deadline.c 73.4 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0
2 3 4 5 6 7 8 9 10 11 12 13
/*
 * Deadline Scheduling Class (SCHED_DEADLINE)
 *
 * Earliest Deadline First (EDF) + Constant Bandwidth Server (CBS).
 *
 * Tasks that periodically executes their instances for less than their
 * runtime won't miss any of their deadlines.
 * Tasks that are not periodic or sporadic or that tries to execute more
 * than their reserved bandwidth will be slowed down (and may potentially
 * miss some of their deadlines), and won't affect any other task.
 *
 * Copyright (C) 2012 Dario Faggioli <raistlin@linux.it>,
14
 *                    Juri Lelli <juri.lelli@gmail.com>,
15 16 17 18 19
 *                    Michael Trimarchi <michael@amarulasolutions.com>,
 *                    Fabio Checconi <fchecconi@gmail.com>
 */
#include "sched.h"

20
#include <linux/slab.h>
21
#include <uapi/linux/sched/types.h>
22

23 24
struct dl_bandwidth def_dl_bandwidth;

25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
static inline struct task_struct *dl_task_of(struct sched_dl_entity *dl_se)
{
	return container_of(dl_se, struct task_struct, dl);
}

static inline struct rq *rq_of_dl_rq(struct dl_rq *dl_rq)
{
	return container_of(dl_rq, struct rq, dl);
}

static inline struct dl_rq *dl_rq_of_se(struct sched_dl_entity *dl_se)
{
	struct task_struct *p = dl_task_of(dl_se);
	struct rq *rq = task_rq(p);

	return &rq->dl;
}

static inline int on_dl_rq(struct sched_dl_entity *dl_se)
{
	return !RB_EMPTY_NODE(&dl_se->rb_node);
}

48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
#ifdef CONFIG_SMP
static inline struct dl_bw *dl_bw_of(int i)
{
	RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(),
			 "sched RCU must be held");
	return &cpu_rq(i)->rd->dl_bw;
}

static inline int dl_bw_cpus(int i)
{
	struct root_domain *rd = cpu_rq(i)->rd;
	int cpus = 0;

	RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(),
			 "sched RCU must be held");
	for_each_cpu_and(i, rd->span, cpu_active_mask)
		cpus++;

	return cpus;
}
#else
static inline struct dl_bw *dl_bw_of(int i)
{
	return &cpu_rq(i)->dl.dl_bw;
}

static inline int dl_bw_cpus(int i)
{
	return 1;
}
#endif

80
static inline
81
void __add_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
82 83 84 85 86 87
{
	u64 old = dl_rq->running_bw;

	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
	dl_rq->running_bw += dl_bw;
	SCHED_WARN_ON(dl_rq->running_bw < old); /* overflow */
88
	SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw);
89 90
	/* kick cpufreq (see the comment in kernel/sched/sched.h). */
	cpufreq_update_util(rq_of_dl_rq(dl_rq), SCHED_CPUFREQ_DL);
91 92 93
}

static inline
94
void __sub_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
95 96 97 98 99 100 101 102
{
	u64 old = dl_rq->running_bw;

	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
	dl_rq->running_bw -= dl_bw;
	SCHED_WARN_ON(dl_rq->running_bw > old); /* underflow */
	if (dl_rq->running_bw > old)
		dl_rq->running_bw = 0;
103 104
	/* kick cpufreq (see the comment in kernel/sched/sched.h). */
	cpufreq_update_util(rq_of_dl_rq(dl_rq), SCHED_CPUFREQ_DL);
105 106
}

107
static inline
108
void __add_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
109 110 111 112 113 114 115 116 117
{
	u64 old = dl_rq->this_bw;

	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
	dl_rq->this_bw += dl_bw;
	SCHED_WARN_ON(dl_rq->this_bw < old); /* overflow */
}

static inline
118
void __sub_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
119 120 121 122 123 124 125 126 127 128 129
{
	u64 old = dl_rq->this_bw;

	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
	dl_rq->this_bw -= dl_bw;
	SCHED_WARN_ON(dl_rq->this_bw > old); /* underflow */
	if (dl_rq->this_bw > old)
		dl_rq->this_bw = 0;
	SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw);
}

130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157
static inline
void add_rq_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
	if (!dl_entity_is_special(dl_se))
		__add_rq_bw(dl_se->dl_bw, dl_rq);
}

static inline
void sub_rq_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
	if (!dl_entity_is_special(dl_se))
		__sub_rq_bw(dl_se->dl_bw, dl_rq);
}

static inline
void add_running_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
	if (!dl_entity_is_special(dl_se))
		__add_running_bw(dl_se->dl_bw, dl_rq);
}

static inline
void sub_running_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
	if (!dl_entity_is_special(dl_se))
		__sub_running_bw(dl_se->dl_bw, dl_rq);
}

158 159
void dl_change_utilization(struct task_struct *p, u64 new_bw)
{
160
	struct rq *rq;
161

162 163
	BUG_ON(p->dl.flags & SCHED_FLAG_SUGOV);

164
	if (task_on_rq_queued(p))
165 166
		return;

167 168
	rq = task_rq(p);
	if (p->dl.dl_non_contending) {
169
		sub_running_bw(&p->dl, &rq->dl);
170 171 172 173 174 175 176 177 178 179 180
		p->dl.dl_non_contending = 0;
		/*
		 * If the timer handler is currently running and the
		 * timer cannot be cancelled, inactive_task_timer()
		 * will see that dl_not_contending is not set, and
		 * will not touch the rq's active utilization,
		 * so we are still safe.
		 */
		if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1)
			put_task_struct(p);
	}
181 182
	__sub_rq_bw(p->dl.dl_bw, &rq->dl);
	__add_rq_bw(new_bw, &rq->dl);
183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
}

/*
 * The utilization of a task cannot be immediately removed from
 * the rq active utilization (running_bw) when the task blocks.
 * Instead, we have to wait for the so called "0-lag time".
 *
 * If a task blocks before the "0-lag time", a timer (the inactive
 * timer) is armed, and running_bw is decreased when the timer
 * fires.
 *
 * If the task wakes up again before the inactive timer fires,
 * the timer is cancelled, whereas if the task wakes up after the
 * inactive timer fired (and running_bw has been decreased) the
 * task's utilization has to be added to running_bw again.
 * A flag in the deadline scheduling entity (dl_non_contending)
 * is used to avoid race conditions between the inactive timer handler
 * and task wakeups.
 *
 * The following diagram shows how running_bw is updated. A task is
 * "ACTIVE" when its utilization contributes to running_bw; an
 * "ACTIVE contending" task is in the TASK_RUNNING state, while an
 * "ACTIVE non contending" task is a blocked task for which the "0-lag time"
 * has not passed yet. An "INACTIVE" task is a task for which the "0-lag"
 * time already passed, which does not contribute to running_bw anymore.
 *                              +------------------+
 *             wakeup           |    ACTIVE        |
 *          +------------------>+   contending     |
 *          | add_running_bw    |                  |
 *          |                   +----+------+------+
 *          |                        |      ^
 *          |                dequeue |      |
 * +--------+-------+                |      |
 * |                |   t >= 0-lag   |      | wakeup
 * |    INACTIVE    |<---------------+      |
 * |                | sub_running_bw |      |
 * +--------+-------+                |      |
 *          ^                        |      |
 *          |              t < 0-lag |      |
 *          |                        |      |
 *          |                        V      |
 *          |                   +----+------+------+
 *          | sub_running_bw    |    ACTIVE        |
 *          +-------------------+                  |
 *            inactive timer    |  non contending  |
 *            fired             +------------------+
 *
 * The task_non_contending() function is invoked when a task
 * blocks, and checks if the 0-lag time already passed or
 * not (in the first case, it directly updates running_bw;
 * in the second case, it arms the inactive timer).
 *
 * The task_contending() function is invoked when a task wakes
 * up, and checks if the task is still in the "ACTIVE non contending"
 * state or not (in the second case, it updates running_bw).
 */
static void task_non_contending(struct task_struct *p)
{
	struct sched_dl_entity *dl_se = &p->dl;
	struct hrtimer *timer = &dl_se->inactive_timer;
	struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
	struct rq *rq = rq_of_dl_rq(dl_rq);
	s64 zerolag_time;

	/*
	 * If this is a non-deadline task that has been boosted,
	 * do nothing
	 */
	if (dl_se->dl_runtime == 0)
		return;

254 255 256
	if (dl_entity_is_special(dl_se))
		return;

257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275
	WARN_ON(hrtimer_active(&dl_se->inactive_timer));
	WARN_ON(dl_se->dl_non_contending);

	zerolag_time = dl_se->deadline -
		 div64_long((dl_se->runtime * dl_se->dl_period),
			dl_se->dl_runtime);

	/*
	 * Using relative times instead of the absolute "0-lag time"
	 * allows to simplify the code
	 */
	zerolag_time -= rq_clock(rq);

	/*
	 * If the "0-lag time" already passed, decrease the active
	 * utilization now, instead of starting a timer
	 */
	if (zerolag_time < 0) {
		if (dl_task(p))
276
			sub_running_bw(dl_se, dl_rq);
277 278 279
		if (!dl_task(p) || p->state == TASK_DEAD) {
			struct dl_bw *dl_b = dl_bw_of(task_cpu(p));

280
			if (p->state == TASK_DEAD)
281
				sub_rq_bw(&p->dl, &rq->dl);
282
			raw_spin_lock(&dl_b->lock);
283
			__dl_sub(dl_b, p->dl.dl_bw, dl_bw_cpus(task_cpu(p)));
284
			__dl_clear_params(p);
285 286
			raw_spin_unlock(&dl_b->lock);
		}
287 288 289 290 291 292 293 294 295

		return;
	}

	dl_se->dl_non_contending = 1;
	get_task_struct(p);
	hrtimer_start(timer, ns_to_ktime(zerolag_time), HRTIMER_MODE_REL);
}

296
static void task_contending(struct sched_dl_entity *dl_se, int flags)
297 298 299 300 301 302 303 304 305 306
{
	struct dl_rq *dl_rq = dl_rq_of_se(dl_se);

	/*
	 * If this is a non-deadline task that has been boosted,
	 * do nothing
	 */
	if (dl_se->dl_runtime == 0)
		return;

307
	if (flags & ENQUEUE_MIGRATED)
308
		add_rq_bw(dl_se, dl_rq);
309

310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328
	if (dl_se->dl_non_contending) {
		dl_se->dl_non_contending = 0;
		/*
		 * If the timer handler is currently running and the
		 * timer cannot be cancelled, inactive_task_timer()
		 * will see that dl_not_contending is not set, and
		 * will not touch the rq's active utilization,
		 * so we are still safe.
		 */
		if (hrtimer_try_to_cancel(&dl_se->inactive_timer) == 1)
			put_task_struct(dl_task_of(dl_se));
	} else {
		/*
		 * Since "dl_non_contending" is not set, the
		 * task's utilization has already been removed from
		 * active utilization (either when the task blocked,
		 * when the "inactive timer" fired).
		 * So, add it back.
		 */
329
		add_running_bw(dl_se, dl_rq);
330 331 332
	}
}

333 334 335 336
static inline int is_leftmost(struct task_struct *p, struct dl_rq *dl_rq)
{
	struct sched_dl_entity *dl_se = &p->dl;

337
	return dl_rq->root.rb_leftmost == &dl_se->rb_node;
338 339
}

340 341 342 343 344 345 346 347 348 349 350
void init_dl_bandwidth(struct dl_bandwidth *dl_b, u64 period, u64 runtime)
{
	raw_spin_lock_init(&dl_b->dl_runtime_lock);
	dl_b->dl_period = period;
	dl_b->dl_runtime = runtime;
}

void init_dl_bw(struct dl_bw *dl_b)
{
	raw_spin_lock_init(&dl_b->lock);
	raw_spin_lock(&def_dl_bandwidth.dl_runtime_lock);
351
	if (global_rt_runtime() == RUNTIME_INF)
352 353
		dl_b->bw = -1;
	else
354
		dl_b->bw = to_ratio(global_rt_period(), global_rt_runtime());
355 356 357 358
	raw_spin_unlock(&def_dl_bandwidth.dl_runtime_lock);
	dl_b->total_bw = 0;
}

