sched.h 89.6 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.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
#include <linux/cgroup-defs.h>
62 63

#include <asm/processor.h>
H
H. J. Lu 已提交
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 99 100 101 102 103 104 105
#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.
106 107 108 109
 *
 * 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/.
110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128
 */
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;
};

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;
136
struct nameidata;
L
Linus Torvalds 已提交
137

D
Davidlohr Bueso 已提交
138 139 140 141
#define VMACACHE_BITS 2
#define VMACACHE_SIZE (1U << VMACACHE_BITS)
#define VMACACHE_MASK (VMACACHE_SIZE - 1)

L
Linus Torvalds 已提交
142 143 144 145 146 147 148 149 150 151 152
/*
 * 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 */
153
extern void get_avenrun(unsigned long *loads, unsigned long offset, int shift);
L
Linus Torvalds 已提交
154 155 156

#define FSHIFT		11		/* nr of bits of precision */
#define FIXED_1		(1<<FSHIFT)	/* 1.0 as fixed-point */
157
#define LOAD_FREQ	(5*HZ+1)	/* 5 sec intervals */
L
Linus Torvalds 已提交
158 159 160 161 162 163 164 165 166 167 168 169 170 171
#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);
172
extern bool single_task_running(void);
L
Linus Torvalds 已提交
173
extern unsigned long nr_iowait(void);
174
extern unsigned long nr_iowait_cpu(int cpu);
175
extern void get_iowait_load(unsigned long *nr_waiters, unsigned long *load);
176

177
extern void calc_global_load(unsigned long ticks);
178 179

#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON)
180
extern void update_cpu_load_nohz(void);
181 182 183
#else
static inline void update_cpu_load_nohz(void) { }
#endif
L
Linus Torvalds 已提交
184

185 186
extern unsigned long get_parent_ip(unsigned long addr);

187 188
extern void dump_cpu_task(int cpu);

I
Ingo Molnar 已提交
189 190
struct seq_file;
struct cfs_rq;
191
struct task_group;
I
Ingo Molnar 已提交
192 193 194 195
#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);
#endif
L
Linus Torvalds 已提交
196

197 198 199 200 201 202 203 204 205 206
/*
 * 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
207 208 209
#define TASK_RUNNING		0
#define TASK_INTERRUPTIBLE	1
#define TASK_UNINTERRUPTIBLE	2
M
Matthew Wilcox 已提交
210 211
#define __TASK_STOPPED		4
#define __TASK_TRACED		8
212
/* in tsk->exit_state */
213 214
#define EXIT_DEAD		16
#define EXIT_ZOMBIE		32
215
#define EXIT_TRACE		(EXIT_ZOMBIE | EXIT_DEAD)
216
/* in tsk->state again */
217
#define TASK_DEAD		64
M
Matthew Wilcox 已提交
218
#define TASK_WAKEKILL		128
P
Peter Zijlstra 已提交
219
#define TASK_WAKING		256
220
#define TASK_PARKED		512
221 222
#define TASK_NOLOAD		1024
#define TASK_STATE_MAX		2048
M
Matthew Wilcox 已提交
223

224
#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWPN"
225

226 227
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
228 229 230 231 232

/* 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
233

234 235
#define TASK_IDLE		(TASK_UNINTERRUPTIBLE | TASK_NOLOAD)

236 237
/* Convenience macros for the sake of wake_up */
#define TASK_NORMAL		(TASK_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE)
M
Matthew Wilcox 已提交
238
#define TASK_ALL		(TASK_NORMAL | __TASK_STOPPED | __TASK_TRACED)
239 240 241

/* get_task_state() */
#define TASK_REPORT		(TASK_RUNNING | TASK_INTERRUPTIBLE | \
M
Matthew Wilcox 已提交
242
				 TASK_UNINTERRUPTIBLE | __TASK_STOPPED | \
243
				 __TASK_TRACED | EXIT_ZOMBIE | EXIT_DEAD)
244

M
Matthew Wilcox 已提交
245 246
#define task_is_traced(task)	((task->state & __TASK_TRACED) != 0)
#define task_is_stopped(task)	((task->state & __TASK_STOPPED) != 0)
247
#define task_is_stopped_or_traced(task)	\
M
Matthew Wilcox 已提交
248
			((task->state & (__TASK_STOPPED | __TASK_TRACED)) != 0)
249
#define task_contributes_to_load(task)	\
250
				((task->state & TASK_UNINTERRUPTIBLE) != 0 && \
251 252
				 (task->flags & PF_FROZEN) == 0 && \
				 (task->state & TASK_NOLOAD) == 0)
L
Linus Torvalds 已提交
253

P
Peter Zijlstra 已提交
254 255 256 257 258 259 260 261 262 263
#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_;		\
264
		smp_store_mb((tsk)->state, (state_value));		\
P
Peter Zijlstra 已提交
265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285
	} 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_;		\
286
		smp_store_mb(current->state, (state_value));		\
P
Peter Zijlstra 已提交
287 288 289 290
	} while (0)

#else

L
Linus Torvalds 已提交
291 292 293
#define __set_task_state(tsk, state_value)		\
	do { (tsk)->state = (state_value); } while (0)
#define set_task_state(tsk, state_value)		\
294
	smp_store_mb((tsk)->state, (state_value))
L
Linus Torvalds 已提交
295

296 297 298 299 300 301 302 303 304 305 306
/*
 * 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
307
#define __set_current_state(state_value)		\
L
Linus Torvalds 已提交
308
	do { current->state = (state_value); } while (0)
P
Peter Zijlstra 已提交
309
#define set_current_state(state_value)			\
310
	smp_store_mb(current->state, (state_value))
L
Linus Torvalds 已提交
311

P
Peter Zijlstra 已提交
312 313
#endif

L
Linus Torvalds 已提交
314 315 316 317 318 319 320 321 322 323 324 325 326 327
/* 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;

328
struct task_struct;
L
Linus Torvalds 已提交
329

330 331 332 333
#ifdef CONFIG_PROVE_RCU
extern int lockdep_tasklist_lock_is_held(void);
#endif /* #ifdef CONFIG_PROVE_RCU */

L
Linus Torvalds 已提交
334 335
extern void sched_init(void);
extern void sched_init_smp(void);
336
extern asmlinkage void schedule_tail(struct task_struct *prev);
337
extern void init_idle(struct task_struct *idle, int cpu);
I
Ingo Molnar 已提交
338
extern void init_idle_bootup_task(struct task_struct *idle);
L
Linus Torvalds 已提交
339

340 341
extern cpumask_var_t cpu_isolated_map;

342
extern int runqueue_is_locked(int cpu);
I
Ingo Molnar 已提交
343

344
#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON)
345
extern void nohz_balance_enter_idle(int cpu);
346
extern void set_cpu_sd_state_idle(void);
347
extern int get_nohz_timer_target(void);
348
#else
349
static inline void nohz_balance_enter_idle(int cpu) { }
350
static inline void set_cpu_sd_state_idle(void) { }
351
#endif
L
Linus Torvalds 已提交
352

I
Ingo Molnar 已提交
353
/*
I
Ingo Molnar 已提交
354
 * Only dump TASK_* tasks. (0 for all tasks)
I
Ingo Molnar 已提交
355 356 357 358 359
 */
extern void show_state_filter(unsigned long state_filter);

static inline void show_state(void)
{
I
Ingo Molnar 已提交
360
	show_state_filter(0);
I
Ingo Molnar 已提交
361 362
}

L
Linus Torvalds 已提交
363 364 365 366 367 368 369 370 371 372 373 374 375 376
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);

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

377 378
extern void sched_show_task(struct task_struct *p);

379
#ifdef CONFIG_LOCKUP_DETECTOR
I
Ingo Molnar 已提交
380
extern void touch_softlockup_watchdog(void);
381
extern void touch_softlockup_watchdog_sync(void);
382
extern void touch_all_softlockup_watchdogs(void);
383 384 385
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
386
extern unsigned int  softlockup_panic;
387
void lockup_detector_init(void);
I
Ingo Molnar 已提交
388 389 390 391
#else
static inline void touch_softlockup_watchdog(void)
{
}
392 393 394
static inline void touch_softlockup_watchdog_sync(void)
{
}
395 396 397
static inline void touch_all_softlockup_watchdogs(void)
{
}
398 399 400
static inline void lockup_detector_init(void)
{
}
I
Ingo Molnar 已提交
401 402
#endif

403 404 405 406 407 408 409 410
#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
411 412
/* Attach to any functions which should be ignored in wchan output. */
#define __sched		__attribute__((__section__(".sched.text")))
413 414 415 416

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

L
Linus Torvalds 已提交
417 418 419 420
/* Is this address in the __sched functions? */
extern int in_sched_functions(unsigned long addr);

#define	MAX_SCHEDULE_TIMEOUT	LONG_MAX
421
extern signed long schedule_timeout(signed long timeout);
422
extern signed long schedule_timeout_interruptible(signed long timeout);
M
Matthew Wilcox 已提交
423
extern signed long schedule_timeout_killable(signed long timeout);
424
extern signed long schedule_timeout_uninterruptible(signed long timeout);
L
Linus Torvalds 已提交
425
asmlinkage void schedule(void);
426
extern void schedule_preempt_disabled(void);
L
Linus Torvalds 已提交
427

428 429 430 431 432 433 434
extern long io_schedule_timeout(long timeout);

static inline void io_schedule(void)
{
	io_schedule_timeout(MAX_SCHEDULE_TIMEOUT);
}

S
Serge E. Hallyn 已提交
435
struct nsproxy;
436
struct user_namespace;
L
Linus Torvalds 已提交
437

438 439
#ifdef CONFIG_MMU
extern void arch_pick_mmap_layout(struct mm_struct *mm);
L
Linus Torvalds 已提交
440 441 442 443 444 445 446
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);
447 448 449
#else
static inline void arch_pick_mmap_layout(struct mm_struct *mm) {}
#endif
L
Linus Torvalds 已提交
450

451 452 453 454
#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 */

455
/* mm flags */
H
Hugh Dickins 已提交
456

457
/* for SUID_DUMP_* above */
458
#define MMF_DUMPABLE_BITS 2
H
Hugh Dickins 已提交
459
#define MMF_DUMPABLE_MASK ((1 << MMF_DUMPABLE_BITS) - 1)
460

461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477
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);
}

478 479 480 481 482
/* 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
483
#define MMF_DUMP_ELF_HEADERS	6
484 485
#define MMF_DUMP_HUGETLB_PRIVATE 7
#define MMF_DUMP_HUGETLB_SHARED  8
H
Hugh Dickins 已提交
486

487
#define MMF_DUMP_FILTER_SHIFT	MMF_DUMPABLE_BITS
488
#define MMF_DUMP_FILTER_BITS	7
489 490 491
#define MMF_DUMP_FILTER_MASK \
	(((1 << MMF_DUMP_FILTER_BITS) - 1) << MMF_DUMP_FILTER_SHIFT)
#define MMF_DUMP_FILTER_DEFAULT \
492
	((1 << MMF_DUMP_ANON_PRIVATE) |	(1 << MMF_DUMP_ANON_SHARED) |\
493 494 495 496 497 498 499
	 (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
500 501
					/* leave room for more dump flags */
#define MMF_VM_MERGEABLE	16	/* KSM may merge identical pages */
A
Andrea Arcangeli 已提交
502
#define MMF_VM_HUGEPAGE		17	/* set when VM_HUGEPAGE is set on vma */
503
#define MMF_EXE_FILE_CHANGED	18	/* see prctl_set_mm_exe_file() */
H
Hugh Dickins 已提交
504

