softirq.c 21.4 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5
/*
 *	linux/kernel/softirq.c
 *
 *	Copyright (C) 1992 Linus Torvalds
 *
P
Pavel Machek 已提交
6 7 8
 *	Distribute under GPLv2.
 *
 *	Rewritten. Old one was good in 2.2, but in 2.3 it was immoral. --ANK (990903)
9 10
 *
 *	Remote softirq infrastructure is by Jens Axboe.
L
Linus Torvalds 已提交
11 12 13 14 15 16 17 18 19 20
 */

#include <linux/module.h>
#include <linux/kernel_stat.h>
#include <linux/interrupt.h>
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/notifier.h>
#include <linux/percpu.h>
#include <linux/cpu.h>
21
#include <linux/freezer.h>
L
Linus Torvalds 已提交
22 23
#include <linux/kthread.h>
#include <linux/rcupdate.h>
24
#include <linux/ftrace.h>
25
#include <linux/smp.h>
26
#include <linux/tick.h>
27 28

#define CREATE_TRACE_POINTS
29
#include <trace/events/irq.h>
L
Linus Torvalds 已提交
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

#include <asm/irq.h>
/*
   - No shared variables, all the data are CPU local.
   - If a softirq needs serialization, let it serialize itself
     by its own spinlocks.
   - Even if softirq is serialized, only local cpu is marked for
     execution. Hence, we get something sort of weak cpu binding.
     Though it is still not clear, will it result in better locality
     or will not.

   Examples:
   - NET RX softirq. It is multithreaded and does not require
     any global serialization.
   - NET TX softirq. It kicks software netdevice queues, hence
     it is logically serialized per device, but this serialization
     is invisible to common code.
   - Tasklets: serialized wrt itself.
 */

#ifndef __ARCH_IRQ_STAT
irq_cpustat_t irq_stat[NR_CPUS] ____cacheline_aligned;
EXPORT_SYMBOL(irq_stat);
#endif

A
Alexey Dobriyan 已提交
55
static struct softirq_action softirq_vec[NR_SOFTIRQS] __cacheline_aligned_in_smp;
L
Linus Torvalds 已提交
56 57 58

static DEFINE_PER_CPU(struct task_struct *, ksoftirqd);

59
char *softirq_to_name[NR_SOFTIRQS] = {
60 61
	"HI", "TIMER", "NET_TX", "NET_RX", "BLOCK",
	"TASKLET", "SCHED", "HRTIMER",	"RCU"
62 63
};

L
Linus Torvalds 已提交
64 65 66 67 68 69
/*
 * we cannot loop indefinitely here to avoid userspace starvation,
 * but we also don't want to introduce a worst case 1/HZ latency
 * to the pending events, so lets the scheduler to balance
 * the softirq load for us.
 */
70
void wakeup_softirqd(void)
L
Linus Torvalds 已提交
71 72 73 74 75 76 77 78
{
	/* Interrupts are disabled: no need to stop preemption */
	struct task_struct *tsk = __get_cpu_var(ksoftirqd);

	if (tsk && tsk->state != TASK_RUNNING)
		wake_up_process(tsk);
}

79 80 81 82
/*
 * This one is for softirq.c-internal use,
 * where hardirqs are disabled legitimately:
 */
83
#ifdef CONFIG_TRACE_IRQFLAGS
84 85 86 87 88 89 90
static void __local_bh_disable(unsigned long ip)
{
	unsigned long flags;

	WARN_ON_ONCE(in_irq());

	raw_local_irq_save(flags);
91 92 93 94 95 96 97 98
	/*
	 * The preempt tracer hooks into add_preempt_count and will break
	 * lockdep because it calls back into lockdep after SOFTIRQ_OFFSET
	 * is set and before current->softirq_enabled is cleared.
	 * We must manually increment preempt_count here and manually
	 * call the trace_preempt_off later.
	 */
	preempt_count() += SOFTIRQ_OFFSET;
99 100 101 102 103 104
	/*
	 * Were softirqs turned off above:
	 */
	if (softirq_count() == SOFTIRQ_OFFSET)
		trace_softirqs_off(ip);
	raw_local_irq_restore(flags);
105 106 107

	if (preempt_count() == SOFTIRQ_OFFSET)
		trace_preempt_off(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1));
108
}
109 110 111 112 113 114 115
#else /* !CONFIG_TRACE_IRQFLAGS */
static inline void __local_bh_disable(unsigned long ip)
{
	add_preempt_count(SOFTIRQ_OFFSET);
	barrier();
}
#endif /* CONFIG_TRACE_IRQFLAGS */
116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140

void local_bh_disable(void)
{
	__local_bh_disable((unsigned long)__builtin_return_address(0));
}