359
void init_dl_rq(struct dl_rq *dl_rq)
360
{
361
	dl_rq->root = RB_ROOT_CACHED;
362 363 364 365 366 367 368

#ifdef CONFIG_SMP
	/* zero means no -deadline tasks */
	dl_rq->earliest_dl.curr = dl_rq->earliest_dl.next = 0;

	dl_rq->dl_nr_migratory = 0;
	dl_rq->overloaded = 0;
369
	dl_rq->pushable_dl_tasks_root = RB_ROOT_CACHED;
370 371
#else
	init_dl_bw(&dl_rq->dl_bw);
372
#endif
373 374

	dl_rq->running_bw = 0;
375
	dl_rq->this_bw = 0;
376
	init_dl_rq_bw_ratio(dl_rq);
377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412
}

#ifdef CONFIG_SMP

static inline int dl_overloaded(struct rq *rq)
{
	return atomic_read(&rq->rd->dlo_count);
}

static inline void dl_set_overload(struct rq *rq)
{
	if (!rq->online)
		return;

	cpumask_set_cpu(rq->cpu, rq->rd->dlo_mask);
	/*
	 * Must be visible before the overload count is
	 * set (as in sched_rt.c).
	 *
	 * Matched by the barrier in pull_dl_task().
	 */
	smp_wmb();
	atomic_inc(&rq->rd->dlo_count);
}

static inline void dl_clear_overload(struct rq *rq)
{
	if (!rq->online)
		return;

	atomic_dec(&rq->rd->dlo_count);
	cpumask_clear_cpu(rq->cpu, rq->rd->dlo_mask);
}

static void update_dl_migration(struct dl_rq *dl_rq)
{
413
	if (dl_rq->dl_nr_migratory && dl_rq->dl_nr_running > 1) {
414 415 416 417 418 419 420 421 422 423 424 425 426 427
		if (!dl_rq->overloaded) {
			dl_set_overload(rq_of_dl_rq(dl_rq));
			dl_rq->overloaded = 1;
		}
	} else if (dl_rq->overloaded) {
		dl_clear_overload(rq_of_dl_rq(dl_rq));
		dl_rq->overloaded = 0;
	}
}

static void inc_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
	struct task_struct *p = dl_task_of(dl_se);

428
	if (p->nr_cpus_allowed > 1)
429 430 431 432 433 434 435 436 437
		dl_rq->dl_nr_migratory++;

	update_dl_migration(dl_rq);
}

static void dec_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
	struct task_struct *p = dl_task_of(dl_se);

438
	if (p->nr_cpus_allowed > 1)
439 440 441 442 443 444 445 446 447 448 449 450
		dl_rq->dl_nr_migratory--;

	update_dl_migration(dl_rq);
}

/*
 * The list of pushable -deadline task is not a plist, like in
 * sched_rt.c, it is an rb-tree with tasks ordered by deadline.
 */
static void enqueue_pushable_dl_task(struct rq *rq, struct task_struct *p)
{
	struct dl_rq *dl_rq = &rq->dl;
451
	struct rb_node **link = &dl_rq->pushable_dl_tasks_root.rb_root.rb_node;
452 453
	struct rb_node *parent = NULL;
	struct task_struct *entry;
454
	bool leftmost = true;
455 456 457 458 459 460 461 462 463 464 465

	BUG_ON(!RB_EMPTY_NODE(&p->pushable_dl_tasks));

	while (*link) {
		parent = *link;
		entry = rb_entry(parent, struct task_struct,
				 pushable_dl_tasks);
		if (dl_entity_preempt(&p->dl, &entry->dl))
			link = &parent->rb_left;
		else {
			link = &parent->rb_right;
466
			leftmost = false;
467 468 469
		}
	}

470
	if (leftmost)
471
		dl_rq->earliest_dl.next = p->dl.deadline;
472 473

	rb_link_node(&p->pushable_dl_tasks, parent, link);
474 475
	rb_insert_color_cached(&p->pushable_dl_tasks,
			       &dl_rq->pushable_dl_tasks_root, leftmost);
476 477
}

478 479 480 481 482 483 484
static void dequeue_pushable_dl_task(struct rq *rq, struct task_struct *p)
{
	struct dl_rq *dl_rq = &rq->dl;

	if (RB_EMPTY_NODE(&p->pushable_dl_tasks))
		return;

485
	if (dl_rq->pushable_dl_tasks_root.rb_leftmost == &p->pushable_dl_tasks) {
486 487 488
		struct rb_node *next_node;

		next_node = rb_next(&p->pushable_dl_tasks);
489 490 491 492
		if (next_node) {
			dl_rq->earliest_dl.next = rb_entry(next_node,
				struct task_struct, pushable_dl_tasks)->dl.deadline;
		}
493 494
	}

495
	rb_erase_cached(&p->pushable_dl_tasks, &dl_rq->pushable_dl_tasks_root);
496 497 498 499 500
	RB_CLEAR_NODE(&p->pushable_dl_tasks);
}

static inline int has_pushable_dl_tasks(struct rq *rq)
{
501
	return !RB_EMPTY_ROOT(&rq->dl.pushable_dl_tasks_root.rb_root);
502 503 504 505
}

static int push_dl_task(struct rq *rq);

P
Peter Zijlstra 已提交
506 507 508 509 510
static inline bool need_pull_dl_task(struct rq *rq, struct task_struct *prev)
{
	return dl_task(prev);
}

511 512
static DEFINE_PER_CPU(struct callback_head, dl_push_head);
static DEFINE_PER_CPU(struct callback_head, dl_pull_head);
513 514

static void push_dl_tasks(struct rq *);
515
static void pull_dl_task(struct rq *);
516 517

static inline void queue_push_tasks(struct rq *rq)
P
Peter Zijlstra 已提交
518
{
519 520 521
	if (!has_pushable_dl_tasks(rq))
		return;

522 523 524 525 526 527
	queue_balance_callback(rq, &per_cpu(dl_push_head, rq->cpu), push_dl_tasks);
}

static inline void queue_pull_task(struct rq *rq)
{
	queue_balance_callback(rq, &per_cpu(dl_pull_head, rq->cpu), pull_dl_task);
P
Peter Zijlstra 已提交
528 529
}

530 531
static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq);

532
static struct rq *dl_task_offline_migration(struct rq *rq, struct task_struct *p)
533 534 535 536 537 538 539 540 541 542 543
{
	struct rq *later_rq = NULL;

	later_rq = find_lock_later_rq(p, rq);
	if (!later_rq) {
		int cpu;

		/*
		 * If we cannot preempt any rq, fall back to pick any
		 * online cpu.
		 */
544
		cpu = cpumask_any_and(cpu_active_mask, &p->cpus_allowed);
545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563
		if (cpu >= nr_cpu_ids) {
			/*
			 * Fail to find any suitable cpu.
			 * The task will never come back!
			 */
			BUG_ON(dl_bandwidth_enabled());

			/*
			 * If admission control is disabled we
			 * try a little harder to let the task
			 * run.
			 */
			cpu = cpumask_any(cpu_active_mask);
		}
		later_rq = cpu_rq(cpu);
		double_lock_balance(rq, later_rq);
	}

	set_task_cpu(p, later_rq->cpu);
564 565 566
	double_unlock_balance(later_rq, rq);

	return later_rq;
567 568
}

569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590
#else

static inline
void enqueue_pushable_dl_task(struct rq *rq, struct task_struct *p)
{
}

static inline
void dequeue_pushable_dl_task(struct rq *rq, struct task_struct *p)
{
}

static inline
void inc_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
}

static inline
void dec_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
}

P
Peter Zijlstra 已提交
591 592 593 594 595
static inline bool need_pull_dl_task(struct rq *rq, struct task_struct *prev)
{
	return false;
}

596
static inline void pull_dl_task(struct rq *rq)
P
Peter Zijlstra 已提交
597 598 599
{
}

600
static inline void queue_push_tasks(struct rq *rq)
P
Peter Zijlstra 已提交
601 602 603
{
}

604
static inline void queue_pull_task(struct rq *rq)
P
Peter Zijlstra 已提交
605 606
{
}
607 608
#endif /* CONFIG_SMP */

609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625
static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags);
static void __dequeue_task_dl(struct rq *rq, struct task_struct *p, int flags);
static void check_preempt_curr_dl(struct rq *rq, struct task_struct *p,
				  int flags);

/*
 * We are being explicitly informed that a new instance is starting,
 * and this means that:
 *  - the absolute deadline of the entity has to be placed at
 *    current time + relative deadline;
 *  - the runtime of the entity has to be set to the maximum value.
 *
 * The capability of specifying such event is useful whenever a -deadline
 * entity wants to (try to!) synchronize its behaviour with the scheduler's
 * one, and to (try to!) reconcile itself with its own scheduling
 * parameters.
 */
626
static inline void setup_new_dl_entity(struct sched_dl_entity *dl_se)
627 628 629 630
{
	struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
	struct rq *rq = rq_of_dl_rq(dl_rq);

631
	WARN_ON(dl_se->dl_boosted);
632 633 634 635 636 637 638 639 640
	WARN_ON(dl_time_before(rq_clock(rq), dl_se->deadline));

	/*
	 * We are racing with the deadline timer. So, do nothing because
	 * the deadline timer handler will take care of properly recharging
	 * the runtime and postponing the deadline
	 */
	if (dl_se->dl_throttled)
		return;
641 642 643 644 645 646

	/*
	 * We use the regular wall clock time to set deadlines in the
	 * future; in fact, we must consider execution overheads (time
	 * spent on hardirq context, etc.).
	 */
647 648
	dl_se->deadline = rq_clock(rq) + dl_se->dl_deadline;
	dl_se->runtime = dl_se->dl_runtime;
649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666
}

/*
 * Pure Earliest Deadline First (EDF) scheduling does not deal with the
 * possibility of a entity lasting more than what it declared, and thus
 * exhausting its runtime.
 *
 * Here we are interested in making runtime overrun possible, but we do
 * not want a entity which is misbehaving to affect the scheduling of all
 * other entities.
 * Therefore, a budgeting strategy called Constant Bandwidth Server (CBS)
 * is used, in order to confine each entity within its own bandwidth.
 *
 * This function deals exactly with that, and ensures that when the runtime
 * of a entity is replenished, its deadline is also postponed. That ensures
 * the overrunning entity can't interfere with other entity in the system and
 * can't make them miss their deadlines. Reasons why this kind of overruns
 * could happen are, typically, a entity voluntarily trying to overcome its
667
 * runtime, or it just underestimated it during sched_setattr().
668
 */
669 670
static void replenish_dl_entity(struct sched_dl_entity *dl_se,
				struct sched_dl_entity *pi_se)
671 672 673 674
{
	struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
	struct rq *rq = rq_of_dl_rq(dl_rq);

675 676 677 678 679 680 681 682 683 684 685
	BUG_ON(pi_se->dl_runtime <= 0);

	/*
	 * This could be the case for a !-dl task that is boosted.
	 * Just go with full inherited parameters.
	 */
	if (dl_se->dl_deadline == 0) {
		dl_se->deadline = rq_clock(rq) + pi_se->dl_deadline;
		dl_se->runtime = pi_se->dl_runtime;
	}

686 687 688
	if (dl_se->dl_yielded && dl_se->runtime > 0)
		dl_se->runtime = 0;

689 690 691 692 693 694 695
	/*
	 * We keep moving the deadline away until we get some
	 * available runtime for the entity. This ensures correct
	 * handling of situations where the runtime overrun is
	 * arbitrary large.
	 */
	while (dl_se->runtime <= 0) {
696 697
		dl_se->deadline += pi_se->dl_period;
		dl_se->runtime += pi_se->dl_runtime;
698 699 700 701 702 703 704 705 706 707 708 709
	}

	/*
	 * At this point, the deadline really should be "in
	 * the future" with respect to rq->clock. If it's
	 * not, we are, for some reason, lagging too much!
	 * Anyway, after having warn userspace abut that,
	 * we still try to keep the things running by
	 * resetting the deadline and the budget of the
	 * entity.
	 */
	if (dl_time_before(dl_se->deadline, rq_clock(rq))) {
710
		printk_deferred_once("sched: DL replenish lagged too much\n");
711 712
		dl_se->deadline = rq_clock(rq) + pi_se->dl_deadline;
		dl_se->runtime = pi_se->dl_runtime;
713
	}
714 715 716 717 718

	if (dl_se->dl_yielded)
		dl_se->dl_yielded = 0;
	if (dl_se->dl_throttled)
		dl_se->dl_throttled = 0;
719 720 721 722 723 724 725 726 727 728 729 730 731
}