505 506
#define MMF_HAS_UPROBES		19	/* has uprobes */
#define MMF_RECALC_UPROBES	20	/* MMF_HAS_UPROBES can be wrong */
507

H
Hugh Dickins 已提交
508
#define MMF_INIT_MASK		(MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK)
509

L
Linus Torvalds 已提交
510 511 512 513
struct sighand_struct {
	atomic_t		count;
	struct k_sigaction	action[_NSIG];
	spinlock_t		siglock;
D
Davide Libenzi 已提交
514
	wait_queue_head_t	signalfd_wqh;
L
Linus Torvalds 已提交
515 516
};

517
struct pacct_struct {
518 519
	int			ac_flag;
	long			ac_exitcode;
520
	unsigned long		ac_mem;
521 522
	cputime_t		ac_utime, ac_stime;
	unsigned long		ac_minflt, ac_majflt;
523 524
};

525 526 527
struct cpu_itimer {
	cputime_t expires;
	cputime_t incr;
528 529
	u32 error;
	u32 incr_error;
530 531
};

532 533 534 535 536 537 538 539 540 541 542 543
/**
 * 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;
};

544 545 546 547 548
/**
 * 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
549
 *
550 551 552 553 554
 * 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
555 556 557 558 559 560 561 562 563 564 565 566 567
 * 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

568 569
#define INIT_CPUTIME	\
	(struct task_cputime) {					\
570 571
		.utime = 0,					\
		.stime = 0,					\
572 573 574
		.sum_exec_runtime = 0,				\
	}

575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591
/*
 * This is the atomic variant of task_cputime, which can be used for
 * storing and updating task_cputime statistics without locking.
 */
struct task_cputime_atomic {
	atomic64_t utime;
	atomic64_t stime;
	atomic64_t sum_exec_runtime;
};

#define INIT_CPUTIME_ATOMIC \
	(struct task_cputime_atomic) {				\
		.utime = ATOMIC64_INIT(0),			\
		.stime = ATOMIC64_INIT(0),			\
		.sum_exec_runtime = ATOMIC64_INIT(0),		\
	}

592 593 594 595 596 597
#ifdef CONFIG_PREEMPT_COUNT
#define PREEMPT_DISABLED	(1 + PREEMPT_ENABLED)
#else
#define PREEMPT_DISABLED	PREEMPT_ENABLED
#endif

P
Peter Zijlstra 已提交
598 599 600
/*
 * Disable preemption until the scheduler is running.
 * Reset by start_kernel()->sched_init()->init_idle().
P
Peter Zijlstra 已提交
601 602 603
 *
 * We include PREEMPT_ACTIVE to avoid cond_resched() from working
 * before the scheduler is active -- see should_resched().
P
Peter Zijlstra 已提交
604
 */
605
#define INIT_PREEMPT_COUNT	(PREEMPT_DISABLED + PREEMPT_ACTIVE)
P
Peter Zijlstra 已提交
606

607
/**
608
 * struct thread_group_cputimer - thread group interval timer counts
609
 * @cputime_atomic:	atomic thread group interval timers.
610 611
 * @running:		non-zero when there are timers running and
 * 			@cputime receives updates.
612 613
 *
 * This structure contains the version of task_cputime, above, that is
614
 * used for thread group CPU timer calculations.
615
 */
616
struct thread_group_cputimer {
617
	struct task_cputime_atomic cputime_atomic;
618
	int running;
619 620
};

621
#include <linux/rwsem.h>
622 623
struct autogroup;

L
Linus Torvalds 已提交
624
/*
625
 * NOTE! "signal_struct" does not have its own
L
Linus Torvalds 已提交
626 627 628 629 630 631
 * 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 {
632
	atomic_t		sigcnt;
L
Linus Torvalds 已提交
633
	atomic_t		live;
634
	int			nr_threads;
635
	struct list_head	thread_head;
L
Linus Torvalds 已提交
636 637 638 639

	wait_queue_head_t	wait_chldexit;	/* for wait4() */

	/* current thread group signal load-balancing target: */
640
	struct task_struct	*curr_target;
L
Linus Torvalds 已提交
641 642 643 644 645 646 647 648 649 650 651 652

	/* 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;
653
	struct task_struct	*group_exit_task;
L
Linus Torvalds 已提交
654 655 656 657 658

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

659 660 661 662 663 664 665 666 667 668 669 670
	/*
	 * 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
671
	/* POSIX.1b Interval Timers */
672 673
	int			posix_timer_id;
	struct list_head	posix_timers;
L
Linus Torvalds 已提交
674 675

	/* ITIMER_REAL timer for the process */
676
	struct hrtimer real_timer;
677
	struct pid *leader_pid;
678
	ktime_t it_real_incr;
L
Linus Torvalds 已提交
679

680 681 682 683 684 685
	/*
	 * 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
686

687
	/*
688 689
	 * Thread group totals for process CPU timers.
	 * See thread_group_cputimer(), et al, for details.
690
	 */
691
	struct thread_group_cputimer cputimer;
692 693 694 695 696 697

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

	struct list_head cpu_timers[3];

698
	struct pid *tty_old_pgrp;
699

L
Linus Torvalds 已提交
700 701 702 703 704
	/* boolean value for session group leader */
	int leader;

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

705 706 707
#ifdef CONFIG_SCHED_AUTOGROUP
	struct autogroup *autogroup;
#endif
L
Linus Torvalds 已提交
708 709 710 711 712 713
	/*
	 * 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.
	 */
714
	seqlock_t stats_lock;
715
	cputime_t utime, stime, cutime, cstime;
716 717
	cputime_t gtime;
	cputime_t cgtime;
718
#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
719
	struct cputime prev_cputime;
720
#endif
L
Linus Torvalds 已提交
721 722
	unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw;
	unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt;
723
	unsigned long inblock, oublock, cinblock, coublock;
J
Jiri Pirko 已提交
724
	unsigned long maxrss, cmaxrss;
725
	struct task_io_accounting ioac;
L
Linus Torvalds 已提交
726

727 728 729 730 731 732 733 734
	/*
	 * 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
735 736 737 738 739 740 741 742 743 744 745
	/*
	 * 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];

746 747 748
#ifdef CONFIG_BSD_PROCESS_ACCT
	struct pacct_struct pacct;	/* per-process accounting information */
#endif
749 750 751
#ifdef CONFIG_TASKSTATS
	struct taskstats *stats;
#endif
M
Miloslav Trmac 已提交
752 753
#ifdef CONFIG_AUDIT
	unsigned audit_tty;
754
	unsigned audit_tty_log_passwd;
M
Miloslav Trmac 已提交
755 756
	struct tty_audit_buf *tty_audit_buf;
#endif
757

758
	oom_flags_t oom_flags;
759 760 761
	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. */
762 763 764 765

	struct mutex cred_guard_mutex;	/* guard against foreign influences on
					 * credential calculations
					 * (notably. ptrace) */
L
Linus Torvalds 已提交
766 767 768 769 770 771
};

/*
 * Bits in flags field of signal_struct.
 */
#define SIGNAL_STOP_STOPPED	0x00000001 /* job control stop in effect */
772 773
#define SIGNAL_STOP_CONTINUED	0x00000002 /* SIGCONT since WCONTINUED reap */
#define SIGNAL_GROUP_EXIT	0x00000004 /* group exit in progress */
774
#define SIGNAL_GROUP_COREDUMP	0x00000008 /* coredump in progress */
775 776 777 778 779 780
/*
 * 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
781

782 783
#define SIGNAL_UNKILLABLE	0x00000040 /* for init: ignore fatal signals */

784 785 786 787 788 789 790
/* 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
791 792 793 794 795 796 797
/*
 * 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? */
798
#ifdef CONFIG_INOTIFY_USER
R
Robert Love 已提交
799 800 801
	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
802 803 804
#ifdef CONFIG_FANOTIFY
	atomic_t fanotify_listeners;
#endif
805
#ifdef CONFIG_EPOLL
806
	atomic_long_t epoll_watches; /* The number of file descriptors currently watched */
807
#endif
A
Alexey Dobriyan 已提交
808
#ifdef CONFIG_POSIX_MQUEUE
L
Linus Torvalds 已提交
809 810
	/* protected by mq_lock	*/
	unsigned long mq_bytes;	/* How many bytes can be allocated to mqueue? */
A
Alexey Dobriyan 已提交
811
#endif
L
Linus Torvalds 已提交
812 813 814 815 816 817 818 819
	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
820
	struct hlist_node uidhash_node;
821
	kuid_t uid;
822

823
#ifdef CONFIG_PERF_EVENTS
824 825
	atomic_long_t locked_vm;
#endif
L
Linus Torvalds 已提交
826 827
};

828
extern int uids_sysfs_init(void);
829

830
extern struct user_struct *find_user(kuid_t);
L
Linus Torvalds 已提交
831 832 833 834

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

835

L
Linus Torvalds 已提交
836 837 838
struct backing_dev_info;
struct reclaim_state;

839
#ifdef CONFIG_SCHED_INFO
L
Linus Torvalds 已提交
840 841
struct sched_info {
	/* cumulative counters */
842
	unsigned long pcount;	      /* # of times run on this cpu */
843
	unsigned long long run_delay; /* time spent waiting on a runqueue */
L
Linus Torvalds 已提交
844 845

	/* timestamps */
846 847
	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
848
};
849
#endif /* CONFIG_SCHED_INFO */
L
Linus Torvalds 已提交
850

851 852 853 854 855 856 857 858 859 860 861 862 863 864
#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).
	 */
865 866 867 868 869 870

	/*
	 * 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.
	 */
871
	u64 blkio_start;	/* Shared by blkio, swapin */
872 873 874 875 876 877
	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 */
878

879
	u64 freepages_start;
880 881
	u64 freepages_delay;	/* wait for memory reclaim */
	u32 freepages_count;	/* total count of memory reclaim */
882
};
883 884 885 886 887 888 889 890 891 892 893
#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;
894
#endif
895
}
896

I
Ingo Molnar 已提交
897 898 899 900 901
enum cpu_idle_type {
	CPU_IDLE,
	CPU_NOT_IDLE,
	CPU_NEWLY_IDLE,
	CPU_MAX_IDLE_TYPES
L
Linus Torvalds 已提交
902 903
};

904
/*
905
 * Increase resolution of cpu_capacity calculations
906
 */
907 908
#define SCHED_CAPACITY_SHIFT	10
#define SCHED_CAPACITY_SCALE	(1L << SCHED_CAPACITY_SHIFT)
L
Linus Torvalds 已提交
909

910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953
/*
 * Wake-queues are lists of tasks with a pending wakeup, whose
 * callers have already marked the task as woken internally,
 * and can thus carry on. A common use case is being able to
 * do the wakeups once the corresponding user lock as been
 * released.
 *
 * We hold reference to each task in the list across the wakeup,
 * thus guaranteeing that the memory is still valid by the time
 * the actual wakeups are performed in wake_up_q().
 *
 * One per task suffices, because there's never a need for a task to be
 * in two wake queues simultaneously; it is forbidden to abandon a task
 * in a wake queue (a call to wake_up_q() _must_ follow), so if a task is
 * already in a wake queue, the wakeup will happen soon and the second
 * waker can just skip it.
 *
 * The WAKE_Q macro declares and initializes the list head.
 * wake_up_q() does NOT reinitialize the list; it's expected to be
 * called near the end of a function, where the fact that the queue is
 * not used again will be easy to see by inspection.
 *
 * Note that this can cause spurious wakeups. schedule() callers
 * must ensure the call is done inside a loop, confirming that the
 * wakeup condition has in fact occurred.
 */
struct wake_q_node {
	struct wake_q_node *next;
};

struct wake_q_head {
	struct wake_q_node *first;
	struct wake_q_node **lastp;
};