EXPORT_SYMBOL(local_bh_disable);

/*
 * Special-case - softirqs can safely be enabled in
 * cond_resched_softirq(), or by __do_softirq(),
 * without processing still-pending softirqs:
 */
void _local_bh_enable(void)
{
	WARN_ON_ONCE(in_irq());
	WARN_ON_ONCE(!irqs_disabled());

	if (softirq_count() == SOFTIRQ_OFFSET)
		trace_softirqs_on((unsigned long)__builtin_return_address(0));
	sub_preempt_count(SOFTIRQ_OFFSET);
}

EXPORT_SYMBOL(_local_bh_enable);

141
static inline void _local_bh_enable_ip(unsigned long ip)
142
{
143
	WARN_ON_ONCE(in_irq() || irqs_disabled());
144
#ifdef CONFIG_TRACE_IRQFLAGS
145
	local_irq_disable();
146
#endif
147 148 149 150
	/*
	 * Are softirqs going to be turned on now:
	 */
	if (softirq_count() == SOFTIRQ_OFFSET)
151
		trace_softirqs_on(ip);
152 153 154 155 156 157 158 159 160 161
	/*
	 * Keep preemption disabled until we are done with
	 * softirq processing:
 	 */
 	sub_preempt_count(SOFTIRQ_OFFSET - 1);

	if (unlikely(!in_interrupt() && local_softirq_pending()))
		do_softirq();

	dec_preempt_count();
162
#ifdef CONFIG_TRACE_IRQFLAGS
163
	local_irq_enable();
164
#endif
165 166
	preempt_check_resched();
}
167 168 169 170 171

void local_bh_enable(void)
{
	_local_bh_enable_ip((unsigned long)__builtin_return_address(0));
}
172 173 174 175
EXPORT_SYMBOL(local_bh_enable);

void local_bh_enable_ip(unsigned long ip)
{
176
	_local_bh_enable_ip(ip);
177 178 179
}
EXPORT_SYMBOL(local_bh_enable_ip);

L
Linus Torvalds 已提交
180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
/*
 * We restart softirq processing MAX_SOFTIRQ_RESTART times,
 * and we fall back to softirqd after that.
 *
 * This number has been established via experimentation.
 * The two things to balance is latency against fairness -
 * we want to handle softirqs as soon as possible, but they
 * should not be able to lock up the box.
 */
#define MAX_SOFTIRQ_RESTART 10

asmlinkage void __do_softirq(void)
{
	struct softirq_action *h;
	__u32 pending;
	int max_restart = MAX_SOFTIRQ_RESTART;
	int cpu;

	pending = local_softirq_pending();
199 200
	account_system_vtime(current);

201
	__local_bh_disable((unsigned long)__builtin_return_address(0));
202
	lockdep_softirq_enter();
L
Linus Torvalds 已提交
203 204 205 206

	cpu = smp_processor_id();
restart:
	/* Reset the pending bitmask before enabling irqs */
207
	set_softirq_pending(0);
L
Linus Torvalds 已提交
208

209
	local_irq_enable();
L
Linus Torvalds 已提交
210 211 212 213 214

	h = softirq_vec;

	do {
		if (pending & 1) {
215
			int prev_count = preempt_count();
216
			kstat_incr_softirqs_this_cpu(h - softirq_vec);
217