/*
 * Here we check if --at time t-- an entity (which is probably being
 * [re]activated or, in general, enqueued) can use its remaining runtime
 * and its current deadline _without_ exceeding the bandwidth it is
 * assigned (function returns true if it can't). We are in fact applying
 * one of the CBS rules: when a task wakes up, if the residual runtime
 * over residual deadline fits within the allocated bandwidth, then we
 * can keep the current (absolute) deadline and residual budget without
 * disrupting the schedulability of the system. Otherwise, we should
 * refill the runtime and set the deadline a period in the future,
 * because keeping the current (absolute) deadline of the task would
732 733
 * result in breaking guarantees promised to other tasks (refer to
 * Documentation/scheduler/sched-deadline.txt for more informations).
734 735 736
 *
 * This function returns true if:
 *
737
 *   runtime / (deadline - t) > dl_runtime / dl_deadline ,
738 739
 *
 * IOW we can't recycle current parameters.
740
 *
741
 * Notice that the bandwidth check is done against the deadline. For
742
 * task with deadline equal to period this is the same of using
743
 * dl_period instead of dl_deadline in the equation above.
744
 */
745 746
static bool dl_entity_overflow(struct sched_dl_entity *dl_se,
			       struct sched_dl_entity *pi_se, u64 t)
747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767
{
	u64 left, right;

	/*
	 * left and right are the two sides of the equation above,
	 * after a bit of shuffling to use multiplications instead
	 * of divisions.
	 *
	 * Note that none of the time values involved in the two
	 * multiplications are absolute: dl_deadline and dl_runtime
	 * are the relative deadline and the maximum runtime of each
	 * instance, runtime is the runtime left for the last instance
	 * and (deadline - t), since t is rq->clock, is the time left
	 * to the (absolute) deadline. Even if overflowing the u64 type
	 * is very unlikely to occur in both cases, here we scale down
	 * as we want to avoid that risk at all. Scaling down by 10
	 * means that we reduce granularity to 1us. We are fine with it,
	 * since this is only a true/false check and, anyway, thinking
	 * of anything below microseconds resolution is actually fiction
	 * (but still we want to give the user that illusion >;).
	 */
768
	left = (pi_se->dl_deadline >> DL_SCALE) * (dl_se->runtime >> DL_SCALE);
769 770
	right = ((dl_se->deadline - t) >> DL_SCALE) *
		(pi_se->dl_runtime >> DL_SCALE);
771 772 773 774 775

	return dl_time_before(right, left);
}

/*
776 777 778 779
 * Revised wakeup rule [1]: For self-suspending tasks, rather then
 * re-initializing task's runtime and deadline, the revised wakeup
 * rule adjusts the task's runtime to avoid the task to overrun its
 * density.
780
 *
781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853
 * Reasoning: a task may overrun the density if:
 *    runtime / (deadline - t) > dl_runtime / dl_deadline
 *
 * Therefore, runtime can be adjusted to:
 *     runtime = (dl_runtime / dl_deadline) * (deadline - t)
 *
 * In such way that runtime will be equal to the maximum density
 * the task can use without breaking any rule.
 *
 * [1] Luca Abeni, Giuseppe Lipari, and Juri Lelli. 2015. Constant
 * bandwidth server revisited. SIGBED Rev. 11, 4 (January 2015), 19-24.
 */
static void
update_dl_revised_wakeup(struct sched_dl_entity *dl_se, struct rq *rq)
{
	u64 laxity = dl_se->deadline - rq_clock(rq);

	/*
	 * If the task has deadline < period, and the deadline is in the past,
	 * it should already be throttled before this check.
	 *
	 * See update_dl_entity() comments for further details.
	 */
	WARN_ON(dl_time_before(dl_se->deadline, rq_clock(rq)));

	dl_se->runtime = (dl_se->dl_density * laxity) >> BW_SHIFT;
}

/*
 * Regarding the deadline, a task with implicit deadline has a relative
 * deadline == relative period. A task with constrained deadline has a
 * relative deadline <= relative period.
 *
 * We support constrained deadline tasks. However, there are some restrictions
 * applied only for tasks which do not have an implicit deadline. See
 * update_dl_entity() to know more about such restrictions.
 *
 * The dl_is_implicit() returns true if the task has an implicit deadline.
 */
static inline bool dl_is_implicit(struct sched_dl_entity *dl_se)
{
	return dl_se->dl_deadline == dl_se->dl_period;
}

/*
 * When a deadline entity is placed in the runqueue, its runtime and deadline
 * might need to be updated. This is done by a CBS wake up rule. There are two
 * different rules: 1) the original CBS; and 2) the Revisited CBS.
 *
 * When the task is starting a new period, the Original CBS is used. In this
 * case, the runtime is replenished and a new absolute deadline is set.
 *
 * When a task is queued before the begin of the next period, using the
 * remaining runtime and deadline could make the entity to overflow, see
 * dl_entity_overflow() to find more about runtime overflow. When such case
 * is detected, the runtime and deadline need to be updated.
 *
 * If the task has an implicit deadline, i.e., deadline == period, the Original
 * CBS is applied. the runtime is replenished and a new absolute deadline is
 * set, as in the previous cases.
 *
 * However, the Original CBS does not work properly for tasks with
 * deadline < period, which are said to have a constrained deadline. By
 * applying the Original CBS, a constrained deadline task would be able to run
 * runtime/deadline in a period. With deadline < period, the task would
 * overrun the runtime/period allowed bandwidth, breaking the admission test.
 *
 * In order to prevent this misbehave, the Revisited CBS is used for
 * constrained deadline tasks when a runtime overflow is detected. In the
 * Revisited CBS, rather than replenishing & setting a new absolute deadline,
 * the remaining runtime of the task is reduced to avoid runtime overflow.
 * Please refer to the comments update_dl_revised_wakeup() function to find
 * more about the Revised CBS rule.
854
 */
855 856
static void update_dl_entity(struct sched_dl_entity *dl_se,
			     struct sched_dl_entity *pi_se)
857 858 859 860 861
{
	struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
	struct rq *rq = rq_of_dl_rq(dl_rq);

	if (dl_time_before(dl_se->deadline, rq_clock(rq)) ||
862
	    dl_entity_overflow(dl_se, pi_se, rq_clock(rq))) {
863 864 865 866 867 868 869 870

		if (unlikely(!dl_is_implicit(dl_se) &&
			     !dl_time_before(dl_se->deadline, rq_clock(rq)) &&
			     !dl_se->dl_boosted)){
			update_dl_revised_wakeup(dl_se, rq);
			return;
		}

871 872
		dl_se->deadline = rq_clock(rq) + pi_se->dl_deadline;
		dl_se->runtime = pi_se->dl_runtime;
873 874 875
	}
}

876 877 878 879 880
static inline u64 dl_next_period(struct sched_dl_entity *dl_se)
{
	return dl_se->deadline - dl_se->dl_deadline + dl_se->dl_period;
}

881 882 883
/*
 * If the entity depleted all its runtime, and if we want it to sleep
 * while waiting for some new execution time to become available, we
884
 * set the bandwidth replenishment timer to the replenishment instant
885 886 887 888 889 890
 * and try to activate it.
 *
 * Notice that it is important for the caller to know if the timer
 * actually started or not (i.e., the replenishment instant is in
 * the future or in the past).
 */
891
static int start_dl_timer(struct task_struct *p)
892
{
893 894 895
	struct sched_dl_entity *dl_se = &p->dl;
	struct hrtimer *timer = &dl_se->dl_timer;
	struct rq *rq = task_rq(p);
896 897 898
	ktime_t now, act;
	s64 delta;

899 900
	lockdep_assert_held(&rq->lock);

901 902 903 904 905
	/*
	 * We want the timer to fire at the deadline, but considering
	 * that it is actually coming from rq->clock and not from
	 * hrtimer's time base reading.
	 */
906
	act = ns_to_ktime(dl_next_period(dl_se));
907
	now = hrtimer_cb_get_time(timer);
908 909 910 911 912 913 914 915 916 917 918
	delta = ktime_to_ns(now) - rq_clock(rq);
	act = ktime_add_ns(act, delta);

	/*
	 * If the expiry time already passed, e.g., because the value
	 * chosen as the deadline is too small, don't even try to
	 * start the timer in the past!
	 */
	if (ktime_us_delta(act, now) < 0)
		return 0;

919 920 921 922 923 924 925 926 927 928 929 930 931
	/*
	 * !enqueued will guarantee another callback; even if one is already in
	 * progress. This ensures a balanced {get,put}_task_struct().
	 *
	 * The race against __run_timer() clearing the enqueued state is
	 * harmless because we're holding task_rq()->lock, therefore the timer
	 * expiring after we've done the check will wait on its task_rq_lock()
	 * and observe our state.
	 */
	if (!hrtimer_is_queued(timer)) {
		get_task_struct(p);
		hrtimer_start(timer, act, HRTIMER_MODE_ABS);
	}
932

933
	return 1;
934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954
}

/*
 * This is the bandwidth enforcement timer callback. If here, we know
 * a task is not on its dl_rq, since the fact that the timer was running
 * means the task is throttled and needs a runtime replenishment.
 *
 * However, what we actually do depends on the fact the task is active,
 * (it is on its rq) or has been removed from there by a call to
 * dequeue_task_dl(). In the former case we must issue the runtime
 * replenishment and add the task back to the dl_rq; in the latter, we just
 * do nothing but clearing dl_throttled, so that runtime and deadline
 * updating (and the queueing back to dl_rq) will be done by the
 * next call to enqueue_task_dl().
 */
static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
{
	struct sched_dl_entity *dl_se = container_of(timer,
						     struct sched_dl_entity,
						     dl_timer);
	struct task_struct *p = dl_task_of(dl_se);
955
	struct rq_flags rf;
956
	struct rq *rq;
957

958
	rq = task_rq_lock(p, &rf);
959

960
	/*
961
	 * The task might have changed its scheduling policy to something
962
	 * different than SCHED_DEADLINE (through switched_from_dl()).
963
	 */
964
	if (!dl_task(p))
965 966 967 968 969 970 971 972
		goto unlock;

	/*
	 * The task might have been boosted by someone else and might be in the
	 * boosting/deboosting path, its not throttled.
	 */
	if (dl_se->dl_boosted)
		goto unlock;
973

974
	/*
975 976
	 * Spurious timer due to start_dl_timer() race; or we already received
	 * a replenishment from rt_mutex_setprio().
977
	 */
978
	if (!dl_se->dl_throttled)
979
		goto unlock;
980 981 982

	sched_clock_tick();
	update_rq_clock(rq);
983

984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002
	/*
	 * If the throttle happened during sched-out; like:
	 *
	 *   schedule()
	 *     deactivate_task()
	 *       dequeue_task_dl()
	 *         update_curr_dl()
	 *           start_dl_timer()
	 *         __dequeue_task_dl()
	 *     prev->on_rq = 0;
	 *
	 * We can be both throttled and !queued. Replenish the counter
	 * but do not enqueue -- wait for our wakeup to do that.
	 */
	if (!task_on_rq_queued(p)) {
		replenish_dl_entity(dl_se, dl_se);
		goto unlock;
	}

1003
#ifdef CONFIG_SMP
1004
	if (unlikely(!rq->online)) {
1005 1006 1007 1008
		/*
		 * If the runqueue is no longer available, migrate the
		 * task elsewhere. This necessarily changes rq.
		 */
1009
		lockdep_unpin_lock(&rq->lock, rf.cookie);
1010
		rq = dl_task_offline_migration(rq, p);
1011
		rf.cookie = lockdep_pin_lock(&rq->lock);
1012
		update_rq_clock(rq);
1013 1014 1015 1016 1017 1018

		/*
		 * Now that the task has been migrated to the new RQ and we
		 * have that locked, proceed as normal and enqueue the task
		 * there.
		 */
1019
	}
1020
#endif
1021

1022 1023 1024 1025 1026
	enqueue_task_dl(rq, p, ENQUEUE_REPLENISH);
	if (dl_task(rq->curr))
		check_preempt_curr_dl(rq, p, 0);
	else
		resched_curr(rq);
1027