#define WAKE_Q_TAIL ((struct wake_q_node *) 0x01)

#define WAKE_Q(name)					\
	struct wake_q_head name = { WAKE_Q_TAIL, &name.first }

extern void wake_q_add(struct wake_q_head *head,
		       struct task_struct *task);
extern void wake_up_q(struct wake_q_head *head);

954 955 956
/*
 * sched-domains (multiprocessor balancing) declarations:
 */
957
#ifdef CONFIG_SMP
P
Peter Zijlstra 已提交
958 959 960 961
#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 */
962
#define SD_BALANCE_WAKE		0x0010  /* Balance on wakeup */
P
Peter Zijlstra 已提交
963
#define SD_WAKE_AFFINE		0x0020	/* Wake task to waking CPU */
964
#define SD_SHARE_CPUCAPACITY	0x0080	/* Domain members share cpu power */
965
#define SD_SHARE_POWERDOMAIN	0x0100	/* Domain members share power domain */
P
Peter Zijlstra 已提交
966 967
#define SD_SHARE_PKG_RESOURCES	0x0200	/* Domain members share cpu pkg resources */
#define SD_SERIALIZE		0x0400	/* Only a single load balancing instance */
968
#define SD_ASYM_PACKING		0x0800  /* Place busy groups earlier in the domain */
P
Peter Zijlstra 已提交
969
#define SD_PREFER_SIBLING	0x1000	/* Prefer to place tasks in a sibling domain */
970
#define SD_OVERLAP		0x2000	/* sched_domains of this level overlap */
971
#define SD_NUMA			0x4000	/* cross-node balancing */
972

973
#ifdef CONFIG_SCHED_SMT
G
Guenter Roeck 已提交
974
static inline int cpu_smt_flags(void)
975
{
976
	return SD_SHARE_CPUCAPACITY | SD_SHARE_PKG_RESOURCES;
977 978 979 980
}
#endif

#ifdef CONFIG_SCHED_MC
G
Guenter Roeck 已提交
981
static inline int cpu_core_flags(void)
982 983 984 985 986 987
{
	return SD_SHARE_PKG_RESOURCES;
}
#endif

#ifdef CONFIG_NUMA
G
Guenter Roeck 已提交
988
static inline int cpu_numa_flags(void)
989 990 991 992
{
	return SD_NUMA;
}
#endif
993

994 995 996 997 998 999 1000 1001
struct sched_domain_attr {
	int relax_domain_level;
};

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

1002 1003
extern int sched_domain_level_max;

1004 1005
struct sched_group;

L
Linus Torvalds 已提交
1006 1007 1008
struct sched_domain {
	/* These fields must be setup */
	struct sched_domain *parent;	/* top domain must be null terminated */
1009
	struct sched_domain *child;	/* bottom domain must be null terminated */
L
Linus Torvalds 已提交
1010 1011 1012 1013 1014 1015
	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
1016 1017 1018 1019
	unsigned int busy_idx;
	unsigned int idle_idx;
	unsigned int newidle_idx;
	unsigned int wake_idx;
N
Nick Piggin 已提交
1020
	unsigned int forkexec_idx;
P
Peter Zijlstra 已提交
1021
	unsigned int smt_gain;
V
Vincent Guittot 已提交
1022 1023

	int nohz_idle;			/* NOHZ IDLE status */
L
Linus Torvalds 已提交
1024
	int flags;			/* See SD_* */
1025
	int level;
L
Linus Torvalds 已提交
1026 1027 1028 1029 1030 1031

	/* 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 */

1032
	/* idle_balance() stats */
1033
	u64 max_newidle_lb_cost;
1034
	unsigned long next_decay_max_lb_cost;
P
Peter Zijlstra 已提交
1035

L
Linus Torvalds 已提交
1036 1037
#ifdef CONFIG_SCHEDSTATS
	/* load_balance() stats */
1038 1039 1040 1041 1042 1043 1044 1045
	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
1046 1047

	/* Active load balancing */
1048 1049 1050
	unsigned int alb_count;
	unsigned int alb_failed;
	unsigned int alb_pushed;
L
Linus Torvalds 已提交
1051

1052
	/* SD_BALANCE_EXEC stats */
1053 1054 1055
	unsigned int sbe_count;
	unsigned int sbe_balanced;
	unsigned int sbe_pushed;
L
Linus Torvalds 已提交
1056

1057
	/* SD_BALANCE_FORK stats */
1058 1059 1060
	unsigned int sbf_count;
	unsigned int sbf_balanced;
	unsigned int sbf_pushed;
1061

L
Linus Torvalds 已提交
1062
	/* try_to_wake_up() stats */
1063 1064 1065
	unsigned int ttwu_wake_remote;
	unsigned int ttwu_move_affine;
	unsigned int ttwu_move_balance;
L
Linus Torvalds 已提交
1066
#endif
1067 1068 1069
#ifdef CONFIG_SCHED_DEBUG
	char *name;
#endif
1070 1071 1072 1073
	union {
		void *private;		/* used during construction */
		struct rcu_head rcu;	/* used during destruction */
	};
1074

1075
	unsigned int span_weight;
1076 1077 1078 1079 1080 1081 1082 1083
	/*
	 * 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
1084 1085
};

1086 1087
static inline struct cpumask *sched_domain_span(struct sched_domain *sd)
{
1088
	return to_cpumask(sd->span);
1089 1090
}

1091
extern void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[],
1092
				    struct sched_domain_attr *dattr_new);
P
Paul Jackson 已提交
1093

1094 1095 1096 1097
/* 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);

1098 1099
bool cpus_share_cache(int this_cpu, int that_cpu);

1100
typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
G
Guenter Roeck 已提交
1101
typedef int (*sched_domain_flags_f)(void);
1102 1103 1104 1105 1106 1107

#define SDTL_OVERLAP	0x01

struct sd_data {
	struct sched_domain **__percpu sd;
	struct sched_group **__percpu sg;
1108
	struct sched_group_capacity **__percpu sgc;
1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124
};

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);
1125
extern void wake_up_if_idle(int cpu);
1126 1127 1128 1129 1130 1131 1132

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

1133
#else /* CONFIG_SMP */
L
Linus Torvalds 已提交
1134

1135
struct sched_domain_attr;
1136

1137
static inline void
1138
partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[],
1139 1140
			struct sched_domain_attr *dattr_new)
{
1141
}
1142 1143 1144 1145 1146 1147

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

1148
#endif	/* !CONFIG_SMP */
L
Linus Torvalds 已提交
1149

1150

L
Linus Torvalds 已提交
1151 1152 1153
struct io_context;			/* See blkdev.h */


1154
#ifdef ARCH_HAS_PREFETCH_SWITCH_STACK
1155
extern void prefetch_stack(struct task_struct *t);
1156 1157 1158
#else
static inline void prefetch_stack(struct task_struct *t) { }
#endif
L
Linus Torvalds 已提交
1159 1160 1161

struct audit_context;		/* See audit.c */
struct mempolicy;
1162
struct pipe_inode_info;
1163
struct uts_namespace;
L
Linus Torvalds 已提交
1164

I
Ingo Molnar 已提交
1165
struct load_weight {
1166 1167
	unsigned long weight;
	u32 inv_weight;
I
Ingo Molnar 已提交
1168 1169
};

1170
struct sched_avg {
1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181
	u64 last_runnable_update;
	s64 decay_count;
	/*
	 * utilization_avg_contrib describes the amount of time that a
	 * sched_entity is running on a CPU. It is based on running_avg_sum
	 * and is scaled in the range [0..SCHED_LOAD_SCALE].
	 * load_avg_contrib described the amount of time that a sched_entity
	 * is runnable on a rq. It is based on both runnable_avg_sum and the
	 * weight of the task.
	 */
	unsigned long load_avg_contrib, utilization_avg_contrib;
1182 1183
	/*
	 * These sums represent an infinite geometric series and so are bound
1184
	 * above by 1024/(1-y).  Thus we only need a u32 to store them for all
1185
	 * choices of y < 1-2^(-32)*1024.
1186 1187 1188 1189 1190
	 * running_avg_sum reflects the time that the sched_entity is
	 * effectively running on the CPU.
	 * runnable_avg_sum represents the amount of time a sched_entity is on
	 * a runqueue which includes the running time that is monitored by
	 * running_avg_sum.
1191
	 */
1192
	u32 runnable_avg_sum, avg_period, running_avg_sum;
1193 1194
};

1195
#ifdef CONFIG_SCHEDSTATS
1196
struct sched_statistics {
I
Ingo Molnar 已提交
1197
	u64			wait_start;
1198
	u64			wait_max;
1199 1200
	u64			wait_count;
	u64			wait_sum;
1201 1202
	u64			iowait_count;
	u64			iowait_sum;
1203

I
Ingo Molnar 已提交
1204 1205
	u64			sleep_start;
	u64			sleep_max;
1206 1207 1208
	s64			sum_sleep_runtime;

	u64			block_start;
I
Ingo Molnar 已提交
1209 1210
	u64			block_max;
	u64			exec_max;
I
Ingo Molnar 已提交
1211
	u64			slice_max;
1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227

	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;
1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245
};
#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;
1246 1247
#endif

I
Ingo Molnar 已提交
1248
#ifdef CONFIG_FAIR_GROUP_SCHED
P
Peter Zijlstra 已提交
1249
	int			depth;
I
Ingo Molnar 已提交
1250 1251 1252 1253 1254 1255
	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
1256

1257
#ifdef CONFIG_SMP
1258
	/* Per-entity load-tracking */
1259 1260
	struct sched_avg	avg;
#endif
I
Ingo Molnar 已提交
1261
};
1262

P
Peter Zijlstra 已提交
1263 1264
struct sched_rt_entity {
	struct list_head run_list;
1265
	unsigned long timeout;
1266
	unsigned long watchdog_stamp;
1267
	unsigned int time_slice;
P
Peter Zijlstra 已提交
1268

1269
	struct sched_rt_entity *back;
1270
#ifdef CONFIG_RT_GROUP_SCHED
P
Peter Zijlstra 已提交
1271 1272 1273 1274 1275 1276
	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
1277 1278
};

1279 1280 1281 1282 1283
struct sched_dl_entity {
	struct rb_node	rb_node;

	/*
	 * Original scheduling parameters. Copied here from sched_attr
1284 1285
	 * during sched_setattr(), they will remain the same until
	 * the next sched_setattr().
1286 1287 1288
	 */
	u64 dl_runtime;		/* maximum runtime for each instance	*/
	u64 dl_deadline;	/* relative deadline of each instance	*/
1289
	u64 dl_period;		/* separation of two instances (period) */
1290
	u64 dl_bw;		/* dl_runtime / dl_deadline		*/
1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310

	/*
	 * 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;
1311 1312 1313
	 *
	 * @dl_boosted tells if we are boosted due to DI. If so we are
	 * outside bandwidth enforcement mechanism (but only until we
1314 1315 1316 1317
	 * exit the critical section);
	 *
	 * @dl_yielded tells if task gave up the cpu before consuming
	 * all its available runtime during the last job.
1318
	 */
1319
	int dl_throttled, dl_new, dl_boosted, dl_yielded;
1320 1321 1322 1323 1324 1325 1326