218
			trace_softirq_entry(h, softirq_vec);
L
Linus Torvalds 已提交
219
			h->action(h);
220
			trace_softirq_exit(h, softirq_vec);
221
			if (unlikely(prev_count != preempt_count())) {
222
				printk(KERN_ERR "huh, entered softirq %td %s %p"
223 224
				       "with preempt_count %08x,"
				       " exited with %08x?\n", h - softirq_vec,
225
				       softirq_to_name[h - softirq_vec],
226 227 228 229
				       h->action, prev_count, preempt_count());
				preempt_count() = prev_count;
			}

230
			rcu_bh_qs(cpu);
L
Linus Torvalds 已提交
231 232 233 234 235
		}
		h++;
		pending >>= 1;
	} while (pending);

236
	local_irq_disable();
L
Linus Torvalds 已提交
237 238 239 240 241 242 243 244

	pending = local_softirq_pending();
	if (pending && --max_restart)
		goto restart;

	if (pending)
		wakeup_softirqd();

245
	lockdep_softirq_exit();
246 247

	account_system_vtime(current);
248
	_local_bh_enable();
L
Linus Torvalds 已提交
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272
}

#ifndef __ARCH_HAS_DO_SOFTIRQ

asmlinkage void do_softirq(void)
{
	__u32 pending;
	unsigned long flags;

	if (in_interrupt())
		return;

	local_irq_save(flags);

	pending = local_softirq_pending();

	if (pending)
		__do_softirq();

	local_irq_restore(flags);
}

#endif

I
Ingo Molnar 已提交
273 274 275 276 277
/*
 * Enter an interrupt context.
 */
void irq_enter(void)
{
278
	int cpu = smp_processor_id();
279

280
	rcu_irq_enter();
281 282
	if (idle_cpu(cpu) && !in_interrupt()) {
		__irq_enter();
283
		tick_check_idle(cpu);
284 285
	} else
		__irq_enter();
I
Ingo Molnar 已提交
286 287
}

L
Linus Torvalds 已提交
288 289 290 291 292 293 294 295 296 297 298 299
#ifdef __ARCH_IRQ_EXIT_IRQS_DISABLED
# define invoke_softirq()	__do_softirq()
#else
# define invoke_softirq()	do_softirq()
#endif

/*
 * Exit an interrupt context. Process softirqs if needed and possible:
 */
void irq_exit(void)
{
	account_system_vtime(current);
300
	trace_hardirq_exit();
L
Linus Torvalds 已提交
301 302 303
	sub_preempt_count(IRQ_EXIT_OFFSET);
	if (!in_interrupt() && local_softirq_pending())
		invoke_softirq();
304 305 306

#ifdef CONFIG_NO_HZ
	/* Make sure that timer wheel updates are propagated */
307
	rcu_irq_exit();
308 309
	if (idle_cpu(smp_processor_id()) && !in_interrupt() && !need_resched())
		tick_nohz_stop_sched_tick(0);
310
#endif
L
Linus Torvalds 已提交
311 312 313 314 315 316
	preempt_enable_no_resched();
}

/*
 * This function must run with irqs disabled!
 */
317
inline void raise_softirq_irqoff(unsigned int nr)
L
Linus Torvalds 已提交
318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333
{
	__raise_softirq_irqoff(nr);

	/*
	 * If we're in an interrupt or softirq, we're done
	 * (this also catches softirq-disabled code). We will
	 * actually run the softirq once we return from
	 * the irq or softirq.
	 *
	 * Otherwise we wake up ksoftirqd to make sure we
	 * schedule the softirq soon.
	 */
	if (!in_interrupt())
		wakeup_softirqd();
}

334
void raise_softirq(unsigned int nr)
L
Linus Torvalds 已提交
335 336 337 338 339 340 341 342
{
	unsigned long flags;

	local_irq_save(flags);
	raise_softirq_irqoff(nr);
	local_irq_restore(flags);
}

343
void open_softirq(int nr, void (*action)(struct softirq_action *))
L
Linus Torvalds 已提交
344 345 346 347
{
	softirq_vec[nr].action = action;
}

348 349 350
/*
 * Tasklets
 */
L
Linus Torvalds 已提交
351 352
struct tasklet_head
{
353 354
	struct tasklet_struct *head;
	struct tasklet_struct **tail;
L
Linus Torvalds 已提交
355 356
};

