smpboot.c 13.1 KB
Newer Older
1 2 3
/*
 * Common SMP CPU bringup/teardown functions
 */
4
#include <linux/cpu.h>
5 6
#include <linux/err.h>
#include <linux/smp.h>
7
#include <linux/delay.h>
8
#include <linux/init.h>
9 10
#include <linux/list.h>
#include <linux/slab.h>
11
#include <linux/sched.h>
12
#include <linux/sched/task.h>
13
#include <linux/export.h>
14
#include <linux/percpu.h>
15 16
#include <linux/kthread.h>
#include <linux/smpboot.h>
17 18 19

#include "smpboot.h"

20 21
#ifdef CONFIG_SMP

22 23 24 25 26 27 28
#ifdef CONFIG_GENERIC_SMP_IDLE_THREAD
/*
 * For the hotplug case we keep the task structs around and reuse
 * them.
 */
static DEFINE_PER_CPU(struct task_struct *, idle_threads);

29
struct task_struct *idle_thread_get(unsigned int cpu)
30 31 32 33
{
	struct task_struct *tsk = per_cpu(idle_threads, cpu);

	if (!tsk)
34
		return ERR_PTR(-ENOMEM);
35 36 37 38
	init_idle(tsk, cpu);
	return tsk;
}

39
void __init idle_thread_set_boot_cpu(void)
40
{
41
	per_cpu(idle_threads, smp_processor_id()) = current;
42 43
}

44 45 46 47 48 49
/**
 * idle_init - Initialize the idle thread for a cpu
 * @cpu:	The cpu for which the idle thread should be initialized
 *
 * Creates the thread if it does not exist.
 */
50
static inline void idle_init(unsigned int cpu)
51
{
52 53 54 55 56 57 58 59 60
	struct task_struct *tsk = per_cpu(idle_threads, cpu);

	if (!tsk) {
		tsk = fork_idle(cpu);
		if (IS_ERR(tsk))
			pr_err("SMP: fork_idle() failed for CPU %u\n", cpu);
		else
			per_cpu(idle_threads, cpu) = tsk;
	}
61 62 63
}

/**
64
 * idle_threads_init - Initialize idle threads for all cpus
65
 */
66
void __init idle_threads_init(void)
67
{
68 69 70
	unsigned int cpu, boot_cpu;

	boot_cpu = smp_processor_id();
71

72
	for_each_possible_cpu(cpu) {
73
		if (cpu != boot_cpu)
74
			idle_init(cpu);
75 76 77
	}
}
#endif
78

79 80
#endif /* #ifdef CONFIG_SMP */

81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114
static LIST_HEAD(hotplug_threads);
static DEFINE_MUTEX(smpboot_threads_lock);

struct smpboot_thread_data {
	unsigned int			cpu;
	unsigned int			status;
	struct smp_hotplug_thread	*ht;
};

enum {
	HP_THREAD_NONE = 0,
	HP_THREAD_ACTIVE,
	HP_THREAD_PARKED,
};

/**
 * smpboot_thread_fn - percpu hotplug thread loop function
 * @data:	thread data pointer
 *
 * Checks for thread stop and park conditions. Calls the necessary
 * setup, cleanup, park and unpark functions for the registered
 * thread.
 *
 * Returns 1 when the thread should exit, 0 otherwise.
 */
static int smpboot_thread_fn(void *data)
{
	struct smpboot_thread_data *td = data;
	struct smp_hotplug_thread *ht = td->ht;

	while (1) {
		set_current_state(TASK_INTERRUPTIBLE);
		preempt_disable();
		if (kthread_should_stop()) {
115
			__set_current_state(TASK_RUNNING);
116
			preempt_enable();
117 118
			/* cleanup must mirror setup */
			if (ht->cleanup && td->status != HP_THREAD_NONE)
119 120 121 122 123 124 125
				ht->cleanup(td->cpu, cpu_online(td->cpu));
			kfree(td);
			return 0;
		}

		if (kthread_should_park()) {
			__set_current_state(TASK_RUNNING);
126
			preempt_enable();
127 128 129 130 131 132 133 134 135 136
			if (ht->park && td->status == HP_THREAD_ACTIVE) {
				BUG_ON(td->cpu != smp_processor_id());
				ht->park(td->cpu);
				td->status = HP_THREAD_PARKED;
			}
			kthread_parkme();
			/* We might have been woken for stop */
			continue;
		}

137
		BUG_ON(td->cpu != smp_processor_id());
138 139 140 141

		/* Check for state change setup */
		switch (td->status) {
		case HP_THREAD_NONE:
142
			__set_current_state(TASK_RUNNING);
143 144 145 146
			preempt_enable();
			if (ht->setup)
				ht->setup(td->cpu);
			td->status = HP_THREAD_ACTIVE;
147 148
			continue;

