sched.h 102.9 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 43
#include <linux/signal.h>
#include <linux/compiler.h>
#include <linux/completion.h>
#include <linux/pid.h>
#include <linux/percpu.h>
#include <linux/topology.h>
#include <linux/seccomp.h>
I
Ingo Molnar 已提交
44
#include <linux/rcupdate.h>
45
#include <linux/rculist.h>
I
Ingo Molnar 已提交
46
#include <linux/rtmutex.h>
L
Linus Torvalds 已提交
47

48 49 50 51 52
#include <linux/time.h>
#include <linux/param.h>
#include <linux/resource.h>
#include <linux/timer.h>
#include <linux/hrtimer.h>
D
Dmitry Vyukov 已提交
53
#include <linux/kcov.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)
180 181
extern void cpu_load_update_nohz_start(void);
extern void cpu_load_update_nohz_stop(void);
182
#else
183 184
static inline void cpu_load_update_nohz_start(void) { }
static inline void cpu_load_update_nohz_stop(void) { }
185
#endif
L
Linus Torvalds 已提交
186

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
#define TASK_NOLOAD		1024
222 223
#define TASK_NEW		2048
#define TASK_STATE_MAX		4096
M
Matthew Wilcox 已提交
224

225
#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWPNn"
226

227 228
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
229 230 231 232 233

/* 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
234

235 236
#define TASK_IDLE		(TASK_UNINTERRUPTIBLE | TASK_NOLOAD)

237 238
/* Convenience macros for the sake of wake_up */
#define TASK_NORMAL		(TASK_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE)
M
Matthew Wilcox 已提交
239
#define TASK_ALL		(TASK_NORMAL | __TASK_STOPPED | __TASK_TRACED)
240 241 242

/* get_task_state() */
#define TASK_REPORT		(TASK_RUNNING | TASK_INTERRUPTIBLE | \
M
Matthew Wilcox 已提交
243
				 TASK_UNINTERRUPTIBLE | __TASK_STOPPED | \
244
				 __TASK_TRACED | EXIT_ZOMBIE | EXIT_DEAD)
245

M
Matthew Wilcox 已提交
246 247
#define task_is_traced(task)	((task->state & __TASK_TRACED) != 0)
#define task_is_stopped(task)	((task->state & __TASK_STOPPED) != 0)
248
#define task_is_stopped_or_traced(task)	\
M
Matthew Wilcox 已提交
249
			((task->state & (__TASK_STOPPED | __TASK_TRACED)) != 0)
250
#define task_contributes_to_load(task)	\
251
				((task->state & TASK_UNINTERRUPTIBLE) != 0 && \
252 253
				 (task->flags & PF_FROZEN) == 0 && \
				 (task->state & TASK_NOLOAD) == 0)
L
Linus Torvalds 已提交
254

P
Peter Zijlstra 已提交
255 256 257 258 259 260 261 262 263 264
#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_;		\
265
		smp_store_mb((tsk)->state, (state_value));	\
P
Peter Zijlstra 已提交
266 267 268 269 270 271 272 273 274 275
	} while (0)

#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_;		\
276
		smp_store_mb(current->state, (state_value));	\
P
Peter Zijlstra 已提交
277 278 279 280
	} while (0)

#else

281 282 283 284 285 286 287 288
/*
 * @tsk had better be current, or you get to keep the pieces.
 *
 * The only reason is that computing current can be more expensive than
 * using a pointer that's already available.
 *
 * Therefore, see set_current_state().
 */
L
Linus Torvalds 已提交
289 290 291
#define __set_task_state(tsk, state_value)		\
	do { (tsk)->state = (state_value); } while (0)
#define set_task_state(tsk, state_value)		\
292
	smp_store_mb((tsk)->state, (state_value))
L
Linus Torvalds 已提交
293

294 295 296 297 298
/*
 * set_current_state() includes a barrier so that the write of current->state
 * is correctly serialised wrt the caller's subsequent test of whether to
 * actually sleep:
 *
299
 *   for (;;) {
300
 *	set_current_state(TASK_UNINTERRUPTIBLE);
301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324
 *	if (!need_sleep)
 *		break;
 *
 *	schedule();
 *   }
 *   __set_current_state(TASK_RUNNING);
 *
 * If the caller does not need such serialisation (because, for instance, the
 * condition test and condition change and wakeup are under the same lock) then
 * use __set_current_state().
 *
 * The above is typically ordered against the wakeup, which does:
 *
 *	need_sleep = false;
 *	wake_up_state(p, TASK_UNINTERRUPTIBLE);
 *
 * Where wake_up_state() (and all other wakeup primitives) imply enough
 * barriers to order the store of the variable against wakeup.
 *
 * Wakeup will do: if (@state & p->state) p->state = TASK_RUNNING, that is,
 * once it observes the TASK_UNINTERRUPTIBLE store the waking CPU can issue a
 * TASK_RUNNING store which can collide with __set_current_state(TASK_RUNNING).
 *
 * This is obviously fine, since they both store the exact same value.
325
 *
326
 * Also see the comments of try_to_wake_up().
327
 */
P
Peter Zijlstra 已提交
328
#define __set_current_state(state_value)		\
L
Linus Torvalds 已提交
329
	do { current->state = (state_value); } while (0)
P
Peter Zijlstra 已提交
330
#define set_current_state(state_value)			\
331
	smp_store_mb(current->state, (state_value))
L
Linus Torvalds 已提交
332

P
Peter Zijlstra 已提交
333 334
#endif

L
Linus Torvalds 已提交
335 336 337 338 339 340 341 342 343 344 345 346 347 348
/* 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;

349
struct task_struct;
L
Linus Torvalds 已提交
350

351 352 353 354
#ifdef CONFIG_PROVE_RCU
extern int lockdep_tasklist_lock_is_held(void);
#endif /* #ifdef CONFIG_PROVE_RCU */

L
Linus Torvalds 已提交
355 356
extern void sched_init(void);
extern void sched_init_smp(void);
357
extern asmlinkage void schedule_tail(struct task_struct *prev);
358
extern void init_idle(struct task_struct *idle, int cpu);
I
Ingo Molnar 已提交
359
extern void init_idle_bootup_task(struct task_struct *idle);
L
Linus Torvalds 已提交
360

361 362
extern cpumask_var_t cpu_isolated_map;

363
extern int runqueue_is_locked(int cpu);
I
Ingo Molnar 已提交
364

365
#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON)
366
extern void nohz_balance_enter_idle(int cpu);
367
extern void set_cpu_sd_state_idle(void);
368
extern int get_nohz_timer_target(void);
369
#else
370
static inline void nohz_balance_enter_idle(int cpu) { }
371
static inline void set_cpu_sd_state_idle(void) { }
372
#endif
L
Linus Torvalds 已提交
373

I
Ingo Molnar 已提交
374
/*
I
Ingo Molnar 已提交
375
 * Only dump TASK_* tasks. (0 for all tasks)
I
Ingo Molnar 已提交
376 377 378 379 380
 */
extern void show_state_filter(unsigned long state_filter);

static inline void show_state(void)
{
I
Ingo Molnar 已提交
381
	show_state_filter(0);
I
Ingo Molnar 已提交
382 383
}

L
Linus Torvalds 已提交
384 385 386 387 388 389 390 391 392 393 394 395 396
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);
397
extern int sched_cpu_starting(unsigned int cpu);
398 399
extern int sched_cpu_activate(unsigned int cpu);
extern int sched_cpu_deactivate(unsigned int cpu);
L
Linus Torvalds 已提交
400

401 402 403 404 405
#ifdef CONFIG_HOTPLUG_CPU
extern int sched_cpu_dying(unsigned int cpu);
#else
# define sched_cpu_dying	NULL
#endif
L
Linus Torvalds 已提交
406

407 408
extern void sched_show_task(struct task_struct *p);

409
#ifdef CONFIG_LOCKUP_DETECTOR
410
extern void touch_softlockup_watchdog_sched(void);
I
Ingo Molnar 已提交
411
extern void touch_softlockup_watchdog(void);
412
extern void touch_softlockup_watchdog_sync(void);
413
extern void touch_all_softlockup_watchdogs(void);
414 415 416
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
417
extern unsigned int  softlockup_panic;
418
extern unsigned int  hardlockup_panic;
419
void lockup_detector_init(void);
I
Ingo Molnar 已提交
420
#else
421 422 423
static inline void touch_softlockup_watchdog_sched(void)
{
}
I
Ingo Molnar 已提交
424 425 426
static inline void touch_softlockup_watchdog(void)
{
}
427 428 429
static inline void touch_softlockup_watchdog_sync(void)
{
}
430 431 432
static inline void touch_all_softlockup_watchdogs(void)
{
}
433 434 435
static inline void lockup_detector_init(void)
{
}
I
Ingo Molnar 已提交
436 437
#endif

438 439 440 441 442 443 444 445
#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
446 447
/* Attach to any functions which should be ignored in wchan output. */
#define __sched		__attribute__((__section__(".sched.text")))
448 449 450 451

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

L
Linus Torvalds 已提交
452 453 454 455
/* Is this address in the __sched functions? */
extern int in_sched_functions(unsigned long addr);

#define	MAX_SCHEDULE_TIMEOUT	LONG_MAX
456
extern signed long schedule_timeout(signed long timeout);
457
extern signed long schedule_timeout_interruptible(signed long timeout);
M
Matthew Wilcox 已提交
458
extern signed long schedule_timeout_killable(signed long timeout);
459
extern signed long schedule_timeout_uninterruptible(signed long timeout);
460
extern signed long schedule_timeout_idle(signed long timeout);
L
Linus Torvalds 已提交
461
asmlinkage void schedule(void);
462
extern void schedule_preempt_disabled(void);
L
Linus Torvalds 已提交
463

464 465 466 467 468 469 470
extern long io_schedule_timeout(long timeout);

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

471 472
void __noreturn do_task_dead(void);

S
Serge E. Hallyn 已提交
473
struct nsproxy;
474
struct user_namespace;
L
Linus Torvalds 已提交
475

476 477
#ifdef CONFIG_MMU
extern void arch_pick_mmap_layout(struct mm_struct *mm);
L
Linus Torvalds 已提交
478 479 480 481 482 483 484
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);
485 486 487
#else
static inline void arch_pick_mmap_layout(struct mm_struct *mm) {}
#endif
L
Linus Torvalds 已提交
488

489 490 491 492
#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 */

493
/* mm flags */
H
Hugh Dickins 已提交
494

495
/* for SUID_DUMP_* above */
496
#define MMF_DUMPABLE_BITS 2
H
Hugh Dickins 已提交
497
#define MMF_DUMPABLE_MASK ((1 << MMF_DUMPABLE_BITS) - 1)
498

499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515
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);
}

516 517 518 519 520
/* 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
521
#define MMF_DUMP_ELF_HEADERS	6
522 523
#define MMF_DUMP_HUGETLB_PRIVATE 7
#define MMF_DUMP_HUGETLB_SHARED  8
524 525
#define MMF_DUMP_DAX_PRIVATE	9
#define MMF_DUMP_DAX_SHARED	10
H
Hugh Dickins 已提交
526

527
#define MMF_DUMP_FILTER_SHIFT	MMF_DUMPABLE_BITS
528
#define MMF_DUMP_FILTER_BITS	9
529 530 531
#define MMF_DUMP_FILTER_MASK \
	(((1 << MMF_DUMP_FILTER_BITS) - 1) << MMF_DUMP_FILTER_SHIFT)
#define MMF_DUMP_FILTER_DEFAULT \
532
	((1 << MMF_DUMP_ANON_PRIVATE) |	(1 << MMF_DUMP_ANON_SHARED) |\
533 534 535 536 537 538 539
	 (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
540 541
					/* leave room for more dump flags */
#define MMF_VM_MERGEABLE	16	/* KSM may merge identical pages */
A
Andrea Arcangeli 已提交
542
#define MMF_VM_HUGEPAGE		17	/* set when VM_HUGEPAGE is set on vma */
543
#define MMF_EXE_FILE_CHANGED	18	/* see prctl_set_mm_exe_file() */
H
Hugh Dickins 已提交
544

545 546
#define MMF_HAS_UPROBES		19	/* has uprobes */
#define MMF_RECALC_UPROBES	20	/* MMF_HAS_UPROBES can be wrong */
547
#define MMF_OOM_SKIP		21	/* mm is of no interest for the OOM killer */
548
#define MMF_UNSTABLE		22	/* mm is unstable for copy_from_user */
549
#define MMF_HUGE_ZERO_PAGE	23      /* mm has ever used the global huge zero page */
550

H
Hugh Dickins 已提交
551
#define MMF_INIT_MASK		(MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK)
552

L
Linus Torvalds 已提交
553 554 555 556
struct sighand_struct {
	atomic_t		count;
	struct k_sigaction	action[_NSIG];
	spinlock_t		siglock;
D
Davide Libenzi 已提交
557
	wait_queue_head_t	signalfd_wqh;
L
Linus Torvalds 已提交
558 559
};

560
struct pacct_struct {
561 562
	int			ac_flag;
	long			ac_exitcode;
563
	unsigned long		ac_mem;
564 565
	cputime_t		ac_utime, ac_stime;
	unsigned long		ac_minflt, ac_majflt;
566 567
};

568 569 570
struct cpu_itimer {
	cputime_t expires;
	cputime_t incr;
571 572
	u32 error;
	u32 incr_error;
573 574
};

