sched.h 50.3 KB
Newer Older
1 2

#include <linux/sched.h>
3
#include <linux/sched/autogroup.h>
4
#include <linux/sched/sysctl.h>
5
#include <linux/sched/topology.h>
6
#include <linux/sched/rt.h>
7
#include <linux/sched/deadline.h>
8
#include <linux/sched/clock.h>
9
#include <linux/sched/wake_q.h>
10
#include <linux/sched/signal.h>
11
#include <linux/sched/numa_balancing.h>
12
#include <linux/sched/mm.h>
13
#include <linux/sched/cpufreq.h>
14
#include <linux/sched/stat.h>
15
#include <linux/sched/nohz.h>
16
#include <linux/sched/debug.h>
17
#include <linux/sched/hotplug.h>
18
#include <linux/sched/task.h>
19
#include <linux/sched/task_stack.h>
20
#include <linux/sched/cputime.h>
21
#include <linux/sched/init.h>
22

23
#include <linux/u64_stats_sync.h>
24
#include <linux/kernel_stat.h>
25
#include <linux/binfmts.h>
26 27 28
#include <linux/mutex.h>
#include <linux/spinlock.h>
#include <linux/stop_machine.h>
29
#include <linux/irq_work.h>
30
#include <linux/tick.h>
31
#include <linux/slab.h>
32

33 34 35 36
#ifdef CONFIG_PARAVIRT
#include <asm/paravirt.h>
#endif

37
#include "cpupri.h"
38
#include "cpudeadline.h"
39
#include "cpuacct.h"
40

41 42 43 44 45 46
#ifdef CONFIG_SCHED_DEBUG
#define SCHED_WARN_ON(x)	WARN_ONCE(x, #x)
#else
#define SCHED_WARN_ON(x)	((void)(x))
#endif

47
struct rq;
48
struct cpuidle_state;
49

50 51
/* task_struct::on_rq states: */
#define TASK_ON_RQ_QUEUED	1
52
#define TASK_ON_RQ_MIGRATING	2
53

54 55
extern __read_mostly int scheduler_running;

56 57 58
extern unsigned long calc_load_update;
extern atomic_long_t calc_load_tasks;

59
extern void calc_global_load_tick(struct rq *this_rq);
60
extern long calc_load_fold_active(struct rq *this_rq, long adjust);
61 62

#ifdef CONFIG_SMP
63
extern void cpu_load_update_active(struct rq *this_rq);
64
#else
65
static inline void cpu_load_update_active(struct rq *this_rq) { }
66
#endif
67

68 69 70 71 72
/*
 * Helpers for converting nanosecond timing to jiffy resolution
 */
#define NS_TO_JIFFIES(TIME)	((unsigned long)(TIME) / (NSEC_PER_SEC / HZ))

73 74 75 76 77 78 79 80
/*
 * Increase resolution of nice-level calculations for 64-bit architectures.
 * The extra resolution improves shares distribution and load balancing of
 * low-weight task groups (eg. nice +19 on an autogroup), deeper taskgroup
 * hierarchies, especially on larger systems. This is not a user-visible change
 * and does not change the user-interface for setting shares/weights.
 *
 * We increase resolution only if we have enough bits to allow this increased
81 82 83 84 85
 * resolution (i.e. 64bit). The costs for increasing resolution when 32bit are
 * pretty high and the returns do not justify the increased costs.
 *
 * Really only required when CONFIG_FAIR_GROUP_SCHED is also set, but to
 * increase coverage and consistency always enable it on 64bit platforms.
86
 */
87
#ifdef CONFIG_64BIT
88
# define NICE_0_LOAD_SHIFT	(SCHED_FIXEDPOINT_SHIFT + SCHED_FIXEDPOINT_SHIFT)
89 90
# define scale_load(w)		((w) << SCHED_FIXEDPOINT_SHIFT)
# define scale_load_down(w)	((w) >> SCHED_FIXEDPOINT_SHIFT)
91
#else
92
# define NICE_0_LOAD_SHIFT	(SCHED_FIXEDPOINT_SHIFT)
93 94 95 96
# define scale_load(w)		(w)
# define scale_load_down(w)	(w)
#endif

97
/*
98 99 100 101 102 103 104
 * Task weight (visible to users) and its load (invisible to users) have
 * independent resolution, but they should be well calibrated. We use
 * scale_load() and scale_load_down(w) to convert between them. The
 * following must be true:
 *
 *  scale_load(sched_prio_to_weight[USER_PRIO(NICE_TO_PRIO(0))]) == NICE_0_LOAD
 *
105
 */
106
#define NICE_0_LOAD		(1L << NICE_0_LOAD_SHIFT)
107

108 109 110 111 112 113 114
/*
 * Single value that decides SCHED_DEADLINE internal math precision.
 * 10 -> just above 1us
 * 9  -> just above 0.5us
 */
#define DL_SCALE (10)

115 116 117 118 119 120 121 122 123
/*
 * These are the 'tuning knobs' of the scheduler:
 */

/*
 * single value that denotes runtime == period, ie unlimited time.
 */
#define RUNTIME_INF	((u64)~0ULL)

124 125 126 127
static inline int idle_policy(int policy)
{
	return policy == SCHED_IDLE;
}
128 129 130 131 132
static inline int fair_policy(int policy)
{
	return policy == SCHED_NORMAL || policy == SCHED_BATCH;
}

133 134
static inline int rt_policy(int policy)
{
135
	return policy == SCHED_FIFO || policy == SCHED_RR;
136 137
}

138 139 140 141
static inline int dl_policy(int policy)
{
	return policy == SCHED_DEADLINE;
}
142 143 144 145 146
static inline bool valid_policy(int policy)
{
	return idle_policy(policy) || fair_policy(policy) ||
		rt_policy(policy) || dl_policy(policy);
}
147

148 149 150 151 152
static inline int task_has_rt_policy(struct task_struct *p)
{
	return rt_policy(p->policy);
}

153 154 155 156 157
static inline int task_has_dl_policy(struct task_struct *p)
{
	return dl_policy(p->policy);
}

158 159 160
/*
 * Tells if entity @a should preempt entity @b.
 */
161 162
static inline bool
dl_entity_preempt(struct sched_dl_entity *a, struct sched_dl_entity *b)
163 164 165 166
{
	return dl_time_before(a->deadline, b->deadline);
}

167 168 169 170 171 172 173 174 175 176 177 178 179 180
/*
 * This is the priority-queue data structure of the RT scheduling class:
 */
struct rt_prio_array {
	DECLARE_BITMAP(bitmap, MAX_RT_PRIO+1); /* include 1 bit for delimiter */
	struct list_head queue[MAX_RT_PRIO];
};

struct rt_bandwidth {
	/* nests inside the rq lock: */
	raw_spinlock_t		rt_runtime_lock;
	ktime_t			rt_period;
	u64			rt_runtime;
	struct hrtimer		rt_period_timer;
P
Peter Zijlstra 已提交
181
	unsigned int		rt_period_active;
182
};
183 184 185

void __dl_clear_params(struct task_struct *p);

186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217
/*
 * To keep the bandwidth of -deadline tasks and groups under control
 * we need some place where:
 *  - store the maximum -deadline bandwidth of the system (the group);
 *  - cache the fraction of that bandwidth that is currently allocated.
 *
 * This is all done in the data structure below. It is similar to the
 * one used for RT-throttling (rt_bandwidth), with the main difference
 * that, since here we are only interested in admission control, we
 * do not decrease any runtime while the group "executes", neither we
 * need a timer to replenish it.
 *
 * With respect to SMP, the bandwidth is given on a per-CPU basis,
 * meaning that:
 *  - dl_bw (< 100%) is the bandwidth of the system (group) on each CPU;
 *  - dl_total_bw array contains, in the i-eth element, the currently
 *    allocated bandwidth on the i-eth CPU.
 * Moreover, groups consume bandwidth on each CPU, while tasks only
 * consume bandwidth on the CPU they're running on.
 * Finally, dl_total_bw_cpu is used to cache the index of dl_total_bw
 * that will be shown the next time the proc or cgroup controls will
 * be red. It on its turn can be changed by writing on its own
 * control.
 */
struct dl_bandwidth {
	raw_spinlock_t dl_runtime_lock;
	u64 dl_runtime;
	u64 dl_period;
};

static inline int dl_bandwidth_enabled(void)
{
218
	return sysctl_sched_rt_runtime >= 0;
219 220 221 222 223 224 225 226 227
}

extern struct dl_bw *dl_bw_of(int i);

struct dl_bw {
	raw_spinlock_t lock;
	u64 bw, total_bw;
};