357 358
static DEFINE_PER_CPU(struct tasklet_head, tasklet_vec);
static DEFINE_PER_CPU(struct tasklet_head, tasklet_hi_vec);
L
Linus Torvalds 已提交
359

360
void __tasklet_schedule(struct tasklet_struct *t)
L
Linus Torvalds 已提交
361 362 363 364
{
	unsigned long flags;

	local_irq_save(flags);
365 366 367
	t->next = NULL;
	*__get_cpu_var(tasklet_vec).tail = t;
	__get_cpu_var(tasklet_vec).tail = &(t->next);
L
Linus Torvalds 已提交
368 369 370 371 372 373
	raise_softirq_irqoff(TASKLET_SOFTIRQ);
	local_irq_restore(flags);
}

EXPORT_SYMBOL(__tasklet_schedule);

374
void __tasklet_hi_schedule(struct tasklet_struct *t)
L
Linus Torvalds 已提交
375 376 377 378
{
	unsigned long flags;

	local_irq_save(flags);
379 380 381
	t->next = NULL;
	*__get_cpu_var(tasklet_hi_vec).tail = t;
	__get_cpu_var(tasklet_hi_vec).tail = &(t->next);
L
Linus Torvalds 已提交
382 383 384 385 386 387
	raise_softirq_irqoff(HI_SOFTIRQ);
	local_irq_restore(flags);
}

EXPORT_SYMBOL(__tasklet_hi_schedule);

388 389 390 391 392 393 394 395 396 397 398
void __tasklet_hi_schedule_first(struct tasklet_struct *t)
{
	BUG_ON(!irqs_disabled());

	t->next = __get_cpu_var(tasklet_hi_vec).head;
	__get_cpu_var(tasklet_hi_vec).head = t;
	__raise_softirq_irqoff(HI_SOFTIRQ);
}

EXPORT_SYMBOL(__tasklet_hi_schedule_first);

L
Linus Torvalds 已提交
399 400 401 402 403
static void tasklet_action(struct softirq_action *a)
{
	struct tasklet_struct *list;

	local_irq_disable();
404 405 406
	list = __get_cpu_var(tasklet_vec).head;
	__get_cpu_var(tasklet_vec).head = NULL;
	__get_cpu_var(tasklet_vec).tail = &__get_cpu_var(tasklet_vec).head;
L
Linus Torvalds 已提交
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425
	local_irq_enable();

	while (list) {
		struct tasklet_struct *t = list;

		list = list->next;

		if (tasklet_trylock(t)) {
			if (!atomic_read(&t->count)) {
				if (!test_and_clear_bit(TASKLET_STATE_SCHED, &t->state))
					BUG();
				t->func(t->data);
				tasklet_unlock(t);
				continue;
			}
			tasklet_unlock(t);
		}

		local_irq_disable();
426 427 428
		t->next = NULL;
		*__get_cpu_var(tasklet_vec).tail = t;
		__get_cpu_var(tasklet_vec).tail = &(t->next);
L
Linus Torvalds 已提交
429 430 431 432 433 434 435 436 437 438
		__raise_softirq_irqoff(TASKLET_SOFTIRQ);
		local_irq_enable();
	}
}

static void tasklet_hi_action(struct softirq_action *a)
{
	struct tasklet_struct *list;

	local_irq_disable();
439 440 441
	list = __get_cpu_var(tasklet_hi_vec).head;
	__get_cpu_var(tasklet_hi_vec).head = NULL;
	__get_cpu_var(tasklet_hi_vec).tail = &__get_cpu_var(tasklet_hi_vec).head;
L
Linus Torvalds 已提交
442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460
	local_irq_enable();

	while (list) {
		struct tasklet_struct *t = list;

		list = list->next;

		if (tasklet_trylock(t)) {
			if (!atomic_read(&t->count)) {
				if (!test_and_clear_bit(TASKLET_STATE_SCHED, &t->state))
					BUG();
				t->func(t->data);
				tasklet_unlock(t);
				continue;
			}
			tasklet_unlock(t);
		}

		local_irq_disable();
461 462 463
		t->next = NULL;
		*__get_cpu_var(tasklet_hi_vec).tail = t;
		__get_cpu_var(tasklet_hi_vec).tail = &(t->next);
L
Linus Torvalds 已提交
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
		__raise_softirq_irqoff(HI_SOFTIRQ);
		local_irq_enable();
	}
}


void tasklet_init(struct tasklet_struct *t,
		  void (*func)(unsigned long), unsigned long data)
{
	t->next = NULL;
	t->state = 0;
	atomic_set(&t->count, 0);
	t->func = func;
	t->data = data;
}

