p);
}

L
Linus Torvalds 已提交
2797 2798 2799 2800 2801 2802 2803
/*
 * 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.
 */
2804
extern int _cond_resched(void);
2805

2806
#define cond_resched() ({			\
2807
	___might_sleep(__FILE__, __LINE__, 0);	\
2808 2809
	_cond_resched();			\
})
2810

2811 2812
extern int __cond_resched_lock(spinlock_t *lock);

2813
#ifdef CONFIG_PREEMPT_COUNT
2814
#define PREEMPT_LOCK_OFFSET	PREEMPT_OFFSET
2815
#else
2816
#define PREEMPT_LOCK_OFFSET	0
2817
#endif
2818

2819
#define cond_resched_lock(lock) ({				\
2820
	___might_sleep(__FILE__, __LINE__, PREEMPT_LOCK_OFFSET);\
2821 2822 2823 2824 2825
	__cond_resched_lock(lock);				\
})

extern int __cond_resched_softirq(void);

2826
#define cond_resched_softirq() ({					\
2827
	___might_sleep(__FILE__, __LINE__, SOFTIRQ_DISABLE_OFFSET);	\
2828
	__cond_resched_softirq();					\
2829
})
L
Linus Torvalds 已提交
2830

2831 2832 2833 2834 2835 2836 2837 2838 2839
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
2840 2841
/*
 * Does a critical section need to be broken due to another
N
Nick Piggin 已提交
2842 2843
 * task waiting?: (technically does not depend on CONFIG_PREEMPT,
 * but a general need for low latency)
L
Linus Torvalds 已提交
2844
 */
N
Nick Piggin 已提交
2845
static inline int spin_needbreak(spinlock_t *lock)
L
Linus Torvalds 已提交
2846
{
N
Nick Piggin 已提交
2847 2848 2849
#ifdef CONFIG_PREEMPT
	return spin_is_contended(lock);
#else
L
Linus Torvalds 已提交
2850
	return 0;
N
Nick Piggin 已提交
2851
#endif
L
Linus Torvalds 已提交
2852 2853
}

2854 2855
/*
 * Idle thread specific functions to determine the need_resched
2856
 * polling state.
2857
 */
2858
#ifdef TIF_POLLING_NRFLAG
2859 2860 2861 2862
static inline int tsk_is_polling(struct task_struct *p)
{
	return test_tsk_thread_flag(p, TIF_POLLING_NRFLAG);
}
2863 2864

static inline void __current_set_polling(void)
2865 2866 2867 2868
{
	set_thread_flag(TIF_POLLING_NRFLAG);
}

2869 2870 2871 2872 2873 2874
static inline bool __must_check current_set_polling_and_test(void)
{
	__current_set_polling();

	/*
	 * Polling state must be visible before we test NEED_RESCHED,
2875
	 * paired by resched_curr()
2876
	 */
2877
	smp_mb__after_atomic();
2878 2879 2880 2881 2882

	return unlikely(tif_need_resched());
}

static inline void __current_clr_polling(void)
2883 2884 2885
{
	clear_thread_flag(TIF_POLLING_NRFLAG);
}
2886 2887 2888 2889 2890 2891 2892

static inline bool __must_check current_clr_polling_and_test(void)
{
	__current_clr_polling();

	/*
	 * Polling state must be visible before we test NEED_RESCHED,
2893
	 * paired by resched_curr()
2894
	 */
2895
	smp_mb__after_atomic();
2896 2897 2898 2899

	return unlikely(tif_need_resched());
}

2900 2901
#else
static inline int tsk_is_polling(struct task_struct *p) { return 0; }
2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912
static inline void __current_set_polling(void) { }
static inline void __current_clr_polling(void) { }

static inline bool __must_check current_set_polling_and_test(void)
{
	return unlikely(tif_need_resched());
}
static inline bool __must_check current_clr_polling_and_test(void)
{
	return unlikely(tif_need_resched());
}
2913 2914
#endif

2915 2916 2917 2918 2919 2920 2921 2922 2923 2924
static inline void current_clr_polling(void)
{
	__current_clr_polling();

	/*
	 * Ensure we check TIF_NEED_RESCHED after we clear the polling bit.
	 * Once the bit is cleared, we'll get IPIs with every new
	 * TIF_NEED_RESCHED and the IPI handler, scheduler_ipi(), will also
	 * fold.
	 */
2925
	smp_mb(); /* paired with resched_curr() */
2926 2927 2928 2929

	preempt_fold_need_resched();
}

2930 2931 2932 2933 2934
static __always_inline bool need_resched(void)
{
	return unlikely(tif_need_resched());
}