1028
#ifdef CONFIG_SMP
1029 1030 1031
	/*
	 * Queueing this task back might have overloaded rq, check if we need
	 * to kick someone away.
1032
	 */
1033 1034 1035 1036 1037
	if (has_pushable_dl_tasks(rq)) {
		/*
		 * Nothing relies on rq->lock after this, so its safe to drop
		 * rq->lock.
		 */
1038
		rq_unpin_lock(rq, &rf);
1039
		push_dl_task(rq);
1040
		rq_repin_lock(rq, &rf);
1041
	}
1042
#endif
1043

1044
unlock:
1045
	task_rq_unlock(rq, p, &rf);
1046

1047 1048 1049 1050 1051 1052
	/*
	 * This can free the task_struct, including this hrtimer, do not touch
	 * anything related to that after this.
	 */
	put_task_struct(p);

1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063
	return HRTIMER_NORESTART;
}

void init_dl_task_timer(struct sched_dl_entity *dl_se)
{
	struct hrtimer *timer = &dl_se->dl_timer;

	hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
	timer->function = dl_task_timer;
}

1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091
/*
 * During the activation, CBS checks if it can reuse the current task's
 * runtime and period. If the deadline of the task is in the past, CBS
 * cannot use the runtime, and so it replenishes the task. This rule
 * works fine for implicit deadline tasks (deadline == period), and the
 * CBS was designed for implicit deadline tasks. However, a task with
 * constrained deadline (deadine < period) might be awakened after the
 * deadline, but before the next period. In this case, replenishing the
 * task would allow it to run for runtime / deadline. As in this case
 * deadline < period, CBS enables a task to run for more than the
 * runtime / period. In a very loaded system, this can cause a domino
 * effect, making other tasks miss their deadlines.
 *
 * To avoid this problem, in the activation of a constrained deadline
 * task after the deadline but before the next period, throttle the
 * task and set the replenishing timer to the begin of the next period,
 * unless it is boosted.
 */
static inline void dl_check_constrained_dl(struct sched_dl_entity *dl_se)
{
	struct task_struct *p = dl_task_of(dl_se);
	struct rq *rq = rq_of_dl_rq(dl_rq_of_se(dl_se));

	if (dl_time_before(dl_se->deadline, rq_clock(rq)) &&
	    dl_time_before(rq_clock(rq), dl_next_period(dl_se))) {
		if (unlikely(dl_se->dl_boosted || !start_dl_timer(p)))
			return;
		dl_se->dl_throttled = 1;
1092 1093
		if (dl_se->runtime > 0)
			dl_se->runtime = 0;
1094 1095 1096
	}
}

1097
static
1098
int dl_runtime_exceeded(struct sched_dl_entity *dl_se)
1099
{
1100
	return (dl_se->runtime <= 0);
1101 1102
}

1103 1104
extern bool sched_rt_bandwidth_account(struct rt_rq *rt_rq);

1105 1106 1107
/*
 * This function implements the GRUB accounting rule:
 * according to the GRUB reclaiming algorithm, the runtime is
1108 1109 1110 1111 1112 1113 1114
 * not decreased as "dq = -dt", but as
 * "dq = -max{u / Umax, (1 - Uinact - Uextra)} dt",
 * where u is the utilization of the task, Umax is the maximum reclaimable
 * utilization, Uinact is the (per-runqueue) inactive utilization, computed
 * as the difference between the "total runqueue utilization" and the
 * runqueue active utilization, and Uextra is the (per runqueue) extra
 * reclaimable utilization.
1115
 * Since rq->dl.running_bw and rq->dl.this_bw contain utilizations
1116 1117 1118 1119 1120 1121 1122
 * multiplied by 2^BW_SHIFT, the result has to be shifted right by
 * BW_SHIFT.
 * Since rq->dl.bw_ratio contains 1 / Umax multipled by 2^RATIO_SHIFT,
 * dl_bw is multiped by rq->dl.bw_ratio and shifted right by RATIO_SHIFT.
 * Since delta is a 64 bit variable, to have an overflow its value
 * should be larger than 2^(64 - 20 - 8), which is more than 64 seconds.
 * So, overflow is not an issue here.
1123
 */
1124
u64 grub_reclaim(u64 delta, struct rq *rq, struct sched_dl_entity *dl_se)
1125
{
1126 1127
	u64 u_inact = rq->dl.this_bw - rq->dl.running_bw; /* Utot - Uact */
	u64 u_act;
1128
	u64 u_act_min = (dl_se->dl_bw * rq->dl.bw_ratio) >> RATIO_SHIFT;
1129

1130
	/*
1131 1132 1133 1134 1135 1136
	 * Instead of computing max{u * bw_ratio, (1 - u_inact - u_extra)},
	 * we compare u_inact + rq->dl.extra_bw with
	 * 1 - (u * rq->dl.bw_ratio >> RATIO_SHIFT), because
	 * u_inact + rq->dl.extra_bw can be larger than
	 * 1 * (so, 1 - u_inact - rq->dl.extra_bw would be negative
	 * leading to wrong results)
1137
	 */
1138 1139
	if (u_inact + rq->dl.extra_bw > BW_UNIT - u_act_min)
		u_act = u_act_min;
1140
	else
1141
		u_act = BW_UNIT - u_inact - rq->dl.extra_bw;
1142 1143

	return (delta * u_act) >> BW_SHIFT;
1144 1145
}

1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167
/*
 * Update the current task's runtime statistics (provided it is still
 * a -deadline task and has not been removed from the dl_rq).
 */
static void update_curr_dl(struct rq *rq)
{
	struct task_struct *curr = rq->curr;
	struct sched_dl_entity *dl_se = &curr->dl;
	u64 delta_exec;

	if (!dl_task(curr) || !on_dl_rq(dl_se))
		return;

	/*
	 * Consumed budget is computed considering the time as
	 * observed by schedulable tasks (excluding time spent
	 * in hardirq context, etc.). Deadlines are instead
	 * computed using hard walltime. This seems to be the more
	 * natural solution, but the full ramifications of this
	 * approach need further study.
	 */
	delta_exec = rq_clock_task(rq) - curr->se.exec_start;
1168 1169 1170
	if (unlikely((s64)delta_exec <= 0)) {
		if (unlikely(dl_se->dl_yielded))
			goto throttle;
1171
		return;
1172
	}
1173 1174 1175 1176 1177 1178 1179 1180

	schedstat_set(curr->se.statistics.exec_max,
		      max(curr->se.statistics.exec_max, delta_exec));

	curr->se.sum_exec_runtime += delta_exec;
	account_group_exec_runtime(curr, delta_exec);

	curr->se.exec_start = rq_clock_task(rq);
1181
	cgroup_account_cputime(curr, delta_exec);
1182

1183 1184
	sched_rt_avg_update(rq, delta_exec);

1185 1186 1187
	if (dl_entity_is_special(dl_se))
		return;

1188
	if (unlikely(dl_se->flags & SCHED_FLAG_RECLAIM))
1189
		delta_exec = grub_reclaim(delta_exec, rq, &curr->dl);
1190 1191 1192 1193
	dl_se->runtime -= delta_exec;

throttle:
	if (dl_runtime_exceeded(dl_se) || dl_se->dl_yielded) {
1194
		dl_se->dl_throttled = 1;
1195 1196 1197 1198 1199 1200

		/* If requested, inform the user about runtime overruns. */
		if (dl_runtime_exceeded(dl_se) &&
		    (dl_se->flags & SCHED_FLAG_DL_OVERRUN))
			dl_se->dl_overrun = 1;

1201
		__dequeue_task_dl(rq, curr, 0);
1202
		if (unlikely(dl_se->dl_boosted || !start_dl_timer(curr)))
1203 1204 1205
			enqueue_task_dl(rq, curr, ENQUEUE_REPLENISH);

		if (!is_leftmost(curr, &rq->dl))
1206
			resched_curr(rq);
1207
	}
1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225

	/*
	 * Because -- for now -- we share the rt bandwidth, we need to
	 * account our runtime there too, otherwise actual rt tasks
	 * would be able to exceed the shared quota.
	 *
	 * Account to the root rt group for now.
	 *
	 * The solution we're working towards is having the RT groups scheduled
	 * using deadline servers -- however there's a few nasties to figure
	 * out before that can happen.
	 */
	if (rt_bandwidth_enabled()) {
		struct rt_rq *rt_rq = &rq->rt;

		raw_spin_lock(&rt_rq->rt_runtime_lock);
		/*
		 * We'll let actual RT tasks worry about the overflow here, we
1226 1227
		 * have our own CBS to keep us inline; only account when RT
		 * bandwidth is relevant.
1228
		 */
1229 1230
		if (sched_rt_bandwidth_account(rt_rq))
			rt_rq->rt_time += delta_exec;
1231 1232
		raw_spin_unlock(&rt_rq->rt_runtime_lock);
	}
1233 1234
}

1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246
static enum hrtimer_restart inactive_task_timer(struct hrtimer *timer)
{
	struct sched_dl_entity *dl_se = container_of(timer,
						     struct sched_dl_entity,
						     inactive_timer);
	struct task_struct *p = dl_task_of(dl_se);
	struct rq_flags rf;
	struct rq *rq;

	rq = task_rq_lock(p, &rf);

	if (!dl_task(p) || p->state == TASK_DEAD) {
1247 1248
		struct dl_bw *dl_b = dl_bw_of(task_cpu(p));

1249
		if (p->state == TASK_DEAD && dl_se->dl_non_contending) {
1250 1251
			sub_running_bw(&p->dl, dl_rq_of_se(&p->dl));
			sub_rq_bw(&p->dl, dl_rq_of_se(&p->dl));
1252 1253
			dl_se->dl_non_contending = 0;
		}
1254 1255

		raw_spin_lock(&dl_b->lock);
1256
		__dl_sub(dl_b, p->dl.dl_bw, dl_bw_cpus(task_cpu(p)));
1257
		raw_spin_unlock(&dl_b->lock);
1258 1259 1260 1261 1262 1263 1264 1265 1266 1267
		__dl_clear_params(p);

		goto unlock;
	}
	if (dl_se->dl_non_contending == 0)
		goto unlock;

	sched_clock_tick();
	update_rq_clock(rq);

1268
	sub_running_bw(dl_se, &rq->dl);
1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284
	dl_se->dl_non_contending = 0;
unlock:
	task_rq_unlock(rq, p, &rf);
	put_task_struct(p);

	return HRTIMER_NORESTART;
}

void init_dl_inactive_task_timer(struct sched_dl_entity *dl_se)
{
	struct hrtimer *timer = &dl_se->inactive_timer;

	hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
	timer->function = inactive_task_timer;
}

1285 1286 1287 1288 1289 1290 1291 1292 1293
#ifdef CONFIG_SMP

static void inc_dl_deadline(struct dl_rq *dl_rq, u64 deadline)
{
	struct rq *rq = rq_of_dl_rq(dl_rq);

	if (dl_rq->earliest_dl.curr == 0 ||
	    dl_time_before(deadline, dl_rq->earliest_dl.curr)) {
		dl_rq->earliest_dl.curr = deadline;
1294
		cpudl_set(&rq->rd->cpudl, rq->cpu, deadline);
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308
	}
}

static void dec_dl_deadline(struct dl_rq *dl_rq, u64 deadline)
{
	struct rq *rq = rq_of_dl_rq(dl_rq);

	/*
	 * Since we may have removed our earliest (and/or next earliest)
	 * task we must recompute them.
	 */
	if (!dl_rq->dl_nr_running) {
		dl_rq->earliest_dl.curr = 0;
		dl_rq->earliest_dl.next = 0;
1309
		cpudl_clear(&rq->rd->cpudl, rq->cpu);
1310
	} else {
1311
		struct rb_node *leftmost = dl_rq->root.rb_leftmost;
1312 1313 1314 1315
		struct sched_dl_entity *entry;

		entry = rb_entry(leftmost, struct sched_dl_entity, rb_node);
		dl_rq->earliest_dl.curr = entry->deadline;
1316
		cpudl_set(&rq->rd->cpudl, rq->cpu, entry->deadline);
1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334
	}
}

#else

static inline void inc_dl_deadline(struct dl_rq *dl_rq, u64 deadline) {}
static inline void dec_dl_deadline(struct dl_rq *dl_rq, u64 deadline) {}