575
/**
576
 * struct prev_cputime - snaphsot of system and user cputime
577 578
 * @utime: time spent in user mode
 * @stime: time spent in system mode
579
 * @lock: protects the above two fields
580
 *
581 582
 * Stores previous user/system time values such that we can guarantee
 * monotonicity.
583
 */
584 585
struct prev_cputime {
#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
586 587
	cputime_t utime;
	cputime_t stime;
588 589
	raw_spinlock_t lock;
#endif
590 591
};

592 593 594 595 596 597 598 599
static inline void prev_cputime_init(struct prev_cputime *prev)
{
#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
	prev->utime = prev->stime = 0;
	raw_spin_lock_init(&prev->lock);
#endif
}

600 601 602 603 604
/**
 * 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
605
 *
606 607 608
 * This structure groups together three kinds of CPU time that are tracked for
 * threads and thread groups.  Most things considering CPU time want to group
 * these counts together and treat all three of them in parallel.
609 610 611 612 613 614
 */
struct task_cputime {
	cputime_t utime;
	cputime_t stime;
	unsigned long long sum_exec_runtime;
};
615

616 617
/* Alternate field names when used to cache expirations. */
#define virt_exp	utime
618
#define prof_exp	stime
619 620
#define sched_exp	sum_exec_runtime

621 622
#define INIT_CPUTIME	\
	(struct task_cputime) {					\
623 624
		.utime = 0,					\
		.stime = 0,					\
625 626 627
		.sum_exec_runtime = 0,				\
	}

628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644
/*
 * 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),		\
	}

645
#define PREEMPT_DISABLED	(PREEMPT_DISABLE_OFFSET + PREEMPT_ENABLED)
646

P
Peter Zijlstra 已提交
647
/*
648 649
 * Disable preemption until the scheduler is running -- use an unconditional
 * value so that it also works on !PREEMPT_COUNT kernels.
P
Peter Zijlstra 已提交
650
 *
651
 * Reset by start_kernel()->sched_init()->init_idle()->init_idle_preempt_count().
P
Peter Zijlstra 已提交
652
 */
653
#define INIT_PREEMPT_COUNT	PREEMPT_OFFSET
654

P
Peter Zijlstra 已提交
655
/*
656 657
 * Initial preempt_count value; reflects the preempt_count schedule invariant
 * which states that during context switches:
P
Peter Zijlstra 已提交
658
 *
659 660 661 662
 *    preempt_count() == 2*PREEMPT_DISABLE_OFFSET
 *
 * Note: PREEMPT_DISABLE_OFFSET is 0 for !PREEMPT_COUNT kernels.
 * Note: See finish_task_switch().
P
Peter Zijlstra 已提交
663
 */
664
#define FORK_PREEMPT_COUNT	(2*PREEMPT_DISABLE_OFFSET + PREEMPT_ENABLED)
P
Peter Zijlstra 已提交
665

666
/**
667
 * struct thread_group_cputimer - thread group interval timer counts
668
 * @cputime_atomic:	atomic thread group interval timers.
669 670
 * @running:		true when there are timers running and
 *			@cputime_atomic receives updates.
671 672
 * @checking_timer:	true when a thread in the group is in the
 *			process of checking for thread group timers.
673 674
 *
 * This structure contains the version of task_cputime, above, that is
675
 * used for thread group CPU timer calculations.
676
 */
677
struct thread_group_cputimer {
678
	struct task_cputime_atomic cputime_atomic;
679
	bool running;
680
	bool checking_timer;
681 682
};

683
#include <linux/rwsem.h>
684 685
struct autogroup;

L
Linus Torvalds 已提交
686
/*
687
 * NOTE! "signal_struct" does not have its own
L
Linus Torvalds 已提交
688 689 690 691 692 693
 * 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 {
694
	atomic_t		sigcnt;
L
Linus Torvalds 已提交
695
	atomic_t		live;
696
	int			nr_threads;
697
	struct list_head	thread_head;
L
Linus Torvalds 已提交
698 699 700 701

	wait_queue_head_t	wait_chldexit;	/* for wait4() */

	/* current thread group signal load-balancing target: */
702
	struct task_struct	*curr_target;
L
Linus Torvalds 已提交
703 704 705 706 707 708 709 710 711 712 713 714

	/* 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;
715
	struct task_struct	*group_exit_task;
L
Linus Torvalds 已提交
716 717 718 719 720

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

721 722 723 724 725 726 727 728 729 730 731 732
	/*
	 * 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
733
	/* POSIX.1b Interval Timers */
734 735
	int			posix_timer_id;
	struct list_head	posix_timers;
L
Linus Torvalds 已提交
736 737

	/* ITIMER_REAL timer for the process */
738
	struct hrtimer real_timer;
739
	struct pid *leader_pid;
740
	ktime_t it_real_incr;
L
Linus Torvalds 已提交
741

742 743 744 745 746 747
	/*
	 * 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
748

749
	/*
750 751
	 * Thread group totals for process CPU timers.
	 * See thread_group_cputimer(), et al, for details.
752
	 */
753
	struct thread_group_cputimer cputimer;
754 755 756 757

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

758
#ifdef CONFIG_NO_HZ_FULL
759
	atomic_t tick_dep_mask;
760 761
#endif

762 763
	struct list_head cpu_timers[3];

764
	struct pid *tty_old_pgrp;
765

L
Linus Torvalds 已提交
766 767 768 769 770
	/* boolean value for session group leader */
	int leader;

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

771 772 773
#ifdef CONFIG_SCHED_AUTOGROUP
	struct autogroup *autogroup;
#endif
L
Linus Torvalds 已提交
774 775 776 777 778 779
	/*
	 * 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.
	 */
780
	seqlock_t stats_lock;
781
	cputime_t utime, stime, cutime, cstime;
782 783
	cputime_t gtime;
	cputime_t cgtime;
784
	struct prev_cputime prev_cputime;
L
Linus Torvalds 已提交
785 786
	unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw;
	unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt;
787
	unsigned long inblock, oublock, cinblock, coublock;
J
Jiri Pirko 已提交
788
	unsigned long maxrss, cmaxrss;
789
	struct task_io_accounting ioac;
L
Linus Torvalds 已提交
790

791 792 793 794 795 796 797 798
	/*
	 * 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
799 800 801 802 803 804 805 806 807 808 809
	/*
	 * 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];

810 811 812
#ifdef CONFIG_BSD_PROCESS_ACCT
	struct pacct_struct pacct;	/* per-process accounting information */
#endif
813 814 815
#ifdef CONFIG_TASKSTATS
	struct taskstats *stats;
#endif
M
Miloslav Trmac 已提交
816 817 818 819
#ifdef CONFIG_AUDIT
	unsigned audit_tty;
	struct tty_audit_buf *tty_audit_buf;
#endif
820

T
Tetsuo Handa 已提交
821 822 823 824 825
	/*
	 * Thread is the potential origin of an oom condition; kill first on
	 * oom
	 */
	bool oom_flag_origin;
826 827 828
	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. */
829 830
	struct mm_struct *oom_mm;	/* recorded mm when the thread group got
					 * killed by the oom killer */
831 832 833 834

	struct mutex cred_guard_mutex;	/* guard against foreign influences on
					 * credential calculations
					 * (notably. ptrace) */
L
Linus Torvalds 已提交
835 836 837 838 839 840
};

/*
 * Bits in flags field of signal_struct.
 */
#define SIGNAL_STOP_STOPPED	0x00000001 /* job control stop in effect */
841 842
#define SIGNAL_STOP_CONTINUED	0x00000002 /* SIGCONT since WCONTINUED reap */
#define SIGNAL_GROUP_EXIT	0x00000004 /* group exit in progress */
843
#define SIGNAL_GROUP_COREDUMP	0x00000008 /* coredump in progress */
844 845 846 847 848 849
/*
 * 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
850

851 852
#define SIGNAL_UNKILLABLE	0x00000040 /* for init: ignore fatal signals */

853 854 855 856 857 858 859
/* 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
860 861 862 863 864 865 866
/*
 * 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? */
867
#ifdef CONFIG_INOTIFY_USER
R
Robert Love 已提交
868 869 870
	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
871 872 873
#ifdef CONFIG_FANOTIFY
	atomic_t fanotify_listeners;
#endif
874
#ifdef CONFIG_EPOLL
875
	atomic_long_t epoll_watches; /* The number of file descriptors currently watched */
876
#endif
A
Alexey Dobriyan 已提交
877
#ifdef CONFIG_POSIX_MQUEUE
L
Linus Torvalds 已提交
878 879
	/* protected by mq_lock	*/
	unsigned long mq_bytes;	/* How many bytes can be allocated to mqueue? */
A
Alexey Dobriyan 已提交
880
#endif
L
Linus Torvalds 已提交
881
	unsigned long locked_shm; /* How many pages of mlocked shm ? */
882
	unsigned long unix_inflight;	/* How many files in flight in unix sockets */
883
	atomic_long_t pipe_bufs;  /* how many pages are allocated in pipe buffers */
L
Linus Torvalds 已提交
884 885 886 887 888 889 890

#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
891
	struct hlist_node uidhash_node;
892
	kuid_t uid;
893

894
#if defined(CONFIG_PERF_EVENTS) || defined(CONFIG_BPF_SYSCALL)
895 896
	atomic_long_t locked_vm;
#endif
L
Linus Torvalds 已提交
897 898
};

899
extern int uids_sysfs_init(void);
900

901
extern struct user_struct *find_user(kuid_t);
L
Linus Torvalds 已提交
902 903 904 905

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

906

L
Linus Torvalds 已提交
907 908 909
struct backing_dev_info;
struct reclaim_state;

910
#ifdef CONFIG_SCHED_INFO
L
Linus Torvalds 已提交
911 912
struct sched_info {
	/* cumulative counters */
913
	unsigned long pcount;	      /* # of times run on this cpu */
914
	unsigned long long run_delay; /* time spent waiting on a runqueue */
L
Linus Torvalds 已提交
915 916

	/* timestamps */
917 918
	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
919
};
920
#endif /* CONFIG_SCHED_INFO */
L
Linus Torvalds 已提交
921

922 923 924 925 926 927 928 929 930 931 932 933 934 935
#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).
	 */
936 937 938 939 940 941

	/*
	 * 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.
	 */
942
	u64 blkio_start;	/* Shared by blkio, swapin */
943 944 945 946 947 948
	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 */
949

950
	u64 freepages_start;
951 952
	u64 freepages_delay;	/* wait for memory reclaim */
	u32 freepages_count;	/* total count of memory reclaim */
953
};
954 955 956 957 958 959 960 961 962 963 964
#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;
965
#endif
966
}
967

968 969 970 971
#ifdef CONFIG_SCHEDSTATS
void force_schedstat_enabled(void);
#endif

I
Ingo Molnar 已提交
972 973 974 975 976
enum cpu_idle_type {
	CPU_IDLE,
	CPU_NOT_IDLE,
	CPU_NEWLY_IDLE,
	CPU_MAX_IDLE_TYPES
L
Linus Torvalds 已提交
977 978
};

979 980 981 982 983 984 985 986 987 988
/*
 * Integer metrics need fixed point arithmetic, e.g., sched/fair
 * has a few: load, load_avg, util_avg, freq, and capacity.
 *
 * We define a basic fixed point arithmetic range, and then formalize
 * all these metrics based on that basic range.
 */
# define SCHED_FIXEDPOINT_SHIFT	10
# define SCHED_FIXEDPOINT_SCALE	(1L << SCHED_FIXEDPOINT_SHIFT)

989
/*
990
 * Increase resolution of cpu_capacity calculations
991
 */
992
#define SCHED_CAPACITY_SHIFT	SCHED_FIXEDPOINT_SHIFT
993
#define SCHED_CAPACITY_SCALE	(1L << SCHED_CAPACITY_SHIFT)
L
Linus Torvalds 已提交
994

995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038
/*
 * 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);

1039 1040 1041
/*
 * sched-domains (multiprocessor balancing) declarations:
 */
1042
#ifdef CONFIG_SMP
P
Peter Zijlstra 已提交
1043 1044 1045 1046
#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 */
1047
#define SD_BALANCE_WAKE		0x0010  /* Balance on wakeup */
P
Peter Zijlstra 已提交
1048
#define SD_WAKE_AFFINE		0x0020	/* Wake task to waking CPU */
1049
#define SD_ASYM_CPUCAPACITY	0x0040  /* Groups have different max cpu capacities */
1050
#define SD_SHARE_CPUCAPACITY	0x0080	/* Domain members share cpu capacity */
1051
#define SD_SHARE_POWERDOMAIN	0x0100	/* Domain members share power domain */
P
Peter Zijlstra 已提交
1052 1053
#define SD_SHARE_PKG_RESOURCES	0x0200	/* Domain members share cpu pkg resources */
#define SD_SERIALIZE		0x0400	/* Only a single load balancing instance */
1054
#define SD_ASYM_PACKING		0x0800  /* Place busy groups earlier in the domain */
P
Peter Zijlstra 已提交
1055
#define SD_PREFER_SIBLING	0x1000	/* Prefer to place tasks in a sibling domain */
1056
#define SD_OVERLAP		0x2000	/* sched_domains of this level overlap */
1057
#define SD_NUMA			0x4000	/* cross-node balancing */
1058

1059
#ifdef CONFIG_SCHED_SMT
G
Guenter Roeck 已提交
1060
static inline int cpu_smt_flags(void)
1061
{
1062
	return SD_SHARE_CPUCAPACITY | SD_SHARE_PKG_RESOURCES;
1063 1064 1065 1066
}
#endif

#ifdef CONFIG_SCHED_MC
G
Guenter Roeck 已提交
1067
static inline int cpu_core_flags(void)
1068 1069 1070 1071 1072 1073
{
	return SD_SHARE_PKG_RESOURCES;
}
#endif