149
		case HP_THREAD_PARKED:
150
			__set_current_state(TASK_RUNNING);
151 152 153 154
			preempt_enable();
			if (ht->unpark)
				ht->unpark(td->cpu);
			td->status = HP_THREAD_ACTIVE;
155
			continue;
156 157 158
		}

		if (!ht->thread_should_run(td->cpu)) {
159
			preempt_enable_no_resched();
160 161
			schedule();
		} else {
162
			__set_current_state(TASK_RUNNING);
163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189
			preempt_enable();
			ht->thread_fn(td->cpu);
		}
	}
}

static int
__smpboot_create_thread(struct smp_hotplug_thread *ht, unsigned int cpu)
{
	struct task_struct *tsk = *per_cpu_ptr(ht->store, cpu);
	struct smpboot_thread_data *td;

	if (tsk)
		return 0;

	td = kzalloc_node(sizeof(*td), GFP_KERNEL, cpu_to_node(cpu));
	if (!td)
		return -ENOMEM;
	td->cpu = cpu;
	td->ht = ht;

	tsk = kthread_create_on_cpu(smpboot_thread_fn, td, cpu,
				    ht->thread_comm);
	if (IS_ERR(tsk)) {
		kfree(td);
		return PTR_ERR(tsk);
	}
190 191 192 193 194
	/*
	 * Park the thread so that it could start right on the CPU
	 * when it is available.
	 */
	kthread_park(tsk);
195 196
	get_task_struct(tsk);
	*per_cpu_ptr(ht->store, cpu) = tsk;
197 198 199 200 201 202 203 204 205 206 207 208
	if (ht->create) {
		/*
		 * Make sure that the task has actually scheduled out
		 * into park position, before calling the create
		 * callback. At least the migration thread callback
		 * requires that the task is off the runqueue.
		 */
		if (!wait_task_inactive(tsk, TASK_PARKED))
			WARN_ON(1);
		else
			ht->create(cpu);
	}
209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230
	return 0;
}

int smpboot_create_threads(unsigned int cpu)
{
	struct smp_hotplug_thread *cur;
	int ret = 0;

	mutex_lock(&smpboot_threads_lock);
	list_for_each_entry(cur, &hotplug_threads, list) {
		ret = __smpboot_create_thread(cur, cpu);
		if (ret)
			break;
	}
	mutex_unlock(&smpboot_threads_lock);
	return ret;
}

static void smpboot_unpark_thread(struct smp_hotplug_thread *ht, unsigned int cpu)
{
	struct task_struct *tsk = *per_cpu_ptr(ht->store, cpu);

231 232
	if (!ht->selfparking)
		kthread_unpark(tsk);
233 234
}

235
int smpboot_unpark_threads(unsigned int cpu)
236 237 238 239 240
{
	struct smp_hotplug_thread *cur;

	mutex_lock(&smpboot_threads_lock);
	list_for_each_entry(cur, &hotplug_threads, list)
241 242
		if (cpumask_test_cpu(cpu, cur->cpumask))
			smpboot_unpark_thread(cur, cpu);
243
	mutex_unlock(&smpboot_threads_lock);
244
	return 0;
245 246 247 248 249 250
}

static void smpboot_park_thread(struct smp_hotplug_thread *ht, unsigned int cpu)
{
	struct task_struct *tsk = *per_cpu_ptr(ht->store, cpu);

251
	if (tsk && !ht->selfparking)
252 253 254
		kthread_park(tsk);
}

255
int smpboot_park_threads(unsigned int cpu)
256 257 258 259 260 261 262
{
	struct smp_hotplug_thread *cur;

	mutex_lock(&smpboot_threads_lock);
	list_for_each_entry_reverse(cur, &hotplug_threads, list)
		smpboot_park_thread(cur, cpu);
	mutex_unlock(&smpboot_threads_lock);
263
	return 0;
264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
}

static void smpboot_destroy_threads(struct smp_hotplug_thread *ht)
{
	unsigned int cpu;

	/* We need to destroy also the parked threads of offline cpus */
	for_each_possible_cpu(cpu) {
		struct task_struct *tsk = *per_cpu_ptr(ht->store, cpu);

		if (tsk) {
			kthread_stop(tsk);
			put_task_struct(tsk);
			*per_cpu_ptr(ht->store, cpu) = NULL;
		}
	}
}