	/*
	 * Bandwidth enforcement timer. Each -deadline task has its
	 * own bandwidth to be enforced, thus we need one timer per task.
	 */
	struct hrtimer dl_timer;
};
1327

1328 1329 1330 1331 1332 1333 1334
union rcu_special {
	struct {
		bool blocked;
		bool need_qs;
	} b;
	short s;
};
1335 1336
struct rcu_node;

P
Peter Zijlstra 已提交
1337 1338 1339
enum perf_event_task_context {
	perf_invalid_context = -1,
	perf_hw_context = 0,
1340
	perf_sw_context,
P
Peter Zijlstra 已提交
1341 1342 1343
	perf_nr_task_contexts,
};

L
Linus Torvalds 已提交
1344 1345
struct task_struct {
	volatile long state;	/* -1 unrunnable, 0 runnable, >0 stopped */
R
Roman Zippel 已提交
1346
	void *stack;
L
Linus Torvalds 已提交
1347
	atomic_t usage;
1348 1349
	unsigned int flags;	/* per process flags, defined below */
	unsigned int ptrace;
L
Linus Torvalds 已提交
1350

1351
#ifdef CONFIG_SMP
P
Peter Zijlstra 已提交
1352
	struct llist_node wake_entry;
P
Peter Zijlstra 已提交
1353
	int on_cpu;
1354 1355 1356
	struct task_struct *last_wakee;
	unsigned long wakee_flips;
	unsigned long wakee_flip_decay_ts;
1357 1358

	int wake_cpu;
1359
#endif
P
Peter Zijlstra 已提交
1360
	int on_rq;
1361

1362
	int prio, static_prio, normal_prio;
1363
	unsigned int rt_priority;
1364
	const struct sched_class *sched_class;
I
Ingo Molnar 已提交
1365
	struct sched_entity se;
P
Peter Zijlstra 已提交
1366
	struct sched_rt_entity rt;
P
Peter Zijlstra 已提交
1367 1368 1369
#ifdef CONFIG_CGROUP_SCHED
	struct task_group *sched_task_group;
#endif
1370
	struct sched_dl_entity dl;
L
Linus Torvalds 已提交
1371

1372 1373 1374 1375 1376
#ifdef CONFIG_PREEMPT_NOTIFIERS
	/* list of struct preempt_notifier: */
	struct hlist_head preempt_notifiers;
#endif

1377
#ifdef CONFIG_BLK_DEV_IO_TRACE
1378
	unsigned int btrace_seq;
1379
#endif
L
Linus Torvalds 已提交
1380

1381
	unsigned int policy;
1382
	int nr_cpus_allowed;
L
Linus Torvalds 已提交
1383 1384
	cpumask_t cpus_allowed;

P
Paul E. McKenney 已提交
1385
#ifdef CONFIG_PREEMPT_RCU
P
Paul E. McKenney 已提交
1386
	int rcu_read_lock_nesting;
1387
	union rcu_special rcu_read_unlock_special;
1388
	struct list_head rcu_node_entry;
P
Paul E. McKenney 已提交
1389
	struct rcu_node *rcu_blocked_node;
1390
#endif /* #ifdef CONFIG_PREEMPT_RCU */
P
Paul E. McKenney 已提交
1391 1392 1393 1394
#ifdef CONFIG_TASKS_RCU
	unsigned long rcu_tasks_nvcsw;
	bool rcu_tasks_holdout;
	struct list_head rcu_tasks_holdout_list;
1395
	int rcu_tasks_idle_cpu;
P
Paul E. McKenney 已提交
1396
#endif /* #ifdef CONFIG_TASKS_RCU */
P
Paul E. McKenney 已提交
1397

1398
#ifdef CONFIG_SCHED_INFO
L
Linus Torvalds 已提交
1399 1400 1401 1402
	struct sched_info sched_info;
#endif

	struct list_head tasks;
1403
#ifdef CONFIG_SMP
1404
	struct plist_node pushable_tasks;
1405
	struct rb_node pushable_dl_tasks;
1406
#endif
L
Linus Torvalds 已提交
1407 1408

	struct mm_struct *mm, *active_mm;
D
Davidlohr Bueso 已提交
1409 1410 1411
	/* per-thread vma caching */
	u32 vmacache_seqnum;
	struct vm_area_struct *vmacache[VMACACHE_SIZE];
1412 1413 1414
#if defined(SPLIT_RSS_COUNTING)
	struct task_rss_stat	rss_stat;
#endif
L
Linus Torvalds 已提交
1415
/* task state */
1416
	int exit_state;
L
Linus Torvalds 已提交
1417 1418
	int exit_code, exit_signal;
	int pdeath_signal;  /*  The signal sent when the parent dies  */
1419
	unsigned long jobctl;	/* JOBCTL_*, siglock protected */
1420 1421

	/* Used for emulating ABI behavior of previous Linux versions */
1422
	unsigned int personality;
1423

1424 1425
	unsigned in_execve:1;	/* Tell the LSMs that the process is doing an
				 * execve */
1426 1427
	unsigned in_iowait:1;

1428 1429
	/* Revert to default priority/policy when forking */
	unsigned sched_reset_on_fork:1;
1430
	unsigned sched_contributes_to_load:1;
1431
	unsigned sched_migrated:1;
1432

1433 1434 1435
#ifdef CONFIG_MEMCG_KMEM
	unsigned memcg_kmem_skip_account:1;
#endif
1436 1437 1438
#ifdef CONFIG_COMPAT_BRK
	unsigned brk_randomized:1;
#endif
1439

1440 1441
	unsigned long atomic_flags; /* Flags needing atomic access. */

1442 1443
	struct restart_block restart_block;

L
Linus Torvalds 已提交
1444 1445
	pid_t pid;
	pid_t tgid;
1446

1447
#ifdef CONFIG_CC_STACKPROTECTOR
1448 1449
	/* Canary value for the -fstack-protector gcc feature */
	unsigned long stack_canary;
1450
#endif
1451
	/*
L
Linus Torvalds 已提交
1452
	 * pointers to (original) parent process, youngest child, younger sibling,
1453
	 * older sibling, respectively.  (p->father can be replaced with
R
Roland McGrath 已提交
1454
	 * p->real_parent->pid)
L
Linus Torvalds 已提交
1455
	 */
1456 1457
	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
1458
	/*
R
Roland McGrath 已提交
1459
	 * children/sibling forms the list of my natural children
L
Linus Torvalds 已提交
1460 1461 1462 1463 1464
	 */
	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
1465 1466 1467 1468 1469 1470 1471 1472
	/*
	 * 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
1473
	/* PID/PID hash table linkage. */
1474
	struct pid_link pids[PIDTYPE_MAX];
O
Oleg Nesterov 已提交
1475
	struct list_head thread_group;
1476
	struct list_head thread_node;
L
Linus Torvalds 已提交
1477 1478 1479 1480 1481

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

1482
	cputime_t utime, stime, utimescaled, stimescaled;
1483
	cputime_t gtime;
1484
#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
1485
	struct cputime prev_cputime;
1486 1487 1488 1489 1490 1491 1492 1493 1494
#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;
1495
#endif
L
Linus Torvalds 已提交
1496
	unsigned long nvcsw, nivcsw; /* context switch counts */
1497
	u64 start_time;		/* monotonic time in nsec */
1498
	u64 real_start_time;	/* boot based time in nsec */
L
Linus Torvalds 已提交
1499 1500 1501
/* mm fault and swap info: this can arguably be seen as either mm-specific or thread-specific */
	unsigned long min_flt, maj_flt;

1502
	struct task_cputime cputime_expires;
L
Linus Torvalds 已提交
1503 1504 1505
	struct list_head cpu_timers[3];

/* process credentials */
A
Arnd Bergmann 已提交
1506
	const struct cred __rcu *real_cred; /* objective and real subjective task
1507
					 * credentials (COW) */
A
Arnd Bergmann 已提交
1508
	const struct cred __rcu *cred;	/* effective (overridable) subjective task
1509
					 * credentials (COW) */
1510 1511 1512
	char comm[TASK_COMM_LEN]; /* executable name excluding path
				     - access with [gs]et_task_comm (which lock
				       it with task_lock())
1513
				     - initialized normally by setup_new_exec */
L
Linus Torvalds 已提交
1514
/* file system info */
1515
	struct nameidata *nameidata;
1516
#ifdef CONFIG_SYSVIPC
L
Linus Torvalds 已提交
1517 1518
/* ipc stuff */
	struct sysv_sem sysvsem;
1519
	struct sysv_shm sysvshm;
1520
#endif
1521
#ifdef CONFIG_DETECT_HUNG_TASK
1522 1523 1524
/* hung task detection */
	unsigned long last_switch_count;
#endif
L
Linus Torvalds 已提交
1525 1526 1527 1528 1529 1530
/* CPU-specific state of this task */
	struct thread_struct thread;
/* filesystem information */
	struct fs_struct *fs;
/* open file information */
	struct files_struct *files;
1531
/* namespaces */
S
Serge E. Hallyn 已提交
1532
	struct nsproxy *nsproxy;
L
Linus Torvalds 已提交
1533 1534 1535 1536 1537
/* signal handlers */
	struct signal_struct *signal;
	struct sighand_struct *sighand;

	sigset_t blocked, real_blocked;
1538
	sigset_t saved_sigmask;	/* restored if set_restore_sigmask() was used */
L
Linus Torvalds 已提交
1539 1540 1541 1542 1543 1544 1545
	struct sigpending pending;

	unsigned long sas_ss_sp;
	size_t sas_ss_size;
	int (*notifier)(void *priv);
	void *notifier_data;
	sigset_t *notifier_mask;
1546
	struct callback_head *task_works;
1547

L
Linus Torvalds 已提交
1548
	struct audit_context *audit_context;
A
Al Viro 已提交
1549
#ifdef CONFIG_AUDITSYSCALL
1550
	kuid_t loginuid;
1551
	unsigned int sessionid;
A
Al Viro 已提交
1552
#endif
1553
	struct seccomp seccomp;
L
Linus Torvalds 已提交
1554 1555 1556 1557

/* Thread group tracking */
   	u32 parent_exec_id;
   	u32 self_exec_id;
1558 1559
/* Protection of (de-)allocation: mm, files, fs, tty, keyrings, mems_allowed,
 * mempolicy */
L
Linus Torvalds 已提交
1560 1561
	spinlock_t alloc_lock;

1562
	/* Protection of the PI data structures: */
1563
	raw_spinlock_t pi_lock;
1564

1565 1566
	struct wake_q_node wake_q;

I
Ingo Molnar 已提交
1567 1568
#ifdef CONFIG_RT_MUTEXES
	/* PI waiters blocked on a rt_mutex held by this task */
1569 1570
	struct rb_root pi_waiters;
	struct rb_node *pi_waiters_leftmost;
I
Ingo Molnar 已提交
1571 1572 1573 1574
	/* Deadlock detection and priority inheritance handling */
	struct rt_mutex_waiter *pi_blocked_on;
#endif