#ifdef CONFIG_NUMA
G
Guenter Roeck 已提交
1074
static inline int cpu_numa_flags(void)
1075 1076 1077 1078
{
	return SD_NUMA;
}
#endif
1079

1080 1081 1082 1083 1084 1085 1086 1087
struct sched_domain_attr {
	int relax_domain_level;
};

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

1088 1089
extern int sched_domain_level_max;

1090 1091
struct sched_group;

1092 1093
struct sched_domain_shared {
	atomic_t	ref;
1094
	atomic_t	nr_busy_cpus;
1095
	int		has_idle_cores;
1096 1097
};

L
Linus Torvalds 已提交
1098 1099 1100
struct sched_domain {
	/* These fields must be setup */
	struct sched_domain *parent;	/* top domain must be null terminated */
1101
	struct sched_domain *child;	/* bottom domain must be null terminated */
L
Linus Torvalds 已提交
1102 1103 1104 1105 1106 1107
	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
1108 1109 1110 1111
	unsigned int busy_idx;
	unsigned int idle_idx;
	unsigned int newidle_idx;
	unsigned int wake_idx;
N
Nick Piggin 已提交
1112
	unsigned int forkexec_idx;
P
Peter Zijlstra 已提交
1113
	unsigned int smt_gain;
V
Vincent Guittot 已提交
1114 1115

	int nohz_idle;			/* NOHZ IDLE status */
L
Linus Torvalds 已提交
1116
	int flags;			/* See SD_* */
1117
	int level;
L
Linus Torvalds 已提交
1118 1119 1120 1121 1122 1123

	/* 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 */

1124
	/* idle_balance() stats */
1125
	u64 max_newidle_lb_cost;
1126
	unsigned long next_decay_max_lb_cost;
P
Peter Zijlstra 已提交
1127

1128 1129
	u64 avg_scan_cost;		/* select_idle_sibling */

L
Linus Torvalds 已提交
1130 1131
#ifdef CONFIG_SCHEDSTATS
	/* load_balance() stats */
1132 1133 1134 1135 1136 1137 1138 1139
	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
1140 1141

	/* Active load balancing */
1142 1143 1144
	unsigned int alb_count;
	unsigned int alb_failed;
	unsigned int alb_pushed;
L
Linus Torvalds 已提交
1145

1146
	/* SD_BALANCE_EXEC stats */
1147 1148 1149
	unsigned int sbe_count;
	unsigned int sbe_balanced;
	unsigned int sbe_pushed;
L
Linus Torvalds 已提交
1150

1151
	/* SD_BALANCE_FORK stats */
1152 1153 1154
	unsigned int sbf_count;
	unsigned int sbf_balanced;
	unsigned int sbf_pushed;
1155

L
Linus Torvalds 已提交
1156
	/* try_to_wake_up() stats */
1157 1158 1159
	unsigned int ttwu_wake_remote;
	unsigned int ttwu_move_affine;
	unsigned int ttwu_move_balance;
L
Linus Torvalds 已提交
1160
#endif
1161 1162 1163
#ifdef CONFIG_SCHED_DEBUG
	char *name;
#endif
1164 1165 1166 1167
	union {
		void *private;		/* used during construction */
		struct rcu_head rcu;	/* used during destruction */
	};
1168
	struct sched_domain_shared *shared;
1169

1170
	unsigned int span_weight;
1171 1172 1173 1174 1175 1176 1177 1178
	/*
	 * 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
1179 1180
};

1181 1182
static inline struct cpumask *sched_domain_span(struct sched_domain *sd)
{
1183
	return to_cpumask(sd->span);
1184 1185
}

1186
extern void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[],
1187
				    struct sched_domain_attr *dattr_new);
P
Paul Jackson 已提交
1188

1189 1190 1191 1192
/* 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);

1193 1194
bool cpus_share_cache(int this_cpu, int that_cpu);

1195
typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
G
Guenter Roeck 已提交
1196
typedef int (*sched_domain_flags_f)(void);
1197 1198 1199 1200 1201

#define SDTL_OVERLAP	0x01

struct sd_data {
	struct sched_domain **__percpu sd;
1202
	struct sched_domain_shared **__percpu sds;
1203
	struct sched_group **__percpu sg;
1204
	struct sched_group_capacity **__percpu sgc;
1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218
};

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 void set_sched_topology(struct sched_domain_topology_level *tl);
1219
extern void wake_up_if_idle(int cpu);
1220 1221 1222 1223 1224 1225 1226

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

1227
#else /* CONFIG_SMP */
L
Linus Torvalds 已提交
1228

1229
struct sched_domain_attr;
1230

1231
static inline void
1232
partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[],
1233 1234
			struct sched_domain_attr *dattr_new)
{
1235
}
1236 1237 1238 1239 1240 1241

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

1242
#endif	/* !CONFIG_SMP */
L
Linus Torvalds 已提交
1243

1244

L
Linus Torvalds 已提交
1245 1246 1247
struct io_context;			/* See blkdev.h */


1248
#ifdef ARCH_HAS_PREFETCH_SWITCH_STACK
1249
extern void prefetch_stack(struct task_struct *t);
1250 1251 1252
#else
static inline void prefetch_stack(struct task_struct *t) { }
#endif
L
Linus Torvalds 已提交
1253 1254 1255

struct audit_context;		/* See audit.c */
struct mempolicy;
1256
struct pipe_inode_info;
1257
struct uts_namespace;
L
Linus Torvalds 已提交
1258

I
Ingo Molnar 已提交
1259
struct load_weight {
1260 1261
	unsigned long weight;
	u32 inv_weight;
I
Ingo Molnar 已提交
1262 1263
};

1264
/*
1265 1266 1267 1268 1269 1270 1271 1272 1273
 * The load_avg/util_avg accumulates an infinite geometric series
 * (see __update_load_avg() in kernel/sched/fair.c).
 *
 * [load_avg definition]
 *
 *   load_avg = runnable% * scale_load_down(load)
 *
 * where runnable% is the time ratio that a sched_entity is runnable.
 * For cfs_rq, it is the aggregated load_avg of all runnable and
1274
 * blocked sched_entities.
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314
 *
 * load_avg may also take frequency scaling into account:
 *
 *   load_avg = runnable% * scale_load_down(load) * freq%
 *
 * where freq% is the CPU frequency normalized to the highest frequency.
 *
 * [util_avg definition]
 *
 *   util_avg = running% * SCHED_CAPACITY_SCALE
 *
 * where running% is the time ratio that a sched_entity is running on
 * a CPU. For cfs_rq, it is the aggregated util_avg of all runnable
 * and blocked sched_entities.
 *
 * util_avg may also factor frequency scaling and CPU capacity scaling:
 *
 *   util_avg = running% * SCHED_CAPACITY_SCALE * freq% * capacity%
 *
 * where freq% is the same as above, and capacity% is the CPU capacity
 * normalized to the greatest capacity (due to uarch differences, etc).
 *
 * N.B., the above ratios (runnable%, running%, freq%, and capacity%)
 * themselves are in the range of [0, 1]. To do fixed point arithmetics,
 * we therefore scale them to as large a range as necessary. This is for
 * example reflected by util_avg's SCHED_CAPACITY_SCALE.
 *
 * [Overflow issue]
 *
 * The 64-bit load_sum can have 4353082796 (=2^64/47742/88761) entities
 * with the highest load (=88761), always runnable on a single cfs_rq,
 * and should not overflow as the number already hits PID_MAX_LIMIT.
 *
 * For all other cases (including 32-bit kernels), struct load_weight's
 * weight will overflow first before we do, because:
 *
 *    Max(load_avg) <= Max(load.weight)
 *
 * Then it is the load_weight's responsibility to consider overflow
 * issues.
1315
 */
1316
struct sched_avg {
1317 1318 1319
	u64 last_update_time, load_sum;
	u32 util_sum, period_contrib;
	unsigned long load_avg, util_avg;
1320 1321
};

1322
#ifdef CONFIG_SCHEDSTATS
1323
struct sched_statistics {
I
Ingo Molnar 已提交
1324
	u64			wait_start;
1325
	u64			wait_max;
1326 1327
	u64			wait_count;
	u64			wait_sum;
1328 1329
	u64			iowait_count;
	u64			iowait_sum;
1330

I
Ingo Molnar 已提交
1331 1332
	u64			sleep_start;
	u64			sleep_max;
1333 1334 1335
	s64			sum_sleep_runtime;

	u64			block_start;
I
Ingo Molnar 已提交
1336 1337
	u64			block_max;
	u64			exec_max;
I
Ingo Molnar 已提交
1338
	u64			slice_max;
1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354

	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;
1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
};
#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;
1373 1374
#endif

I
Ingo Molnar 已提交
1375
#ifdef CONFIG_FAIR_GROUP_SCHED
P
Peter Zijlstra 已提交
1376
	int			depth;
I
Ingo Molnar 已提交
1377 1378 1379 1380 1381 1382
	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
1383

1384
#ifdef CONFIG_SMP
1385 1386 1387 1388 1389 1390 1391
	/*
	 * Per entity load average tracking.
	 *
	 * Put into separate cache line so it does not
	 * collide with read-mostly values above.
	 */
	struct sched_avg	avg ____cacheline_aligned_in_smp;
1392
#endif
I
Ingo Molnar 已提交
1393
};
1394

P
Peter Zijlstra 已提交
1395 1396
struct sched_rt_entity {
	struct list_head run_list;
1397
	unsigned long timeout;
1398
	unsigned long watchdog_stamp;
1399
	unsigned int time_slice;
1400 1401
	unsigned short on_rq;
	unsigned short on_list;
P
Peter Zijlstra 已提交
1402

1403
	struct sched_rt_entity *back;
1404
#ifdef CONFIG_RT_GROUP_SCHED
P
Peter Zijlstra 已提交
1405 1406 1407 1408 1409 1410
	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
1411 1412
};

1413 1414 1415 1416 1417
struct sched_dl_entity {
	struct rb_node	rb_node;

	/*
	 * Original scheduling parameters. Copied here from sched_attr
1418 1419
	 * during sched_setattr(), they will remain the same until
	 * the next sched_setattr().
1420 1421 1422
	 */
	u64 dl_runtime;		/* maximum runtime for each instance	*/
	u64 dl_deadline;	/* relative deadline of each instance	*/
1423
	u64 dl_period;		/* separation of two instances (period) */
1424
	u64 dl_bw;		/* dl_runtime / dl_deadline		*/
1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441

	/*
	 * 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.
	 *
1442 1443
	 * @dl_boosted tells if we are boosted due to DI. If so we are
	 * outside bandwidth enforcement mechanism (but only until we
1444 1445 1446 1447
	 * exit the critical section);
	 *
	 * @dl_yielded tells if task gave up the cpu before consuming
	 * all its available runtime during the last job.
1448
	 */
1449
	int dl_throttled, dl_boosted, dl_yielded;
1450 1451 1452 1453 1454 1455 1456

	/*
	 * Bandwidth enforcement timer. Each -deadline task has its
	 * own bandwidth to be enforced, thus we need one timer per task.
	 */
	struct hrtimer dl_timer;
};
1457

1458 1459
union rcu_special {
	struct {
1460 1461 1462 1463 1464 1465
		u8 blocked;
		u8 need_qs;
		u8 exp_need_qs;
		u8 pad;	/* Otherwise the compiler can store garbage here. */
	} b; /* Bits. */
	u32 s; /* Set of bits. */
1466
};
1467 1468
struct rcu_node;

P
Peter Zijlstra 已提交
1469 1470 1471
enum perf_event_task_context {
	perf_invalid_context = -1,
	perf_hw_context = 0,
1472
	perf_sw_context,
P
Peter Zijlstra 已提交
1473 1474 1475
	perf_nr_task_contexts,
};

1476 1477 1478 1479 1480 1481 1482 1483 1484 1485
/* Track pages that require TLB flushes */
struct tlbflush_unmap_batch {
	/*
	 * Each bit set is a CPU that potentially has a TLB entry for one of
	 * the PFNs being flushed. See set_tlb_ubc_flush_pending().
	 */
	struct cpumask cpumask;

	/* True if any bit in cpumask is set */
	bool flush_required;
1486 1487 1488 1489 1490 1491 1492

	/*
	 * If true then the PTE was dirty when unmapped. The entry must be
	 * flushed before IO is initiated or a stale TLB entry potentially
	 * allows an update without redirtying the page.
	 */
	bool writable;
1493 1494
};

L
Linus Torvalds 已提交
1495
struct task_struct {
1496 1497 1498 1499 1500 1501 1502
#ifdef CONFIG_THREAD_INFO_IN_TASK
	/*
	 * For reasons of header soup (see current_thread_info()), this
	 * must be the first element of task_struct.
	 */
	struct thread_info thread_info;
#endif
L
Linus Torvalds 已提交
1503
	volatile long state;	/* -1 unrunnable, 0 runnable, >0 stopped */
R
Roman Zippel 已提交
1504
	void *stack;
L
Linus Torvalds 已提交
1505
	atomic_t usage;
1506 1507
	unsigned int flags;	/* per process flags, defined below */
	unsigned int ptrace;
L
Linus Torvalds 已提交
1508