228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246
static inline
void __dl_clear(struct dl_bw *dl_b, u64 tsk_bw)
{
	dl_b->total_bw -= tsk_bw;
}

static inline
void __dl_add(struct dl_bw *dl_b, u64 tsk_bw)
{
	dl_b->total_bw += tsk_bw;
}

static inline
bool __dl_overflow(struct dl_bw *dl_b, int cpus, u64 old_bw, u64 new_bw)
{
	return dl_b->bw != -1 &&
	       dl_b->bw * cpus < dl_b->total_bw - old_bw + new_bw;
}

247
extern void init_dl_bw(struct dl_bw *dl_b);
248 249 250 251 252 253 254 255

#ifdef CONFIG_CGROUP_SCHED

#include <linux/cgroup.h>

struct cfs_rq;
struct rt_rq;

256
extern struct list_head task_groups;
257 258 259 260 261 262

struct cfs_bandwidth {
#ifdef CONFIG_CFS_BANDWIDTH
	raw_spinlock_t lock;
	ktime_t period;
	u64 quota, runtime;
263
	s64 hierarchical_quota;
264 265
	u64 runtime_expires;

P
Peter Zijlstra 已提交
266
	int idle, period_active;
267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286
	struct hrtimer period_timer, slack_timer;
	struct list_head throttled_cfs_rq;

	/* statistics */
	int nr_periods, nr_throttled;
	u64 throttled_time;
#endif
};

/* task group related information */
struct task_group {
	struct cgroup_subsys_state css;

#ifdef CONFIG_FAIR_GROUP_SCHED
	/* schedulable entities of this group on each cpu */
	struct sched_entity **se;
	/* runqueue "owned" by this group on each cpu */
	struct cfs_rq **cfs_rq;
	unsigned long shares;

287
#ifdef	CONFIG_SMP
288 289 290 291 292 293
	/*
	 * load_avg can be heavily contended at clock tick time, so put
	 * it in its own cacheline separated from the fields above which
	 * will also be accessed at each tick.
	 */
	atomic_long_t load_avg ____cacheline_aligned;
294
#endif
295
#endif
296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352

#ifdef CONFIG_RT_GROUP_SCHED
	struct sched_rt_entity **rt_se;
	struct rt_rq **rt_rq;

	struct rt_bandwidth rt_bandwidth;
#endif

	struct rcu_head rcu;
	struct list_head list;

	struct task_group *parent;
	struct list_head siblings;
	struct list_head children;

#ifdef CONFIG_SCHED_AUTOGROUP
	struct autogroup *autogroup;
#endif

	struct cfs_bandwidth cfs_bandwidth;
};

#ifdef CONFIG_FAIR_GROUP_SCHED
#define ROOT_TASK_GROUP_LOAD	NICE_0_LOAD

/*
 * A weight of 0 or 1 can cause arithmetics problems.
 * A weight of a cfs_rq is the sum of weights of which entities
 * are queued on this cfs_rq, so a weight of a entity should not be
 * too large, so as the shares value of a task group.
 * (The default weight is 1024 - so there's no practical
 *  limitation from this.)
 */
#define MIN_SHARES	(1UL <<  1)
#define MAX_SHARES	(1UL << 18)
#endif

typedef int (*tg_visitor)(struct task_group *, void *);

extern int walk_tg_tree_from(struct task_group *from,
			     tg_visitor down, tg_visitor up, void *data);

/*
 * Iterate the full tree, calling @down when first entering a node and @up when
 * leaving it for the final time.
 *
 * Caller must hold rcu_lock or sufficient equivalent.
 */
static inline int walk_tg_tree(tg_visitor down, tg_visitor up, void *data)
{
	return walk_tg_tree_from(&root_task_group, down, up, data);
}

extern int tg_nop(struct task_group *tg, void *data);

extern void free_fair_sched_group(struct task_group *tg);
extern int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent);
353
extern void online_fair_sched_group(struct task_group *tg);
354
extern void unregister_fair_sched_group(struct task_group *tg);
355 356 357 358 359 360
extern void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq,
			struct sched_entity *se, int cpu,
			struct sched_entity *parent);
extern void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b);

extern void __refill_cfs_bandwidth_runtime(struct cfs_bandwidth *cfs_b);
P
Peter Zijlstra 已提交
361
extern void start_cfs_bandwidth(struct cfs_bandwidth *cfs_b);
362 363 364 365 366 367 368 369
extern void unthrottle_cfs_rq(struct cfs_rq *cfs_rq);

extern void free_rt_sched_group(struct task_group *tg);
extern int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent);
extern void init_tg_rt_entry(struct task_group *tg, struct rt_rq *rt_rq,
		struct sched_rt_entity *rt_se, int cpu,
		struct sched_rt_entity *parent);

370 371 372 373 374 375 376 377 378 379
extern struct task_group *sched_create_group(struct task_group *parent);
extern void sched_online_group(struct task_group *tg,
			       struct task_group *parent);
extern void sched_destroy_group(struct task_group *tg);
extern void sched_offline_group(struct task_group *tg);

extern void sched_move_task(struct task_struct *tsk);

#ifdef CONFIG_FAIR_GROUP_SCHED
extern int sched_group_set_shares(struct task_group *tg, unsigned long shares);
380 381 382 383 384 385 386 387 388

#ifdef CONFIG_SMP
extern void set_task_rq_fair(struct sched_entity *se,
			     struct cfs_rq *prev, struct cfs_rq *next);
#else /* !CONFIG_SMP */
static inline void set_task_rq_fair(struct sched_entity *se,
			     struct cfs_rq *prev, struct cfs_rq *next) { }
#endif /* CONFIG_SMP */
#endif /* CONFIG_FAIR_GROUP_SCHED */
389

390 391 392 393 394 395 396 397 398
#else /* CONFIG_CGROUP_SCHED */

struct cfs_bandwidth { };

#endif	/* CONFIG_CGROUP_SCHED */

/* CFS-related fields in a runqueue */
struct cfs_rq {
	struct load_weight load;
399
	unsigned int nr_running, h_nr_running;
400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419

	u64 exec_clock;
	u64 min_vruntime;
#ifndef CONFIG_64BIT
	u64 min_vruntime_copy;
#endif

	struct rb_root tasks_timeline;
	struct rb_node *rb_leftmost;

	/*
	 * 'curr' points to currently running entity on this cfs_rq.
	 * It is set to NULL otherwise (i.e when none are currently running).
	 */
	struct sched_entity *curr, *next, *last, *skip;

#ifdef	CONFIG_SCHED_DEBUG
	unsigned int nr_spread_over;
#endif

420 421
#ifdef CONFIG_SMP
	/*
422
	 * CFS load tracking
423
	 */
424
	struct sched_avg avg;
425 426
	u64 runnable_load_sum;
	unsigned long runnable_load_avg;
427
#ifdef CONFIG_FAIR_GROUP_SCHED
428
	unsigned long tg_load_avg_contrib;
429
	unsigned long propagate_avg;
430 431 432 433 434
#endif
	atomic_long_t removed_load_avg, removed_util_avg;
#ifndef CONFIG_64BIT
	u64 load_last_update_time_copy;
#endif
435

436
#ifdef CONFIG_FAIR_GROUP_SCHED
437 438 439 440 441 442 443
	/*
	 *   h_load = weight * f(tg)
	 *
	 * Where f(tg) is the recursive weight fraction assigned to
	 * this group.
	 */
	unsigned long h_load;
444 445 446
	u64 last_h_load_update;
	struct sched_entity *h_load_next;
#endif /* CONFIG_FAIR_GROUP_SCHED */
447 448
#endif /* CONFIG_SMP */

449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468
#ifdef CONFIG_FAIR_GROUP_SCHED
	struct rq *rq;	/* cpu runqueue to which this cfs_rq is attached */

	/*
	 * leaf cfs_rqs are those that hold tasks (lowest schedulable entity in
	 * a hierarchy). Non-leaf lrqs hold other higher schedulable entities
	 * (like users, containers etc.)
	 *
	 * leaf_cfs_rq_list ties together list of leaf cfs_rq's in a cpu. This
	 * list is used during load balance.
	 */
	int on_list;
	struct list_head leaf_cfs_rq_list;
	struct task_group *tg;	/* group that "owns" this runqueue */

#ifdef CONFIG_CFS_BANDWIDTH
	int runtime_enabled;
	u64 runtime_expires;
	s64 runtime_remaining;