#endif /* CONFIG_SMP */

static inline
void inc_dl_tasks(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
	int prio = dl_task_of(dl_se)->prio;
	u64 deadline = dl_se->deadline;

	WARN_ON(!dl_prio(prio));
	dl_rq->dl_nr_running++;
1335
	add_nr_running(rq_of_dl_rq(dl_rq), 1);
1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348

	inc_dl_deadline(dl_rq, deadline);
	inc_dl_migration(dl_se, dl_rq);
}

static inline
void dec_dl_tasks(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
{
	int prio = dl_task_of(dl_se)->prio;

	WARN_ON(!dl_prio(prio));
	WARN_ON(!dl_rq->dl_nr_running);
	dl_rq->dl_nr_running--;
1349
	sub_nr_running(rq_of_dl_rq(dl_rq), 1);
1350 1351 1352 1353 1354

	dec_dl_deadline(dl_rq, dl_se->deadline);
	dec_dl_migration(dl_se, dl_rq);
}

1355 1356 1357
static void __enqueue_dl_entity(struct sched_dl_entity *dl_se)
{
	struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
1358
	struct rb_node **link = &dl_rq->root.rb_root.rb_node;
1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376
	struct rb_node *parent = NULL;
	struct sched_dl_entity *entry;
	int leftmost = 1;

	BUG_ON(!RB_EMPTY_NODE(&dl_se->rb_node));

	while (*link) {
		parent = *link;
		entry = rb_entry(parent, struct sched_dl_entity, rb_node);
		if (dl_time_before(dl_se->deadline, entry->deadline))
			link = &parent->rb_left;
		else {
			link = &parent->rb_right;
			leftmost = 0;
		}
	}

	rb_link_node(&dl_se->rb_node, parent, link);
1377
	rb_insert_color_cached(&dl_se->rb_node, &dl_rq->root, leftmost);
1378

1379
	inc_dl_tasks(dl_se, dl_rq);
1380 1381 1382 1383 1384 1385 1386 1387 1388
}

static void __dequeue_dl_entity(struct sched_dl_entity *dl_se)
{
	struct dl_rq *dl_rq = dl_rq_of_se(dl_se);

	if (RB_EMPTY_NODE(&dl_se->rb_node))
		return;

1389
	rb_erase_cached(&dl_se->rb_node, &dl_rq->root);
1390 1391
	RB_CLEAR_NODE(&dl_se->rb_node);

1392
	dec_dl_tasks(dl_se, dl_rq);
1393 1394 1395
}

static void
1396 1397
enqueue_dl_entity(struct sched_dl_entity *dl_se,
		  struct sched_dl_entity *pi_se, int flags)
1398 1399 1400 1401 1402 1403 1404 1405
{
	BUG_ON(on_dl_rq(dl_se));

	/*
	 * If this is a wakeup or a new instance, the scheduling
	 * parameters of the task might need updating. Otherwise,
	 * we want a replenishment of its runtime.
	 */
1406
	if (flags & ENQUEUE_WAKEUP) {
1407
		task_contending(dl_se, flags);
1408
		update_dl_entity(dl_se, pi_se);
1409
	} else if (flags & ENQUEUE_REPLENISH) {
1410
		replenish_dl_entity(dl_se, pi_se);
1411 1412 1413 1414
	} else if ((flags & ENQUEUE_RESTORE) &&
		  dl_time_before(dl_se->deadline,
				 rq_clock(rq_of_dl_rq(dl_rq_of_se(dl_se))))) {
		setup_new_dl_entity(dl_se);
1415
	}
1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426

	__enqueue_dl_entity(dl_se);
}

static void dequeue_dl_entity(struct sched_dl_entity *dl_se)
{
	__dequeue_dl_entity(dl_se);
}

static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags)
{
1427 1428 1429 1430
	struct task_struct *pi_task = rt_mutex_get_top_task(p);
	struct sched_dl_entity *pi_se = &p->dl;

	/*
1431 1432 1433 1434 1435 1436
	 * Use the scheduling parameters of the top pi-waiter task if:
	 * - we have a top pi-waiter which is a SCHED_DEADLINE task AND
	 * - our dl_boosted is set (i.e. the pi-waiter's (absolute) deadline is
	 *   smaller than our deadline OR we are a !SCHED_DEADLINE task getting
	 *   boosted due to a SCHED_DEADLINE pi-waiter).
	 * Otherwise we keep our runtime and deadline.
1437
	 */
1438
	if (pi_task && dl_prio(pi_task->normal_prio) && p->dl.dl_boosted) {
1439
		pi_se = &pi_task->dl;
1440 1441 1442
	} else if (!dl_prio(p->normal_prio)) {
		/*
		 * Special case in which we have a !SCHED_DEADLINE task
1443
		 * that is going to be deboosted, but exceeds its
1444 1445 1446 1447 1448 1449 1450
		 * runtime while doing so. No point in replenishing
		 * it, as it's going to return back to its original
		 * scheduling class after this.
		 */
		BUG_ON(!p->dl.dl_boosted || flags != ENQUEUE_REPLENISH);
		return;
	}
1451

1452 1453 1454 1455 1456 1457
	/*
	 * Check if a constrained deadline task was activated
	 * after the deadline but before the next period.
	 * If that is the case, the task will be throttled and
	 * the replenishment timer will be set to the next period.
	 */
1458
	if (!p->dl.dl_throttled && !dl_is_implicit(&p->dl))
1459 1460
		dl_check_constrained_dl(&p->dl);

1461
	if (p->on_rq == TASK_ON_RQ_MIGRATING || flags & ENQUEUE_RESTORE) {
1462 1463
		add_rq_bw(&p->dl, &rq->dl);
		add_running_bw(&p->dl, &rq->dl);
1464
	}
1465

1466
	/*
1467
	 * If p is throttled, we do not enqueue it. In fact, if it exhausted
1468 1469 1470
	 * its budget it needs a replenishment and, since it now is on
	 * its rq, the bandwidth timer callback (which clearly has not
	 * run yet) will take care of this.
1471 1472 1473 1474 1475 1476
	 * However, the active utilization does not depend on the fact
	 * that the task is on the runqueue or not (but depends on the
	 * task's state - in GRUB parlance, "inactive" vs "active contending").
	 * In other words, even if a task is throttled its utilization must
	 * be counted in the active utilization; hence, we need to call
	 * add_running_bw().
1477
	 */
1478
	if (p->dl.dl_throttled && !(flags & ENQUEUE_REPLENISH)) {
1479
		if (flags & ENQUEUE_WAKEUP)
1480
			task_contending(&p->dl, flags);
1481

1482
		return;
1483
	}
1484

1485
	enqueue_dl_entity(&p->dl, pi_se, flags);
1486

1487
	if (!task_current(rq, p) && p->nr_cpus_allowed > 1)
1488
		enqueue_pushable_dl_task(rq, p);
1489 1490 1491 1492 1493
}

static void __dequeue_task_dl(struct rq *rq, struct task_struct *p, int flags)
{
	dequeue_dl_entity(&p->dl);
1494
	dequeue_pushable_dl_task(rq, p);
1495 1496 1497 1498 1499 1500
}

static void dequeue_task_dl(struct rq *rq, struct task_struct *p, int flags)
{
	update_curr_dl(rq);
	__dequeue_task_dl(rq, p, flags);
1501

1502
	if (p->on_rq == TASK_ON_RQ_MIGRATING || flags & DEQUEUE_SAVE) {
1503 1504
		sub_running_bw(&p->dl, &rq->dl);
		sub_rq_bw(&p->dl, &rq->dl);
1505
	}
1506 1507

	/*
1508 1509
	 * This check allows to start the inactive timer (or to immediately
	 * decrease the active utilization, if needed) in two cases:
1510 1511 1512 1513 1514 1515 1516
	 * when the task blocks and when it is terminating
	 * (p->state == TASK_DEAD). We can handle the two cases in the same
	 * way, because from GRUB's point of view the same thing is happening
	 * (the task moves from "active contending" to "active non contending"
	 * or "inactive")
	 */
	if (flags & DEQUEUE_SLEEP)
1517
		task_non_contending(p);
1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535
}

/*
 * Yield task semantic for -deadline tasks is:
 *
 *   get off from the CPU until our next instance, with
 *   a new runtime. This is of little use now, since we
 *   don't have a bandwidth reclaiming mechanism. Anyway,
 *   bandwidth reclaiming is planned for the future, and
 *   yield_task_dl will indicate that some spare budget
 *   is available for other task instances to use it.
 */
static void yield_task_dl(struct rq *rq)
{
	/*
	 * We make the task go to sleep until its current deadline by
	 * forcing its runtime to zero. This way, update_curr_dl() stops
	 * it and the bandwidth timer will wake it up and will give it
1536
	 * new scheduling parameters (thanks to dl_yielded=1).
1537
	 */
1538 1539
	rq->curr->dl.dl_yielded = 1;

1540
	update_rq_clock(rq);
1541
	update_curr_dl(rq);
1542 1543 1544 1545 1546 1547
	/*
	 * Tell update_rq_clock() that we've just updated,
	 * so we don't do microscopic update in schedule()
	 * and double the fastpath cost.
	 */
	rq_clock_skip_update(rq, true);
1548 1549
}

1550 1551 1552 1553 1554 1555 1556 1557 1558 1559
#ifdef CONFIG_SMP

static int find_later_rq(struct task_struct *task);

static int
select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags)
{
	struct task_struct *curr;
	struct rq *rq;

1560
	if (sd_flag != SD_BALANCE_WAKE)
1561 1562 1563 1564 1565
		goto out;

	rq = cpu_rq(cpu);

	rcu_read_lock();
1566
	curr = READ_ONCE(rq->curr); /* unlocked access */
1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577

	/*
	 * If we are dealing with a -deadline task, we must
	 * decide where to wake it up.
	 * If it has a later deadline and the current task
	 * on this rq can't move (provided the waking task
	 * can!) we prefer to send it somewhere else. On the
	 * other hand, if it has a shorter deadline, we
	 * try to make it stay here, it might be important.
	 */
	if (unlikely(dl_task(curr)) &&
1578
	    (curr->nr_cpus_allowed < 2 ||
1579
	     !dl_entity_preempt(&p->dl, &curr->dl)) &&
1580
	    (p->nr_cpus_allowed > 1)) {
1581 1582
		int target = find_later_rq(p);

1583
		if (target != -1 &&
1584 1585 1586
				(dl_time_before(p->dl.deadline,
					cpu_rq(target)->dl.earliest_dl.curr) ||
				(cpu_rq(target)->dl.dl_nr_running == 0)))
1587 1588 1589 1590 1591 1592 1593 1594
			cpu = target;
	}
	rcu_read_unlock();

out:
	return cpu;
}

1595 1596 1597 1598
static void migrate_task_rq_dl(struct task_struct *p)
{
	struct rq *rq;

1599
	if (p->state != TASK_WAKING)
1600 1601 1602 1603 1604 1605 1606 1607 1608
		return;

	rq = task_rq(p);
	/*
	 * Since p->state == TASK_WAKING, set_task_cpu() has been called
	 * from try_to_wake_up(). Hence, p->pi_lock is locked, but
	 * rq->lock is not... So, lock it
	 */
	raw_spin_lock(&rq->lock);
1609
	if (p->dl.dl_non_contending) {
1610
		sub_running_bw(&p->dl, &rq->dl);
1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621
		p->dl.dl_non_contending = 0;
		/*
		 * If the timer handler is currently running and the
		 * timer cannot be cancelled, inactive_task_timer()
		 * will see that dl_not_contending is not set, and
		 * will not touch the rq's active utilization,
		 * so we are still safe.
		 */
		if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1)
			put_task_struct(p);
	}
1622
	sub_rq_bw(&p->dl, &rq->dl);
1623 1624 1625
	raw_spin_unlock(&rq->lock);
}

1626 1627 1628 1629 1630 1631
static void check_preempt_equal_dl(struct rq *rq, struct task_struct *p)
{
	/*
	 * Current can't be migrated, useless to reschedule,
	 * let's hope p can move out.
	 */
1632
	if (rq->curr->nr_cpus_allowed == 1 ||
1633
	    !cpudl_find(&rq->rd->cpudl, rq->curr, NULL))
1634 1635 1636 1637 1638 1639
		return;