1509
#ifdef CONFIG_SMP
P
Peter Zijlstra 已提交
1510
	struct llist_node wake_entry;
P
Peter Zijlstra 已提交
1511
	int on_cpu;
1512 1513 1514
#ifdef CONFIG_THREAD_INFO_IN_TASK
	unsigned int cpu;	/* current CPU */
#endif
M
Mike Galbraith 已提交
1515
	unsigned int wakee_flips;
1516
	unsigned long wakee_flip_decay_ts;
M
Mike Galbraith 已提交
1517
	struct task_struct *last_wakee;
1518 1519

	int wake_cpu;
1520
#endif
P
Peter Zijlstra 已提交
1521
	int on_rq;
1522

1523
	int prio, static_prio, normal_prio;
1524
	unsigned int rt_priority;
1525
	const struct sched_class *sched_class;
I
Ingo Molnar 已提交
1526
	struct sched_entity se;
P
Peter Zijlstra 已提交
1527
	struct sched_rt_entity rt;
P
Peter Zijlstra 已提交
1528 1529 1530
#ifdef CONFIG_CGROUP_SCHED
	struct task_group *sched_task_group;
#endif
1531
	struct sched_dl_entity dl;
L
Linus Torvalds 已提交
1532

1533 1534 1535 1536 1537
#ifdef CONFIG_PREEMPT_NOTIFIERS
	/* list of struct preempt_notifier: */
	struct hlist_head preempt_notifiers;
#endif

1538
#ifdef CONFIG_BLK_DEV_IO_TRACE
1539
	unsigned int btrace_seq;
1540
#endif
L
Linus Torvalds 已提交
1541

1542
	unsigned int policy;
1543
	int nr_cpus_allowed;
L
Linus Torvalds 已提交
1544 1545
	cpumask_t cpus_allowed;

P
Paul E. McKenney 已提交
1546
#ifdef CONFIG_PREEMPT_RCU
P
Paul E. McKenney 已提交
1547
	int rcu_read_lock_nesting;
1548
	union rcu_special rcu_read_unlock_special;
1549
	struct list_head rcu_node_entry;
P
Paul E. McKenney 已提交
1550
	struct rcu_node *rcu_blocked_node;
1551
#endif /* #ifdef CONFIG_PREEMPT_RCU */
P
Paul E. McKenney 已提交
1552 1553 1554 1555
#ifdef CONFIG_TASKS_RCU
	unsigned long rcu_tasks_nvcsw;
	bool rcu_tasks_holdout;
	struct list_head rcu_tasks_holdout_list;
1556
	int rcu_tasks_idle_cpu;
P
Paul E. McKenney 已提交
1557
#endif /* #ifdef CONFIG_TASKS_RCU */
P
Paul E. McKenney 已提交
1558

1559
#ifdef CONFIG_SCHED_INFO
L
Linus Torvalds 已提交
1560 1561 1562 1563
	struct sched_info sched_info;
#endif

	struct list_head tasks;
1564
#ifdef CONFIG_SMP
1565
	struct plist_node pushable_tasks;
1566
	struct rb_node pushable_dl_tasks;
1567
#endif
L
Linus Torvalds 已提交
1568 1569

	struct mm_struct *mm, *active_mm;
D
Davidlohr Bueso 已提交
1570 1571 1572
	/* per-thread vma caching */
	u32 vmacache_seqnum;
	struct vm_area_struct *vmacache[VMACACHE_SIZE];
1573 1574 1575
#if defined(SPLIT_RSS_COUNTING)
	struct task_rss_stat	rss_stat;
#endif
L
Linus Torvalds 已提交
1576
/* task state */
1577
	int exit_state;
L
Linus Torvalds 已提交
1578 1579
	int exit_code, exit_signal;
	int pdeath_signal;  /*  The signal sent when the parent dies  */
1580
	unsigned long jobctl;	/* JOBCTL_*, siglock protected */
1581 1582

	/* Used for emulating ABI behavior of previous Linux versions */
1583
	unsigned int personality;
1584

1585
	/* scheduler bits, serialized by scheduler locks */
1586
	unsigned sched_reset_on_fork:1;
1587
	unsigned sched_contributes_to_load:1;
1588
	unsigned sched_migrated:1;
P
Peter Zijlstra 已提交
1589
	unsigned sched_remote_wakeup:1;
1590 1591 1592 1593 1594
	unsigned :0; /* force alignment to the next boundary */

	/* unserialized, strictly 'current' */
	unsigned in_execve:1; /* bit to tell LSMs we're in execve */
	unsigned in_iowait:1;
1595 1596 1597
#if !defined(TIF_RESTORE_SIGMASK)
	unsigned restore_sigmask:1;
#endif
T
Tejun Heo 已提交
1598 1599
#ifdef CONFIG_MEMCG
	unsigned memcg_may_oom:1;
1600
#ifndef CONFIG_SLOB
1601 1602
	unsigned memcg_kmem_skip_account:1;
#endif
1603
#endif
1604 1605 1606
#ifdef CONFIG_COMPAT_BRK
	unsigned brk_randomized:1;
#endif
1607

1608 1609
	unsigned long atomic_flags; /* Flags needing atomic access. */

1610 1611
	struct restart_block restart_block;

L
Linus Torvalds 已提交
1612 1613
	pid_t pid;
	pid_t tgid;
1614

1615
#ifdef CONFIG_CC_STACKPROTECTOR
1616 1617
	/* Canary value for the -fstack-protector gcc feature */
	unsigned long stack_canary;
1618
#endif
1619
	/*
L
Linus Torvalds 已提交
1620
	 * pointers to (original) parent process, youngest child, younger sibling,
1621
	 * older sibling, respectively.  (p->father can be replaced with
R
Roland McGrath 已提交
1622
	 * p->real_parent->pid)
L
Linus Torvalds 已提交
1623
	 */
1624 1625
	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
1626
	/*
R
Roland McGrath 已提交
1627
	 * children/sibling forms the list of my natural children
L
Linus Torvalds 已提交
1628 1629 1630 1631 1632
	 */
	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
1633 1634 1635 1636 1637 1638 1639 1640
	/*
	 * 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
1641
	/* PID/PID hash table linkage. */
1642
	struct pid_link pids[PIDTYPE_MAX];
O
Oleg Nesterov 已提交
1643
	struct list_head thread_group;
1644
	struct list_head thread_node;
L
Linus Torvalds 已提交
1645 1646 1647 1648 1649

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

1650
	cputime_t utime, stime, utimescaled, stimescaled;
1651
	cputime_t gtime;
1652
	struct prev_cputime prev_cputime;
1653
#ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN
1654
	seqcount_t vtime_seqcount;
1655 1656
	unsigned long long vtime_snap;
	enum {
1657 1658 1659
		/* Task is sleeping or running in a CPU with VTIME inactive */
		VTIME_INACTIVE = 0,
		/* Task runs in userspace in a CPU with VTIME active */
1660
		VTIME_USER,
1661
		/* Task runs in kernelspace in a CPU with VTIME active */
1662 1663
		VTIME_SYS,
	} vtime_snap_whence;
1664
#endif
1665 1666

#ifdef CONFIG_NO_HZ_FULL
1667
	atomic_t tick_dep_mask;
1668
#endif
L
Linus Torvalds 已提交
1669
	unsigned long nvcsw, nivcsw; /* context switch counts */
1670
	u64 start_time;		/* monotonic time in nsec */
1671
	u64 real_start_time;	/* boot based time in nsec */
L
Linus Torvalds 已提交
1672 1673 1674
/* mm fault and swap info: this can arguably be seen as either mm-specific or thread-specific */
	unsigned long min_flt, maj_flt;

1675
	struct task_cputime cputime_expires;
L
Linus Torvalds 已提交
1676 1677 1678
	struct list_head cpu_timers[3];

/* process credentials */
A
Arnd Bergmann 已提交
1679
	const struct cred __rcu *real_cred; /* objective and real subjective task
1680
					 * credentials (COW) */
A
Arnd Bergmann 已提交
1681
	const struct cred __rcu *cred;	/* effective (overridable) subjective task
1682
					 * credentials (COW) */
1683 1684 1685
	char comm[TASK_COMM_LEN]; /* executable name excluding path
				     - access with [gs]et_task_comm (which lock
				       it with task_lock())
1686
				     - initialized normally by setup_new_exec */
L
Linus Torvalds 已提交
1687
/* file system info */
1688
	struct nameidata *nameidata;
1689
#ifdef CONFIG_SYSVIPC
L
Linus Torvalds 已提交
1690 1691
/* ipc stuff */
	struct sysv_sem sysvsem;
1692
	struct sysv_shm sysvshm;
1693
#endif
1694
#ifdef CONFIG_DETECT_HUNG_TASK
1695 1696 1697
/* hung task detection */
	unsigned long last_switch_count;
#endif
L
Linus Torvalds 已提交
1698 1699 1700 1701
/* filesystem information */
	struct fs_struct *fs;
/* open file information */
	struct files_struct *files;
1702
/* namespaces */
S
Serge E. Hallyn 已提交
1703
	struct nsproxy *nsproxy;
L
Linus Torvalds 已提交
1704 1705 1706 1707 1708
/* signal handlers */
	struct signal_struct *signal;
	struct sighand_struct *sighand;

	sigset_t blocked, real_blocked;
1709
	sigset_t saved_sigmask;	/* restored if set_restore_sigmask() was used */
L
Linus Torvalds 已提交
1710 1711 1712 1713
	struct sigpending pending;

	unsigned long sas_ss_sp;
	size_t sas_ss_size;
1714
	unsigned sas_ss_flags;
1715

1716
	struct callback_head *task_works;
1717

L
Linus Torvalds 已提交
1718
	struct audit_context *audit_context;
A
Al Viro 已提交
1719
#ifdef CONFIG_AUDITSYSCALL
1720
	kuid_t loginuid;
1721
	unsigned int sessionid;
A
Al Viro 已提交
1722
#endif
1723
	struct seccomp seccomp;
L
Linus Torvalds 已提交
1724 1725 1726 1727

/* Thread group tracking */
   	u32 parent_exec_id;
   	u32 self_exec_id;
1728 1729
/* Protection of (de-)allocation: mm, files, fs, tty, keyrings, mems_allowed,
 * mempolicy */
L
Linus Torvalds 已提交
1730 1731
	spinlock_t alloc_lock;

1732
	/* Protection of the PI data structures: */
1733
	raw_spinlock_t pi_lock;
1734

1735 1736
	struct wake_q_node wake_q;

I
Ingo Molnar 已提交
1737 1738
#ifdef CONFIG_RT_MUTEXES
	/* PI waiters blocked on a rt_mutex held by this task */
1739 1740
	struct rb_root pi_waiters;
	struct rb_node *pi_waiters_leftmost;
I
Ingo Molnar 已提交
1741 1742 1743 1744
	/* Deadlock detection and priority inheritance handling */
	struct rt_mutex_waiter *pi_blocked_on;
#endif

1745 1746 1747 1748
#ifdef CONFIG_DEBUG_MUTEXES
	/* mutex deadlock detection */
	struct mutex_waiter *blocked_on;
#endif
1749 1750 1751 1752
#ifdef CONFIG_TRACE_IRQFLAGS
	unsigned int irq_events;
	unsigned long hardirq_enable_ip;
	unsigned long hardirq_disable_ip;
1753
	unsigned int hardirq_enable_event;
1754
	unsigned int hardirq_disable_event;
1755 1756
	int hardirqs_enabled;
	int hardirq_context;
1757 1758
	unsigned long softirq_disable_ip;
	unsigned long softirq_enable_ip;
1759
	unsigned int softirq_disable_event;
1760
	unsigned int softirq_enable_event;
1761
	int softirqs_enabled;
1762 1763
	int softirq_context;
#endif
I
Ingo Molnar 已提交
1764
#ifdef CONFIG_LOCKDEP
1765
# define MAX_LOCK_DEPTH 48UL
I
Ingo Molnar 已提交
1766 1767 1768
	u64 curr_chain_key;
	int lockdep_depth;
	unsigned int lockdep_recursion;
1769
	struct held_lock held_locks[MAX_LOCK_DEPTH];
1770
	gfp_t lockdep_reclaim_gfp;
I
Ingo Molnar 已提交
1771
#endif
1772 1773 1774
#ifdef CONFIG_UBSAN
	unsigned int in_ubsan;
#endif
1775

L
Linus Torvalds 已提交
1776 1777 1778
/* journalling filesystem info */
	void *journal_info;

1779
/* stacked block device info */
1780
	struct bio_list *bio_list;
1781

1782 1783 1784 1785 1786
#ifdef CONFIG_BLOCK
/* stack plugging */
	struct blk_plug *plug;
#endif