469 470
	u64 throttled_clock, throttled_clock_task;
	u64 throttled_clock_task_time;
471
	int throttled, throttle_count;
472 473 474 475 476 477 478 479 480 481
	struct list_head throttled_list;
#endif /* CONFIG_CFS_BANDWIDTH */
#endif /* CONFIG_FAIR_GROUP_SCHED */
};

static inline int rt_bandwidth_enabled(void)
{
	return sysctl_sched_rt_runtime >= 0;
}

482 483 484 485 486
/* RT IPI pull logic requires IRQ_WORK */
#ifdef CONFIG_IRQ_WORK
# define HAVE_RT_PUSH_IPI
#endif

487 488 489
/* Real-Time classes' related field in a runqueue: */
struct rt_rq {
	struct rt_prio_array active;
490
	unsigned int rt_nr_running;
F
Frederic Weisbecker 已提交
491
	unsigned int rr_nr_running;
492 493 494 495 496 497 498 499 500 501 502 503 504
#if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED
	struct {
		int curr; /* highest queued rt task prio */
#ifdef CONFIG_SMP
		int next; /* next highest */
#endif
	} highest_prio;
#endif
#ifdef CONFIG_SMP
	unsigned long rt_nr_migratory;
	unsigned long rt_nr_total;
	int overloaded;
	struct plist_head pushable_tasks;
505 506 507 508 509
#ifdef HAVE_RT_PUSH_IPI
	int push_flags;
	int push_cpu;
	struct irq_work push_work;
	raw_spinlock_t push_lock;
510
#endif
511
#endif /* CONFIG_SMP */
512 513
	int rt_queued;

514 515 516 517 518 519 520 521 522 523 524 525 526 527
	int rt_throttled;
	u64 rt_time;
	u64 rt_runtime;
	/* Nests inside the rq lock: */
	raw_spinlock_t rt_runtime_lock;

#ifdef CONFIG_RT_GROUP_SCHED
	unsigned long rt_nr_boosted;

	struct rq *rq;
	struct task_group *tg;
#endif
};

528 529 530 531 532 533 534
/* Deadline class' related fields in a runqueue */
struct dl_rq {
	/* runqueue is an rbtree, ordered by deadline */
	struct rb_root rb_root;
	struct rb_node *rb_leftmost;

	unsigned long dl_nr_running;
535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557

#ifdef CONFIG_SMP
	/*
	 * Deadline values of the currently executing and the
	 * earliest ready task on this rq. Caching these facilitates
	 * the decision wether or not a ready but not running task
	 * should migrate somewhere else.
	 */
	struct {
		u64 curr;
		u64 next;
	} earliest_dl;

	unsigned long dl_nr_migratory;
	int overloaded;

	/*
	 * Tasks on this rq that can be pushed away. They are kept in
	 * an rb-tree, ordered by tasks' deadlines, with caching
	 * of the leftmost (earliest deadline) element.
	 */
	struct rb_root pushable_dl_tasks_root;
	struct rb_node *pushable_dl_tasks_leftmost;
558 559
#else
	struct dl_bw dl_bw;
560
#endif
561 562
};

563 564
#ifdef CONFIG_SMP

T
Tim Chen 已提交
565 566 567 568 569
static inline bool sched_asym_prefer(int a, int b)
{
	return arch_asym_cpu_priority(a) > arch_asym_cpu_priority(b);
}

570 571 572 573 574 575 576 577 578 579 580 581 582 583 584
/*
 * We add the notion of a root-domain which will be used to define per-domain
 * variables. Each exclusive cpuset essentially defines an island domain by
 * fully partitioning the member cpus from any other cpuset. Whenever a new
 * exclusive cpuset is created, we also create and attach a new root-domain
 * object.
 *
 */
struct root_domain {
	atomic_t refcount;
	atomic_t rto_count;
	struct rcu_head rcu;
	cpumask_var_t span;
	cpumask_var_t online;

585 586 587
	/* Indicate more than one runnable task for any CPU */
	bool overload;

588 589 590 591 592 593
	/*
	 * The bit corresponding to a CPU gets set here if such CPU has more
	 * than one runnable -deadline task (as it is below for RT tasks).
	 */
	cpumask_var_t dlo_mask;
	atomic_t dlo_count;
594
	struct dl_bw dl_bw;
595
	struct cpudl cpudl;
596

597 598 599 600 601 602
	/*
	 * The "RT overload" flag: it gets set if a CPU has more than
	 * one runnable RT task.
	 */
	cpumask_var_t rto_mask;
	struct cpupri cpupri;
603 604

	unsigned long max_cpu_capacity;
605 606 607
};

extern struct root_domain def_root_domain;
608 609 610 611 612 613 614
extern struct mutex sched_domains_mutex;
extern cpumask_var_t fallback_doms;
extern cpumask_var_t sched_domains_tmpmask;

extern void init_defrootdomain(void);
extern int init_sched_domains(const struct cpumask *cpu_map);
extern void rq_attach_root(struct rq *rq, struct root_domain *rd);
615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632

#endif /* CONFIG_SMP */

/*
 * This is the main, per-CPU runqueue data structure.
 *
 * Locking rule: those places that want to lock multiple runqueues
 * (such as the load balancing or the thread migration code), lock
 * acquire operations must be ordered by ascending &runqueue.
 */
struct rq {
	/* runqueue lock: */
	raw_spinlock_t lock;

	/*
	 * nr_running and cpu_load should be in the same cacheline because
	 * remote CPUs use both these fields when doing load calculation.
	 */
633
	unsigned int nr_running;
634 635 636 637
#ifdef CONFIG_NUMA_BALANCING
	unsigned int nr_numa_running;
	unsigned int nr_preferred_running;
#endif
638 639
	#define CPU_LOAD_IDX_MAX 5
	unsigned long cpu_load[CPU_LOAD_IDX_MAX];
640
#ifdef CONFIG_NO_HZ_COMMON
641 642 643
#ifdef CONFIG_SMP
	unsigned long last_load_update_tick;
#endif /* CONFIG_SMP */
644
	unsigned long nohz_flags;
645
#endif /* CONFIG_NO_HZ_COMMON */
646 647
#ifdef CONFIG_NO_HZ_FULL
	unsigned long last_sched_tick;
648 649 650 651 652 653 654 655
#endif
	/* capture load from *all* tasks on this cpu: */
	struct load_weight load;
	unsigned long nr_load_updates;
	u64 nr_switches;

	struct cfs_rq cfs;
	struct rt_rq rt;
656
	struct dl_rq dl;
657 658 659 660

#ifdef CONFIG_FAIR_GROUP_SCHED
	/* list of leaf cfs_rq on this cpu: */
	struct list_head leaf_cfs_rq_list;
661
	struct list_head *tmp_alone_branch;
662 663
#endif /* CONFIG_FAIR_GROUP_SCHED */

664 665 666 667 668 669 670 671 672 673 674 675
	/*
	 * This is part of a global counter where only the total sum
	 * over all CPUs matters. A task can increase this counter on
	 * one CPU and if it got migrated afterwards it may decrease
	 * it on another CPU. Always updated under the runqueue lock:
	 */
	unsigned long nr_uninterruptible;

	struct task_struct *curr, *idle, *stop;
	unsigned long next_balance;
	struct mm_struct *prev_mm;

676
	unsigned int clock_update_flags;
677 678 679 680 681 682 683 684 685
	u64 clock;
	u64 clock_task;

	atomic_t nr_iowait;

#ifdef CONFIG_SMP
	struct root_domain *rd;
	struct sched_domain *sd;

686
	unsigned long cpu_capacity;
687
	unsigned long cpu_capacity_orig;
688

689 690
	struct callback_head *balance_callback;

691 692 693 694 695 696 697 698 699
	unsigned char idle_balance;
	/* For active balancing */
	int active_balance;
	int push_cpu;
	struct cpu_stop_work active_balance_work;
	/* cpu of this runqueue: */
	int cpu;
	int online;

700 701
	struct list_head cfs_tasks;

702 703 704 705
	u64 rt_avg;
	u64 age_stamp;
	u64 idle_stamp;
	u64 avg_idle;
706 707 708

	/* This is used to determine avg_idle's max value */
	u64 max_idle_balance_cost;
709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753
#endif

#ifdef CONFIG_IRQ_TIME_ACCOUNTING
	u64 prev_irq_time;
#endif
#ifdef CONFIG_PARAVIRT
	u64 prev_steal_time;
#endif
#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING
	u64 prev_steal_time_rq;
#endif

	/* calc_load related fields */
	unsigned long calc_load_update;
	long calc_load_active;

#ifdef CONFIG_SCHED_HRTICK
#ifdef CONFIG_SMP
	int hrtick_csd_pending;
	struct call_single_data hrtick_csd;
#endif
	struct hrtimer hrtick_timer;
#endif