	/*
	 * p is migratable, so let's not schedule it and
	 * see if it is pushed or pulled somewhere else.
	 */
1640
	if (p->nr_cpus_allowed != 1 &&
1641
	    cpudl_find(&rq->rd->cpudl, p, NULL))
1642 1643
		return;

1644
	resched_curr(rq);
1645 1646 1647 1648
}

#endif /* CONFIG_SMP */

1649 1650 1651 1652 1653 1654 1655
/*
 * Only called when both the current and waking task are -deadline
 * tasks.
 */
static void check_preempt_curr_dl(struct rq *rq, struct task_struct *p,
				  int flags)
{
1656
	if (dl_entity_preempt(&p->dl, &rq->curr->dl)) {
1657
		resched_curr(rq);
1658 1659 1660 1661 1662 1663 1664 1665
		return;
	}

#ifdef CONFIG_SMP
	/*
	 * In the unlikely case current and p have the same deadline
	 * let us try to decide what's the best thing to do...
	 */
1666 1667
	if ((p->dl.deadline == rq->curr->dl.deadline) &&
	    !test_tsk_need_resched(rq->curr))
1668 1669
		check_preempt_equal_dl(rq, p);
#endif /* CONFIG_SMP */
1670 1671 1672 1673 1674
}

#ifdef CONFIG_SCHED_HRTICK
static void start_hrtick_dl(struct rq *rq, struct task_struct *p)
{
1675
	hrtick_start(rq, p->dl.runtime);
1676
}
1677 1678 1679 1680
#else /* !CONFIG_SCHED_HRTICK */
static void start_hrtick_dl(struct rq *rq, struct task_struct *p)
{
}
1681 1682 1683 1684 1685
#endif

static struct sched_dl_entity *pick_next_dl_entity(struct rq *rq,
						   struct dl_rq *dl_rq)
{
1686
	struct rb_node *left = rb_first_cached(&dl_rq->root);
1687 1688 1689 1690 1691 1692 1693

	if (!left)
		return NULL;

	return rb_entry(left, struct sched_dl_entity, rb_node);
}

1694
static struct task_struct *
1695
pick_next_task_dl(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
1696 1697 1698 1699 1700 1701 1702
{
	struct sched_dl_entity *dl_se;
	struct task_struct *p;
	struct dl_rq *dl_rq;

	dl_rq = &rq->dl;

1703
	if (need_pull_dl_task(rq, prev)) {
1704 1705 1706 1707 1708 1709
		/*
		 * This is OK, because current is on_cpu, which avoids it being
		 * picked for load-balance and preemption/IRQs are still
		 * disabled avoiding further scheduler activity on it and we're
		 * being very careful to re-start the picking loop.
		 */
1710
		rq_unpin_lock(rq, rf);
1711
		pull_dl_task(rq);
1712
		rq_repin_lock(rq, rf);
1713
		/*
1714
		 * pull_dl_task() can drop (and re-acquire) rq->lock; this
1715 1716 1717
		 * means a stop task can slip in, in which case we need to
		 * re-start task selection.
		 */
1718
		if (rq->stop && task_on_rq_queued(rq->stop))
1719 1720 1721
			return RETRY_TASK;
	}

1722 1723 1724 1725 1726 1727
	/*
	 * When prev is DL, we may throttle it in put_prev_task().
	 * So, we update time before we check for dl_nr_running.
	 */
	if (prev->sched_class == &dl_sched_class)
		update_curr_dl(rq);
1728

1729 1730 1731
	if (unlikely(!dl_rq->dl_nr_running))
		return NULL;

1732
	put_prev_task(rq, prev);
1733

1734 1735 1736 1737 1738
	dl_se = pick_next_dl_entity(rq, dl_rq);
	BUG_ON(!dl_se);

	p = dl_task_of(dl_se);
	p->se.exec_start = rq_clock_task(rq);
1739 1740

	/* Running task will never be pushed. */
1741
       dequeue_pushable_dl_task(rq, p);
1742

1743 1744
	if (hrtick_enabled(rq))
		start_hrtick_dl(rq, p);
1745

1746
	queue_push_tasks(rq);
1747

1748 1749 1750 1751 1752 1753
	return p;
}

static void put_prev_task_dl(struct rq *rq, struct task_struct *p)
{
	update_curr_dl(rq);
1754

1755
	if (on_dl_rq(&p->dl) && p->nr_cpus_allowed > 1)
1756
		enqueue_pushable_dl_task(rq, p);
1757 1758 1759 1760 1761 1762
}

static void task_tick_dl(struct rq *rq, struct task_struct *p, int queued)
{
	update_curr_dl(rq);

1763 1764 1765 1766 1767 1768 1769
	/*
	 * Even when we have runtime, update_curr_dl() might have resulted in us
	 * not being the leftmost task anymore. In that case NEED_RESCHED will
	 * be set and schedule() will start a new hrtick for the next task.
	 */
	if (hrtick_enabled(rq) && queued && p->dl.runtime > 0 &&
	    is_leftmost(p, &rq->dl))
1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785
		start_hrtick_dl(rq, p);
}

static void task_fork_dl(struct task_struct *p)
{
	/*
	 * SCHED_DEADLINE tasks cannot fork and this is achieved through
	 * sched_fork()
	 */
}

static void set_curr_task_dl(struct rq *rq)
{
	struct task_struct *p = rq->curr;

	p->se.exec_start = rq_clock_task(rq);
1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798

	/* You can't push away the running task */
	dequeue_pushable_dl_task(rq, p);
}

#ifdef CONFIG_SMP

/* Only try algorithms three times */
#define DL_MAX_TRIES 3

static int pick_dl_task(struct rq *rq, struct task_struct *p, int cpu)
{
	if (!task_running(rq, p) &&
1799
	    cpumask_test_cpu(cpu, &p->cpus_allowed))
1800 1801 1802 1803
		return 1;
	return 0;
}

1804 1805 1806 1807 1808 1809
/*
 * Return the earliest pushable rq's task, which is suitable to be executed
 * on the CPU, NULL otherwise:
 */
static struct task_struct *pick_earliest_pushable_dl_task(struct rq *rq, int cpu)
{
1810
	struct rb_node *next_node = rq->dl.pushable_dl_tasks_root.rb_leftmost;
1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829
	struct task_struct *p = NULL;

	if (!has_pushable_dl_tasks(rq))
		return NULL;

next_node:
	if (next_node) {
		p = rb_entry(next_node, struct task_struct, pushable_dl_tasks);

		if (pick_dl_task(rq, p, cpu))
			return p;

		next_node = rb_next(next_node);
		goto next_node;
	}

	return NULL;
}

1830 1831 1832 1833 1834
static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask_dl);

static int find_later_rq(struct task_struct *task)
{
	struct sched_domain *sd;
1835
	struct cpumask *later_mask = this_cpu_cpumask_var_ptr(local_cpu_mask_dl);
1836
	int this_cpu = smp_processor_id();
1837
	int cpu = task_cpu(task);
1838 1839 1840 1841 1842

	/* Make sure the mask is initialized first */
	if (unlikely(!later_mask))
		return -1;

1843
	if (task->nr_cpus_allowed == 1)
1844 1845
		return -1;

1846 1847 1848 1849
	/*
	 * We have to consider system topology and task affinity
	 * first, then we can look for a suitable cpu.
	 */
1850
	if (!cpudl_find(&task_rq(task)->rd->cpudl, task, later_mask))
1851 1852 1853
		return -1;

	/*
1854 1855 1856 1857
	 * If we are here, some targets have been found, including
	 * the most suitable which is, among the runqueues where the
	 * current tasks have later deadlines than the task's one, the
	 * rq with the latest possible one.
1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876
	 *
	 * Now we check how well this matches with task's
	 * affinity and system topology.
	 *
	 * The last cpu where the task run is our first
	 * guess, since it is most likely cache-hot there.
	 */
	if (cpumask_test_cpu(cpu, later_mask))
		return cpu;
	/*
	 * Check if this_cpu is to be skipped (i.e., it is
	 * not in the mask) or not.
	 */
	if (!cpumask_test_cpu(this_cpu, later_mask))
		this_cpu = -1;

	rcu_read_lock();
	for_each_domain(cpu, sd) {
		if (sd->flags & SD_WAKE_AFFINE) {
1877
			int best_cpu;
1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888

			/*
			 * If possible, preempting this_cpu is
			 * cheaper than migrating.
			 */
			if (this_cpu != -1 &&
			    cpumask_test_cpu(this_cpu, sched_domain_span(sd))) {
				rcu_read_unlock();
				return this_cpu;
			}

1889 1890
			best_cpu = cpumask_first_and(later_mask,
							sched_domain_span(sd));
1891
			/*
1892 1893 1894 1895
			 * Last chance: if a cpu being in both later_mask
			 * and current sd span is valid, that becomes our
			 * choice. Of course, the latest possible cpu is
			 * already under consideration through later_mask.
1896
			 */
1897
			if (best_cpu < nr_cpu_ids) {
1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933
				rcu_read_unlock();
				return best_cpu;
			}
		}
	}
	rcu_read_unlock();

	/*
	 * At this point, all our guesses failed, we just return
	 * 'something', and let the caller sort the things out.
	 */
	if (this_cpu != -1)
		return this_cpu;

	cpu = cpumask_any(later_mask);
	if (cpu < nr_cpu_ids)
		return cpu;

	return -1;
}

/* Locks the rq it finds */
static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq)
{
	struct rq *later_rq = NULL;
	int tries;
	int cpu;

	for (tries = 0; tries < DL_MAX_TRIES; tries++) {
		cpu = find_later_rq(task);

		if ((cpu == -1) || (cpu == rq->cpu))
			break;

		later_rq = cpu_rq(cpu);

1934 1935
		if (later_rq->dl.dl_nr_running &&
		    !dl_time_before(task->dl.deadline,
1936 1937 1938 1939 1940 1941 1942 1943 1944 1945
					later_rq->dl.earliest_dl.curr)) {
			/*
			 * Target rq has tasks of equal or earlier deadline,
			 * retrying does not release any lock and is unlikely
			 * to yield a different result.
			 */
			later_rq = NULL;
			break;
		}

1946 1947 1948
		/* Retry if something changed. */
		if (double_lock_balance(rq, later_rq)) {
			if (unlikely(task_rq(task) != rq ||
1949
				     !cpumask_test_cpu(later_rq->cpu, &task->cpus_allowed) ||
1950
				     task_running(rq, task) ||
1951
				     !dl_task(task) ||
1952
				     !task_on_rq_queued(task))) {
1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983
				double_unlock_balance(rq, later_rq);
				later_rq = NULL;
				break;
			}
		}

		/*
		 * If the rq we found has no -deadline task, or
		 * its earliest one has a later deadline than our
		 * task, the rq is a good one.
		 */
		if (!later_rq->dl.dl_nr_running ||
		    dl_time_before(task->dl.deadline,
				   later_rq->dl.earliest_dl.curr))
			break;

		/* Otherwise we try again. */
		double_unlock_balance(rq, later_rq);
		later_rq = NULL;
	}

	return later_rq;
}

static struct task_struct *pick_next_pushable_dl_task(struct rq *rq)
{
	struct task_struct *p;

	if (!has_pushable_dl_tasks(rq))
		return NULL;

1984
	p = rb_entry(rq->dl.pushable_dl_tasks_root.rb_leftmost,
1985 1986 1987 1988
		     struct task_struct, pushable_dl_tasks);

	BUG_ON(rq->cpu != task_cpu(p));
	BUG_ON(task_current(rq, p));
1989
	BUG_ON(p->nr_cpus_allowed <= 1);
1990

1991
	BUG_ON(!task_on_rq_queued(p));
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
	BUG_ON(!dl_task(p));

	return p;
}

/*
 * See if the non running -deadline tasks on this rq
 * can be sent to some other CPU where they can preempt
 * and start executing.
 */
static int push_dl_task(struct rq *rq)
{
	struct task_struct *next_task;
	struct rq *later_rq;
2006
	int ret = 0;
2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027

	if (!rq->dl.overloaded)
		return 0;

	next_task = pick_next_pushable_dl_task(rq);
	if (!next_task)
		return 0;

retry:
	if (unlikely(next_task == rq->curr)) {
		WARN_ON(1);
		return 0;
	}

	/*
	 * If next_task preempts rq->curr, and rq->curr
	 * can move away, it makes sense to just reschedule
	 * without going further in pushing next_task.
	 */
	if (dl_task(rq->curr) &&
	    dl_time_before(next_task->dl.deadline, rq->curr->dl.deadline) &&
2028
	    rq->curr->nr_cpus_allowed > 1) {
2029
		resched_curr(rq);
2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046
		return 0;
	}