L
Linus Torvalds 已提交
1787 1788 1789 1790 1791 1792 1793 1794 1795
/* 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.  */
1796
	struct task_io_accounting ioac;
1797
#if defined(CONFIG_TASK_XACCT)
L
Linus Torvalds 已提交
1798 1799
	u64 acct_rss_mem1;	/* accumulated rss usage */
	u64 acct_vm_mem1;	/* accumulated virtual memory usage */
1800
	cputime_t acct_timexpd;	/* stime + utime since last update */
L
Linus Torvalds 已提交
1801 1802
#endif
#ifdef CONFIG_CPUSETS
1803
	nodemask_t mems_allowed;	/* Protected by alloc_lock */
1804
	seqcount_t mems_allowed_seq;	/* Seqence no to catch updates */
1805
	int cpuset_mem_spread_rotor;
1806
	int cpuset_slab_spread_rotor;
L
Linus Torvalds 已提交
1807
#endif
1808
#ifdef CONFIG_CGROUPS
1809
	/* Control Group info protected by css_set_lock */
A
Arnd Bergmann 已提交
1810
	struct css_set __rcu *cgroups;
1811 1812
	/* cg_list protected by css_set_lock and tsk->alloc_lock */
	struct list_head cg_list;
1813
#endif
1814
#ifdef CONFIG_FUTEX
1815
	struct robust_list_head __user *robust_list;
1816 1817 1818
#ifdef CONFIG_COMPAT
	struct compat_robust_list_head __user *compat_robust_list;
#endif
1819 1820
	struct list_head pi_state_list;
	struct futex_pi_state *pi_state_cache;
1821
#endif
1822
#ifdef CONFIG_PERF_EVENTS
P
Peter Zijlstra 已提交
1823
	struct perf_event_context *perf_event_ctxp[perf_nr_task_contexts];
1824 1825
	struct mutex perf_event_mutex;
	struct list_head perf_event_list;
1826
#endif
1827 1828 1829
#ifdef CONFIG_DEBUG_PREEMPT
	unsigned long preempt_disable_ip;
#endif
1830
#ifdef CONFIG_NUMA
1831
	struct mempolicy *mempolicy;	/* Protected by alloc_lock */
1832
	short il_next;
1833
	short pref_node_fork;
1834
#endif
1835 1836 1837
#ifdef CONFIG_NUMA_BALANCING
	int numa_scan_seq;
	unsigned int numa_scan_period;
1838
	unsigned int numa_scan_period_max;
1839
	int numa_preferred_nid;
1840
	unsigned long numa_migrate_retry;
1841
	u64 node_stamp;			/* migration stamp  */
1842 1843
	u64 last_task_numa_placement;
	u64 last_sum_exec_runtime;
1844
	struct callback_head numa_work;
1845

1846 1847 1848
	struct list_head numa_entry;
	struct numa_group *numa_group;

1849
	/*
1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861
	 * 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.
1862
	 */
1863
	unsigned long *numa_faults;
1864
	unsigned long total_numa_faults;
1865

1866 1867
	/*
	 * numa_faults_locality tracks if faults recorded during the last
1868 1869 1870
	 * 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
1871
	 */
1872
	unsigned long numa_faults_locality[3];
1873

I
Ingo Molnar 已提交
1874
	unsigned long numa_pages_migrated;
1875 1876
#endif /* CONFIG_NUMA_BALANCING */

1877 1878 1879 1880
#ifdef CONFIG_ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
	struct tlbflush_unmap_batch tlb_ubc;
#endif

I
Ingo Molnar 已提交
1881
	struct rcu_head rcu;
1882 1883 1884 1885 1886

	/*
	 * cache last used pipe for splice
	 */
	struct pipe_inode_info *splice_pipe;
1887 1888 1889

	struct page_frag task_frag;

1890 1891
#ifdef	CONFIG_TASK_DELAY_ACCT
	struct task_delay_info *delays;
1892 1893 1894
#endif
#ifdef CONFIG_FAULT_INJECTION
	int make_it_fail;
1895
#endif
1896 1897 1898 1899 1900 1901
	/*
	 * 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;
1902
	unsigned long dirty_paused_when; /* start of a write-and-pause period */
1903

A
Arjan van de Ven 已提交
1904 1905 1906 1907
#ifdef CONFIG_LATENCYTOP
	int latency_record_count;
	struct latency_record latency_record[LT_SAVECOUNT];
#endif
1908 1909 1910 1911
	/*
	 * time slack values; these are used to round up poll() and
	 * select() etc timeout values. These are in nanoseconds.
	 */
1912 1913
	u64 timer_slack_ns;
	u64 default_timer_slack_ns;
1914

1915 1916 1917
#ifdef CONFIG_KASAN
	unsigned int kasan_depth;
#endif
1918
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
D
Daniel Mack 已提交
1919
	/* Index of current stored address in ret_stack */
1920 1921 1922
	int curr_ret_stack;
	/* Stack of return addresses for return function tracing */
	struct ftrace_ret_stack	*ret_stack;
1923 1924
	/* time stamp for last schedule */
	unsigned long long ftrace_timestamp;
1925 1926 1927 1928 1929
	/*
	 * Number of functions that haven't been traced
	 * because of depth overrun.
	 */
	atomic_t trace_overrun;
1930 1931
	/* Pause for the tracing */
	atomic_t tracing_graph_pause;
1932
#endif
1933 1934 1935
#ifdef CONFIG_TRACING
	/* state flags for use by tracers */
	unsigned long trace;
1936
	/* bitmask and counter of trace recursion */
1937 1938
	unsigned long trace_recursion;
#endif /* CONFIG_TRACING */
D
Dmitry Vyukov 已提交
1939 1940 1941 1942 1943 1944 1945 1946 1947 1948
#ifdef CONFIG_KCOV
	/* Coverage collection mode enabled for this task (0 if disabled). */
	enum kcov_mode kcov_mode;
	/* Size of the kcov_area. */
	unsigned	kcov_size;
	/* Buffer for coverage collection. */
	void		*kcov_area;
	/* kcov desciptor wired with this task or NULL. */
	struct kcov	*kcov;
#endif
1949
#ifdef CONFIG_MEMCG
T
Tejun Heo 已提交
1950 1951 1952
	struct mem_cgroup *memcg_in_oom;
	gfp_t memcg_oom_gfp_mask;
	int memcg_oom_order;
1953 1954 1955

	/* number of pages to reclaim on returning to userland */
	unsigned int memcg_nr_pages_over_high;
1956
#endif
1957 1958 1959
#ifdef CONFIG_UPROBES
	struct uprobe_task *utask;
#endif
K
Kent Overstreet 已提交
1960 1961 1962 1963
#if defined(CONFIG_BCACHE) || defined(CONFIG_BCACHE_MODULE)
	unsigned int	sequential_io;
	unsigned int	sequential_io_avg;
#endif
P
Peter Zijlstra 已提交
1964 1965 1966
#ifdef CONFIG_DEBUG_ATOMIC_SLEEP
	unsigned long	task_state_change;
#endif
1967
	int pagefault_disabled;
1968
#ifdef CONFIG_MMU
1969
	struct task_struct *oom_reaper_list;
1970
#endif
1971 1972 1973
#ifdef CONFIG_VMAP_STACK
	struct vm_struct *stack_vm_area;
#endif
1974 1975 1976 1977
#ifdef CONFIG_THREAD_INFO_IN_TASK
	/* A live task holds one reference. */
	atomic_t stack_refcount;
#endif
1978 1979 1980 1981 1982 1983 1984 1985
/* CPU-specific state of this task */
	struct thread_struct thread;
/*
 * WARNING: on x86, 'thread_struct' contains a variable-sized
 * structure.  It *MUST* be at the end of 'task_struct'.
 *
 * Do not put anything below here!
 */
L
Linus Torvalds 已提交
1986 1987
};

1988 1989 1990 1991 1992
#ifdef CONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT
extern int arch_task_struct_size __read_mostly;
#else
# define arch_task_struct_size (sizeof(struct task_struct))
#endif
1993

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
#ifdef CONFIG_VMAP_STACK
static inline struct vm_struct *task_stack_vm_area(const struct task_struct *t)
{
	return t->stack_vm_area;
}
#else
static inline struct vm_struct *task_stack_vm_area(const struct task_struct *t)
{
	return NULL;
}
#endif

2006
/* Future-safe accessor for struct task_struct's cpus_allowed. */
2007
#define tsk_cpus_allowed(tsk) (&(tsk)->cpus_allowed)
2008

2009 2010 2011 2012 2013
static inline int tsk_nr_cpus_allowed(struct task_struct *p)
{
	return p->nr_cpus_allowed;
}

2014 2015
#define TNF_MIGRATED	0x01
#define TNF_NO_GROUP	0x02
2016
#define TNF_SHARED	0x04
2017
#define TNF_FAULT_LOCAL	0x08
2018
#define TNF_MIGRATE_FAIL 0x10
2019

2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045
static inline bool in_vfork(struct task_struct *tsk)
{
	bool ret;

	/*
	 * need RCU to access ->real_parent if CLONE_VM was used along with
	 * CLONE_PARENT.
	 *
	 * We check real_parent->mm == tsk->mm because CLONE_VFORK does not
	 * imply CLONE_VM
	 *
	 * CLONE_VFORK can be used with CLONE_PARENT/CLONE_THREAD and thus
	 * ->real_parent is not necessarily the task doing vfork(), so in
	 * theory we can't rely on task_lock() if we want to dereference it.
	 *
	 * And in this case we can't trust the real_parent->mm == tsk->mm
	 * check, it can be false negative. But we do not care, if init or
	 * another oom-unkillable task does this it should blame itself.
	 */
	rcu_read_lock();
	ret = tsk->vfork_done && tsk->real_parent->mm == tsk->mm;
	rcu_read_unlock();

	return ret;
}

2046
#ifdef CONFIG_NUMA_BALANCING
2047
extern void task_numa_fault(int last_node, int node, int pages, int flags);
2048
extern pid_t task_numa_group_id(struct task_struct *p);
2049
extern void set_numabalancing_state(bool enabled);
2050
extern void task_numa_free(struct task_struct *p);
2051 2052
extern bool should_numa_migrate_memory(struct task_struct *p, struct page *page,
					int src_nid, int dst_cpu);
2053
#else
2054
static inline void task_numa_fault(int last_node, int node, int pages,
2055
				   int flags)
2056 2057
{
}
2058 2059 2060 2061
static inline pid_t task_numa_group_id(struct task_struct *p)
{
	return 0;
}
2062 2063 2064
static inline void set_numabalancing_state(bool enabled)
{
}
2065 2066 2067
static inline void task_numa_free(struct task_struct *p)
{
}
2068 2069 2070 2071 2072
static inline bool should_numa_migrate_memory(struct task_struct *p,
				struct page *page, int src_nid, int dst_cpu)
{
	return true;
}
2073 2074
#endif

A
Alexey Dobriyan 已提交
2075
static inline struct pid *task_pid(struct task_struct *task)
2076 2077 2078 2079
{
	return task->pids[PIDTYPE_PID].pid;
}

A
Alexey Dobriyan 已提交
2080
static inline struct pid *task_tgid(struct task_struct *task)
2081 2082 2083 2084
{
	return task->group_leader->pids[PIDTYPE_PID].pid;
}

2085 2086 2087 2088 2089
/*
 * 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
2090
static inline struct pid *task_pgrp(struct task_struct *task)
2091 2092 2093 2094
{
	return task->group_leader->pids[PIDTYPE_PGID].pid;
}

A
Alexey Dobriyan 已提交
2095
static inline struct pid *task_session(struct task_struct *task)
2096 2097 2098 2099
{
	return task->group_leader->pids[PIDTYPE_SID].pid;
}

2100 2101 2102 2103 2104 2105 2106
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
2107 2108
 * task_xid_vnr()    : virtual id, i.e. the id seen from the pid namespace of
 *                     current.
2109 2110 2111 2112 2113 2114
 * 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
 */
2115 2116
pid_t __task_pid_nr_ns(struct task_struct *task, enum pid_type type,
			struct pid_namespace *ns);
2117

A
Alexey Dobriyan 已提交
2118
static inline pid_t task_pid_nr(struct task_struct *tsk)
2119 2120 2121 2122
{
	return tsk->pid;
}

2123 2124 2125 2126 2127
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);
}
2128 2129 2130

static inline pid_t task_pid_vnr(struct task_struct *tsk)
{
2131
	return __task_pid_nr_ns(tsk, PIDTYPE_PID, NULL);
2132 2133 2134
}


A
Alexey Dobriyan 已提交
2135
static inline pid_t task_tgid_nr(struct task_struct *tsk)
2136 2137 2138 2139
{
	return tsk->tgid;
}

2140
pid_t task_tgid_nr_ns(struct task_struct *tsk, struct pid_namespace *ns);
2141 2142 2143 2144 2145 2146 2147

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


2148
static inline int pid_alive(const struct task_struct *p);
2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165
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);
}

2166 2167
static inline pid_t task_pgrp_nr_ns(struct task_struct *tsk,
					struct pid_namespace *ns)
2168
{
2169
	return __task_pid_nr_ns(tsk, PIDTYPE_PGID, ns);
2170 2171 2172 2173
}

static inline pid_t task_pgrp_vnr(struct task_struct *tsk)
{
2174
	return __task_pid_nr_ns(tsk, PIDTYPE_PGID, NULL);
2175 2176 2177
}


2178 2179
static inline pid_t task_session_nr_ns(struct task_struct *tsk,
					struct pid_namespace *ns)
2180
{
2181
	return __task_pid_nr_ns(tsk, PIDTYPE_SID, ns);
2182 2183 2184 2185
}

static inline pid_t task_session_vnr(struct task_struct *tsk)
{
2186
	return __task_pid_nr_ns(tsk, PIDTYPE_SID, NULL);
2187 2188
}

2189 2190 2191 2192 2193
/* obsolete, do not use */
static inline pid_t task_pgrp_nr(struct task_struct *tsk)
{
	return task_pgrp_nr_ns(tsk, &init_pid_ns);
}
2194

L
Linus Torvalds 已提交
2195 2196 2197 2198 2199 2200 2201
/**
 * 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.
2202 2203
 *
 * Return: 1 if the process is alive. 0 otherwise.
L
Linus Torvalds 已提交
2204
 */
2205
static inline int pid_alive(const struct task_struct *p)
L
Linus Torvalds 已提交
2206
{
2207
	return p->pids[PIDTYPE_PID].pid != NULL;
L
Linus Torvalds 已提交
2208 2209
}

2210
/**
2211 2212
 * is_global_init - check if a task structure is init. Since init
 * is free to have sub-threads we need to check tgid.
2213 2214 2215
 * @tsk: Task structure to be checked.
 *
 * Check if a task structure is the first user space task the kernel created.
2216 2217
 *
 * Return: 1 if the task structure is init. 0 otherwise.
2218
 */
A
Alexey Dobriyan 已提交
2219
static inline int is_global_init(struct task_struct *tsk)
2220
{
2221
	return task_tgid_nr(tsk) == 1;
2222
}
2223