#ifdef CONFIG_SCHEDSTATS
	/* latency stats */
	struct sched_info rq_sched_info;
	unsigned long long rq_cpu_time;
	/* could above be rq->cfs_rq.exec_clock + rq->rt_rq.rt_runtime ? */

	/* sys_sched_yield() stats */
	unsigned int yld_count;

	/* schedule() stats */
	unsigned int sched_count;
	unsigned int sched_goidle;

	/* try_to_wake_up() stats */
	unsigned int ttwu_count;
	unsigned int ttwu_local;
#endif

#ifdef CONFIG_SMP
	struct llist_head wake_list;
#endif
754 755 756 757 758

#ifdef CONFIG_CPU_IDLE
	/* Must be inspected within a rcu lock section */
	struct cpuidle_state *idle_state;
#endif
759 760 761 762 763 764 765 766 767 768 769
};

static inline int cpu_of(struct rq *rq)
{
#ifdef CONFIG_SMP
	return rq->cpu;
#else
	return 0;
#endif
}

P
Peter Zijlstra 已提交
770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786

#ifdef CONFIG_SCHED_SMT

extern struct static_key_false sched_smt_present;

extern void __update_idle_core(struct rq *rq);

static inline void update_idle_core(struct rq *rq)
{
	if (static_branch_unlikely(&sched_smt_present))
		__update_idle_core(rq);
}

#else
static inline void update_idle_core(struct rq *rq) { }
#endif

787
DECLARE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues);
788

789
#define cpu_rq(cpu)		(&per_cpu(runqueues, (cpu)))
790
#define this_rq()		this_cpu_ptr(&runqueues)
791 792
#define task_rq(p)		cpu_rq(task_cpu(p))
#define cpu_curr(cpu)		(cpu_rq(cpu)->curr)
793
#define raw_rq()		raw_cpu_ptr(&runqueues)
794

795 796
static inline u64 __rq_clock_broken(struct rq *rq)
{
797
	return READ_ONCE(rq->clock);
798 799
}

800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835
/*
 * rq::clock_update_flags bits
 *
 * %RQCF_REQ_SKIP - will request skipping of clock update on the next
 *  call to __schedule(). This is an optimisation to avoid
 *  neighbouring rq clock updates.
 *
 * %RQCF_ACT_SKIP - is set from inside of __schedule() when skipping is
 *  in effect and calls to update_rq_clock() are being ignored.
 *
 * %RQCF_UPDATED - is a debug flag that indicates whether a call has been
 *  made to update_rq_clock() since the last time rq::lock was pinned.
 *
 * If inside of __schedule(), clock_update_flags will have been
 * shifted left (a left shift is a cheap operation for the fast path
 * to promote %RQCF_REQ_SKIP to %RQCF_ACT_SKIP), so you must use,
 *
 *	if (rq-clock_update_flags >= RQCF_UPDATED)
 *
 * to check if %RQCF_UPADTED is set. It'll never be shifted more than
 * one position though, because the next rq_unpin_lock() will shift it
 * back.
 */
#define RQCF_REQ_SKIP	0x01
#define RQCF_ACT_SKIP	0x02
#define RQCF_UPDATED	0x04

static inline void assert_clock_updated(struct rq *rq)
{
	/*
	 * The only reason for not seeing a clock update since the
	 * last rq_pin_lock() is if we're currently skipping updates.
	 */
	SCHED_WARN_ON(rq->clock_update_flags < RQCF_ACT_SKIP);
}

836 837
static inline u64 rq_clock(struct rq *rq)
{
838
	lockdep_assert_held(&rq->lock);
839 840
	assert_clock_updated(rq);

841 842 843 844 845
	return rq->clock;
}

static inline u64 rq_clock_task(struct rq *rq)
{
846
	lockdep_assert_held(&rq->lock);
847 848
	assert_clock_updated(rq);

849 850 851
	return rq->clock_task;
}

852 853 854 855
static inline void rq_clock_skip_update(struct rq *rq, bool skip)
{
	lockdep_assert_held(&rq->lock);
	if (skip)
856
		rq->clock_update_flags |= RQCF_REQ_SKIP;
857
	else
858
		rq->clock_update_flags &= ~RQCF_REQ_SKIP;
859 860
}

861 862 863
struct rq_flags {
	unsigned long flags;
	struct pin_cookie cookie;
864 865 866 867 868 869 870 871
#ifdef CONFIG_SCHED_DEBUG
	/*
	 * A copy of (rq::clock_update_flags & RQCF_UPDATED) for the
	 * current pin context is stashed here in case it needs to be
	 * restored in rq_repin_lock().
	 */
	unsigned int clock_update_flags;
#endif
872 873 874 875 876
};

static inline void rq_pin_lock(struct rq *rq, struct rq_flags *rf)
{
	rf->cookie = lockdep_pin_lock(&rq->lock);
877 878 879 880 881

#ifdef CONFIG_SCHED_DEBUG
	rq->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP);
	rf->clock_update_flags = 0;
#endif
882 883 884 885
}

static inline void rq_unpin_lock(struct rq *rq, struct rq_flags *rf)
{
886 887 888 889 890
#ifdef CONFIG_SCHED_DEBUG
	if (rq->clock_update_flags > RQCF_ACT_SKIP)
		rf->clock_update_flags = RQCF_UPDATED;
#endif

891 892 893 894 895 896
	lockdep_unpin_lock(&rq->lock, rf->cookie);
}

static inline void rq_repin_lock(struct rq *rq, struct rq_flags *rf)
{
	lockdep_repin_lock(&rq->lock, rf->cookie);
897 898 899 900 901 902 903

#ifdef CONFIG_SCHED_DEBUG
	/*
	 * Restore the value we stashed in @rf for this pin context.
	 */
	rq->clock_update_flags |= rf->clock_update_flags;
#endif
904 905
}

906
#ifdef CONFIG_NUMA
907 908 909 910 911 912
enum numa_topology_type {
	NUMA_DIRECT,
	NUMA_GLUELESS_MESH,
	NUMA_BACKPLANE,
};
extern enum numa_topology_type sched_numa_topology_type;
913 914 915 916
extern int sched_max_numa_distance;
extern bool find_numa_distance(int distance);
#endif

917 918 919 920 921 922 923 924 925 926
#ifdef CONFIG_NUMA
extern void sched_init_numa(void);
extern void sched_domains_numa_masks_set(unsigned int cpu);
extern void sched_domains_numa_masks_clear(unsigned int cpu);
#else
static inline void sched_init_numa(void) { }
static inline void sched_domains_numa_masks_set(unsigned int cpu) { }
static inline void sched_domains_numa_masks_clear(unsigned int cpu) { }
#endif

927
#ifdef CONFIG_NUMA_BALANCING
928 929 930 931 932 933 934
/* The regions in numa_faults array from task_struct */
enum numa_faults_stats {
	NUMA_MEM = 0,
	NUMA_CPU,
	NUMA_MEMBUF,
	NUMA_CPUBUF
};
935
extern void sched_setnuma(struct task_struct *p, int node);
936
extern int migrate_task_to(struct task_struct *p, int cpu);
937
extern int migrate_swap(struct task_struct *, struct task_struct *);
938 939
#endif /* CONFIG_NUMA_BALANCING */

940 941
#ifdef CONFIG_SMP

942 943 944 945 946 947 948 949 950 951 952 953 954 955 956
static inline void
queue_balance_callback(struct rq *rq,
		       struct callback_head *head,
		       void (*func)(struct rq *rq))
{
	lockdep_assert_held(&rq->lock);

	if (unlikely(head->next))
		return;

	head->func = (void (*)(struct callback_head *))func;
	head->next = rq->balance_callback;
	rq->balance_callback = head;
}

957 958
extern void sched_ttwu_pending(void);

959 960 961 962 963 964 965 966 967 968 969 970
#define rcu_dereference_check_sched_domain(p) \
	rcu_dereference_check((p), \
			      lockdep_is_held(&sched_domains_mutex))

/*
 * The domain tree (rq->sd) is protected by RCU's quiescent state transition.
 * See detach_destroy_domains: synchronize_sched for details.
 *
 * The domain tree of any CPU may only be accessed from within
 * preempt-disabled sections.
 */
#define for_each_domain(cpu, __sd) \
971 972
	for (__sd = rcu_dereference_check_sched_domain(cpu_rq(cpu)->sd); \
			__sd; __sd = __sd->parent)
973

974 975
#define for_each_lower_domain(sd) for (; sd; sd = sd->child)