	/* We might release rq lock */
	get_task_struct(next_task);

	/* Will lock the rq it'll find */
	later_rq = find_lock_later_rq(next_task, rq);
	if (!later_rq) {
		struct task_struct *task;

		/*
		 * We must check all this again, since
		 * find_lock_later_rq releases rq->lock and it is
		 * then possible that next_task has migrated.
		 */
		task = pick_next_pushable_dl_task(rq);
2047
		if (task == next_task) {
2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064
			/*
			 * The task is still there. We don't try
			 * again, some other cpu will pull it when ready.
			 */
			goto out;
		}

		if (!task)
			/* No more tasks */
			goto out;

		put_task_struct(next_task);
		next_task = task;
		goto retry;
	}

	deactivate_task(rq, next_task, 0);
2065 2066
	sub_running_bw(&next_task->dl, &rq->dl);
	sub_rq_bw(&next_task->dl, &rq->dl);
2067
	set_task_cpu(next_task, later_rq->cpu);
2068 2069
	add_rq_bw(&next_task->dl, &later_rq->dl);
	add_running_bw(&next_task->dl, &later_rq->dl);
2070
	activate_task(later_rq, next_task, 0);
2071
	ret = 1;
2072

2073
	resched_curr(later_rq);
2074 2075 2076 2077 2078 2079

	double_unlock_balance(rq, later_rq);

out:
	put_task_struct(next_task);

2080
	return ret;
2081 2082 2083 2084
}

static void push_dl_tasks(struct rq *rq)
{
2085
	/* push_dl_task() will return true if it moved a -deadline task */
2086 2087
	while (push_dl_task(rq))
		;
2088 2089
}

2090
static void pull_dl_task(struct rq *this_rq)
2091
{
2092
	int this_cpu = this_rq->cpu, cpu;
2093
	struct task_struct *p;
2094
	bool resched = false;
2095 2096 2097 2098
	struct rq *src_rq;
	u64 dmin = LONG_MAX;

	if (likely(!dl_overloaded(this_rq)))
2099
		return;
2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131

	/*
	 * Match the barrier from dl_set_overloaded; this guarantees that if we
	 * see overloaded we must also see the dlo_mask bit.
	 */
	smp_rmb();

	for_each_cpu(cpu, this_rq->rd->dlo_mask) {
		if (this_cpu == cpu)
			continue;

		src_rq = cpu_rq(cpu);

		/*
		 * It looks racy, abd it is! However, as in sched_rt.c,
		 * we are fine with this.
		 */
		if (this_rq->dl.dl_nr_running &&
		    dl_time_before(this_rq->dl.earliest_dl.curr,
				   src_rq->dl.earliest_dl.next))
			continue;

		/* Might drop this_rq->lock */
		double_lock_balance(this_rq, src_rq);

		/*
		 * If there are no more pullable tasks on the
		 * rq, we're done with it.
		 */
		if (src_rq->dl.dl_nr_running <= 1)
			goto skip;

2132
		p = pick_earliest_pushable_dl_task(src_rq, this_cpu);
2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143

		/*
		 * We found a task to be pulled if:
		 *  - it preempts our current (if there's one),
		 *  - it will preempt the last one we pulled (if any).
		 */
		if (p && dl_time_before(p->dl.deadline, dmin) &&
		    (!this_rq->dl.dl_nr_running ||
		     dl_time_before(p->dl.deadline,
				    this_rq->dl.earliest_dl.curr))) {
			WARN_ON(p == src_rq->curr);
2144
			WARN_ON(!task_on_rq_queued(p));
2145 2146 2147 2148 2149 2150 2151 2152 2153

			/*
			 * Then we pull iff p has actually an earlier
			 * deadline than the current task of its runqueue.
			 */
			if (dl_time_before(p->dl.deadline,
					   src_rq->curr->dl.deadline))
				goto skip;

2154
			resched = true;
2155 2156

			deactivate_task(src_rq, p, 0);
2157 2158
			sub_running_bw(&p->dl, &src_rq->dl);
			sub_rq_bw(&p->dl, &src_rq->dl);
2159
			set_task_cpu(p, this_cpu);
2160 2161
			add_rq_bw(&p->dl, &this_rq->dl);
			add_running_bw(&p->dl, &this_rq->dl);
2162 2163 2164 2165 2166 2167 2168 2169 2170
			activate_task(this_rq, p, 0);
			dmin = p->dl.deadline;

			/* Is there any other task even earlier? */
		}
skip:
		double_unlock_balance(this_rq, src_rq);
	}

2171 2172
	if (resched)
		resched_curr(this_rq);
2173 2174 2175 2176 2177 2178 2179 2180 2181 2182
}

/*
 * Since the task is not running and a reschedule is not going to happen
 * anytime soon on its runqueue, we try pushing it away now.
 */
static void task_woken_dl(struct rq *rq, struct task_struct *p)
{
	if (!task_running(rq, p) &&
	    !test_tsk_need_resched(rq->curr) &&
2183
	    p->nr_cpus_allowed > 1 &&
2184
	    dl_task(rq->curr) &&
2185
	    (rq->curr->nr_cpus_allowed < 2 ||
2186
	     !dl_entity_preempt(&p->dl, &rq->curr->dl))) {
2187 2188 2189 2190 2191 2192 2193
		push_dl_tasks(rq);
	}
}

static void set_cpus_allowed_dl(struct task_struct *p,
				const struct cpumask *new_mask)
{
2194
	struct root_domain *src_rd;
2195
	struct rq *rq;
2196 2197 2198

	BUG_ON(!dl_task(p));

2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216
	rq = task_rq(p);
	src_rd = rq->rd;
	/*
	 * Migrating a SCHED_DEADLINE task between exclusive
	 * cpusets (different root_domains) entails a bandwidth
	 * update. We already made space for us in the destination
	 * domain (see cpuset_can_attach()).
	 */
	if (!cpumask_intersects(src_rd->span, new_mask)) {
		struct dl_bw *src_dl_b;

		src_dl_b = dl_bw_of(cpu_of(rq));
		/*
		 * We now free resources of the root_domain we are migrating
		 * off. In the worst case, sched_setattr() may temporary fail
		 * until we complete the update.
		 */
		raw_spin_lock(&src_dl_b->lock);
2217
		__dl_sub(src_dl_b, p->dl.dl_bw, dl_bw_cpus(task_cpu(p)));
2218 2219 2220
		raw_spin_unlock(&src_dl_b->lock);
	}

2221
	set_cpus_allowed_common(p, new_mask);
2222 2223 2224 2225 2226 2227 2228
}

/* Assumes rq->lock is held */
static void rq_online_dl(struct rq *rq)
{
	if (rq->dl.overloaded)
		dl_set_overload(rq);
2229

2230
	cpudl_set_freecpu(&rq->rd->cpudl, rq->cpu);
2231
	if (rq->dl.dl_nr_running > 0)
2232
		cpudl_set(&rq->rd->cpudl, rq->cpu, rq->dl.earliest_dl.curr);
2233 2234 2235 2236 2237 2238 2239
}

/* Assumes rq->lock is held */
static void rq_offline_dl(struct rq *rq)
{
	if (rq->dl.overloaded)
		dl_clear_overload(rq);
2240

2241
	cpudl_clear(&rq->rd->cpudl, rq->cpu);
2242
	cpudl_clear_freecpu(&rq->rd->cpudl, rq->cpu);
2243 2244
}

2245
void __init init_sched_dl_class(void)
2246 2247 2248 2249 2250 2251 2252 2253 2254 2255
{
	unsigned int i;

	for_each_possible_cpu(i)
		zalloc_cpumask_var_node(&per_cpu(local_cpu_mask_dl, i),
					GFP_KERNEL, cpu_to_node(i));
}

#endif /* CONFIG_SMP */

2256 2257
static void switched_from_dl(struct rq *rq, struct task_struct *p)
{
2258
	/*
2259 2260 2261 2262 2263 2264
	 * task_non_contending() can start the "inactive timer" (if the 0-lag
	 * time is in the future). If the task switches back to dl before
	 * the "inactive timer" fires, it can continue to consume its current
	 * runtime using its current deadline. If it stays outside of
	 * SCHED_DEADLINE until the 0-lag time passes, inactive_task_timer()
	 * will reset the task parameters.
2265
	 */
2266 2267 2268
	if (task_on_rq_queued(p) && p->dl.dl_runtime)
		task_non_contending(p);

2269
	if (!task_on_rq_queued(p))
2270
		sub_rq_bw(&p->dl, &rq->dl);
2271

2272 2273 2274 2275 2276 2277 2278
	/*
	 * We cannot use inactive_task_timer() to invoke sub_running_bw()
	 * at the 0-lag time, because the task could have been migrated
	 * while SCHED_OTHER in the meanwhile.
	 */
	if (p->dl.dl_non_contending)
		p->dl.dl_non_contending = 0;
2279

2280 2281 2282 2283 2284
	/*
	 * Since this might be the only -deadline task on the rq,
	 * this is the right place to try to pull some other one
	 * from an overloaded cpu, if any.
	 */
2285 2286 2287
	if (!task_on_rq_queued(p) || rq->dl.dl_nr_running)
		return;

2288
	queue_pull_task(rq);
2289 2290
}

2291 2292 2293 2294
/*
 * When switching to -deadline, we may overload the rq, then
 * we try to push someone off, if possible.
 */
2295 2296
static void switched_to_dl(struct rq *rq, struct task_struct *p)
{
2297 2298
	if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1)
		put_task_struct(p);
2299 2300

	/* If p is not queued we will update its parameters at next wakeup. */
2301
	if (!task_on_rq_queued(p)) {
2302
		add_rq_bw(&p->dl, &rq->dl);
2303

2304 2305
		return;
	}
2306

2307
	if (rq->curr != p) {
2308
#ifdef CONFIG_SMP
2309
		if (p->nr_cpus_allowed > 1 && rq->dl.overloaded)
2310
			queue_push_tasks(rq);
2311
#endif
2312 2313 2314 2315
		if (dl_task(rq->curr))
			check_preempt_curr_dl(rq, p, 0);
		else
			resched_curr(rq);
2316 2317 2318
	}
}

2319 2320 2321 2322
/*
 * If the scheduling parameters of a -deadline task changed,
 * a push or pull operation might be needed.
 */
2323 2324 2325
static void prio_changed_dl(struct rq *rq, struct task_struct *p,
			    int oldprio)
{
2326
	if (task_on_rq_queued(p) || rq->curr == p) {
2327
#ifdef CONFIG_SMP
2328 2329 2330 2331 2332 2333 2334
		/*
		 * This might be too much, but unfortunately
		 * we don't have the old deadline value, and
		 * we can't argue if the task is increasing
		 * or lowering its prio, so...
		 */
		if (!rq->dl.overloaded)
2335
			queue_pull_task(rq);
2336 2337 2338 2339 2340 2341

		/*
		 * If we now have a earlier deadline task than p,
		 * then reschedule, provided p is still on this
		 * runqueue.
		 */
2342
		if (dl_time_before(rq->dl.earliest_dl.curr, p->dl.deadline))
2343
			resched_curr(rq);
2344 2345 2346 2347 2348 2349
#else
		/*
		 * Again, we don't know if p has a earlier
		 * or later deadline, so let's blindly set a
		 * (maybe not needed) rescheduling point.
		 */
2350
		resched_curr(rq);
2351
#endif /* CONFIG_SMP */
2352
	}
2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367
}

const struct sched_class dl_sched_class = {
	.next			= &rt_sched_class,
	.enqueue_task		= enqueue_task_dl,
	.dequeue_task		= dequeue_task_dl,
	.yield_task		= yield_task_dl,