2224 2225
extern struct pid *cad_pid;

L
Linus Torvalds 已提交
2226 2227
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
2228

2229
extern void __put_task_struct(struct task_struct *t);
I
Ingo Molnar 已提交
2230 2231 2232 2233

static inline void put_task_struct(struct task_struct *t)
{
	if (atomic_dec_and_test(&t->usage))
2234
		__put_task_struct(t);
I
Ingo Molnar 已提交
2235
}
L
Linus Torvalds 已提交
2236

2237 2238 2239
struct task_struct *task_rcu_dereference(struct task_struct **ptask);
struct task_struct *try_get_task_struct(struct task_struct **ptask);

2240 2241 2242 2243 2244 2245 2246
#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
2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264
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;
}
2265 2266 2267 2268 2269 2270

static inline cputime_t task_gtime(struct task_struct *t)
{
	return t->gtime;
}
#endif
2271 2272
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);
2273

L
Linus Torvalds 已提交
2274 2275 2276 2277
/*
 * Per process flags
 */
#define PF_EXITING	0x00000004	/* getting shut down */
2278
#define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
2279
#define PF_VCPU		0x00000010	/* I'm a virtual CPU */
T
Tejun Heo 已提交
2280
#define PF_WQ_WORKER	0x00000020	/* I'm a workqueue worker */
L
Linus Torvalds 已提交
2281
#define PF_FORKNOEXEC	0x00000040	/* forked but didn't exec */
2282
#define PF_MCE_PROCESS  0x00000080      /* process policy on mce errors */
L
Linus Torvalds 已提交
2283 2284 2285 2286
#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 */
2287
#define PF_NPROC_EXCEEDED 0x00001000	/* set_user noticed that RLIMIT_NPROC was exceeded */
L
Linus Torvalds 已提交
2288
#define PF_USED_MATH	0x00002000	/* if unset the fpu must be initialized before use */
2289
#define PF_USED_ASYNC	0x00004000	/* used async_schedule*(), used by module init */
L
Linus Torvalds 已提交
2290 2291 2292 2293
#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 */
2294
#define PF_MEMALLOC_NOIO 0x00080000	/* Allocating memory without IO involved */
L
Linus Torvalds 已提交
2295
#define PF_LESS_THROTTLE 0x00100000	/* Throttle me less: I clean memory */
2296
#define PF_KTHREAD	0x00200000	/* I am a kernel thread */
J
Jens Axboe 已提交
2297 2298
#define PF_RANDOMIZE	0x00400000	/* randomize virtual address space */
#define PF_SWAPWRITE	0x00800000	/* Allowed to write to swap */
2299
#define PF_NO_SETAFFINITY 0x04000000	/* Userland is not allowed to meddle with cpus_allowed */
2300
#define PF_MCE_EARLY    0x08000000      /* Early kill for mce process policy */
2301
#define PF_MUTEX_TESTER	0x20000000	/* Thread belongs to the rt mutex tester */
2302
#define PF_FREEZER_SKIP	0x40000000	/* Freezer should not count it as freezable */
2303
#define PF_SUSPEND_TASK 0x80000000      /* this thread called freeze_processes and should not be frozen */
L
Linus Torvalds 已提交
2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329

/*
 * 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)

2330 2331 2332
/* __GFP_IO isn't allowed if PF_MEMALLOC_NOIO is set in current->flags
 * __GFP_FS is also cleared as it implies __GFP_IO.
 */
2333 2334 2335
static inline gfp_t memalloc_noio_flags(gfp_t flags)
{
	if (unlikely(current->flags & PF_MEMALLOC_NOIO))
2336
		flags &= ~(__GFP_IO | __GFP_FS);
2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351
	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;
}

2352
/* Per-process atomic flags. */
2353
#define PFA_NO_NEW_PRIVS 0	/* May not gain new privileges. */
2354 2355
#define PFA_SPREAD_PAGE  1      /* Spread page cache over cpuset */
#define PFA_SPREAD_SLAB  2      /* Spread some slab caches over cpuset */
2356
#define PFA_LMK_WAITING  3      /* Lowmemorykiller is waiting */
2357

2358

2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370
#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)
2371

2372 2373 2374 2375 2376 2377 2378
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)
2379

2380 2381 2382
TASK_PFA_TEST(LMK_WAITING, lmk_waiting)
TASK_PFA_SET(LMK_WAITING, lmk_waiting)

2383
/*
2384
 * task->jobctl flags
2385
 */
2386
#define JOBCTL_STOP_SIGMASK	0xffff	/* signr of the last group stop */
2387

2388 2389 2390
#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 */
2391
#define JOBCTL_TRAP_STOP_BIT	19	/* trap for STOP */
2392
#define JOBCTL_TRAP_NOTIFY_BIT	20	/* trap for NOTIFY */
2393
#define JOBCTL_TRAPPING_BIT	21	/* switching to TRACED */
T
Tejun Heo 已提交
2394
#define JOBCTL_LISTENING_BIT	22	/* ptracer is listening for events */
2395

2396 2397 2398 2399 2400 2401 2402
#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)
2403

2404
#define JOBCTL_TRAP_MASK	(JOBCTL_TRAP_STOP | JOBCTL_TRAP_NOTIFY)
2405
#define JOBCTL_PENDING_MASK	(JOBCTL_STOP_PENDING | JOBCTL_TRAP_MASK)
2406

2407
extern bool task_set_jobctl_pending(struct task_struct *task,
2408
				    unsigned long mask);
2409
extern void task_clear_jobctl_trapping(struct task_struct *task);
2410
extern void task_clear_jobctl_pending(struct task_struct *task,
2411
				      unsigned long mask);
2412

2413 2414
static inline void rcu_copy_process(struct task_struct *p)
{
P
Paul E. McKenney 已提交
2415
#ifdef CONFIG_PREEMPT_RCU
2416
	p->rcu_read_lock_nesting = 0;
2417
	p->rcu_read_unlock_special.s = 0;
2418
	p->rcu_blocked_node = NULL;
2419
	INIT_LIST_HEAD(&p->rcu_node_entry);
P
Paul E. McKenney 已提交
2420 2421 2422 2423
#endif /* #ifdef CONFIG_PREEMPT_RCU */
#ifdef CONFIG_TASKS_RCU
	p->rcu_tasks_holdout = false;
	INIT_LIST_HEAD(&p->rcu_tasks_holdout_list);
2424
	p->rcu_tasks_idle_cpu = -1;
P
Paul E. McKenney 已提交
2425
#endif /* #ifdef CONFIG_TASKS_RCU */
2426 2427
}

2428 2429 2430 2431 2432 2433 2434
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;
}

2435 2436
extern int cpuset_cpumask_can_shrink(const struct cpumask *cur,
				     const struct cpumask *trial);
2437 2438
extern int task_can_attach(struct task_struct *p,
			   const struct cpumask *cs_cpus_allowed);
L
Linus Torvalds 已提交
2439
#ifdef CONFIG_SMP
2440 2441 2442
extern void do_set_cpus_allowed(struct task_struct *p,
			       const struct cpumask *new_mask);

2443
extern int set_cpus_allowed_ptr(struct task_struct *p,
2444
				const struct cpumask *new_mask);
L
Linus Torvalds 已提交
2445
#else
2446 2447 2448 2449
static inline void do_set_cpus_allowed(struct task_struct *p,
				      const struct cpumask *new_mask)
{
}
2450
static inline int set_cpus_allowed_ptr(struct task_struct *p,
2451
				       const struct cpumask *new_mask)
L
Linus Torvalds 已提交
2452
{
2453
	if (!cpumask_test_cpu(0, new_mask))
L
Linus Torvalds 已提交
2454 2455 2456 2457
		return -EINVAL;
	return 0;
}
#endif
2458

2459
#ifdef CONFIG_NO_HZ_COMMON
2460 2461 2462 2463 2464
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) { }
2465
#endif /* CONFIG_NO_HZ_COMMON */
2466

2467
/*
2468 2469 2470 2471 2472 2473
 * 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.
2474
 */
2475
extern unsigned long long notrace sched_clock(void);
2476
/*
2477
 * See the comment in kernel/sched/clock.c
2478
 */
2479
extern u64 running_clock(void);
2480 2481
extern u64 sched_clock_cpu(int cpu);

2482

2483
extern void sched_clock_init(void);
2484

2485
#ifndef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK
2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496
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)
{
}
2497 2498 2499 2500 2501 2502 2503 2504 2505 2506

static inline u64 cpu_clock(int cpu)
{
	return sched_clock();
}

static inline u64 local_clock(void)
{
	return sched_clock();
}
2507
#else
2508 2509 2510 2511 2512 2513
/*
 * 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:
 */
2514 2515 2516
extern int sched_clock_stable(void);
extern void set_sched_clock_stable(void);
extern void clear_sched_clock_stable(void);
2517

2518 2519 2520
extern void sched_clock_tick(void);
extern void sched_clock_idle_sleep_event(void);
extern void sched_clock_idle_wakeup_event(u64 delta_ns);
2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540

/*
 * As outlined in clock.c, provides a fast, high resolution, nanosecond
 * time source that is monotonic per cpu argument and has bounded drift
 * between cpus.
 *
 * ######################### BIG FAT WARNING ##########################
 * # when comparing cpu_clock(i) to cpu_clock(j) for i != j, time can #
 * # go backwards !!                                                  #
 * ####################################################################
 */
static inline u64 cpu_clock(int cpu)
{
	return sched_clock_cpu(cpu);
}

static inline u64 local_clock(void)
{
	return sched_clock_cpu(raw_smp_processor_id());
}
2541 2542
#endif

2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555
#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

2556
extern unsigned long long
2557
task_sched_runtime(struct task_struct *task);
L
Linus Torvalds 已提交
2558 2559 2560 2561 2562 2563 2564 2565

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

2566 2567
extern void sched_clock_idle_sleep_event(void);
extern void sched_clock_idle_wakeup_event(u64 delta_ns);
2568

L
Linus Torvalds 已提交
2569 2570 2571 2572 2573 2574
#ifdef CONFIG_HOTPLUG_CPU
extern void idle_task_exit(void);
#else
static inline void idle_task_exit(void) {}
#endif

2575
#if defined(CONFIG_NO_HZ_COMMON) && defined(CONFIG_SMP)
2576
extern void wake_up_nohz_cpu(int cpu);
2577
#else
2578
static inline void wake_up_nohz_cpu(int cpu) { }
2579 2580
#endif

2581
#ifdef CONFIG_NO_HZ_FULL
2582
extern u64 scheduler_tick_max_deferment(void);
2583 2584
#endif

2585 2586 2587 2588 2589 2590 2591
#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);
2592
extern int proc_sched_autogroup_set_nice(struct task_struct *p, int nice);
2593 2594 2595 2596 2597 2598 2599 2600
#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

2601
extern int yield_to(struct task_struct *p, bool preempt);
2602 2603
extern void set_user_nice(struct task_struct *p, long nice);
extern int task_prio(const struct task_struct *p);
2604 2605 2606 2607 2608 2609 2610 2611 2612 2613
/**
 * 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);
}
2614 2615
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
2616
extern int idle_cpu(int cpu);
2617 2618
extern int sched_setscheduler(struct task_struct *, int,
			      const struct sched_param *);
2619
extern int sched_setscheduler_nocheck(struct task_struct *, int,
2620
				      const struct sched_param *);
2621 2622
extern int sched_setattr(struct task_struct *,
			 const struct sched_attr *);
2623
extern struct task_struct *idle_task(int cpu);
2624 2625
/**
 * is_idle_task - is the specified task an idle task?
2626
 * @p: the task in question.
2627 2628
 *
 * Return: 1 if @p is an idle task. 0 otherwise.
2629
 */
2630
static inline bool is_idle_task(const struct task_struct *p)
2631 2632 2633
{
	return p->pid == 0;
}
2634
extern struct task_struct *curr_task(int cpu);
2635
extern void ia64_set_curr_task(int cpu, struct task_struct *p);
L
Linus Torvalds 已提交
2636 2637 2638 2639

void yield(void);

union thread_union {
2640
#ifndef CONFIG_THREAD_INFO_IN_TASK
L
Linus Torvalds 已提交
2641
	struct thread_info thread_info;
2642
#endif
L
Linus Torvalds 已提交
2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660
	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;

2661 2662 2663 2664 2665 2666 2667
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
2668 2669
 * find_task_by_vpid():
 *      finds a task by its virtual pid
2670
 *
2671
 * see also find_vpid() etc in include/linux/pid.h
2672 2673
 */

2674 2675 2676
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);
2677

L
Linus Torvalds 已提交
2678
/* per-UID process charging. */
2679
extern struct user_struct * alloc_uid(kuid_t);
L
Linus Torvalds 已提交
2680 2681 2682 2683 2684 2685 2686 2687 2688
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
2689
extern void xtime_update(unsigned long ticks);
L
Linus Torvalds 已提交
2690

2691 2692
extern int wake_up_state(struct task_struct *tsk, unsigned int state);
extern int wake_up_process(struct task_struct *tsk);
2693
extern void wake_up_new_task(struct task_struct *tsk);
L
Linus Torvalds 已提交
2694 2695 2696 2697 2698
#ifdef CONFIG_SMP
 extern void kick_process(struct task_struct *tsk);
#else
 static inline void kick_process(struct task_struct *tsk) { }
#endif
2699
extern int sched_fork(unsigned long clone_flags, struct task_struct *p);
2700
extern void sched_dead(struct task_struct *p);
L
Linus Torvalds 已提交
2701 2702 2703

extern void proc_caches_init(void);
extern void flush_signals(struct task_struct *);
2704
extern void ignore_signals(struct task_struct *);
L
Linus Torvalds 已提交
2705 2706 2707
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);