976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997
/**
 * highest_flag_domain - Return highest sched_domain containing flag.
 * @cpu:	The cpu whose highest level of sched domain is to
 *		be returned.
 * @flag:	The flag to check for the highest sched_domain
 *		for the given cpu.
 *
 * Returns the highest sched_domain of a cpu which contains the given flag.
 */
static inline struct sched_domain *highest_flag_domain(int cpu, int flag)
{
	struct sched_domain *sd, *hsd = NULL;

	for_each_domain(cpu, sd) {
		if (!(sd->flags & flag))
			break;
		hsd = sd;
	}

	return hsd;
}

998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009
static inline struct sched_domain *lowest_flag_domain(int cpu, int flag)
{
	struct sched_domain *sd;

	for_each_domain(cpu, sd) {
		if (sd->flags & flag)
			break;
	}

	return sd;
}

1010
DECLARE_PER_CPU(struct sched_domain *, sd_llc);
1011
DECLARE_PER_CPU(int, sd_llc_size);
1012
DECLARE_PER_CPU(int, sd_llc_id);
1013
DECLARE_PER_CPU(struct sched_domain_shared *, sd_llc_shared);
1014
DECLARE_PER_CPU(struct sched_domain *, sd_numa);
1015
DECLARE_PER_CPU(struct sched_domain *, sd_asym);
1016

1017
struct sched_group_capacity {
1018 1019
	atomic_t ref;
	/*
1020
	 * CPU capacity of this group, SCHED_CAPACITY_SCALE being max capacity
1021
	 * for a single CPU.
1022
	 */
1023 1024
	unsigned long capacity;
	unsigned long min_capacity; /* Min per-CPU capacity in group */
1025
	unsigned long next_update;
1026
	int imbalance; /* XXX unrelated to capacity but shared group state */
1027 1028 1029 1030 1031 1032 1033 1034 1035

	unsigned long cpumask[0]; /* iteration mask */
};

struct sched_group {
	struct sched_group *next;	/* Must be a circular list */
	atomic_t ref;

	unsigned int group_weight;
1036
	struct sched_group_capacity *sgc;
T
Tim Chen 已提交
1037
	int asym_prefer_cpu;		/* cpu of highest priority in group */
1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059

	/*
	 * The CPUs this group covers.
	 *
	 * 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 cpumask[0];
};

static inline struct cpumask *sched_group_cpus(struct sched_group *sg)
{
	return to_cpumask(sg->cpumask);
}

/*
 * cpumask masking which cpus in the group are allowed to iterate up the domain
 * tree.
 */
static inline struct cpumask *sched_group_mask(struct sched_group *sg)
{
1060
	return to_cpumask(sg->sgc->cpumask);
1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071
}

/**
 * group_first_cpu - Returns the first cpu in the cpumask of a sched_group.
 * @group: The group whose first cpu is to be returned.
 */
static inline unsigned int group_first_cpu(struct sched_group *group)
{
	return cpumask_first(sched_group_cpus(group));
}

P
Peter Zijlstra 已提交
1072 1073
extern int group_balance_cpu(struct sched_group *sg);

1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085
#if defined(CONFIG_SCHED_DEBUG) && defined(CONFIG_SYSCTL)
void register_sched_domain_sysctl(void);
void unregister_sched_domain_sysctl(void);
#else
static inline void register_sched_domain_sysctl(void)
{
}
static inline void unregister_sched_domain_sysctl(void)
{
}
#endif

1086 1087 1088 1089
#else

static inline void sched_ttwu_pending(void) { }

1090
#endif /* CONFIG_SMP */
1091

1092
#include "stats.h"
1093
#include "autogroup.h"
1094 1095 1096 1097 1098 1099

#ifdef CONFIG_CGROUP_SCHED

/*
 * Return the group to which this tasks belongs.
 *
1100 1101 1102
 * We cannot use task_css() and friends because the cgroup subsystem
 * changes that value before the cgroup_subsys::attach() method is called,
 * therefore we cannot pin it and might observe the wrong value.
P
Peter Zijlstra 已提交
1103 1104 1105 1106 1107 1108
 *
 * The same is true for autogroup's p->signal->autogroup->tg, the autogroup
 * core changes this before calling sched_move_task().
 *
 * Instead we use a 'copy' which is updated from sched_move_task() while
 * holding both task_struct::pi_lock and rq::lock.
1109 1110 1111
 */
static inline struct task_group *task_group(struct task_struct *p)
{
P
Peter Zijlstra 已提交
1112
	return p->sched_task_group;
1113 1114 1115 1116 1117 1118 1119 1120 1121 1122
}

/* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */
static inline void set_task_rq(struct task_struct *p, unsigned int cpu)
{
#if defined(CONFIG_FAIR_GROUP_SCHED) || defined(CONFIG_RT_GROUP_SCHED)
	struct task_group *tg = task_group(p);
#endif

#ifdef CONFIG_FAIR_GROUP_SCHED
1123
	set_task_rq_fair(&p->se, p->se.cfs_rq, tg->cfs_rq[cpu]);
1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153
	p->se.cfs_rq = tg->cfs_rq[cpu];
	p->se.parent = tg->se[cpu];
#endif

#ifdef CONFIG_RT_GROUP_SCHED
	p->rt.rt_rq  = tg->rt_rq[cpu];
	p->rt.parent = tg->rt_se[cpu];
#endif
}

#else /* CONFIG_CGROUP_SCHED */

static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { }
static inline struct task_group *task_group(struct task_struct *p)
{
	return NULL;
}

#endif /* CONFIG_CGROUP_SCHED */

static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
{
	set_task_rq(p, cpu);
#ifdef CONFIG_SMP
	/*
	 * After ->cpu is set up to a new value, task_rq_lock(p, ...) can be
	 * successfuly executed on another CPU. We must ensure that updates of
	 * per-task data have been completed by this moment.
	 */
	smp_wmb();
1154 1155 1156
#ifdef CONFIG_THREAD_INFO_IN_TASK
	p->cpu = cpu;
#else
1157
	task_thread_info(p)->cpu = cpu;
1158
#endif
1159
	p->wake_cpu = cpu;
1160 1161 1162 1163 1164 1165 1166
#endif
}

/*
 * Tunables that become constants when CONFIG_SCHED_DEBUG is off:
 */
#ifdef CONFIG_SCHED_DEBUG
1167
# include <linux/static_key.h>
1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178
# define const_debug __read_mostly
#else
# define const_debug const
#endif

extern const_debug unsigned int sysctl_sched_features;

#define SCHED_FEAT(name, enabled)	\
	__SCHED_FEAT_##name ,

enum {
1179
#include "features.h"
1180
	__SCHED_FEAT_NR,
1181 1182 1183 1184
};

#undef SCHED_FEAT

1185 1186
#if defined(CONFIG_SCHED_DEBUG) && defined(HAVE_JUMP_LABEL)
#define SCHED_FEAT(name, enabled)					\
1187
static __always_inline bool static_branch_##name(struct static_key *key) \
1188
{									\
1189
	return static_key_##enabled(key);				\
1190 1191 1192 1193 1194 1195
}

#include "features.h"

#undef SCHED_FEAT

1196
extern struct static_key sched_feat_keys[__SCHED_FEAT_NR];
1197 1198
#define sched_feat(x) (static_branch_##x(&sched_feat_keys[__SCHED_FEAT_##x]))
#else /* !(SCHED_DEBUG && HAVE_JUMP_LABEL) */
1199
#define sched_feat(x) (sysctl_sched_features & (1UL << __SCHED_FEAT_##x))
1200
#endif /* SCHED_DEBUG && HAVE_JUMP_LABEL */
1201

1202
extern struct static_key_false sched_numa_balancing;
1203
extern struct static_key_false sched_schedstats;
1204

1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231
static inline u64 global_rt_period(void)
{
	return (u64)sysctl_sched_rt_period * NSEC_PER_USEC;
}

static inline u64 global_rt_runtime(void)
{
	if (sysctl_sched_rt_runtime < 0)
		return RUNTIME_INF;

	return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC;
}

static inline int task_current(struct rq *rq, struct task_struct *p)
{
	return rq->curr == p;
}

static inline int task_running(struct rq *rq, struct task_struct *p)
{
#ifdef CONFIG_SMP
	return p->on_cpu;
#else
	return task_current(rq, p);
#endif
}

1232 1233 1234 1235
static inline int task_on_rq_queued(struct task_struct *p)
{
	return p->on_rq == TASK_ON_RQ_QUEUED;
}
1236

1237 1238 1239 1240 1241
static inline int task_on_rq_migrating(struct task_struct *p)
{
	return p->on_rq == TASK_ON_RQ_MIGRATING;
}