EXPORT_SYMBOL(tasklet_init);

void tasklet_kill(struct tasklet_struct *t)
{
	if (in_interrupt())
		printk("Attempt to kill tasklet from interrupt\n");

	while (test_and_set_bit(TASKLET_STATE_SCHED, &t->state)) {
488
		do {
L
Linus Torvalds 已提交
489
			yield();
490
		} while (test_bit(TASKLET_STATE_SCHED, &t->state));
L
Linus Torvalds 已提交
491 492 493 494 495 496 497
	}
	tasklet_unlock_wait(t);
	clear_bit(TASKLET_STATE_SCHED, &t->state);
}

EXPORT_SYMBOL(tasklet_kill);

498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557
/*
 * tasklet_hrtimer
 */

/*
 * The trampoline is called when the hrtimer expires. If this is
 * called from the hrtimer interrupt then we schedule the tasklet as
 * the timer callback function expects to run in softirq context. If
 * it's called in softirq context anyway (i.e. high resolution timers
 * disabled) then the hrtimer callback is called right away.
 */
static enum hrtimer_restart __hrtimer_tasklet_trampoline(struct hrtimer *timer)
{
	struct tasklet_hrtimer *ttimer =
		container_of(timer, struct tasklet_hrtimer, timer);

	if (hrtimer_is_hres_active(timer)) {
		tasklet_hi_schedule(&ttimer->tasklet);
		return HRTIMER_NORESTART;
	}
	return ttimer->function(timer);
}

/*
 * Helper function which calls the hrtimer callback from
 * tasklet/softirq context
 */
static void __tasklet_hrtimer_trampoline(unsigned long data)
{
	struct tasklet_hrtimer *ttimer = (void *)data;
	enum hrtimer_restart restart;

	restart = ttimer->function(&ttimer->timer);
	if (restart != HRTIMER_NORESTART)
		hrtimer_restart(&ttimer->timer);
}

/**
 * tasklet_hrtimer_init - Init a tasklet/hrtimer combo for softirq callbacks
 * @ttimer:	 tasklet_hrtimer which is initialized
 * @function:	 hrtimer callback funtion which gets called from softirq context
 * @which_clock: clock id (CLOCK_MONOTONIC/CLOCK_REALTIME)
 * @mode:	 hrtimer mode (HRTIMER_MODE_ABS/HRTIMER_MODE_REL)
 */
void tasklet_hrtimer_init(struct tasklet_hrtimer *ttimer,
			  enum hrtimer_restart (*function)(struct hrtimer *),
			  clockid_t which_clock, enum hrtimer_mode mode)
{
	hrtimer_init(&ttimer->timer, which_clock, mode);
	ttimer->timer.function = __hrtimer_tasklet_trampoline;
	tasklet_init(&ttimer->tasklet, __tasklet_hrtimer_trampoline,
		     (unsigned long)ttimer);
	ttimer->function = function;
}
EXPORT_SYMBOL_GPL(tasklet_hrtimer_init);

/*
 * Remote softirq bits
 */

558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593
DEFINE_PER_CPU(struct list_head [NR_SOFTIRQS], softirq_work_list);
EXPORT_PER_CPU_SYMBOL(softirq_work_list);

static void __local_trigger(struct call_single_data *cp, int softirq)
{
	struct list_head *head = &__get_cpu_var(softirq_work_list[softirq]);

	list_add_tail(&cp->list, head);

	/* Trigger the softirq only if the list was previously empty.  */
	if (head->next == &cp->list)
		raise_softirq_irqoff(softirq);
}