2708
static inline int kernel_dequeue_signal(siginfo_t *info)
L
Linus Torvalds 已提交
2709
{
2710 2711
	struct task_struct *tsk = current;
	siginfo_t __info;
L
Linus Torvalds 已提交
2712 2713
	int ret;

2714 2715 2716
	spin_lock_irq(&tsk->sighand->siglock);
	ret = dequeue_signal(tsk, &tsk->blocked, info ?: &__info);
	spin_unlock_irq(&tsk->sighand->siglock);
L
Linus Torvalds 已提交
2717 2718

	return ret;
2719
}
L
Linus Torvalds 已提交
2720

2721 2722 2723 2724 2725 2726 2727 2728 2729 2730
static inline void kernel_signal_stop(void)
{
	spin_lock_irq(&current->sighand->siglock);
	if (current->jobctl & JOBCTL_STOP_DEQUEUED)
		__set_current_state(TASK_STOPPED);
	spin_unlock_irq(&current->sighand->siglock);

	schedule();
}

L
Linus Torvalds 已提交
2731 2732 2733 2734
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 *);
2735 2736
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);
2737 2738
extern int kill_pid_info_as_cred(int, struct siginfo *, struct pid *,
				const struct cred *, u32);
2739 2740
extern int kill_pgrp(struct pid *pid, int sig, int priv);
extern int kill_pid(struct pid *pid, int sig, int priv);
2741
extern int kill_proc_info(int, struct siginfo *, pid_t);
2742
extern __must_check bool do_notify_parent(struct task_struct *, int);
2743
extern void __wake_up_parent(struct task_struct *p, struct task_struct *parent);
L
Linus Torvalds 已提交
2744 2745
extern void force_sig(int, struct task_struct *);
extern int send_sig(int, struct task_struct *, int);
2746
extern int zap_other_threads(struct task_struct *p);
L
Linus Torvalds 已提交
2747 2748
extern struct sigqueue *sigqueue_alloc(void);
extern void sigqueue_free(struct sigqueue *);
2749
extern int send_sigqueue(struct sigqueue *,  struct task_struct *, int group);
2750
extern int do_sigaction(int, struct k_sigaction *, struct k_sigaction *);
L
Linus Torvalds 已提交
2751

2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811
#ifdef TIF_RESTORE_SIGMASK
/*
 * Legacy restore_sigmask accessors.  These are inefficient on
 * SMP architectures because they require atomic operations.
 */

/**
 * set_restore_sigmask() - make sure saved_sigmask processing gets done
 *
 * This sets TIF_RESTORE_SIGMASK and ensures that the arch signal code
 * will run before returning to user mode, to process the flag.  For
 * all callers, TIF_SIGPENDING is already set or it's no harm to set
 * it.  TIF_RESTORE_SIGMASK need not be in the set of bits that the
 * arch code will notice on return to user mode, in case those bits
 * are scarce.  We set TIF_SIGPENDING here to ensure that the arch
 * signal code always gets run when TIF_RESTORE_SIGMASK is set.
 */
static inline void set_restore_sigmask(void)
{
	set_thread_flag(TIF_RESTORE_SIGMASK);
	WARN_ON(!test_thread_flag(TIF_SIGPENDING));
}
static inline void clear_restore_sigmask(void)
{
	clear_thread_flag(TIF_RESTORE_SIGMASK);
}
static inline bool test_restore_sigmask(void)
{
	return test_thread_flag(TIF_RESTORE_SIGMASK);
}
static inline bool test_and_clear_restore_sigmask(void)
{
	return test_and_clear_thread_flag(TIF_RESTORE_SIGMASK);
}

#else	/* TIF_RESTORE_SIGMASK */

/* Higher-quality implementation, used if TIF_RESTORE_SIGMASK doesn't exist. */
static inline void set_restore_sigmask(void)
{
	current->restore_sigmask = true;
	WARN_ON(!test_thread_flag(TIF_SIGPENDING));
}
static inline void clear_restore_sigmask(void)
{
	current->restore_sigmask = false;
}
static inline bool test_restore_sigmask(void)
{
	return current->restore_sigmask;
}
static inline bool test_and_clear_restore_sigmask(void)
{
	if (!current->restore_sigmask)
		return false;
	current->restore_sigmask = false;
	return true;
}
#endif

A
Al Viro 已提交
2812 2813 2814
static inline void restore_saved_sigmask(void)
{
	if (test_and_clear_restore_sigmask())
2815
		__set_current_blocked(&current->saved_sigmask);
A
Al Viro 已提交
2816 2817
}

A
Al Viro 已提交
2818 2819 2820 2821 2822 2823 2824 2825
static inline sigset_t *sigmask_to_save(void)
{
	sigset_t *res = &current->blocked;
	if (unlikely(test_restore_sigmask()))
		res = &current->saved_sigmask;
	return res;
}

2826 2827 2828 2829 2830
static inline int kill_cad_pid(int sig, int priv)
{
	return kill_pid(cad_pid, sig, priv);
}

L
Linus Torvalds 已提交
2831 2832 2833 2834 2835
/* 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)

2836 2837 2838
/*
 * True if we are on the alternate signal stack.
 */
L
Linus Torvalds 已提交
2839 2840
static inline int on_sig_stack(unsigned long sp)
{
2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852
	/*
	 * If the signal stack is SS_AUTODISARM then, by construction, we
	 * can't be on the signal stack unless user code deliberately set
	 * SS_AUTODISARM when we were already on it.
	 *
	 * This improves reliability: if user state gets corrupted such that
	 * the stack pointer points very close to the end of the signal stack,
	 * then this check will enable the signal to be handled anyway.
	 */
	if (current->sas_ss_flags & SS_AUTODISARM)
		return 0;

2853 2854 2855 2856 2857 2858 2859
#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
2860 2861 2862 2863
}

static inline int sas_ss_flags(unsigned long sp)
{
2864 2865 2866 2867
	if (!current->sas_ss_size)
		return SS_DISABLE;

	return on_sig_stack(sp) ? SS_ONSTACK : 0;
L
Linus Torvalds 已提交
2868 2869
}

2870 2871 2872 2873 2874 2875 2876
static inline void sas_ss_reset(struct task_struct *p)
{
	p->sas_ss_sp = 0;
	p->sas_ss_size = 0;
	p->sas_ss_flags = SS_DISABLE;
}

A
Al Viro 已提交
2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887
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
2888 2889 2890 2891 2892 2893
/*
 * Routines for handling mm_structs
 */
extern struct mm_struct * mm_alloc(void);

/* mmdrop drops the mm and the page tables */
2894
extern void __mmdrop(struct mm_struct *);
2895
static inline void mmdrop(struct mm_struct *mm)
L
Linus Torvalds 已提交
2896
{
I
Ingo Molnar 已提交
2897
	if (unlikely(atomic_dec_and_test(&mm->mm_count)))
L
Linus Torvalds 已提交
2898 2899 2900
		__mmdrop(mm);
}

2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914
static inline void mmdrop_async_fn(struct work_struct *work)
{
	struct mm_struct *mm = container_of(work, struct mm_struct, async_put_work);
	__mmdrop(mm);
}

static inline void mmdrop_async(struct mm_struct *mm)
{
	if (unlikely(atomic_dec_and_test(&mm->mm_count))) {
		INIT_WORK(&mm->async_put_work, mmdrop_async_fn);
		schedule_work(&mm->async_put_work);
	}
}

2915 2916 2917 2918 2919
static inline bool mmget_not_zero(struct mm_struct *mm)
{
	return atomic_inc_not_zero(&mm->mm_users);
}

L
Linus Torvalds 已提交
2920 2921
/* mmput gets rid of the mappings and all user-space */
extern void mmput(struct mm_struct *);
M
Michal Hocko 已提交
2922 2923
#ifdef CONFIG_MMU
/* same as above but performs the slow path from the async context. Can
2924 2925 2926
 * be called from the atomic context as well
 */
extern void mmput_async(struct mm_struct *);
M
Michal Hocko 已提交
2927
#endif
2928

L
Linus Torvalds 已提交
2929 2930
/* 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);
2931 2932 2933 2934 2935 2936
/*
 * 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
2937 2938 2939
/* Remove the current tasks stale references to the old mm_struct */
extern void mm_release(struct task_struct *, struct mm_struct *);

2940 2941 2942 2943
#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
2944
extern int copy_thread(unsigned long, unsigned long, unsigned long,
2945
			struct task_struct *);
2946 2947 2948 2949 2950 2951 2952 2953 2954 2955

/* 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
2956
extern void flush_thread(void);
J
Jiri Slaby 已提交
2957 2958

#ifdef CONFIG_HAVE_EXIT_THREAD
2959
extern void exit_thread(struct task_struct *tsk);
J
Jiri Slaby 已提交
2960
#else
2961
static inline void exit_thread(struct task_struct *tsk)
J
Jiri Slaby 已提交
2962 2963 2964
{
}
#endif
L
Linus Torvalds 已提交
2965 2966

extern void exit_files(struct task_struct *);
2967
extern void __cleanup_sighand(struct sighand_struct *);
2968

L
Linus Torvalds 已提交
2969
extern void exit_itimers(struct signal_struct *);
2970
extern void flush_itimer_signals(void);
L
Linus Torvalds 已提交
2971

2972
extern void do_group_exit(int);
L
Linus Torvalds 已提交
2973

2974
extern int do_execve(struct filename *,
2975
		     const char __user * const __user *,
2976
		     const char __user * const __user *);
2977 2978 2979 2980
extern int do_execveat(int, struct filename *,
		       const char __user * const __user *,
		       const char __user * const __user *,
		       int);
2981
extern long _do_fork(unsigned long, unsigned long, unsigned long, int __user *, int __user *, unsigned long);
2982
extern long do_fork(unsigned long, unsigned long, unsigned long, int __user *, int __user *);
2983
struct task_struct *fork_idle(int);
2984
extern pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
L
Linus Torvalds 已提交
2985

2986 2987 2988 2989 2990
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);
}
2991
extern char *get_task_comm(char *to, struct task_struct *tsk);
L
Linus Torvalds 已提交
2992 2993

#ifdef CONFIG_SMP
2994
void scheduler_ipi(void);
R
Roland McGrath 已提交
2995
extern unsigned long wait_task_inactive(struct task_struct *, long match_state);
L
Linus Torvalds 已提交
2996
#else
2997
static inline void scheduler_ipi(void) { }
R
Roland McGrath 已提交
2998 2999 3000 3001 3002
static inline unsigned long wait_task_inactive(struct task_struct *p,
					       long match_state)
{
	return 1;
}
L
Linus Torvalds 已提交
3003 3004
#endif

3005 3006 3007
#define tasklist_empty() \
	list_empty(&init_task.tasks)

3008 3009
#define next_task(p) \
	list_entry_rcu((p)->tasks.next, struct task_struct, tasks)
L
Linus Torvalds 已提交
3010 3011 3012 3013

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

3014
extern bool current_is_single_threaded(void);
D
David Howells 已提交
3015

L
Linus Torvalds 已提交
3016 3017 3018 3019 3020 3021 3022 3023 3024 3025
/*
 * 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)

3026 3027 3028 3029 3030 3031 3032 3033 3034 3035
#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)

3036 3037
static inline int get_nr_threads(struct task_struct *tsk)
{
3038
	return tsk->signal->nr_threads;
3039 3040
}

3041 3042 3043 3044
static inline bool thread_group_leader(struct task_struct *p)
{
	return p->exit_signal >= 0;
}
L
Linus Torvalds 已提交
3045

3046 3047 3048 3049 3050 3051
/* 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.
 */
3052
static inline bool has_group_leader_pid(struct task_struct *p)
3053
{
3054
	return task_pid(p) == p->signal->leader_pid;
3055 3056
}

3057
static inline
3058
bool same_thread_group(struct task_struct *p1, struct task_struct *p2)
3059
{
3060
	return p1->signal == p2->signal;
3061 3062
}

3063
static inline struct task_struct *next_thread(const struct task_struct *p)
O
Oleg Nesterov 已提交
3064
{
3065 3066
	return list_entry_rcu(p->thread_group.next,
			      struct task_struct, thread_group);
O
Oleg Nesterov 已提交
3067 3068
}

A
Alexey Dobriyan 已提交
3069
static inline int thread_group_empty(struct task_struct *p)
L
Linus Torvalds 已提交
3070
{
O
Oleg Nesterov 已提交
3071
	return list_empty(&p->thread_group);
L
Linus Torvalds 已提交
3072 3073 3074 3075 3076 3077
}

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

/*
3078
 * Protects ->fs, ->files, ->mm, ->group_info, ->comm, keyring
3079
 * subscriptions and synchronises with wait4().  Also used in procfs.  Also
3080
 * pins the final release of task.io_context.  Also protects ->cpuset and
O
Oleg Nesterov 已提交
3081
 * ->cgroup.subsys[]. And ->vfork_done.
L
Linus Torvalds 已提交
3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096
 *
 * 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);
}

3097
extern struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
3098 3099
							unsigned long *flags);

3100 3101 3102 3103 3104 3105 3106 3107 3108
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;
}
3109

3110 3111 3112 3113 3114 3115
static inline void unlock_task_sighand(struct task_struct *tsk,
						unsigned long *flags)
{
	spin_unlock_irqrestore(&tsk->sighand->siglock, *flags);
}

