Blame view
kernel/softirq.c
22.1 KB
1da177e4c
|
1 2 3 4 5 |
/* * linux/kernel/softirq.c * * Copyright (C) 1992 Linus Torvalds * |
b10db7f0d
|
6 7 8 |
* Distribute under GPLv2. * * Rewritten. Old one was good in 2.2, but in 2.3 it was immoral. --ANK (990903) |
54514a70a
|
9 10 |
* * Remote softirq infrastructure is by Jens Axboe. |
1da177e4c
|
11 |
*/ |
9984de1a5
|
12 |
#include <linux/export.h> |
1da177e4c
|
13 14 15 16 17 18 19 |
#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> |
831441862
|
20 |
#include <linux/freezer.h> |
1da177e4c
|
21 22 |
#include <linux/kthread.h> #include <linux/rcupdate.h> |
7e49fcce1
|
23 |
#include <linux/ftrace.h> |
78eef01b0
|
24 |
#include <linux/smp.h> |
79bf2bb33
|
25 |
#include <linux/tick.h> |
a0e39ed37
|
26 27 |
#define CREATE_TRACE_POINTS |
ad8d75fff
|
28 |
#include <trace/events/irq.h> |
1da177e4c
|
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
#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 |
978b0116c
|
53 |
static struct softirq_action softirq_vec[NR_SOFTIRQS] __cacheline_aligned_in_smp; |
1da177e4c
|
54 |
|
4dd53d891
|
55 |
DEFINE_PER_CPU(struct task_struct *, ksoftirqd); |
1da177e4c
|
56 |
|
5d592b44b
|
57 |
char *softirq_to_name[NR_SOFTIRQS] = { |
5dd4de587
|
58 |
"HI", "TIMER", "NET_TX", "NET_RX", "BLOCK", "BLOCK_IOPOLL", |
09223371d
|
59 |
"TASKLET", "SCHED", "HRTIMER", "RCU" |
5d592b44b
|
60 |
}; |
1da177e4c
|
61 62 63 64 65 66 |
/* * 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. */ |
676cb02dc
|
67 |
static void wakeup_softirqd(void) |
1da177e4c
|
68 69 |
{ /* Interrupts are disabled: no need to stop preemption */ |
909ea9646
|
70 |
struct task_struct *tsk = __this_cpu_read(ksoftirqd); |
1da177e4c
|
71 72 73 74 75 76 |
if (tsk && tsk->state != TASK_RUNNING) wake_up_process(tsk); } /* |
75e1056f5
|
77 78 79 80 81 82 83 84 85 86 |
* preempt_count and SOFTIRQ_OFFSET usage: * - preempt_count is changed by SOFTIRQ_OFFSET on entering or leaving * softirq processing. * - preempt_count is changed by SOFTIRQ_DISABLE_OFFSET (= 2 * SOFTIRQ_OFFSET) * on local_bh_disable or local_bh_enable. * This lets us distinguish between whether we are currently processing * softirq and whether we just have bh disabled. */ /* |
de30a2b35
|
87 88 89 |
* This one is for softirq.c-internal use, * where hardirqs are disabled legitimately: */ |
3c829c367
|
90 |
#ifdef CONFIG_TRACE_IRQFLAGS |
75e1056f5
|
91 |
static void __local_bh_disable(unsigned long ip, unsigned int cnt) |
de30a2b35
|
92 93 94 95 96 97 |
{ unsigned long flags; WARN_ON_ONCE(in_irq()); raw_local_irq_save(flags); |
7e49fcce1
|
98 99 100 101 102 103 104 |
/* * 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. */ |
75e1056f5
|
105 |
preempt_count() += cnt; |
de30a2b35
|
106 107 108 |
/* * Were softirqs turned off above: */ |
75e1056f5
|
109 |
if (softirq_count() == cnt) |
de30a2b35
|
110 111 |
trace_softirqs_off(ip); raw_local_irq_restore(flags); |
7e49fcce1
|
112 |
|
75e1056f5
|
113 |
if (preempt_count() == cnt) |
7e49fcce1
|
114 |
trace_preempt_off(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1)); |
de30a2b35
|
115 |
} |
3c829c367
|
116 |
#else /* !CONFIG_TRACE_IRQFLAGS */ |
75e1056f5
|
117 |
static inline void __local_bh_disable(unsigned long ip, unsigned int cnt) |