1575 1576 1577 1578
#ifdef CONFIG_DEBUG_MUTEXES
	/* mutex deadlock detection */
	struct mutex_waiter *blocked_on;
#endif
1579 1580 1581 1582
#ifdef CONFIG_TRACE_IRQFLAGS
	unsigned int irq_events;
	unsigned long hardirq_enable_ip;
	unsigned long hardirq_disable_ip;
1583
	unsigned int hardirq_enable_event;
1584
	unsigned int hardirq_disable_event;
1585 1586
	int hardirqs_enabled;
	int hardirq_context;
1587 1588
	unsigned long softirq_disable_ip;
	unsigned long softirq_enable_ip;
1589
	unsigned int softirq_disable_event;
1590
	unsigned int softirq_enable_event;
1591
	int softirqs_enabled;
1592 1593
	int softirq_context;
#endif
I
Ingo Molnar 已提交
1594
#ifdef CONFIG_LOCKDEP
1595
# define MAX_LOCK_DEPTH 48UL
I
Ingo Molnar 已提交
1596 1597 1598
	u64 curr_chain_key;
	int lockdep_depth;
	unsigned int lockdep_recursion;
1599
	struct held_lock held_locks[MAX_LOCK_DEPTH];
1600
	gfp_t lockdep_reclaim_gfp;
I
Ingo Molnar 已提交
1601
#endif
1602

L
Linus Torvalds 已提交
1603 1604 1605
/* journalling filesystem info */
	void *journal_info;

1606
/* stacked block device info */
1607
	struct bio_list *bio_list;
1608

1609 1610 1611 1612 1613
#ifdef CONFIG_BLOCK
/* stack plugging */
	struct blk_plug *plug;
#endif

L
Linus Torvalds 已提交
1614 1615 1616 1617 1618 1619 1620 1621 1622
/* 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.  */
1623
	struct task_io_accounting ioac;
1624
#if defined(CONFIG_TASK_XACCT)
L
Linus Torvalds 已提交
1625 1626
	u64 acct_rss_mem1;	/* accumulated rss usage */
	u64 acct_vm_mem1;	/* accumulated virtual memory usage */
1627
	cputime_t acct_timexpd;	/* stime + utime since last update */
L
Linus Torvalds 已提交
1628 1629
#endif
#ifdef CONFIG_CPUSETS
1630
	nodemask_t mems_allowed;	/* Protected by alloc_lock */
1631
	seqcount_t mems_allowed_seq;	/* Seqence no to catch updates */
1632
	int cpuset_mem_spread_rotor;
1633
	int cpuset_slab_spread_rotor;
L
Linus Torvalds 已提交
1634
#endif
1635
#ifdef CONFIG_CGROUPS
1636
	/* Control Group info protected by css_set_lock */
A
Arnd Bergmann 已提交
1637
	struct css_set __rcu *cgroups;
1638 1639
	/* cg_list protected by css_set_lock and tsk->alloc_lock */
	struct list_head cg_list;
1640
#endif
1641
#ifdef CONFIG_FUTEX
1642
	struct robust_list_head __user *robust_list;
1643 1644 1645
#ifdef CONFIG_COMPAT
	struct compat_robust_list_head __user *compat_robust_list;
#endif
1646 1647
	struct list_head pi_state_list;
	struct futex_pi_state *pi_state_cache;
1648
#endif
1649
#ifdef CONFIG_PERF_EVENTS
P
Peter Zijlstra 已提交
1650
	struct perf_event_context *perf_event_ctxp[perf_nr_task_contexts];
1651 1652
	struct mutex perf_event_mutex;
	struct list_head perf_event_list;
1653
#endif
1654 1655 1656
#ifdef CONFIG_DEBUG_PREEMPT
	unsigned long preempt_disable_ip;
#endif
1657
#ifdef CONFIG_NUMA
1658
	struct mempolicy *mempolicy;	/* Protected by alloc_lock */
1659
	short il_next;
1660
	short pref_node_fork;
1661
#endif
1662 1663 1664
#ifdef CONFIG_NUMA_BALANCING
	int numa_scan_seq;
	unsigned int numa_scan_period;
1665
	unsigned int numa_scan_period_max;
1666
	int numa_preferred_nid;
1667
	unsigned long numa_migrate_retry;
1668
	u64 node_stamp;			/* migration stamp  */
1669 1670
	u64 last_task_numa_placement;
	u64 last_sum_exec_runtime;
1671
	struct callback_head numa_work;
1672

1673 1674 1675
	struct list_head numa_entry;
	struct numa_group *numa_group;

1676
	/*
1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688
	 * 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.
1689
	 */
1690
	unsigned long *numa_faults;
1691
	unsigned long total_numa_faults;
1692

1693 1694
	/*
	 * numa_faults_locality tracks if faults recorded during the last
1695 1696 1697
	 * scan window were remote/local or failed to migrate. The task scan
	 * period is adapted based on the locality of the faults with different
	 * weights depending on whether they were shared or private faults
1698
	 */
1699
	unsigned long numa_faults_locality[3];
1700

I
Ingo Molnar 已提交
1701
	unsigned long numa_pages_migrated;
1702 1703
#endif /* CONFIG_NUMA_BALANCING */

I
Ingo Molnar 已提交
1704
	struct rcu_head rcu;
1705 1706 1707 1708 1709

	/*
	 * cache last used pipe for splice
	 */
	struct pipe_inode_info *splice_pipe;
1710 1711 1712

	struct page_frag task_frag;

1713 1714
#ifdef	CONFIG_TASK_DELAY_ACCT
	struct task_delay_info *delays;
1715 1716 1717
#endif
#ifdef CONFIG_FAULT_INJECTION
	int make_it_fail;
1718
#endif
1719 1720 1721 1722 1723 1724
	/*
	 * 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;
1725
	unsigned long dirty_paused_when; /* start of a write-and-pause period */
1726

A
Arjan van de Ven 已提交
1727 1728 1729 1730
#ifdef CONFIG_LATENCYTOP
	int latency_record_count;
	struct latency_record latency_record[LT_SAVECOUNT];
#endif
1731 1732 1733 1734 1735 1736
	/*
	 * 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;
1737

1738 1739 1740
#ifdef CONFIG_KASAN
	unsigned int kasan_depth;
#endif
1741
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
D
Daniel Mack 已提交
1742
	/* Index of current stored address in ret_stack */
1743 1744 1745
	int curr_ret_stack;
	/* Stack of return addresses for return function tracing */
	struct ftrace_ret_stack	*ret_stack;
1746 1747
	/* time stamp for last schedule */
	unsigned long long ftrace_timestamp;
1748 1749 1750 1751 1752
	/*
	 * Number of functions that haven't been traced
	 * because of depth overrun.
	 */
	atomic_t trace_overrun;
1753 1754
	/* Pause for the tracing */
	atomic_t tracing_graph_pause;
1755
#endif
1756 1757 1758
#ifdef CONFIG_TRACING
	/* state flags for use by tracers */
	unsigned long trace;
1759
	/* bitmask and counter of trace recursion */
1760 1761
	unsigned long trace_recursion;
#endif /* CONFIG_TRACING */
1762
#ifdef CONFIG_MEMCG
1763
	struct memcg_oom_info {
1764 1765 1766
		struct mem_cgroup *memcg;
		gfp_t gfp_mask;
		int order;
1767 1768
		unsigned int may_oom:1;
	} memcg_oom;
1769
#endif
1770 1771 1772
#ifdef CONFIG_UPROBES
	struct uprobe_task *utask;
#endif
K
Kent Overstreet 已提交
1773 1774 1775 1776
#if defined(CONFIG_BCACHE) || defined(CONFIG_BCACHE_MODULE)
	unsigned int	sequential_io;
	unsigned int	sequential_io_avg;
#endif
P
Peter Zijlstra 已提交
1777 1778 1779
#ifdef CONFIG_DEBUG_ATOMIC_SLEEP
	unsigned long	task_state_change;
#endif
1780
	int pagefault_disabled;
L
Linus Torvalds 已提交
1781 1782
};

1783
/* Future-safe accessor for struct task_struct's cpus_allowed. */
1784
#define tsk_cpus_allowed(tsk) (&(tsk)->cpus_allowed)
1785

1786 1787
#define TNF_MIGRATED	0x01
#define TNF_NO_GROUP	0x02
1788
#define TNF_SHARED	0x04
1789
#define TNF_FAULT_LOCAL	0x08
1790
#define TNF_MIGRATE_FAIL 0x10
1791

1792
#ifdef CONFIG_NUMA_BALANCING
1793
extern void task_numa_fault(int last_node, int node, int pages, int flags);
1794
extern pid_t task_numa_group_id(struct task_struct *p);
1795
extern void set_numabalancing_state(bool enabled);
1796
extern void task_numa_free(struct task_struct *p);
1797 1798
extern bool should_numa_migrate_memory(struct task_struct *p, struct page *page,
					int src_nid, int dst_cpu);
1799
#else
1800
static inline void task_numa_fault(int last_node, int node, int pages,
1801
				   int flags)
1802 1803
{
}
1804 1805 1806 1807
static inline pid_t task_numa_group_id(struct task_struct *p)
{
	return 0;
}
1808 1809 1810
static inline void set_numabalancing_state(bool enabled)
{
}
1811 1812 1813
static inline void task_numa_free(struct task_struct *p)
{
}
1814 1815 1816 1817 1818
static inline bool should_numa_migrate_memory(struct task_struct *p,
				struct page *page, int src_nid, int dst_cpu)
{
	return true;
}
1819 1820
#endif

A
Alexey Dobriyan 已提交
1821
static inline struct pid *task_pid(struct task_struct *task)
1822 1823 1824 1825
{
	return task->pids[PIDTYPE_PID].pid;
}

A
Alexey Dobriyan 已提交
1826
static inline struct pid *task_tgid(struct task_struct *task)
1827 1828 1829 1830
{
	return task->group_leader->pids[PIDTYPE_PID].pid;
}

1831 1832 1833 1834 1835
/*
 * 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
1836
static inline struct pid *task_pgrp(struct task_struct *task)
1837 1838 1839 1840
{
	return task->group_leader->pids[PIDTYPE_PGID].pid;
}

A
Alexey Dobriyan 已提交
1841
static inline struct pid *task_session(struct task_struct *task)
1842 1843 1844 1845
{
	return task->group_leader->pids[PIDTYPE_SID].pid;
}

1846 1847 1848 1849 1850 1851 1852
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
1853 1854
 * task_xid_vnr()    : virtual id, i.e. the id seen from the pid namespace of
 *                     current.
1855 1856 1857 1858 1859 1860
 * 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
 */
1861 1862
pid_t __task_pid_nr_ns(struct task_struct *task, enum pid_type type,
			struct pid_namespace *ns);
1863

A
Alexey Dobriyan 已提交
1864
static inline pid_t task_pid_nr(struct task_struct *tsk)
1865 1866 1867 1868
{
	return tsk->pid;
}

1869 1870 1871 1872 1873
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);
}
1874 1875 1876

static inline pid_t task_pid_vnr(struct task_struct *tsk)
{
1877
	return __task_pid_nr_ns(tsk, PIDTYPE_PID, NULL);
1878 1879 1880
}


A
Alexey Dobriyan 已提交
1881
static inline pid_t task_tgid_nr(struct task_struct *tsk)
1882 1883 1884 1885
{
	return tsk->tgid;
}

1886
pid_t task_tgid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns);
1887 1888 1889 1890 1891 1892 1893

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


1894
static inline int pid_alive(const struct task_struct *p);
1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911
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);
}