#ifdef CONFIG_USE_GENERIC_SMP_HELPERS
static void remote_softirq_receive(void *data)
{
	struct call_single_data *cp = data;
	unsigned long flags;
	int softirq;

	softirq = cp->priv;

	local_irq_save(flags);
	__local_trigger(cp, softirq);
	local_irq_restore(flags);
}

static int __try_remote_softirq(struct call_single_data *cp, int cpu, int softirq)
{
	if (cpu_online(cpu)) {
		cp->func = remote_softirq_receive;
		cp->info = cp;
		cp->flags = 0;
		cp->priv = softirq;

594
		__smp_call_function_single(cpu, cp, 0);
595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678
		return 0;
	}
	return 1;
}
#else /* CONFIG_USE_GENERIC_SMP_HELPERS */
static int __try_remote_softirq(struct call_single_data *cp, int cpu, int softirq)
{
	return 1;
}
#endif

/**
 * __send_remote_softirq - try to schedule softirq work on a remote cpu
 * @cp: private SMP call function data area
 * @cpu: the remote cpu
 * @this_cpu: the currently executing cpu
 * @softirq: the softirq for the work
 *
 * Attempt to schedule softirq work on a remote cpu.  If this cannot be
 * done, the work is instead queued up on the local cpu.
 *
 * Interrupts must be disabled.
 */
void __send_remote_softirq(struct call_single_data *cp, int cpu, int this_cpu, int softirq)
{
	if (cpu == this_cpu || __try_remote_softirq(cp, cpu, softirq))
		__local_trigger(cp, softirq);
}
EXPORT_SYMBOL(__send_remote_softirq);

/**
 * send_remote_softirq - try to schedule softirq work on a remote cpu
 * @cp: private SMP call function data area
 * @cpu: the remote cpu
 * @softirq: the softirq for the work
 *
 * Like __send_remote_softirq except that disabling interrupts and
 * computing the current cpu is done for the caller.
 */
void send_remote_softirq(struct call_single_data *cp, int cpu, int softirq)
{
	unsigned long flags;
	int this_cpu;

	local_irq_save(flags);
	this_cpu = smp_processor_id();
	__send_remote_softirq(cp, cpu, this_cpu, softirq);
	local_irq_restore(flags);
}
EXPORT_SYMBOL(send_remote_softirq);

static int __cpuinit remote_softirq_cpu_notify(struct notifier_block *self,
					       unsigned long action, void *hcpu)
{
	/*
	 * If a CPU goes away, splice its entries to the current CPU
	 * and trigger a run of the softirq
	 */
	if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) {
		int cpu = (unsigned long) hcpu;
		int i;

		local_irq_disable();
		for (i = 0; i < NR_SOFTIRQS; i++) {
			struct list_head *head = &per_cpu(softirq_work_list[i], cpu);
			struct list_head *local_head;

			if (list_empty(head))
				continue;

			local_head = &__get_cpu_var(softirq_work_list[i]);
			list_splice_init(head, local_head);
			raise_softirq_irqoff(i);
		}
		local_irq_enable();
	}

	return NOTIFY_OK;
}

static struct notifier_block __cpuinitdata remote_softirq_cpu_notifier = {
	.notifier_call	= remote_softirq_cpu_notify,
};

L
Linus Torvalds 已提交
679 680
void __init softirq_init(void)
{
681 682 683
	int cpu;

	for_each_possible_cpu(cpu) {
684 685
		int i;

686 687 688 689
		per_cpu(tasklet_vec, cpu).tail =
			&per_cpu(tasklet_vec, cpu).head;
		per_cpu(tasklet_hi_vec, cpu).tail =
			&per_cpu(tasklet_hi_vec, cpu).head;
690 691
		for (i = 0; i < NR_SOFTIRQS; i++)
			INIT_LIST_HEAD(&per_cpu(softirq_work_list[i], cpu));
692 693
	}

694 695
	register_hotcpu_notifier(&remote_softirq_cpu_notifier);

696 697
	open_softirq(TASKLET_SOFTIRQ, tasklet_action);
	open_softirq(HI_SOFTIRQ, tasklet_hi_action);
L
Linus Torvalds 已提交
698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723
}

static int ksoftirqd(void * __bind_cpu)
{
	set_current_state(TASK_INTERRUPTIBLE);

	while (!kthread_should_stop()) {
		preempt_disable();
		if (!local_softirq_pending()) {
			preempt_enable_no_resched();
			schedule();
			preempt_disable();
		}