1242 1243 1244
#ifndef prepare_arch_switch
# define prepare_arch_switch(next)	do { } while (0)
#endif
1245 1246 1247
#ifndef finish_arch_post_lock_switch
# define finish_arch_post_lock_switch()	do { } while (0)
#endif
1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267

static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next)
{
#ifdef CONFIG_SMP
	/*
	 * We can optimise this out completely for !SMP, because the
	 * SMP rebalancing from interrupt is the only thing that cares
	 * here.
	 */
	next->on_cpu = 1;
#endif
}

static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev)
{
#ifdef CONFIG_SMP
	/*
	 * After ->on_cpu is cleared, the task can be moved to a different CPU.
	 * We must ensure this doesn't happen until the switch is completely
	 * finished.
1268
	 *
1269 1270 1271
	 * In particular, the load of prev->state in finish_task_switch() must
	 * happen before this.
	 *
1272
	 * Pairs with the smp_cond_load_acquire() in try_to_wake_up().
1273
	 */
1274
	smp_store_release(&prev->on_cpu, 0);
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289
#endif
#ifdef CONFIG_DEBUG_SPINLOCK
	/* this is a valid case when another task releases the spinlock */
	rq->lock.owner = current;
#endif
	/*
	 * If we are tracking spinlock dependencies then we have to
	 * fix up the runqueue lock - which gets 'carried over' from
	 * prev into current:
	 */
	spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_);

	raw_spin_unlock_irq(&rq->lock);
}

1290 1291 1292 1293 1294 1295 1296
/*
 * wake flags
 */
#define WF_SYNC		0x01		/* waker goes to sleep after wakeup */
#define WF_FORK		0x02		/* child wakeup after fork */
#define WF_MIGRATED	0x4		/* internal use, task got migrated */

1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308
/*
 * To aid in avoiding the subversion of "niceness" due to uneven distribution
 * of tasks with abnormal "nice" values across CPUs the contribution that
 * each task makes to its run queue's load is weighted according to its
 * scheduling class and "nice" value. For SCHED_NORMAL tasks this is just a
 * scaled version of the new time slice allocation that they receive on time
 * slice expiry etc.
 */

#define WEIGHT_IDLEPRIO                3
#define WMULT_IDLEPRIO         1431655765

1309 1310
extern const int sched_prio_to_weight[40];
extern const u32 sched_prio_to_wmult[40];
1311

1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326
/*
 * {de,en}queue flags:
 *
 * DEQUEUE_SLEEP  - task is no longer runnable
 * ENQUEUE_WAKEUP - task just became runnable
 *
 * SAVE/RESTORE - an otherwise spurious dequeue/enqueue, done to ensure tasks
 *                are in a known state which allows modification. Such pairs
 *                should preserve as much state as possible.
 *
 * MOVE - paired with SAVE/RESTORE, explicitly does not preserve the location
 *        in the runqueue.
 *
 * ENQUEUE_HEAD      - place at front of runqueue (tail if not specified)
 * ENQUEUE_REPLENISH - CBS (replenish runtime and postpone deadline)
1327
 * ENQUEUE_MIGRATED  - the task was migrated during wakeup
1328 1329 1330 1331 1332 1333 1334
 *
 */

#define DEQUEUE_SLEEP		0x01
#define DEQUEUE_SAVE		0x02 /* matches ENQUEUE_RESTORE */
#define DEQUEUE_MOVE		0x04 /* matches ENQUEUE_MOVE */

1335
#define ENQUEUE_WAKEUP		0x01
1336 1337 1338 1339 1340
#define ENQUEUE_RESTORE		0x02
#define ENQUEUE_MOVE		0x04

#define ENQUEUE_HEAD		0x08
#define ENQUEUE_REPLENISH	0x10
1341
#ifdef CONFIG_SMP
1342
#define ENQUEUE_MIGRATED	0x20
1343
#else
1344
#define ENQUEUE_MIGRATED	0x00
1345 1346
#endif

1347 1348
#define RETRY_TASK		((void *)-1UL)

1349 1350 1351 1352 1353 1354 1355 1356 1357 1358
struct sched_class {
	const struct sched_class *next;

	void (*enqueue_task) (struct rq *rq, struct task_struct *p, int flags);
	void (*dequeue_task) (struct rq *rq, struct task_struct *p, int flags);
	void (*yield_task) (struct rq *rq);
	bool (*yield_to_task) (struct rq *rq, struct task_struct *p, bool preempt);

	void (*check_preempt_curr) (struct rq *rq, struct task_struct *p, int flags);

1359 1360 1361 1362
	/*
	 * It is the responsibility of the pick_next_task() method that will
	 * return the next task to call put_prev_task() on the @prev task or
	 * something equivalent.
1363 1364 1365
	 *
	 * May return RETRY_TASK when it finds a higher prio class has runnable
	 * tasks.
1366 1367
	 */
	struct task_struct * (*pick_next_task) (struct rq *rq,
1368
						struct task_struct *prev,
1369
						struct rq_flags *rf);
1370 1371 1372
	void (*put_prev_task) (struct rq *rq, struct task_struct *p);

#ifdef CONFIG_SMP
1373
	int  (*select_task_rq)(struct task_struct *p, int task_cpu, int sd_flag, int flags);
1374
	void (*migrate_task_rq)(struct task_struct *p);
1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387

	void (*task_woken) (struct rq *this_rq, struct task_struct *task);

	void (*set_cpus_allowed)(struct task_struct *p,
				 const struct cpumask *newmask);

	void (*rq_online)(struct rq *rq);
	void (*rq_offline)(struct rq *rq);
#endif

	void (*set_curr_task) (struct rq *rq);
	void (*task_tick) (struct rq *rq, struct task_struct *p, int queued);
	void (*task_fork) (struct task_struct *p);
1388
	void (*task_dead) (struct task_struct *p);
1389

1390 1391 1392 1393 1394
	/*
	 * The switched_from() call is allowed to drop rq->lock, therefore we
	 * cannot assume the switched_from/switched_to pair is serliazed by
	 * rq->lock. They are however serialized by p->pi_lock.
	 */
1395 1396 1397 1398 1399 1400 1401 1402
	void (*switched_from) (struct rq *this_rq, struct task_struct *task);
	void (*switched_to) (struct rq *this_rq, struct task_struct *task);
	void (*prio_changed) (struct rq *this_rq, struct task_struct *task,
			     int oldprio);

	unsigned int (*get_rr_interval) (struct rq *rq,
					 struct task_struct *task);

1403 1404
	void (*update_curr) (struct rq *rq);

1405 1406 1407
#define TASK_SET_GROUP  0
#define TASK_MOVE_GROUP	1

1408
#ifdef CONFIG_FAIR_GROUP_SCHED
1409
	void (*task_change_group) (struct task_struct *p, int type);
1410 1411
#endif
};
1412

1413 1414 1415 1416 1417
static inline void put_prev_task(struct rq *rq, struct task_struct *prev)
{
	prev->sched_class->put_prev_task(rq, prev);
}

1418 1419 1420 1421 1422
static inline void set_curr_task(struct rq *rq, struct task_struct *curr)
{
	curr->sched_class->set_curr_task(rq);
}

1423 1424 1425 1426 1427
#define sched_class_highest (&stop_sched_class)
#define for_each_class(class) \
   for (class = sched_class_highest; class; class = class->next)

extern const struct sched_class stop_sched_class;
1428
extern const struct sched_class dl_sched_class;
1429 1430 1431 1432 1433 1434 1435
extern const struct sched_class rt_sched_class;
extern const struct sched_class fair_sched_class;
extern const struct sched_class idle_sched_class;


#ifdef CONFIG_SMP

1436
extern void update_group_capacity(struct sched_domain *sd, int cpu);
1437

1438
extern void trigger_load_balance(struct rq *rq);
1439

1440 1441
extern void set_cpus_allowed_common(struct task_struct *p, const struct cpumask *new_mask);

1442 1443
#endif

1444 1445 1446 1447 1448 1449 1450 1451 1452
#ifdef CONFIG_CPU_IDLE
static inline void idle_set_state(struct rq *rq,
				  struct cpuidle_state *idle_state)
{
	rq->idle_state = idle_state;
}

static inline struct cpuidle_state *idle_get_state(struct rq *rq)
{
1453
	SCHED_WARN_ON(!rcu_read_lock_held());
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467
	return rq->idle_state;
}
#else
static inline void idle_set_state(struct rq *rq,
				  struct cpuidle_state *idle_state)
{
}

static inline struct cpuidle_state *idle_get_state(struct rq *rq)
{
	return NULL;
}
#endif