1912 1913
static inline pid_t task_pgrp_nr_ns(struct task_struct *tsk,
					struct pid_namespace *ns)
1914
{
1915
	return __task_pid_nr_ns(tsk, PIDTYPE_PGID, ns);
1916 1917 1918 1919
}

static inline pid_t task_pgrp_vnr(struct task_struct *tsk)
{
1920
	return __task_pid_nr_ns(tsk, PIDTYPE_PGID, NULL);
1921 1922 1923
}


1924 1925
static inline pid_t task_session_nr_ns(struct task_struct *tsk,
					struct pid_namespace *ns)
1926
{
1927
	return __task_pid_nr_ns(tsk, PIDTYPE_SID, ns);
1928 1929 1930 1931
}

static inline pid_t task_session_vnr(struct task_struct *tsk)
{
1932
	return __task_pid_nr_ns(tsk, PIDTYPE_SID, NULL);
1933 1934
}

1935 1936 1937 1938 1939
/* obsolete, do not use */
static inline pid_t task_pgrp_nr(struct task_struct *tsk)
{
	return task_pgrp_nr_ns(tsk, &init_pid_ns);
}
1940

L
Linus Torvalds 已提交
1941 1942 1943 1944 1945 1946 1947
/**
 * 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.
1948 1949
 *
 * Return: 1 if the process is alive. 0 otherwise.
L
Linus Torvalds 已提交
1950
 */
1951
static inline int pid_alive(const struct task_struct *p)
L
Linus Torvalds 已提交
1952
{
1953
	return p->pids[PIDTYPE_PID].pid != NULL;
L
Linus Torvalds 已提交
1954 1955
}

1956
/**
1957
 * is_global_init - check if a task structure is init
1958 1959 1960
 * @tsk: Task structure to be checked.
 *
 * Check if a task structure is the first user space task the kernel created.
1961 1962
 *
 * Return: 1 if the task structure is init. 0 otherwise.
1963
 */
A
Alexey Dobriyan 已提交
1964
static inline int is_global_init(struct task_struct *tsk)
1965 1966 1967
{
	return tsk->pid == 1;
}
1968

1969 1970
extern struct pid *cad_pid;

L
Linus Torvalds 已提交
1971 1972
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
1973

1974
extern void __put_task_struct(struct task_struct *t);
I
Ingo Molnar 已提交
1975 1976 1977 1978

static inline void put_task_struct(struct task_struct *t)
{
	if (atomic_dec_and_test(&t->usage))
1979
		__put_task_struct(t);
I
Ingo Molnar 已提交
1980
}
L
Linus Torvalds 已提交
1981

1982 1983 1984 1985 1986 1987 1988
#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
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
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;
}
2007 2008 2009 2010 2011 2012

static inline cputime_t task_gtime(struct task_struct *t)
{
	return t->gtime;
}
#endif
2013 2014
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);
2015

L
Linus Torvalds 已提交
2016 2017 2018 2019
/*
 * Per process flags
 */
#define PF_EXITING	0x00000004	/* getting shut down */
2020
#define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
2021
#define PF_VCPU		0x00000010	/* I'm a virtual CPU */
T
Tejun Heo 已提交
2022
#define PF_WQ_WORKER	0x00000020	/* I'm a workqueue worker */
L
Linus Torvalds 已提交
2023
#define PF_FORKNOEXEC	0x00000040	/* forked but didn't exec */
2024
#define PF_MCE_PROCESS  0x00000080      /* process policy on mce errors */
L
Linus Torvalds 已提交
2025 2026 2027 2028
#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 */
2029
#define PF_NPROC_EXCEEDED 0x00001000	/* set_user noticed that RLIMIT_NPROC was exceeded */
L
Linus Torvalds 已提交
2030
#define PF_USED_MATH	0x00002000	/* if unset the fpu must be initialized before use */
2031
#define PF_USED_ASYNC	0x00004000	/* used async_schedule*(), used by module init */
L
Linus Torvalds 已提交
2032 2033 2034 2035
#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 */
2036
#define PF_MEMALLOC_NOIO 0x00080000	/* Allocating memory without IO involved */
L
Linus Torvalds 已提交
2037
#define PF_LESS_THROTTLE 0x00100000	/* Throttle me less: I clean memory */
2038
#define PF_KTHREAD	0x00200000	/* I am a kernel thread */
J
Jens Axboe 已提交
2039 2040
#define PF_RANDOMIZE	0x00400000	/* randomize virtual address space */
#define PF_SWAPWRITE	0x00800000	/* Allowed to write to swap */
2041
#define PF_NO_SETAFFINITY 0x04000000	/* Userland is not allowed to meddle with cpus_allowed */
2042
#define PF_MCE_EARLY    0x08000000      /* Early kill for mce process policy */
2043
#define PF_MUTEX_TESTER	0x20000000	/* Thread belongs to the rt mutex tester */
2044
#define PF_FREEZER_SKIP	0x40000000	/* Freezer should not count it as freezable */
2045
#define PF_SUSPEND_TASK 0x80000000      /* this thread called freeze_processes and should not be frozen */
L
Linus Torvalds 已提交
2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071

/*
 * 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)

2072 2073 2074
/* __GFP_IO isn't allowed if PF_MEMALLOC_NOIO is set in current->flags
 * __GFP_FS is also cleared as it implies __GFP_IO.
 */
2075 2076 2077
static inline gfp_t memalloc_noio_flags(gfp_t flags)
{
	if (unlikely(current->flags & PF_MEMALLOC_NOIO))
2078
		flags &= ~(__GFP_IO | __GFP_FS);
2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093
	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;
}

2094
/* Per-process atomic flags. */
2095
#define PFA_NO_NEW_PRIVS 0	/* May not gain new privileges. */
2096 2097 2098
#define PFA_SPREAD_PAGE  1      /* Spread page cache over cpuset */
#define PFA_SPREAD_SLAB  2      /* Spread some slab caches over cpuset */

2099

2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111
#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)
2112

2113 2114 2115 2116 2117 2118 2119
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)
2120

2121
/*
2122
 * task->jobctl flags
2123
 */
2124
#define JOBCTL_STOP_SIGMASK	0xffff	/* signr of the last group stop */
2125

2126 2127 2128
#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 */
2129
#define JOBCTL_TRAP_STOP_BIT	19	/* trap for STOP */
2130
#define JOBCTL_TRAP_NOTIFY_BIT	20	/* trap for NOTIFY */
2131
#define JOBCTL_TRAPPING_BIT	21	/* switching to TRACED */
T
Tejun Heo 已提交
2132
#define JOBCTL_LISTENING_BIT	22	/* ptracer is listening for events */
2133

2134 2135 2136 2137 2138 2139 2140
#define JOBCTL_STOP_DEQUEUED	(1UL << JOBCTL_STOP_DEQUEUED_BIT)
#define JOBCTL_STOP_PENDING	(1UL << JOBCTL_STOP_PENDING_BIT)
#define JOBCTL_STOP_CONSUME	(1UL << JOBCTL_STOP_CONSUME_BIT)
#define JOBCTL_TRAP_STOP	(1UL << JOBCTL_TRAP_STOP_BIT)
#define JOBCTL_TRAP_NOTIFY	(1UL << JOBCTL_TRAP_NOTIFY_BIT)
#define JOBCTL_TRAPPING		(1UL << JOBCTL_TRAPPING_BIT)
#define JOBCTL_LISTENING	(1UL << JOBCTL_LISTENING_BIT)
2141

2142
#define JOBCTL_TRAP_MASK	(JOBCTL_TRAP_STOP | JOBCTL_TRAP_NOTIFY)
2143
#define JOBCTL_PENDING_MASK	(JOBCTL_STOP_PENDING | JOBCTL_TRAP_MASK)
2144

2145
extern bool task_set_jobctl_pending(struct task_struct *task,
2146
				    unsigned long mask);
2147
extern void task_clear_jobctl_trapping(struct task_struct *task);
2148
extern void task_clear_jobctl_pending(struct task_struct *task,
2149
				      unsigned long mask);
2150

2151 2152
static inline void rcu_copy_process(struct task_struct *p)
{
P
Paul E. McKenney 已提交
2153
#ifdef CONFIG_PREEMPT_RCU
2154
	p->rcu_read_lock_nesting = 0;
2155
	p->rcu_read_unlock_special.s = 0;
2156
	p->rcu_blocked_node = NULL;
2157
	INIT_LIST_HEAD(&p->rcu_node_entry);
P
Paul E. McKenney 已提交
2158 2159 2160 2161
#endif /* #ifdef CONFIG_PREEMPT_RCU */
#ifdef CONFIG_TASKS_RCU
	p->rcu_tasks_holdout = false;
	INIT_LIST_HEAD(&p->rcu_tasks_holdout_list);
2162
	p->rcu_tasks_idle_cpu = -1;
P
Paul E. McKenney 已提交
2163
#endif /* #ifdef CONFIG_TASKS_RCU */
2164 2165
}

2166 2167 2168 2169 2170 2171 2172
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;
}

2173 2174
extern int cpuset_cpumask_can_shrink(const struct cpumask *cur,
				     const struct cpumask *trial);
2175 2176
extern int task_can_attach(struct task_struct *p,
			   const struct cpumask *cs_cpus_allowed);
L
Linus Torvalds 已提交
2177
#ifdef CONFIG_SMP
2178 2179 2180
extern void do_set_cpus_allowed(struct task_struct *p,
			       const struct cpumask *new_mask);

2181
extern int set_cpus_allowed_ptr(struct task_struct *p,
2182
				const struct cpumask *new_mask);
L
Linus Torvalds 已提交
2183
#else
2184 2185 2186 2187
static inline void do_set_cpus_allowed(struct task_struct *p,
				      const struct cpumask *new_mask)
{
}
2188
static inline int set_cpus_allowed_ptr(struct task_struct *p,
2189
				       const struct cpumask *new_mask)
L
Linus Torvalds 已提交
2190
{
2191
	if (!cpumask_test_cpu(0, new_mask))
L
Linus Torvalds 已提交
2192 2193 2194 2195
		return -EINVAL;
	return 0;
}
#endif
2196

2197
#ifdef CONFIG_NO_HZ_COMMON
2198 2199 2200 2201 2202
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) { }
2203
#endif /* CONFIG_NO_HZ_COMMON */
2204

2205
#ifndef CONFIG_CPUMASK_OFFSTACK
2206 2207 2208 2209
static inline int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask)
{
	return set_cpus_allowed_ptr(p, &new_mask);
}
2210
#endif
L
Linus Torvalds 已提交
2211

2212
/*
2213 2214 2215 2216 2217 2218
 * 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.
2219
 */
2220
extern unsigned long long notrace sched_clock(void);
2221
/*
2222
 * See the comment in kernel/sched/clock.c
2223 2224 2225
 */
extern u64 cpu_clock(int cpu);
extern u64 local_clock(void);
2226
extern u64 running_clock(void);
2227 2228
extern u64 sched_clock_cpu(int cpu);

2229

2230
extern void sched_clock_init(void);
2231

2232
#ifndef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244
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
2245 2246 2247 2248 2249 2250
/*
 * 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:
 */
2251 2252 2253
extern int sched_clock_stable(void);
extern void set_sched_clock_stable(void);
extern void clear_sched_clock_stable(void);
2254

2255 2256 2257 2258 2259
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

2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272
#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

2273
extern unsigned long long
2274
task_sched_runtime(struct task_struct *task);
L
Linus Torvalds 已提交
2275 2276 2277 2278 2279 2280 2281 2282

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

2283 2284
extern void sched_clock_idle_sleep_event(void);
extern void sched_clock_idle_wakeup_event(u64 delta_ns);
2285