/**
283 284
 * smpboot_register_percpu_thread_cpumask - Register a per_cpu thread related
 * 					    to hotplug
285
 * @plug_thread:	Hotplug thread descriptor
286
 * @cpumask:		The cpumask where threads run
287 288 289
 *
 * Creates and starts the threads on all online cpus.
 */
290 291
int smpboot_register_percpu_thread_cpumask(struct smp_hotplug_thread *plug_thread,
					   const struct cpumask *cpumask)
292 293 294 295
{
	unsigned int cpu;
	int ret = 0;

296 297
	if (!alloc_cpumask_var(&plug_thread->cpumask, GFP_KERNEL))
		return -ENOMEM;
298
	cpumask_copy(plug_thread->cpumask, cpumask);
299

300
	get_online_cpus();
301 302 303 304 305
	mutex_lock(&smpboot_threads_lock);
	for_each_online_cpu(cpu) {
		ret = __smpboot_create_thread(plug_thread, cpu);
		if (ret) {
			smpboot_destroy_threads(plug_thread);
306
			free_cpumask_var(plug_thread->cpumask);
307 308
			goto out;
		}
309 310
		if (cpumask_test_cpu(cpu, cpumask))
			smpboot_unpark_thread(plug_thread, cpu);
311 312 313 314
	}
	list_add(&plug_thread->list, &hotplug_threads);
out:
	mutex_unlock(&smpboot_threads_lock);
315
	put_online_cpus();
316 317
	return ret;
}
318
EXPORT_SYMBOL_GPL(smpboot_register_percpu_thread_cpumask);
319 320 321 322 323 324 325 326 327 328 329 330 331 332 333

/**
 * smpboot_unregister_percpu_thread - Unregister a per_cpu thread related to hotplug
 * @plug_thread:	Hotplug thread descriptor
 *
 * Stops all threads on all possible cpus.
 */
void smpboot_unregister_percpu_thread(struct smp_hotplug_thread *plug_thread)
{
	get_online_cpus();
	mutex_lock(&smpboot_threads_lock);
	list_del(&plug_thread->list);
	smpboot_destroy_threads(plug_thread);
	mutex_unlock(&smpboot_threads_lock);
	put_online_cpus();
334
	free_cpumask_var(plug_thread->cpumask);
335 336
}
EXPORT_SYMBOL_GPL(smpboot_unregister_percpu_thread);
337

338 339 340 341 342 343 344
/**
 * smpboot_update_cpumask_percpu_thread - Adjust which per_cpu hotplug threads stay parked
 * @plug_thread:	Hotplug thread descriptor
 * @new:		Revised mask to use
 *
 * The cpumask field in the smp_hotplug_thread must not be updated directly
 * by the client, but only by calling this function.
345
 * This function can only be called on a registered smp_hotplug_thread.
346
 */
347 348
void smpboot_update_cpumask_percpu_thread(struct smp_hotplug_thread *plug_thread,
					  const struct cpumask *new)
349 350
{
	struct cpumask *old = plug_thread->cpumask;
351
	static struct cpumask tmp;
352 353
	unsigned int cpu;

354
	lockdep_assert_cpus_held();
355 356 357
	mutex_lock(&smpboot_threads_lock);

	/* Park threads that were exclusively enabled on the old mask. */
358 359
	cpumask_andnot(&tmp, old, new);
	for_each_cpu_and(cpu, &tmp, cpu_online_mask)
360 361 362
		smpboot_park_thread(plug_thread, cpu);

	/* Unpark threads that are exclusively enabled on the new mask. */
363 364
	cpumask_andnot(&tmp, new, old);
	for_each_cpu_and(cpu, &tmp, cpu_online_mask)
365 366 367 368 369 370 371
		smpboot_unpark_thread(plug_thread, cpu);

	cpumask_copy(old, new);

	mutex_unlock(&smpboot_threads_lock);
}

372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525
static DEFINE_PER_CPU(atomic_t, cpu_hotplug_state) = ATOMIC_INIT(CPU_POST_DEAD);

/*
 * Called to poll specified CPU's state, for example, when waiting for
 * a CPU to come online.
 */
int cpu_report_state(int cpu)
{
	return atomic_read(&per_cpu(cpu_hotplug_state, cpu));
}

/*
 * If CPU has died properly, set its state to CPU_UP_PREPARE and
 * return success.  Otherwise, return -EBUSY if the CPU died after
 * cpu_wait_death() timed out.  And yet otherwise again, return -EAGAIN
 * if cpu_wait_death() timed out and the CPU still hasn't gotten around
 * to dying.  In the latter two cases, the CPU might not be set up
 * properly, but it is up to the arch-specific code to decide.
 * Finally, -EIO indicates an unanticipated problem.
 *
 * Note that it is permissible to omit this call entirely, as is
 * done in architectures that do no CPU-hotplug error checking.
 */
int cpu_check_up_prepare(int cpu)
{
	if (!IS_ENABLED(CONFIG_HOTPLUG_CPU)) {
		atomic_set(&per_cpu(cpu_hotplug_state, cpu), CPU_UP_PREPARE);
		return 0;
	}

	switch (atomic_read(&per_cpu(cpu_hotplug_state, cpu))) {

	case CPU_POST_DEAD:

		/* The CPU died properly, so just start it up again. */
		atomic_set(&per_cpu(cpu_hotplug_state, cpu), CPU_UP_PREPARE);
		return 0;

	case CPU_DEAD_FROZEN:

		/*
		 * Timeout during CPU death, so let caller know.
		 * The outgoing CPU completed its processing, but after
		 * cpu_wait_death() timed out and reported the error. The
		 * caller is free to proceed, in which case the state
		 * will be reset properly by cpu_set_state_online().
		 * Proceeding despite this -EBUSY return makes sense
		 * for systems where the outgoing CPUs take themselves
		 * offline, with no post-death manipulation required from
		 * a surviving CPU.
		 */
		return -EBUSY;

	case CPU_BROKEN:

		/*
		 * The most likely reason we got here is that there was
		 * a timeout during CPU death, and the outgoing CPU never
		 * did complete its processing.  This could happen on
		 * a virtualized system if the outgoing VCPU gets preempted
		 * for more than five seconds, and the user attempts to
		 * immediately online that same CPU.  Trying again later
		 * might return -EBUSY above, hence -EAGAIN.
		 */
		return -EAGAIN;

	default:

		/* Should not happen.  Famous last words. */
		return -EIO;
	}
}

/*
 * Mark the specified CPU online.
 *
 * Note that it is permissible to omit this call entirely, as is
 * done in architectures that do no CPU-hotplug error checking.
 */
void cpu_set_state_online(int cpu)
{
	(void)atomic_xchg(&per_cpu(cpu_hotplug_state, cpu), CPU_ONLINE);
}

#ifdef CONFIG_HOTPLUG_CPU

/*
 * Wait for the specified CPU to exit the idle loop and die.
 */
bool cpu_wait_death(unsigned int cpu, int seconds)
{
	int jf_left = seconds * HZ;
	int oldstate;
	bool ret = true;
	int sleep_jf = 1;

	might_sleep();

	/* The outgoing CPU will normally get done quite quickly. */
	if (atomic_read(&per_cpu(cpu_hotplug_state, cpu)) == CPU_DEAD)
		goto update_state;
	udelay(5);

	/* But if the outgoing CPU dawdles, wait increasingly long times. */
	while (atomic_read(&per_cpu(cpu_hotplug_state, cpu)) != CPU_DEAD) {
		schedule_timeout_uninterruptible(sleep_jf);
		jf_left -= sleep_jf;
		if (jf_left <= 0)
			break;
		sleep_jf = DIV_ROUND_UP(sleep_jf * 11, 10);
	}
update_state:
	oldstate = atomic_read(&per_cpu(cpu_hotplug_state, cpu));
	if (oldstate == CPU_DEAD) {
		/* Outgoing CPU died normally, update state. */
		smp_mb(); /* atomic_read() before update. */
		atomic_set(&per_cpu(cpu_hotplug_state, cpu), CPU_POST_DEAD);
	} else {
		/* Outgoing CPU still hasn't died, set state accordingly. */
		if (atomic_cmpxchg(&per_cpu(cpu_hotplug_state, cpu),
				   oldstate, CPU_BROKEN) != oldstate)
			goto update_state;
		ret = false;
	}
	return ret;
}

/*
 * Called by the outgoing CPU to report its successful death.  Return
 * false if this report follows the surviving CPU's timing out.
 *
 * A separate "CPU_DEAD_FROZEN" is used when the surviving CPU
 * timed out.  This approach allows architectures to omit calls to
 * cpu_check_up_prepare() and cpu_set_state_online() without defeating
 * the next cpu_wait_death()'s polling loop.
 */
bool cpu_report_death(void)
{
	int oldstate;
	int newstate;
	int cpu = smp_processor_id();

	do {
		oldstate = atomic_read(&per_cpu(cpu_hotplug_state, cpu));
		if (oldstate != CPU_BROKEN)
			newstate = CPU_DEAD;
		else
			newstate = CPU_DEAD_FROZEN;
	} while (atomic_cmpxchg(&per_cpu(cpu_hotplug_state, cpu),
				oldstate, newstate) != oldstate);
	return newstate == CPU_DEAD;
}

#endif /* #ifdef CONFIG_HOTPLUG_CPU */