3116
/**
3117 3118
 * threadgroup_change_begin - mark the beginning of changes to a threadgroup
 * @tsk: task causing the changes
3119
 *
3120 3121 3122 3123 3124 3125
 * 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.
3126
 */
3127
static inline void threadgroup_change_begin(struct task_struct *tsk)
3128
{
3129 3130
	might_sleep();
	cgroup_threadgroup_change_begin(tsk);
3131
}
3132 3133

/**
3134 3135
 * threadgroup_change_end - mark the end of changes to a threadgroup
 * @tsk: task causing the changes
3136
 *
3137
 * See threadgroup_change_begin().
3138
 */
3139
static inline void threadgroup_change_end(struct task_struct *tsk)
3140
{
3141
	cgroup_threadgroup_change_end(tsk);
3142 3143
}

3144 3145 3146 3147 3148 3149
#ifdef CONFIG_THREAD_INFO_IN_TASK

static inline struct thread_info *task_thread_info(struct task_struct *task)
{
	return &task->thread_info;
}
3150 3151 3152 3153 3154 3155

/*
 * When accessing the stack of a non-current task that might exit, use
 * try_get_task_stack() instead.  task_stack_page will return a pointer
 * that could get freed out from under you.
 */
3156 3157 3158 3159
static inline void *task_stack_page(const struct task_struct *task)
{
	return task->stack;
}
3160

3161
#define setup_thread_stack(new,old)	do { } while(0)
3162

3163 3164 3165 3166 3167 3168
static inline unsigned long *end_of_stack(const struct task_struct *task)
{
	return task->stack;
}

#elif !defined(__HAVE_THREAD_FUNCTIONS)
A
Al Viro 已提交
3169

R
Roman Zippel 已提交
3170
#define task_thread_info(task)	((struct thread_info *)(task)->stack)
3171
#define task_stack_page(task)	((void *)(task)->stack)
A
Al Viro 已提交
3172

3173 3174 3175 3176 3177 3178
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;
}

3179 3180 3181 3182 3183 3184 3185 3186 3187
/*
 * 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.
 */
3188 3189
static inline unsigned long *end_of_stack(struct task_struct *p)
{
3190 3191 3192
#ifdef CONFIG_STACK_GROWSUP
	return (unsigned long *)((unsigned long)task_thread_info(p) + THREAD_SIZE) - 1;
#else
R
Roman Zippel 已提交
3193
	return (unsigned long *)(task_thread_info(p) + 1);
3194
#endif
3195 3196
}

A
Al Viro 已提交
3197
#endif
3198

3199 3200 3201 3202 3203 3204 3205 3206 3207
#ifdef CONFIG_THREAD_INFO_IN_TASK
static inline void *try_get_task_stack(struct task_struct *tsk)
{
	return atomic_inc_not_zero(&tsk->stack_refcount) ?
		task_stack_page(tsk) : NULL;
}

extern void put_task_stack(struct task_struct *tsk);
#else
3208 3209 3210 3211 3212 3213
static inline void *try_get_task_stack(struct task_struct *tsk)
{
	return task_stack_page(tsk);
}

static inline void put_task_stack(struct task_struct *tsk) {}
3214
#endif
3215

3216 3217
#define task_stack_end_corrupted(task) \
		(*(end_of_stack(task)) != STACK_END_MAGIC)
A
Al Viro 已提交
3218

3219 3220 3221 3222 3223 3224 3225
static inline int object_is_on_stack(void *obj)
{
	void *stack = task_stack_page(current);

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

3226
extern void thread_stack_cache_init(void);
3227

3228 3229 3230 3231 3232 3233
#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 */
3234 3235 3236
# ifdef CONFIG_STACK_GROWSUP
		n--;
# else
3237
		n++;
3238
# endif
3239 3240
	} while (!*n);

3241 3242 3243
# ifdef CONFIG_STACK_GROWSUP
	return (unsigned long)end_of_stack(p) - (unsigned long)n;
# else
3244
	return (unsigned long)n - (unsigned long)end_of_stack(p);
3245
# endif
3246 3247
}
#endif
3248
extern void set_task_stack_end_magic(struct task_struct *tsk);
3249

L
Linus Torvalds 已提交
3250 3251 3252 3253 3254
/* 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
3255
	set_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
3256 3257 3258 3259
}

static inline void clear_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
3260
	clear_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
3261 3262 3263 3264
}

static inline int test_and_set_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
3265
	return test_and_set_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
3266 3267 3268 3269
}

static inline int test_and_clear_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
3270
	return test_and_clear_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
3271 3272 3273 3274
}

static inline int test_tsk_thread_flag(struct task_struct *tsk, int flag)
{
A
Al Viro 已提交
3275
	return test_ti_thread_flag(task_thread_info(tsk), flag);
L
Linus Torvalds 已提交
3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287
}

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);
}

3288 3289 3290 3291 3292
static inline int test_tsk_need_resched(struct task_struct *tsk)
{
	return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED));
}

3293 3294 3295 3296 3297 3298
static inline int restart_syscall(void)
{
	set_tsk_thread_flag(current, TIF_SIGPENDING);
	return -ERESTARTNOINTR;
}

L
Linus Torvalds 已提交
3299 3300 3301 3302
static inline int signal_pending(struct task_struct *p)
{
	return unlikely(test_tsk_thread_flag(p,TIF_SIGPENDING));
}
M
Matthew Wilcox 已提交
3303

3304 3305 3306 3307
static inline int __fatal_signal_pending(struct task_struct *p)
{
	return unlikely(sigismember(&p->pending.signal, SIGKILL));
}
M
Matthew Wilcox 已提交
3308 3309 3310 3311 3312 3313

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

3314 3315 3316 3317 3318 3319 3320 3321 3322 3323
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
3324 3325 3326 3327 3328 3329 3330
/*
 * 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.
 */
3331
#ifndef CONFIG_PREEMPT
3332
extern int _cond_resched(void);
3333 3334 3335
#else
static inline int _cond_resched(void) { return 0; }
#endif
3336

3337
#define cond_resched() ({			\
3338
	___might_sleep(__FILE__, __LINE__, 0);	\
3339 3340
	_cond_resched();			\
})
3341

3342 3343 3344
extern int __cond_resched_lock(spinlock_t *lock);

#define cond_resched_lock(lock) ({				\
3345
	___might_sleep(__FILE__, __LINE__, PREEMPT_LOCK_OFFSET);\
3346 3347 3348 3349 3350
	__cond_resched_lock(lock);				\
})

extern int __cond_resched_softirq(void);

3351
#define cond_resched_softirq() ({					\
3352
	___might_sleep(__FILE__, __LINE__, SOFTIRQ_DISABLE_OFFSET);	\
3353
	__cond_resched_softirq();					\
3354
})
L
Linus Torvalds 已提交
3355

3356 3357 3358 3359 3360 3361 3362 3363 3364
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
}

3365 3366 3367 3368 3369 3370 3371 3372 3373
static inline unsigned long get_preempt_disable_ip(struct task_struct *p)
{
#ifdef CONFIG_DEBUG_PREEMPT
	return p->preempt_disable_ip;
#else
	return 0;
#endif
}

L
Linus Torvalds 已提交
3374 3375
/*
 * Does a critical section need to be broken due to another
N
Nick Piggin 已提交
3376 3377
 * task waiting?: (technically does not depend on CONFIG_PREEMPT,
 * but a general need for low latency)
L
Linus Torvalds 已提交
3378
 */
N
Nick Piggin 已提交
3379
static inline int spin_needbreak(spinlock_t *lock)
L
Linus Torvalds 已提交
3380
{
N
Nick Piggin 已提交
3381 3382 3383
#ifdef CONFIG_PREEMPT
	return spin_is_contended(lock);
#else
L
Linus Torvalds 已提交
3384
	return 0;
N
Nick Piggin 已提交
3385
#endif
L
Linus Torvalds 已提交
3386 3387
}

3388 3389
/*
 * Idle thread specific functions to determine the need_resched
3390
 * polling state.
3391
 */
3392
#ifdef TIF_POLLING_NRFLAG
3393 3394 3395 3396
static inline int tsk_is_polling(struct task_struct *p)
{
	return test_tsk_thread_flag(p, TIF_POLLING_NRFLAG);
}
3397 3398

static inline void __current_set_polling(void)
3399 3400 3401 3402
{
	set_thread_flag(TIF_POLLING_NRFLAG);
}

3403 3404 3405 3406 3407 3408
static inline bool __must_check current_set_polling_and_test(void)
{
	__current_set_polling();

	/*
	 * Polling state must be visible before we test NEED_RESCHED,
3409
	 * paired by resched_curr()
3410
	 */
3411
	smp_mb__after_atomic();
3412 3413 3414 3415 3416

	return unlikely(tif_need_resched());
}

static inline void __current_clr_polling(void)
3417 3418 3419
{
	clear_thread_flag(TIF_POLLING_NRFLAG);
}
3420 3421 3422 3423 3424 3425 3426

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

	/*
	 * Polling state must be visible before we test NEED_RESCHED,
3427
	 * paired by resched_curr()
3428
	 */
3429
	smp_mb__after_atomic();
3430 3431 3432 3433

	return unlikely(tif_need_resched());
}

3434 3435
#else
static inline int tsk_is_polling(struct task_struct *p) { return 0; }
3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446
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());
}
3447 3448
#endif

3449 3450 3451 3452 3453 3454 3455 3456 3457 3458
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.
	 */
3459
	smp_mb(); /* paired with resched_curr() */
3460 3461 3462 3463

	preempt_fold_need_resched();
}

3464 3465 3466 3467 3468
static __always_inline bool need_resched(void)
{
	return unlikely(tif_need_resched());
}

3469 3470 3471
/*
 * Thread group CPU time accounting.
 */
3472
void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times);
3473
void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times);
3474

R
Roland McGrath 已提交
3475 3476 3477 3478 3479 3480 3481
/*
 * 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
3482 3483
extern void recalc_sigpending(void);

3484 3485 3486 3487 3488 3489 3490 3491 3492 3493
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
3494 3495 3496 3497 3498 3499 3500 3501

/*
 * 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)
{
3502 3503 3504
#ifdef CONFIG_THREAD_INFO_IN_TASK
	return p->cpu;
#else
A
Al Viro 已提交
3505
	return task_thread_info(p)->cpu;
3506
#endif
L
Linus Torvalds 已提交
3507 3508
}

I
Ingo Molnar 已提交
3509 3510 3511 3512 3513
static inline int task_node(const struct task_struct *p)
{
	return cpu_to_node(task_cpu(p));
}

I
Ingo Molnar 已提交
3514
extern void set_task_cpu(struct task_struct *p, unsigned int cpu);
L
Linus Torvalds 已提交
3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528

#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 */

3529 3530
extern long sched_setaffinity(pid_t pid, const struct cpumask *new_mask);
extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
3531

D
Dhaval Giani 已提交
3532
#ifdef CONFIG_CGROUP_SCHED
3533
extern struct task_group root_task_group;
P
Peter Zijlstra 已提交
3534
#endif /* CONFIG_CGROUP_SCHED */
3535

3536 3537 3538
extern int task_can_switch_user(struct user_struct *up,
					struct task_struct *tsk);

3539 3540 3541
#ifdef CONFIG_TASK_XACCT
static inline void add_rchar(struct task_struct *tsk, ssize_t amt)
{
3542
	tsk->ioac.rchar += amt;
3543 3544 3545 3546
}

static inline void add_wchar(struct task_struct *tsk, ssize_t amt)
{
3547
	tsk->ioac.wchar += amt;
3548 3549 3550 3551
}

static inline void inc_syscr(struct task_struct *tsk)
{
3552
	tsk->ioac.syscr++;
3553 3554 3555 3556
}

static inline void inc_syscw(struct task_struct *tsk)
{
3557
	tsk->ioac.syscw++;
3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576
}
#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
3577 3578 3579 3580
#ifndef TASK_SIZE_OF
#define TASK_SIZE_OF(tsk)	TASK_SIZE
#endif

O
Oleg Nesterov 已提交
3581
#ifdef CONFIG_MEMCG
3582 3583 3584 3585 3586
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
3587
#endif /* CONFIG_MEMCG */
3588

3589 3590 3591
static inline unsigned long task_rlimit(const struct task_struct *tsk,
		unsigned int limit)
{
3592
	return READ_ONCE(tsk->signal->rlim[limit].rlim_cur);
3593 3594 3595 3596 3597
}

static inline unsigned long task_rlimit_max(const struct task_struct *tsk,
		unsigned int limit)
{
3598
	return READ_ONCE(tsk->signal->rlim[limit].rlim_max);
3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610
}

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);
}

3611 3612
#define SCHED_CPUFREQ_RT	(1U << 0)
#define SCHED_CPUFREQ_DL	(1U << 1)
3613
#define SCHED_CPUFREQ_IOWAIT	(1U << 2)
3614 3615 3616

#define SCHED_CPUFREQ_RT_DL	(SCHED_CPUFREQ_RT | SCHED_CPUFREQ_DL)

3617 3618
#ifdef CONFIG_CPU_FREQ
struct update_util_data {
3619
       void (*func)(struct update_util_data *data, u64 time, unsigned int flags);
3620 3621
};

3622
void cpufreq_add_update_util_hook(int cpu, struct update_util_data *data,
3623 3624
                       void (*func)(struct update_util_data *data, u64 time,
				    unsigned int flags));
3625
void cpufreq_remove_update_util_hook(int cpu);
3626 3627
#endif /* CONFIG_CPU_FREQ */

L
Linus Torvalds 已提交
3628
#endif