L
Linus Torvalds 已提交
2286 2287 2288 2289 2290 2291
#ifdef CONFIG_HOTPLUG_CPU
extern void idle_task_exit(void);
#else
static inline void idle_task_exit(void) {}
#endif

2292
#if defined(CONFIG_NO_HZ_COMMON) && defined(CONFIG_SMP)
2293
extern void wake_up_nohz_cpu(int cpu);
2294
#else
2295
static inline void wake_up_nohz_cpu(int cpu) { }
2296 2297
#endif

2298 2299
#ifdef CONFIG_NO_HZ_FULL
extern bool sched_can_stop_tick(void);
2300
extern u64 scheduler_tick_max_deferment(void);
2301 2302
#else
static inline bool sched_can_stop_tick(void) { return false; }
2303 2304
#endif

2305 2306 2307 2308 2309 2310 2311
#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);
2312
extern int proc_sched_autogroup_set_nice(struct task_struct *p, int nice);
2313 2314 2315 2316 2317 2318 2319 2320
#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

2321
extern int yield_to(struct task_struct *p, bool preempt);
2322 2323
extern void set_user_nice(struct task_struct *p, long nice);
extern int task_prio(const struct task_struct *p);
2324 2325 2326 2327 2328 2329 2330 2331 2332 2333
/**
 * 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);
}
2334 2335
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
2336
extern int idle_cpu(int cpu);
2337 2338
extern int sched_setscheduler(struct task_struct *, int,
			      const struct sched_param *);
2339
extern int sched_setscheduler_nocheck(struct task_struct *, int,
2340
				      const struct sched_param *);
2341 2342
extern int sched_setattr(struct task_struct *,
			 const struct sched_attr *);
2343
extern struct task_struct *idle_task(int cpu);
2344 2345
/**
 * is_idle_task - is the specified task an idle task?
2346
 * @p: the task in question.
2347 2348
 *
 * Return: 1 if @p is an idle task. 0 otherwise.
2349
 */
2350
static inline bool is_idle_task(const struct task_struct *p)
2351 2352 2353
{
	return p->pid == 0;
}
2354 2355
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
2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378

void yield(void);

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;

2379 2380 2381 2382 2383 2384 2385
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
2386 2387
 * find_task_by_vpid():
 *      finds a task by its virtual pid
2388
 *
2389
 * see also find_vpid() etc in include/linux/pid.h
2390 2391
 */

2392 2393 2394
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);
2395

L
Linus Torvalds 已提交
2396
/* per-UID process charging. */
2397
extern struct user_struct * alloc_uid(kuid_t);
L
Linus Torvalds 已提交
2398 2399 2400 2401 2402 2403 2404 2405 2406
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
2407
extern void xtime_update(unsigned long ticks);
L
Linus Torvalds 已提交
2408

2409 2410
extern int wake_up_state(struct task_struct *tsk, unsigned int state);
extern int wake_up_process(struct task_struct *tsk);
2411
extern void wake_up_new_task(struct task_struct *tsk);
L
Linus Torvalds 已提交
2412 2413 2414 2415 2416
#ifdef CONFIG_SMP
 extern void kick_process(struct task_struct *tsk);
#else
 static inline void kick_process(struct task_struct *tsk) { }
#endif
2417
extern int sched_fork(unsigned long clone_flags, struct task_struct *p);
2418
extern void sched_dead(struct task_struct *p);
L
Linus Torvalds 已提交
2419 2420 2421

extern void proc_caches_init(void);
extern void flush_signals(struct task_struct *);
2422
extern void ignore_signals(struct task_struct *);
L
Linus Torvalds 已提交
2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435
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;
2436
}
L
Linus Torvalds 已提交
2437 2438 2439 2440 2441 2442 2443 2444

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 *);
2445 2446
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);
2447 2448
extern int kill_pid_info_as_cred(int, struct siginfo *, struct pid *,
				const struct cred *, u32);
2449 2450
extern int kill_pgrp(struct pid *pid, int sig, int priv);
extern int kill_pid(struct pid *pid, int sig, int priv);
2451
extern int kill_proc_info(int, struct siginfo *, pid_t);
2452
extern __must_check bool do_notify_parent(struct task_struct *, int);
2453
extern void __wake_up_parent(struct task_struct *p, struct task_struct *parent);
L
Linus Torvalds 已提交
2454 2455
extern void force_sig(int, struct task_struct *);
extern int send_sig(int, struct task_struct *, int);
2456
extern int zap_other_threads(struct task_struct *p);
L
Linus Torvalds 已提交
2457 2458
extern struct sigqueue *sigqueue_alloc(void);
extern void sigqueue_free(struct sigqueue *);
2459
extern int send_sigqueue(struct sigqueue *,  struct task_struct *, int group);
2460
extern int do_sigaction(int, struct k_sigaction *, struct k_sigaction *);
L
Linus Torvalds 已提交
2461

A
Al Viro 已提交
2462 2463 2464
static inline void restore_saved_sigmask(void)
{
	if (test_and_clear_restore_sigmask())
2465
		__set_current_blocked(&current->saved_sigmask);
A
Al Viro 已提交
2466 2467
}

A
Al Viro 已提交
2468 2469 2470 2471 2472 2473 2474 2475
static inline sigset_t *sigmask_to_save(void)
{
	sigset_t *res = &current->blocked;
	if (unlikely(test_restore_sigmask()))
		res = &current->saved_sigmask;
	return res;
}

2476 2477 2478 2479 2480
static inline int kill_cad_pid(int sig, int priv)
{
	return kill_pid(cad_pid, sig, priv);
}

L
Linus Torvalds 已提交
2481 2482 2483 2484 2485
/* 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)

2486 2487 2488
/*
 * True if we are on the alternate signal stack.
 */
L
Linus Torvalds 已提交
2489 2490
static inline int on_sig_stack(unsigned long sp)
{
2491 2492 2493 2494 2495 2496 2497
#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
2498 2499 2500 2501
}

static inline int sas_ss_flags(unsigned long sp)
{
2502 2503 2504 2505
	if (!current->sas_ss_size)
		return SS_DISABLE;

	return on_sig_stack(sp) ? SS_ONSTACK : 0;
L
Linus Torvalds 已提交
2506 2507
}

A
Al Viro 已提交
2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518
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
2519 2520 2521 2522 2523 2524
/*
 * Routines for handling mm_structs
 */
extern struct mm_struct * mm_alloc(void);

/* mmdrop drops the mm and the page tables */
2525
extern void __mmdrop(struct mm_struct *);
L
Linus Torvalds 已提交
2526 2527
static inline void mmdrop(struct mm_struct * mm)
{
I
Ingo Molnar 已提交
2528
	if (unlikely(atomic_dec_and_test(&mm->mm_count)))
L
Linus Torvalds 已提交
2529 2530 2531 2532 2533 2534 2535
		__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);
2536 2537 2538 2539 2540 2541
/*
 * 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
2542 2543 2544
/* Remove the current tasks stale references to the old mm_struct */
extern void mm_release(struct task_struct *, struct mm_struct *);

2545 2546 2547 2548
#ifdef CONFIG_HAVE_COPY_THREAD_TLS
extern int copy_thread_tls(unsigned long, unsigned long, unsigned long,
			struct task_struct *, unsigned long);
#else
A
Alexey Dobriyan 已提交
2549
extern int copy_thread(unsigned long, unsigned long, unsigned long,
2550
			struct task_struct *);
2551 2552 2553 2554 2555 2556 2557 2558 2559 2560

/* Architectures that haven't opted into copy_thread_tls get the tls argument
 * via pt_regs, so ignore the tls argument passed via C. */
static inline int copy_thread_tls(
		unsigned long clone_flags, unsigned long sp, unsigned long arg,
		struct task_struct *p, unsigned long tls)
{
	return copy_thread(clone_flags, sp, arg, p);
}
#endif
L
Linus Torvalds 已提交
2561 2562 2563 2564
extern void flush_thread(void);
extern void exit_thread(void);

extern void exit_files(struct task_struct *);
2565
extern void __cleanup_sighand(struct sighand_struct *);
2566

L
Linus Torvalds 已提交
2567
extern void exit_itimers(struct signal_struct *);
2568
extern void flush_itimer_signals(void);
L
Linus Torvalds 已提交
2569

2570
extern void do_group_exit(int);
L
Linus Torvalds 已提交
2571

2572
extern int do_execve(struct filename *,
2573
		     const char __user * const __user *,
2574
		     const char __user * const __user *);
2575 2576 2577 2578
extern int do_execveat(int, struct filename *,
		       const char __user * const __user *,
		       const char __user * const __user *,
		       int);
2579
extern long _do_fork(unsigned long, unsigned long, unsigned long, int __user *, int __user *, unsigned long);
2580
extern long do_fork(unsigned long, unsigned long, unsigned long, int __user *, int __user *);
2581
struct task_struct *fork_idle(int);
2582
extern pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
L
Linus Torvalds 已提交
2583

2584 2585 2586 2587 2588
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);
}
2589
extern char *get_task_comm(char *to, struct task_struct *tsk);
L
Linus Torvalds 已提交
2590 2591

#ifdef CONFIG_SMP
2592
void scheduler_ipi(void);
R
Roland McGrath 已提交
2593
extern unsigned long wait_task_inactive(struct task_struct *, long match_state);
L
Linus Torvalds 已提交
2594
#else
2595
static inline void scheduler_ipi(void) { }
R
Roland McGrath 已提交
2596 2597 2598 2599 2600
static inline unsigned long wait_task_inactive(struct task_struct *p,
					       long match_state)
{
	return 1;
}
L
Linus Torvalds 已提交
2601 2602
#endif

2603 2604 2605
#define tasklist_empty() \
	list_empty(&init_task.tasks)

2606 2607
#define next_task(p) \
	list_entry_rcu((p)->tasks.next, struct task_struct, tasks)
L
Linus Torvalds 已提交
2608 2609 2610 2611

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

2612
extern bool current_is_single_threaded(void);
D
David Howells 已提交
2613

L
Linus Torvalds 已提交
2614 2615 2616 2617 2618 2619 2620 2621 2622 2623
/*
 * 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)

2624 2625 2626 2627 2628 2629 2630 2631 2632 2633
#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)

2634 2635
static inline int get_nr_threads(struct task_struct *tsk)
{
2636
	return tsk->signal->nr_threads;
2637 2638
}

2639 2640 2641 2642
static inline bool thread_group_leader(struct task_struct *p)
{
	return p->exit_signal >= 0;
}
L
Linus Torvalds 已提交
2643

2644 2645 2646 2647 2648 2649
/* 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.
 */
2650
static inline bool has_group_leader_pid(struct task_struct *p)
2651
{
2652
	return task_pid(p) == p->signal->leader_pid;
2653 2654
}

2655
static inline
2656
bool same_thread_group(struct task_struct *p1, struct task_struct *p2)
2657
{
2658
	return p1->signal == p2->signal;
2659 2660
}

2661
static inline struct task_struct *next_thread(const struct task_struct *p)
O
Oleg Nesterov 已提交
2662
{
2663 2664
	return list_entry_rcu(p->thread_group.next,
			      struct task_struct, thread_group);
O
Oleg Nesterov 已提交
2665 2666
}

A
Alexey Dobriyan 已提交
2667
static inline int thread_group_empty(struct task_struct *p)
L
Linus Torvalds 已提交
2668
{
O
Oleg Nesterov 已提交
2669
	return list_empty(&p->thread_group);
L
Linus Torvalds 已提交
2670 2671 2672 2673 2674 2675
}