	.check_preempt_curr	= check_preempt_curr_dl,

	.pick_next_task		= pick_next_task_dl,
	.put_prev_task		= put_prev_task_dl,

#ifdef CONFIG_SMP
	.select_task_rq		= select_task_rq_dl,
2368
	.migrate_task_rq	= migrate_task_rq_dl,
2369 2370 2371 2372
	.set_cpus_allowed       = set_cpus_allowed_dl,
	.rq_online              = rq_online_dl,
	.rq_offline             = rq_offline_dl,
	.task_woken		= task_woken_dl,
2373 2374 2375 2376 2377 2378 2379 2380 2381
#endif

	.set_curr_task		= set_curr_task_dl,
	.task_tick		= task_tick_dl,
	.task_fork              = task_fork_dl,

	.prio_changed           = prio_changed_dl,
	.switched_from		= switched_from_dl,
	.switched_to		= switched_to_dl,
2382 2383

	.update_curr		= update_curr_dl,
2384
};
2385

2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480
int sched_dl_global_validate(void)
{
	u64 runtime = global_rt_runtime();
	u64 period = global_rt_period();
	u64 new_bw = to_ratio(period, runtime);
	struct dl_bw *dl_b;
	int cpu, ret = 0;
	unsigned long flags;

	/*
	 * Here we want to check the bandwidth not being set to some
	 * value smaller than the currently allocated bandwidth in
	 * any of the root_domains.
	 *
	 * FIXME: Cycling on all the CPUs is overdoing, but simpler than
	 * cycling on root_domains... Discussion on different/better
	 * solutions is welcome!
	 */
	for_each_possible_cpu(cpu) {
		rcu_read_lock_sched();
		dl_b = dl_bw_of(cpu);

		raw_spin_lock_irqsave(&dl_b->lock, flags);
		if (new_bw < dl_b->total_bw)
			ret = -EBUSY;
		raw_spin_unlock_irqrestore(&dl_b->lock, flags);

		rcu_read_unlock_sched();

		if (ret)
			break;
	}

	return ret;
}

void init_dl_rq_bw_ratio(struct dl_rq *dl_rq)
{
	if (global_rt_runtime() == RUNTIME_INF) {
		dl_rq->bw_ratio = 1 << RATIO_SHIFT;
		dl_rq->extra_bw = 1 << BW_SHIFT;
	} else {
		dl_rq->bw_ratio = to_ratio(global_rt_runtime(),
			  global_rt_period()) >> (BW_SHIFT - RATIO_SHIFT);
		dl_rq->extra_bw = to_ratio(global_rt_period(),
						    global_rt_runtime());
	}
}

void sched_dl_do_global(void)
{
	u64 new_bw = -1;
	struct dl_bw *dl_b;
	int cpu;
	unsigned long flags;

	def_dl_bandwidth.dl_period = global_rt_period();
	def_dl_bandwidth.dl_runtime = global_rt_runtime();

	if (global_rt_runtime() != RUNTIME_INF)
		new_bw = to_ratio(global_rt_period(), global_rt_runtime());

	/*
	 * FIXME: As above...
	 */
	for_each_possible_cpu(cpu) {
		rcu_read_lock_sched();
		dl_b = dl_bw_of(cpu);

		raw_spin_lock_irqsave(&dl_b->lock, flags);
		dl_b->bw = new_bw;
		raw_spin_unlock_irqrestore(&dl_b->lock, flags);

		rcu_read_unlock_sched();
		init_dl_rq_bw_ratio(&cpu_rq(cpu)->dl);
	}
}

/*
 * We must be sure that accepting a new task (or allowing changing the
 * parameters of an existing one) is consistent with the bandwidth
 * constraints. If yes, this function also accordingly updates the currently
 * allocated bandwidth to reflect the new situation.
 *
 * This function is called while holding p's rq->lock.
 */
int sched_dl_overflow(struct task_struct *p, int policy,
		      const struct sched_attr *attr)
{
	struct dl_bw *dl_b = dl_bw_of(task_cpu(p));
	u64 period = attr->sched_period ?: attr->sched_deadline;
	u64 runtime = attr->sched_runtime;
	u64 new_bw = dl_policy(policy) ? to_ratio(period, runtime) : 0;
	int cpus, err = -1;

2481 2482 2483
	if (attr->sched_flags & SCHED_FLAG_SUGOV)
		return 0;

2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497
	/* !deadline task may carry old deadline bandwidth */
	if (new_bw == p->dl.dl_bw && task_has_dl_policy(p))
		return 0;

	/*
	 * Either if a task, enters, leave, or stays -deadline but changes
	 * its parameters, we may need to update accordingly the total
	 * allocated bandwidth of the container.
	 */
	raw_spin_lock(&dl_b->lock);
	cpus = dl_bw_cpus(task_cpu(p));
	if (dl_policy(policy) && !task_has_dl_policy(p) &&
	    !__dl_overflow(dl_b, cpus, 0, new_bw)) {
		if (hrtimer_active(&p->dl.inactive_timer))
2498
			__dl_sub(dl_b, p->dl.dl_bw, cpus);
2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509
		__dl_add(dl_b, new_bw, cpus);
		err = 0;
	} else if (dl_policy(policy) && task_has_dl_policy(p) &&
		   !__dl_overflow(dl_b, cpus, p->dl.dl_bw, new_bw)) {
		/*
		 * XXX this is slightly incorrect: when the task
		 * utilization decreases, we should delay the total
		 * utilization change until the task's 0-lag point.
		 * But this would require to set the task's "inactive
		 * timer" when the task is not inactive.
		 */
2510
		__dl_sub(dl_b, p->dl.dl_bw, cpus);
2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569
		__dl_add(dl_b, new_bw, cpus);
		dl_change_utilization(p, new_bw);
		err = 0;
	} else if (!dl_policy(policy) && task_has_dl_policy(p)) {
		/*
		 * Do not decrease the total deadline utilization here,
		 * switched_from_dl() will take care to do it at the correct
		 * (0-lag) time.
		 */
		err = 0;
	}
	raw_spin_unlock(&dl_b->lock);

	return err;
}

/*
 * This function initializes the sched_dl_entity of a newly becoming
 * SCHED_DEADLINE task.
 *
 * Only the static values are considered here, the actual runtime and the
 * absolute deadline will be properly calculated when the task is enqueued
 * for the first time with its new policy.
 */
void __setparam_dl(struct task_struct *p, const struct sched_attr *attr)
{
	struct sched_dl_entity *dl_se = &p->dl;

	dl_se->dl_runtime = attr->sched_runtime;
	dl_se->dl_deadline = attr->sched_deadline;
	dl_se->dl_period = attr->sched_period ?: dl_se->dl_deadline;
	dl_se->flags = attr->sched_flags;
	dl_se->dl_bw = to_ratio(dl_se->dl_period, dl_se->dl_runtime);
	dl_se->dl_density = to_ratio(dl_se->dl_deadline, dl_se->dl_runtime);
}

void __getparam_dl(struct task_struct *p, struct sched_attr *attr)
{
	struct sched_dl_entity *dl_se = &p->dl;

	attr->sched_priority = p->rt_priority;
	attr->sched_runtime = dl_se->dl_runtime;
	attr->sched_deadline = dl_se->dl_deadline;
	attr->sched_period = dl_se->dl_period;
	attr->sched_flags = dl_se->flags;
}

/*
 * This function validates the new parameters of a -deadline task.
 * We ask for the deadline not being zero, and greater or equal
 * than the runtime, as well as the period of being zero or
 * greater than deadline. Furthermore, we have to be sure that
 * user parameters are above the internal resolution of 1us (we
 * check sched_runtime only since it is always the smaller one) and
 * below 2^63 ns (we have to check both sched_deadline and
 * sched_period, as the latter can be zero).
 */
bool __checkparam_dl(const struct sched_attr *attr)
{
2570 2571 2572 2573
	/* special dl tasks don't actually use any parameter */
	if (attr->sched_flags & SCHED_FLAG_SUGOV)
		return true;

2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618
	/* deadline != 0 */
	if (attr->sched_deadline == 0)
		return false;

	/*
	 * Since we truncate DL_SCALE bits, make sure we're at least
	 * that big.
	 */
	if (attr->sched_runtime < (1ULL << DL_SCALE))
		return false;

	/*
	 * Since we use the MSB for wrap-around and sign issues, make
	 * sure it's not set (mind that period can be equal to zero).
	 */
	if (attr->sched_deadline & (1ULL << 63) ||
	    attr->sched_period & (1ULL << 63))
		return false;

	/* runtime <= deadline <= period (if period != 0) */
	if ((attr->sched_period != 0 &&
	     attr->sched_period < attr->sched_deadline) ||
	    attr->sched_deadline < attr->sched_runtime)
		return false;

	return true;
}

/*
 * This function clears the sched_dl_entity static params.
 */
void __dl_clear_params(struct task_struct *p)
{
	struct sched_dl_entity *dl_se = &p->dl;

	dl_se->dl_runtime = 0;
	dl_se->dl_deadline = 0;
	dl_se->dl_period = 0;
	dl_se->flags = 0;
	dl_se->dl_bw = 0;
	dl_se->dl_density = 0;

	dl_se->dl_throttled = 0;
	dl_se->dl_yielded = 0;
	dl_se->dl_non_contending = 0;
2619
	dl_se->dl_overrun = 0;
2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704
}

bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr)
{
	struct sched_dl_entity *dl_se = &p->dl;

	if (dl_se->dl_runtime != attr->sched_runtime ||
	    dl_se->dl_deadline != attr->sched_deadline ||
	    dl_se->dl_period != attr->sched_period ||
	    dl_se->flags != attr->sched_flags)
		return true;

	return false;
}

#ifdef CONFIG_SMP
int dl_task_can_attach(struct task_struct *p, const struct cpumask *cs_cpus_allowed)
{
	unsigned int dest_cpu = cpumask_any_and(cpu_active_mask,
							cs_cpus_allowed);
	struct dl_bw *dl_b;
	bool overflow;
	int cpus, ret;
	unsigned long flags;

	rcu_read_lock_sched();
	dl_b = dl_bw_of(dest_cpu);
	raw_spin_lock_irqsave(&dl_b->lock, flags);
	cpus = dl_bw_cpus(dest_cpu);
	overflow = __dl_overflow(dl_b, cpus, 0, p->dl.dl_bw);
	if (overflow)
		ret = -EBUSY;
	else {
		/*
		 * We reserve space for this task in the destination
		 * root_domain, as we can't fail after this point.
		 * We will free resources in the source root_domain
		 * later on (see set_cpus_allowed_dl()).
		 */
		__dl_add(dl_b, p->dl.dl_bw, cpus);
		ret = 0;
	}
	raw_spin_unlock_irqrestore(&dl_b->lock, flags);
	rcu_read_unlock_sched();
	return ret;
}

int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur,
				 const struct cpumask *trial)
{
	int ret = 1, trial_cpus;
	struct dl_bw *cur_dl_b;
	unsigned long flags;

	rcu_read_lock_sched();
	cur_dl_b = dl_bw_of(cpumask_any(cur));
	trial_cpus = cpumask_weight(trial);

	raw_spin_lock_irqsave(&cur_dl_b->lock, flags);
	if (cur_dl_b->bw != -1 &&
	    cur_dl_b->bw * trial_cpus < cur_dl_b->total_bw)
		ret = 0;
	raw_spin_unlock_irqrestore(&cur_dl_b->lock, flags);
	rcu_read_unlock_sched();
	return ret;
}

bool dl_cpu_busy(unsigned int cpu)
{
	unsigned long flags;
	struct dl_bw *dl_b;
	bool overflow;
	int cpus;

	rcu_read_lock_sched();
	dl_b = dl_bw_of(cpu);
	raw_spin_lock_irqsave(&dl_b->lock, flags);
	cpus = dl_bw_cpus(cpu);
	overflow = __dl_overflow(dl_b, cpus, 0, 0);
	raw_spin_unlock_irqrestore(&dl_b->lock, flags);
	rcu_read_unlock_sched();
	return overflow;
}
#endif

2705 2706 2707 2708 2709 2710 2711 2712
#ifdef CONFIG_SCHED_DEBUG
extern void print_dl_rq(struct seq_file *m, int cpu, struct dl_rq *dl_rq);

void print_dl_stats(struct seq_file *m, int cpu)
{
	print_dl_rq(m, cpu, &cpu_rq(cpu)->dl);
}
#endif /* CONFIG_SCHED_DEBUG */