2935 2936 2937
/*
 * Thread group CPU time accounting.
 */
2938
void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times);
2939
void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times);
2940

2941
static inline void thread_group_cputime_init(struct signal_struct *sig)
2942
{
2943
	raw_spin_lock_init(&sig->cputimer.lock);
2944 2945
}

R
Roland McGrath 已提交
2946 2947 2948 2949 2950 2951 2952
/*
 * Reevaluate whether the task has signals pending delivery.
 * Wake the task if so.
 * This is required every time the blocked sigset_t changes.
 * callers must hold sighand->siglock.
 */
extern void recalc_sigpending_and_wake(struct task_struct *t);
L
Linus Torvalds 已提交
2953 2954
extern void recalc_sigpending(void);

2955 2956 2957 2958 2959 2960 2961 2962 2963 2964
extern void signal_wake_up_state(struct task_struct *t, unsigned int state);

static inline void signal_wake_up(struct task_struct *t, bool resume)
{
	signal_wake_up_state(t, resume ? TASK_WAKEKILL : 0);
}
static inline void ptrace_signal_wake_up(struct task_struct *t, bool resume)
{
	signal_wake_up_state(t, resume ? __TASK_TRACED : 0);
}
L
Linus Torvalds 已提交
2965 2966 2967 2968 2969 2970 2971 2972

/*
 * 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)
{
A
Al Viro 已提交
2973
	return task_thread_info(p)->cpu;
L
Linus Torvalds 已提交
2974 2975
}

I
Ingo Molnar 已提交
2976 2977 2978 2979 2980
static inline int task_node(const struct task_struct *p)
{
	return cpu_to_node(task_cpu(p));
}

I
Ingo Molnar 已提交
2981
extern void set_task_cpu(struct task_struct *p, unsigned int cpu);
L
Linus Torvalds 已提交
2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995

#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 */

2996 2997
extern long sched_setaffinity(pid_t pid, const struct cpumask *new_mask);
extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
2998

D
Dhaval Giani 已提交
2999
#ifdef CONFIG_CGROUP_SCHED
3000
extern struct task_group root_task_group;
P
Peter Zijlstra 已提交
3001
#endif /* CONFIG_CGROUP_SCHED */
3002

3003 3004 3005
extern int task_can_switch_user(struct user_struct *up,
					struct task_struct *tsk);

3006 3007 3008
#ifdef CONFIG_TASK_XACCT
static inline void add_rchar(struct task_struct *tsk, ssize_t amt)
{
3009
	tsk->ioac.rchar += amt;
3010 3011 3012 3013
}

static inline void add_wchar(struct task_struct *tsk, ssize_t amt)
{
3014
	tsk->ioac.wchar += amt;
3015 3016 3017 3018
}

static inline void inc_syscr(struct task_struct *tsk)
{
3019
	tsk->ioac.syscr++;
3020 3021 3022 3023
}

static inline void inc_syscw(struct task_struct *tsk)
{
3024
	tsk->ioac.syscw++;
3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043
}
#else
static inline void add_rchar(struct task_struct *tsk, ssize_t amt)
{
}

static inline void add_wchar(struct task_struct *tsk, ssize_t amt)
{
}

static inline void inc_syscr(struct task_struct *tsk)
{
}

static inline void inc_syscw(struct task_struct *tsk)
{
}
#endif

D
Dave Hansen 已提交
3044 3045 3046 3047
#ifndef TASK_SIZE_OF
#define TASK_SIZE_OF(tsk)	TASK_SIZE
#endif

O
Oleg Nesterov 已提交
3048
#ifdef CONFIG_MEMCG
3049 3050 3051 3052 3053
extern void mm_update_next_owner(struct mm_struct *mm);
#else
static inline void mm_update_next_owner(struct mm_struct *mm)
{
}
O
Oleg Nesterov 已提交
3054
#endif /* CONFIG_MEMCG */
3055

3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077
static inline unsigned long task_rlimit(const struct task_struct *tsk,
		unsigned int limit)
{
	return ACCESS_ONCE(tsk->signal->rlim[limit].rlim_cur);
}

static inline unsigned long task_rlimit_max(const struct task_struct *tsk,
		unsigned int limit)
{
	return ACCESS_ONCE(tsk->signal->rlim[limit].rlim_max);
}

static inline unsigned long rlimit(unsigned int limit)
{
	return task_rlimit(current, limit);
}

static inline unsigned long rlimit_max(unsigned int limit)
{
	return task_rlimit_max(current, limit);
}

L
Linus Torvalds 已提交
3078
#endif