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

/*
2676
 * Protects ->fs, ->files, ->mm, ->group_info, ->comm, keyring
2677
 * subscriptions and synchronises with wait4().  Also used in procfs.  Also
2678
 * pins the final release of task.io_context.  Also protects ->cpuset and
O
Oleg Nesterov 已提交
2679
 * ->cgroup.subsys[]. And ->vfork_done.
L
Linus Torvalds 已提交
2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694
 *
 * 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);
}

2695
extern struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
2696 2697
							unsigned long *flags);

2698 2699 2700 2701 2702 2703 2704 2705 2706
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;
}
2707

2708 2709 2710 2711 2712 2713
static inline void unlock_task_sighand(struct task_struct *tsk,
						unsigned long *flags)
{
	spin_unlock_irqrestore(&tsk->sighand->siglock, *flags);
}

2714
/**
2715 2716
 * threadgroup_change_begin - mark the beginning of changes to a threadgroup
 * @tsk: task causing the changes
2717
 *
2718 2719 2720 2721 2722 2723
 * All operations which modify a threadgroup - a new thread joining the
 * group, death of a member thread (the assertion of PF_EXITING) and
 * exec(2) dethreading the process and replacing the leader - are wrapped
 * by threadgroup_change_{begin|end}().  This is to provide a place which
 * subsystems needing threadgroup stability can hook into for
 * synchronization.
2724
 */
2725
static inline void threadgroup_change_begin(struct task_struct *tsk)
2726
{
2727 2728
	might_sleep();
	cgroup_threadgroup_change_begin(tsk);
2729
}
2730 2731

/**
2732 2733
 * threadgroup_change_end - mark the end of changes to a threadgroup
 * @tsk: task causing the changes
2734
 *
2735
 * See threadgroup_change_begin().
2736
 */
2737
static inline void threadgroup_change_end(struct task_struct *tsk)
2738
{
2739
	cgroup_threadgroup_change_end(tsk);
2740 2741
}

A
Al Viro 已提交
2742 2743
#ifndef __HAVE_THREAD_FUNCTIONS

R
Roman Zippel 已提交
2744 2745
#define task_thread_info(task)	((struct thread_info *)(task)->stack)
#define task_stack_page(task)	((task)->stack)
A
Al Viro 已提交
2746

2747 2748 2749 2750 2751 2752
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;
}

2753 2754 2755 2756 2757 2758 2759 2760 2761
/*
 * 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.
 */
2762 2763
static inline unsigned long *end_of_stack(struct task_struct *p)
{
2764 2765 2766
#ifdef CONFIG_STACK_GROWSUP
	return (unsigned long *)((unsigned long)task_thread_info(p) + THREAD_SIZE) - 1;
#else
R
Roman Zippel 已提交
2767
	return (unsigned long *)(task_thread_info(p) + 1);
2768
#endif
2769 2770
}

A
Al Viro 已提交
2771
#endif
2772 2773
#define task_stack_end_corrupted(task) \
		(*(end_of_stack(task)) != STACK_END_MAGIC)
A
Al Viro 已提交
2774

2775 2776 2777 2778 2779 2780 2781
static inline int object_is_on_stack(void *obj)
{
	void *stack = task_stack_page(current);

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

2782 2783
extern void thread_info_cache_init(void);

2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795
#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
2796
extern void set_task_stack_end_magic(struct task_struct *tsk);
2797

L
Linus Torvalds 已提交
2798 2799 2800 2801 2802
/* 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
2803
	set_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2804 2805 2806 2807
}

static inline void clear_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
2808
	clear_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2809 2810 2811 2812
}

static inline int test_and_set_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
2813
	return test_and_set_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2814 2815 2816 2817
}

static inline int test_and_clear_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
2818
	return test_and_clear_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2819 2820 2821 2822
}

static inline int test_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
2823
	return test_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835
}

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);
}

2836 2837 2838 2839 2840
static inline int test_tsk_need_resched(struct task_struct *tsk)
{
	return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED));
}

2841 2842 2843 2844 2845 2846
static inline int restart_syscall(void)
{
	set_tsk_thread_flag(current, TIF_SIGPENDING);
	return -ERESTARTNOINTR;
}

L
Linus Torvalds 已提交
2847 2848 2849 2850
static inline int signal_pending(struct task_struct *p)
{
	return unlikely(test_tsk_thread_flag(p,TIF_SIGPENDING));
}
M
Matthew Wilcox 已提交
2851

2852 2853 2854 2855
static inline int __fatal_signal_pending(struct task_struct *p)
{
	return unlikely(sigismember(&p->pending.signal, SIGKILL));
}
M
Matthew Wilcox 已提交
2856 2857 2858 2859 2860 2861

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

2862 2863 2864 2865 2866 2867 2868 2869 2870 2871
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
2872 2873 2874 2875 2876 2877 2878
/*
 * 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.
 */
2879
extern int _cond_resched(void);
2880

2881
#define cond_resched() ({			\
2882
	___might_sleep(__FILE__, __LINE__, 0);	\
2883 2884
	_cond_resched();			\
})
2885

2886 2887
extern int __cond_resched_lock(spinlock_t *lock);

2888
#ifdef CONFIG_PREEMPT_COUNT
2889
#define PREEMPT_LOCK_OFFSET	PREEMPT_OFFSET
2890
#else
2891
#define PREEMPT_LOCK_OFFSET	0
2892
#endif
2893

2894
#define cond_resched_lock(lock) ({				\
2895
	___might_sleep(__FILE__, __LINE__, PREEMPT_LOCK_OFFSET);\
2896 2897 2898 2899 2900
	__cond_resched_lock(lock);				\
})

extern int __cond_resched_softirq(void);

2901
#define cond_resched_softirq() ({					\
2902
	___might_sleep(__FILE__, __LINE__, SOFTIRQ_DISABLE_OFFSET);	\
2903
	__cond_resched_softirq();					\
2904
})
L
Linus Torvalds 已提交
2905

2906 2907 2908 2909 2910 2911 2912 2913 2914
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
2915 2916
/*
 * Does a critical section need to be broken due to another
N
Nick Piggin 已提交
2917 2918
 * task waiting?: (technically does not depend on CONFIG_PREEMPT,
 * but a general need for low latency)
L
Linus Torvalds 已提交
2919
 */
N
Nick Piggin 已提交
2920
static inline int spin_needbreak(spinlock_t *lock)
L
Linus Torvalds 已提交
2921
{
N
Nick Piggin 已提交
2922 2923 2924
#ifdef CONFIG_PREEMPT
	return spin_is_contended(lock);
#else
L
Linus Torvalds 已提交
2925
	return 0;
N
Nick Piggin 已提交
2926
#endif
L
Linus Torvalds 已提交
2927 2928
}

2929 2930
/*
 * Idle thread specific functions to determine the need_resched
2931
 * polling state.
2932
 */
2933
#ifdef TIF_POLLING_NRFLAG
2934 2935 2936 2937
static inline int tsk_is_polling(struct task_struct *p)
{
	return test_tsk_thread_flag(p, TIF_POLLING_NRFLAG);
}
2938 2939

static inline void __current_set_polling(void)
2940 2941 2942 2943
{
	set_thread_flag(TIF_POLLING_NRFLAG);
}

2944 2945 2946 2947 2948 2949
static inline bool __must_check current_set_polling_and_test(void)
{
	__current_set_polling();

	/*
	 * Polling state must be visible before we test NEED_RESCHED,
2950
	 * paired by resched_curr()
2951
	 */
2952
	smp_mb__after_atomic();
2953 2954 2955 2956 2957

	return unlikely(tif_need_resched());
}

static inline void __current_clr_polling(void)
2958 2959 2960
{
	clear_thread_flag(TIF_POLLING_NRFLAG);
}
2961 2962 2963 2964 2965 2966 2967

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

	/*
	 * Polling state must be visible before we test NEED_RESCHED,
2968
	 * paired by resched_curr()
2969
	 */
2970
	smp_mb__after_atomic();
2971 2972 2973 2974

	return unlikely(tif_need_resched());
}

2975 2976
#else
static inline int tsk_is_polling(struct task_struct *p) { return 0; }
2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987
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());
}
2988 2989
#endif

2990 2991 2992 2993 2994 2995 2996 2997 2998 2999
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.
	 */
3000
	smp_mb(); /* paired with resched_curr() */
3001 3002 3003 3004

	preempt_fold_need_resched();
}

3005 3006 3007 3008 3009
static __always_inline bool need_resched(void)
{
	return unlikely(tif_need_resched());
}

3010 3011 3012
/*
 * Thread group CPU time accounting.
 */
3013
void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times);
3014
void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times);
3015

R
Roland McGrath 已提交
3016 3017 3018 3019 3020 3021 3022
/*
 * 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
3023 3024
extern void recalc_sigpending(void);

3025 3026 3027 3028 3029 3030 3031 3032 3033 3034
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
3035 3036 3037 3038 3039 3040 3041 3042

/*
 * 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
3043
	return task_thread_info(p)->cpu;
L
Linus Torvalds 已提交
3044 3045
}

I
Ingo Molnar 已提交
3046 3047 3048 3049 3050
static inline int task_node(const struct task_struct *p)
{
	return cpu_to_node(task_cpu(p));
}

I
Ingo Molnar 已提交
3051
extern void set_task_cpu(struct task_struct *p, unsigned int cpu);
L
Linus Torvalds 已提交
3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065

#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 */

3066 3067
extern long sched_setaffinity(pid_t pid, const struct cpumask *new_mask);
extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
3068

D
Dhaval Giani 已提交
3069
#ifdef CONFIG_CGROUP_SCHED
3070
extern struct task_group root_task_group;
P
Peter Zijlstra 已提交
3071
#endif /* CONFIG_CGROUP_SCHED */
3072

3073 3074 3075
extern int task_can_switch_user(struct user_struct *up,
					struct task_struct *tsk);

3076 3077 3078
#ifdef CONFIG_TASK_XACCT
static inline void add_rchar(struct task_struct *tsk, ssize_t amt)
{
3079
	tsk->ioac.rchar += amt;
3080 3081 3082 3083
}

static inline void add_wchar(struct task_struct *tsk, ssize_t amt)
{
3084
	tsk->ioac.wchar += amt;
3085 3086 3087 3088
}

static inline void inc_syscr(struct task_struct *tsk)
{
3089
	tsk->ioac.syscr++;
3090 3091 3092 3093
}

static inline void inc_syscw(struct task_struct *tsk)
{
3094
	tsk->ioac.syscw++;
3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113
}
#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
3114 3115 3116 3117
#ifndef TASK_SIZE_OF
#define TASK_SIZE_OF(tsk)	TASK_SIZE
#endif

O
Oleg Nesterov 已提交
3118
#ifdef CONFIG_MEMCG
3119 3120 3121 3122 3123
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
3124
#endif /* CONFIG_MEMCG */
3125

3126 3127 3128
static inline unsigned long task_rlimit(const struct task_struct *tsk,
		unsigned int limit)
{
3129
	return READ_ONCE(tsk->signal->rlim[limit].rlim_cur);
3130 3131 3132 3133 3134
}

static inline unsigned long task_rlimit_max(const struct task_struct *tsk,
		unsigned int limit)
{
3135
	return READ_ONCE(tsk->signal->rlim[limit].rlim_max);
3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147
}

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
3148
#endif