		__set_current_state(TASK_RUNNING);

		while (local_softirq_pending()) {
			/* Preempt disable stops cpu going offline.
			   If already offline, we'll be on wrong CPU:
			   don't process */
			if (cpu_is_offline((long)__bind_cpu))
				goto wait_to_die;
			do_softirq();
			preempt_enable_no_resched();
			cond_resched();
			preempt_disable();
724
			rcu_sched_qs((long)__bind_cpu);
L
Linus Torvalds 已提交
725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764
		}
		preempt_enable();
		set_current_state(TASK_INTERRUPTIBLE);
	}
	__set_current_state(TASK_RUNNING);
	return 0;

wait_to_die:
	preempt_enable();
	/* Wait for kthread_stop */
	set_current_state(TASK_INTERRUPTIBLE);
	while (!kthread_should_stop()) {
		schedule();
		set_current_state(TASK_INTERRUPTIBLE);
	}
	__set_current_state(TASK_RUNNING);
	return 0;
}

#ifdef CONFIG_HOTPLUG_CPU
/*
 * tasklet_kill_immediate is called to remove a tasklet which can already be
 * scheduled for execution on @cpu.
 *
 * Unlike tasklet_kill, this function removes the tasklet
 * _immediately_, even if the tasklet is in TASKLET_STATE_SCHED state.
 *
 * When this function is called, @cpu must be in the CPU_DEAD state.
 */
void tasklet_kill_immediate(struct tasklet_struct *t, unsigned int cpu)
{
	struct tasklet_struct **i;

	BUG_ON(cpu_online(cpu));
	BUG_ON(test_bit(TASKLET_STATE_RUN, &t->state));

	if (!test_bit(TASKLET_STATE_SCHED, &t->state))
		return;

	/* CPU is dead, so no lock needed. */
765
	for (i = &per_cpu(tasklet_vec, cpu).head; *i; i = &(*i)->next) {
L
Linus Torvalds 已提交
766 767
		if (*i == t) {
			*i = t->next;
768 769 770
			/* If this was the tail element, move the tail ptr */
			if (*i == NULL)
				per_cpu(tasklet_vec, cpu).tail = i;
L
Linus Torvalds 已提交
771 772 773 774 775 776 777 778 779 780 781 782
			return;
		}
	}
	BUG();
}

static void takeover_tasklets(unsigned int cpu)
{
	/* CPU is dead, so no lock needed. */
	local_irq_disable();

	/* Find end, append list for that CPU. */
783 784 785 786 787 788
	if (&per_cpu(tasklet_vec, cpu).head != per_cpu(tasklet_vec, cpu).tail) {
		*(__get_cpu_var(tasklet_vec).tail) = per_cpu(tasklet_vec, cpu).head;
		__get_cpu_var(tasklet_vec).tail = per_cpu(tasklet_vec, cpu).tail;
		per_cpu(tasklet_vec, cpu).head = NULL;
		per_cpu(tasklet_vec, cpu).tail = &per_cpu(tasklet_vec, cpu).head;
	}
L
Linus Torvalds 已提交
789 790
	raise_softirq_irqoff(TASKLET_SOFTIRQ);

791 792 793 794 795 796
	if (&per_cpu(tasklet_hi_vec, cpu).head != per_cpu(tasklet_hi_vec, cpu).tail) {
		*__get_cpu_var(tasklet_hi_vec).tail = per_cpu(tasklet_hi_vec, cpu).head;
		__get_cpu_var(tasklet_hi_vec).tail = per_cpu(tasklet_hi_vec, cpu).tail;
		per_cpu(tasklet_hi_vec, cpu).head = NULL;
		per_cpu(tasklet_hi_vec, cpu).tail = &per_cpu(tasklet_hi_vec, cpu).head;
	}
L
Linus Torvalds 已提交
797 798 799 800 801 802
	raise_softirq_irqoff(HI_SOFTIRQ);