1468 1469 1470
extern void sysrq_sched_debug_show(void);
extern void sched_init_granularity(void);
extern void update_max_interval(void);
1471 1472

extern void init_sched_dl_class(void);
1473 1474 1475
extern void init_sched_rt_class(void);
extern void init_sched_fair_class(void);

1476
extern void resched_curr(struct rq *rq);
1477 1478 1479 1480 1481
extern void resched_cpu(int cpu);

extern struct rt_bandwidth def_rt_bandwidth;
extern void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime);

1482 1483
extern struct dl_bandwidth def_dl_bandwidth;
extern void init_dl_bandwidth(struct dl_bandwidth *dl_b, u64 period, u64 runtime);
1484 1485
extern void init_dl_task_timer(struct sched_dl_entity *dl_se);

1486 1487
unsigned long to_ratio(u64 period, u64 runtime);

1488
extern void init_entity_runnable_average(struct sched_entity *se);
1489
extern void post_init_entity_util_avg(struct sched_entity *se);
1490

1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519
#ifdef CONFIG_NO_HZ_FULL
extern bool sched_can_stop_tick(struct rq *rq);

/*
 * Tick may be needed by tasks in the runqueue depending on their policy and
 * requirements. If tick is needed, lets send the target an IPI to kick it out of
 * nohz mode if necessary.
 */
static inline void sched_update_tick_dependency(struct rq *rq)
{
	int cpu;

	if (!tick_nohz_full_enabled())
		return;

	cpu = cpu_of(rq);

	if (!tick_nohz_full_cpu(cpu))
		return;

	if (sched_can_stop_tick(rq))
		tick_nohz_dep_clear_cpu(cpu, TICK_DEP_BIT_SCHED);
	else
		tick_nohz_dep_set_cpu(cpu, TICK_DEP_BIT_SCHED);
}
#else
static inline void sched_update_tick_dependency(struct rq *rq) { }
#endif

1520
static inline void add_nr_running(struct rq *rq, unsigned count)
1521
{
1522 1523 1524
	unsigned prev_nr = rq->nr_running;

	rq->nr_running = prev_nr + count;
1525

1526
	if (prev_nr < 2 && rq->nr_running >= 2) {
1527 1528 1529 1530 1531
#ifdef CONFIG_SMP
		if (!rq->rd->overload)
			rq->rd->overload = true;
#endif
	}
1532 1533

	sched_update_tick_dependency(rq);
1534 1535
}

1536
static inline void sub_nr_running(struct rq *rq, unsigned count)
1537
{
1538
	rq->nr_running -= count;
1539 1540
	/* Check if we still need preemption */
	sched_update_tick_dependency(rq);
1541 1542
}

1543 1544 1545 1546 1547 1548 1549
static inline void rq_last_tick_reset(struct rq *rq)
{
#ifdef CONFIG_NO_HZ_FULL
	rq->last_sched_tick = jiffies;
#endif
}

1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583
extern void update_rq_clock(struct rq *rq);

extern void activate_task(struct rq *rq, struct task_struct *p, int flags);
extern void deactivate_task(struct rq *rq, struct task_struct *p, int flags);

extern void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags);

extern const_debug unsigned int sysctl_sched_time_avg;
extern const_debug unsigned int sysctl_sched_nr_migrate;
extern const_debug unsigned int sysctl_sched_migration_cost;

static inline u64 sched_avg_period(void)
{
	return (u64)sysctl_sched_time_avg * NSEC_PER_MSEC / 2;
}

#ifdef CONFIG_SCHED_HRTICK

/*
 * Use hrtick when:
 *  - enabled by features
 *  - hrtimer is actually high res
 */
static inline int hrtick_enabled(struct rq *rq)
{
	if (!sched_feat(HRTICK))
		return 0;
	if (!cpu_active(cpu_of(rq)))
		return 0;
	return hrtimer_is_hres_active(&rq->hrtick_timer);
}

void hrtick_start(struct rq *rq, u64 delay);

1584 1585 1586 1587 1588 1589 1590
#else

static inline int hrtick_enabled(struct rq *rq)
{
	return 0;
}

1591 1592 1593 1594
#endif /* CONFIG_SCHED_HRTICK */

#ifdef CONFIG_SMP
extern void sched_avg_update(struct rq *rq);
1595 1596 1597 1598 1599 1600 1601 1602

#ifndef arch_scale_freq_capacity
static __always_inline
unsigned long arch_scale_freq_capacity(struct sched_domain *sd, int cpu)
{
	return SCHED_CAPACITY_SCALE;
}
#endif
1603

1604 1605 1606 1607
#ifndef arch_scale_cpu_capacity
static __always_inline
unsigned long arch_scale_cpu_capacity(struct sched_domain *sd, int cpu)
{
1608
	if (sd && (sd->flags & SD_SHARE_CPUCAPACITY) && (sd->span_weight > 1))
1609 1610 1611 1612 1613 1614
		return sd->smt_gain / sd->span_weight;

	return SCHED_CAPACITY_SCALE;
}
#endif

1615 1616
static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta)
{
1617
	rq->rt_avg += rt_delta * arch_scale_freq_capacity(NULL, cpu_of(rq));
1618 1619 1620 1621 1622 1623 1624
	sched_avg_update(rq);
}
#else
static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { }
static inline void sched_avg_update(struct rq *rq) { }
#endif

1625
struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf)
1626
	__acquires(rq->lock);
1627
struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf)
1628
	__acquires(p->pi_lock)
1629
	__acquires(rq->lock);
1630

1631
static inline void __task_rq_unlock(struct rq *rq, struct rq_flags *rf)
1632 1633
	__releases(rq->lock)
{
1634
	rq_unpin_lock(rq, rf);
1635 1636 1637 1638
	raw_spin_unlock(&rq->lock);
}

static inline void
1639
task_rq_unlock(struct rq *rq, struct task_struct *p, struct rq_flags *rf)
1640 1641 1642
	__releases(rq->lock)
	__releases(p->pi_lock)
{
1643
	rq_unpin_lock(rq, rf);
1644
	raw_spin_unlock(&rq->lock);
1645
	raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags);
1646 1647
}

1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723
#ifdef CONFIG_SMP
#ifdef CONFIG_PREEMPT

static inline void double_rq_lock(struct rq *rq1, struct rq *rq2);

/*
 * fair double_lock_balance: Safely acquires both rq->locks in a fair
 * way at the expense of forcing extra atomic operations in all
 * invocations.  This assures that the double_lock is acquired using the
 * same underlying policy as the spinlock_t on this architecture, which
 * reduces latency compared to the unfair variant below.  However, it
 * also adds more overhead and therefore may reduce throughput.
 */
static inline int _double_lock_balance(struct rq *this_rq, struct rq *busiest)
	__releases(this_rq->lock)
	__acquires(busiest->lock)
	__acquires(this_rq->lock)
{
	raw_spin_unlock(&this_rq->lock);
	double_rq_lock(this_rq, busiest);

	return 1;
}

#else
/*
 * Unfair double_lock_balance: Optimizes throughput at the expense of
 * latency by eliminating extra atomic operations when the locks are
 * already in proper order on entry.  This favors lower cpu-ids and will
 * grant the double lock to lower cpus over higher ids under contention,
 * regardless of entry order into the function.
 */
static inline int _double_lock_balance(struct rq *this_rq, struct rq *busiest)
	__releases(this_rq->lock)
	__acquires(busiest->lock)
	__acquires(this_rq->lock)
{
	int ret = 0;

	if (unlikely(!raw_spin_trylock(&busiest->lock))) {
		if (busiest < this_rq) {
			raw_spin_unlock(&this_rq->lock);
			raw_spin_lock(&busiest->lock);
			raw_spin_lock_nested(&this_rq->lock,
					      SINGLE_DEPTH_NESTING);
			ret = 1;
		} else
			raw_spin_lock_nested(&busiest->lock,
					      SINGLE_DEPTH_NESTING);
	}
	return ret;
}

#endif /* CONFIG_PREEMPT */

/*
 * double_lock_balance - lock the busiest runqueue, this_rq is locked already.
 */
static inline int double_lock_balance(struct rq *this_rq, struct rq *busiest)
{
	if (unlikely(!irqs_disabled())) {
		/* printk() doesn't work good under rq->lock */
		raw_spin_unlock(&this_rq->lock);
		BUG_ON(1);
	}

	return _double_lock_balance(this_rq, busiest);
}

static inline void double_unlock_balance(struct rq *this_rq, struct rq *busiest)
	__releases(busiest->lock)
{
	raw_spin_unlock(&busiest->lock);
	lock_set_subclass(&this_rq->lock.dep_map, 0, _RET_IP_);
}