	local_irq_enable();
}
#endif /* CONFIG_HOTPLUG_CPU */

803
static int __cpuinit cpu_callback(struct notifier_block *nfb,
L
Linus Torvalds 已提交
804 805 806 807 808 809 810 811
				  unsigned long action,
				  void *hcpu)
{
	int hotcpu = (unsigned long)hcpu;
	struct task_struct *p;

	switch (action) {
	case CPU_UP_PREPARE:
812
	case CPU_UP_PREPARE_FROZEN:
L
Linus Torvalds 已提交
813 814 815 816 817 818 819 820 821
		p = kthread_create(ksoftirqd, hcpu, "ksoftirqd/%d", hotcpu);
		if (IS_ERR(p)) {
			printk("ksoftirqd for %i failed\n", hotcpu);
			return NOTIFY_BAD;
		}
		kthread_bind(p, hotcpu);
  		per_cpu(ksoftirqd, hotcpu) = p;
 		break;
	case CPU_ONLINE:
822
	case CPU_ONLINE_FROZEN:
L
Linus Torvalds 已提交
823 824 825 826
		wake_up_process(per_cpu(ksoftirqd, hotcpu));
		break;
#ifdef CONFIG_HOTPLUG_CPU
	case CPU_UP_CANCELED:
827
	case CPU_UP_CANCELED_FROZEN:
828 829
		if (!per_cpu(ksoftirqd, hotcpu))
			break;
L
Linus Torvalds 已提交
830
		/* Unbind so it can run.  Fall thru. */
831
		kthread_bind(per_cpu(ksoftirqd, hotcpu),
832
			     cpumask_any(cpu_online_mask));
L
Linus Torvalds 已提交
833
	case CPU_DEAD:
834 835 836
	case CPU_DEAD_FROZEN: {
		struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };

L
Linus Torvalds 已提交
837 838
		p = per_cpu(ksoftirqd, hotcpu);
		per_cpu(ksoftirqd, hotcpu) = NULL;
839
		sched_setscheduler_nocheck(p, SCHED_FIFO, &param);
L
Linus Torvalds 已提交
840 841 842
		kthread_stop(p);
		takeover_tasklets(hotcpu);
		break;
843
	}
L
Linus Torvalds 已提交
844 845 846 847 848
#endif /* CONFIG_HOTPLUG_CPU */
 	}
	return NOTIFY_OK;
}

849
static struct notifier_block __cpuinitdata cpu_nfb = {
L
Linus Torvalds 已提交
850 851 852
	.notifier_call = cpu_callback
};

853
static __init int spawn_ksoftirqd(void)
L
Linus Torvalds 已提交
854 855
{
	void *cpu = (void *)(long)smp_processor_id();
856 857 858
	int err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu);

	BUG_ON(err == NOTIFY_BAD);
L
Linus Torvalds 已提交
859 860 861 862
	cpu_callback(&cpu_nfb, CPU_ONLINE, cpu);
	register_cpu_notifier(&cpu_nfb);
	return 0;
}
863
early_initcall(spawn_ksoftirqd);
864 865 866 867 868

#ifdef CONFIG_SMP
/*
 * Call a function on all processors
 */
869
int on_each_cpu(void (*func) (void *info), void *info, int wait)
870 871 872 873
{
	int ret = 0;

	preempt_disable();
874
	ret = smp_call_function(func, info, wait);
875 876 877 878 879 880 881 882
	local_irq_disable();
	func(info);
	local_irq_enable();
	preempt_enable();
	return ret;
}
EXPORT_SYMBOL(on_each_cpu);
#endif
883 884 885 886 887 888 889 890 891 892 893

/*
 * [ These __weak aliases are kept in a separate compilation unit, so that
 *   GCC does not inline them incorrectly. ]
 */

int __init __weak early_irq_init(void)
{
	return 0;
}

Y
Yinghai Lu 已提交
894 895 896 897 898
int __init __weak arch_probe_nr_irqs(void)
{
	return 0;
}

899 900 901 902 903
int __init __weak arch_early_irq_init(void)
{
	return 0;
}

904
int __weak arch_init_chip_data(struct irq_desc *desc, int node)
905 906 907
{
	return 0;
}