1724 1725 1726 1727 1728 1729 1730 1731 1732
static inline void double_lock(spinlock_t *l1, spinlock_t *l2)
{
	if (l1 > l2)
		swap(l1, l2);

	spin_lock(l1);
	spin_lock_nested(l2, SINGLE_DEPTH_NESTING);
}

1733 1734 1735 1736 1737 1738 1739 1740 1741
static inline void double_lock_irq(spinlock_t *l1, spinlock_t *l2)
{
	if (l1 > l2)
		swap(l1, l2);

	spin_lock_irq(l1);
	spin_lock_nested(l2, SINGLE_DEPTH_NESTING);
}

1742 1743 1744 1745 1746 1747 1748 1749 1750
static inline void double_raw_lock(raw_spinlock_t *l1, raw_spinlock_t *l2)
{
	if (l1 > l2)
		swap(l1, l2);

	raw_spin_lock(l1);
	raw_spin_lock_nested(l2, SINGLE_DEPTH_NESTING);
}

1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792
/*
 * double_rq_lock - safely lock two runqueues
 *
 * Note this does not disable interrupts like task_rq_lock,
 * you need to do so manually before calling.
 */
static inline void double_rq_lock(struct rq *rq1, struct rq *rq2)
	__acquires(rq1->lock)
	__acquires(rq2->lock)
{
	BUG_ON(!irqs_disabled());
	if (rq1 == rq2) {
		raw_spin_lock(&rq1->lock);
		__acquire(rq2->lock);	/* Fake it out ;) */
	} else {
		if (rq1 < rq2) {
			raw_spin_lock(&rq1->lock);
			raw_spin_lock_nested(&rq2->lock, SINGLE_DEPTH_NESTING);
		} else {
			raw_spin_lock(&rq2->lock);
			raw_spin_lock_nested(&rq1->lock, SINGLE_DEPTH_NESTING);
		}
	}
}

/*
 * double_rq_unlock - safely unlock two runqueues
 *
 * Note this does not restore interrupts like task_rq_unlock,
 * you need to do so manually after calling.
 */
static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2)
	__releases(rq1->lock)
	__releases(rq2->lock)
{
	raw_spin_unlock(&rq1->lock);
	if (rq1 != rq2)
		raw_spin_unlock(&rq2->lock);
	else
		__release(rq2->lock);
}

1793 1794 1795 1796
extern void set_rq_online (struct rq *rq);
extern void set_rq_offline(struct rq *rq);
extern bool sched_smp_initialized;

1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833
#else /* CONFIG_SMP */

/*
 * double_rq_lock - safely lock two runqueues
 *
 * Note this does not disable interrupts like task_rq_lock,
 * you need to do so manually before calling.
 */
static inline void double_rq_lock(struct rq *rq1, struct rq *rq2)
	__acquires(rq1->lock)
	__acquires(rq2->lock)
{
	BUG_ON(!irqs_disabled());
	BUG_ON(rq1 != rq2);
	raw_spin_lock(&rq1->lock);
	__acquire(rq2->lock);	/* Fake it out ;) */
}

/*
 * double_rq_unlock - safely unlock two runqueues
 *
 * Note this does not restore interrupts like task_rq_unlock,
 * you need to do so manually after calling.
 */
static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2)
	__releases(rq1->lock)
	__releases(rq2->lock)
{
	BUG_ON(rq1 != rq2);
	raw_spin_unlock(&rq1->lock);
	__release(rq2->lock);
}

#endif

extern struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq);
extern struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq);
1834 1835

#ifdef	CONFIG_SCHED_DEBUG
1836 1837
extern void print_cfs_stats(struct seq_file *m, int cpu);
extern void print_rt_stats(struct seq_file *m, int cpu);
1838
extern void print_dl_stats(struct seq_file *m, int cpu);
1839 1840
extern void
print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq);
1841 1842 1843 1844 1845 1846 1847 1848
#ifdef CONFIG_NUMA_BALANCING
extern void
show_numa_stats(struct task_struct *p, struct seq_file *m);
extern void
print_numa_stats(struct seq_file *m, int node, unsigned long tsf,
	unsigned long tpf, unsigned long gsf, unsigned long gpf);
#endif /* CONFIG_NUMA_BALANCING */
#endif /* CONFIG_SCHED_DEBUG */
1849 1850

extern void init_cfs_rq(struct cfs_rq *cfs_rq);
1851 1852
extern void init_rt_rq(struct rt_rq *rt_rq);
extern void init_dl_rq(struct dl_rq *dl_rq);
1853

1854 1855
extern void cfs_bandwidth_usage_inc(void);
extern void cfs_bandwidth_usage_dec(void);
1856

1857
#ifdef CONFIG_NO_HZ_COMMON
1858 1859 1860 1861 1862 1863
enum rq_nohz_flag_bits {
	NOHZ_TICK_STOPPED,
	NOHZ_BALANCE_KICK,
};

#define nohz_flags(cpu)	(&cpu_rq(cpu)->nohz_flags)
1864 1865 1866 1867

extern void nohz_balance_exit_idle(unsigned int cpu);
#else
static inline void nohz_balance_exit_idle(unsigned int cpu) { }
1868
#endif
1869 1870

#ifdef CONFIG_IRQ_TIME_ACCOUNTING
1871
struct irqtime {
1872
	u64			total;
1873
	u64			tick_delta;
1874 1875 1876
	u64			irq_start_time;
	struct u64_stats_sync	sync;
};
1877

1878
DECLARE_PER_CPU(struct irqtime, cpu_irqtime);
1879

1880 1881 1882 1883 1884
/*
 * Returns the irqtime minus the softirq time computed by ksoftirqd.
 * Otherwise ksoftirqd's sum_exec_runtime is substracted its own runtime
 * and never move forward.
 */
1885 1886
static inline u64 irq_time_read(int cpu)
{
1887 1888 1889
	struct irqtime *irqtime = &per_cpu(cpu_irqtime, cpu);
	unsigned int seq;
	u64 total;
1890 1891

	do {
1892
		seq = __u64_stats_fetch_begin(&irqtime->sync);
1893
		total = irqtime->total;
1894
	} while (__u64_stats_fetch_retry(&irqtime->sync, seq));
1895

1896
	return total;
1897 1898
}
#endif /* CONFIG_IRQ_TIME_ACCOUNTING */
1899 1900 1901 1902 1903 1904

#ifdef CONFIG_CPU_FREQ
DECLARE_PER_CPU(struct update_util_data *, cpufreq_update_util_data);

/**
 * cpufreq_update_util - Take a note about CPU utilization changes.
1905
 * @rq: Runqueue to carry out the update for.
1906
 * @flags: Update reason flags.
1907
 *
1908 1909
 * This function is called by the scheduler on the CPU whose utilization is
 * being updated.
1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924
 *
 * It can only be called from RCU-sched read-side critical sections.
 *
 * The way cpufreq is currently arranged requires it to evaluate the CPU
 * performance state (frequency/voltage) on a regular basis to prevent it from
 * being stuck in a completely inadequate performance level for too long.
 * That is not guaranteed to happen if the updates are only triggered from CFS,
 * though, because they may not be coming in if RT or deadline tasks are active
 * all the time (or there are RT and DL tasks only).
 *
 * As a workaround for that issue, this function is called by the RT and DL
 * sched classes to trigger extra cpufreq updates to prevent it from stalling,
 * but that really is a band-aid.  Going forward it should be replaced with
 * solutions targeted more specifically at RT and DL tasks.
 */
1925
static inline void cpufreq_update_util(struct rq *rq, unsigned int flags)
1926
{
1927 1928 1929 1930
	struct update_util_data *data;

	data = rcu_dereference_sched(*this_cpu_ptr(&cpufreq_update_util_data));
	if (data)
1931 1932 1933 1934 1935 1936 1937
		data->func(data, rq_clock(rq), flags);
}

static inline void cpufreq_update_this_cpu(struct rq *rq, unsigned int flags)
{
	if (cpu_of(rq) == smp_processor_id())
		cpufreq_update_util(rq, flags);
1938 1939
}
#else
1940 1941
static inline void cpufreq_update_util(struct rq *rq, unsigned int flags) {}
static inline void cpufreq_update_this_cpu(struct rq *rq, unsigned int flags) {}
1942
#endif /* CONFIG_CPU_FREQ */
1943

1944 1945 1946 1947 1948 1949 1950
#ifdef arch_scale_freq_capacity
#ifndef arch_scale_freq_invariant
#define arch_scale_freq_invariant()	(true)
#endif
#else /* arch_scale_freq_capacity */
#define arch_scale_freq_invariant()	(false)
#endif