3c829c367
|
118 |
{ |
75e1056f5
|
119 |
add_preempt_count(cnt); |
3c829c367
|
120 121 122 |
barrier(); } #endif /* CONFIG_TRACE_IRQFLAGS */ |
de30a2b35
|
123 124 125 |
void local_bh_disable(void) { |
75e1056f5
|
126 127 |
__local_bh_disable((unsigned long)__builtin_return_address(0), SOFTIRQ_DISABLE_OFFSET); |
de30a2b35
|
128 129 130 |
} EXPORT_SYMBOL(local_bh_disable); |
75e1056f5
|
131 132 133 134 135 136 137 138 139 |
static void __local_bh_enable(unsigned int cnt) { WARN_ON_ONCE(in_irq()); WARN_ON_ONCE(!irqs_disabled()); if (softirq_count() == cnt) trace_softirqs_on((unsigned long)__builtin_return_address(0)); sub_preempt_count(cnt); } |
de30a2b35
|
140 141 142 143 144 145 146 |
/* * 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) { |
75e1056f5
|
147 |
__local_bh_enable(SOFTIRQ_DISABLE_OFFSET); |
de30a2b35
|
148 149 150 |
} EXPORT_SYMBOL(_local_bh_enable); |
0f476b6d9
|
151 |
static inline void _local_bh_enable_ip(unsigned long ip) |
de30a2b35
|
152 |
{ |
0f476b6d9
|
153 |
WARN_ON_ONCE(in_irq() || irqs_disabled()); |
3c829c367
|
154 |
#ifdef CONFIG_TRACE_IRQFLAGS |
0f476b6d9
|
155 |
local_irq_disable(); |
3c829c367
|
156 |
#endif |
de30a2b35
|
157 158 159 |
/* * Are softirqs going to be turned on now: */ |
75e1056f5
|
160 |
if (softirq_count() == SOFTIRQ_DISABLE_OFFSET) |
0f476b6d9
|
161 |
trace_softirqs_on(ip); |
de30a2b35
|
162 163 164 165 |
/* * Keep preemption disabled until we are done with * softirq processing: */ |
75e1056f5
|
166 |
sub_preempt_count(SOFTIRQ_DISABLE_OFFSET - 1); |
de30a2b35
|
167 168 169 170 171 |
if (unlikely(!in_interrupt() && local_softirq_pending())) do_softirq(); dec_preempt_count(); |
3c829c367
|
172 |
#ifdef CONFIG_TRACE_IRQFLAGS |
0f476b6d9
|
173 |
local_irq_enable(); |
3c829c367
|
174 |
#endif |
de30a2b35
|
175 176 |
preempt_check_resched(); } |
0f476b6d9
|
177 178 179 180 181 |
void local_bh_enable(void) { _local_bh_enable_ip((unsigned long)__builtin_return_address(0)); } |
de30a2b35
|
182 183 184 185 |
EXPORT_SYMBOL(local_bh_enable); void local_bh_enable_ip(unsigned long ip) { |
0f476b6d9
|
186 |
_local_bh_enable_ip(ip); |
de30a2b35
|
187 188 189 190 |
} EXPORT_SYMBOL(local_bh_enable_ip); /* |
1da177e4c
|
191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 |
* 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(); |
829035fd7
|
209 |
account_system_vtime(current); |
75e1056f5
|
210 211 |
__local_bh_disable((unsigned long)__builtin_return_address(0), SOFTIRQ_OFFSET); |
d820ac4c2
|
212 |
lockdep_softirq_enter(); |
1da177e4c
|
213 |
|
1da177e4c
|
214 215 216 |
cpu = smp_processor_id(); restart: /* Reset the pending bitmask before enabling irqs */ |
3f74478b5
|
217 |
set_softirq_pending(0); |
1da177e4c
|
218 |
|
c70f5d661
|
219 |
local_irq_enable(); |
1da177e4c
|
220 221 222 223 224 |
h = softirq_vec; do { if (pending & 1) { |
f4bc6bb2d
|
225 |
unsigned int vec_nr = h - softirq_vec; |
8e85b4b55
|
226 |
int prev_count = preempt_count(); |
f4bc6bb2d
|
227 228 229 |
kstat_incr_softirqs_this_cpu(vec_nr); trace_softirq_entry(vec_nr); |
1da177e4c
|
230 |
h->action(h); |
f4bc6bb2d
|
231 |
trace_softirq_exit(vec_nr); |
8e85b4b55
|
232 |
if (unlikely(prev_count != preempt_count())) { |
f4bc6bb2d
|
233 |
printk(KERN_ERR "huh, entered softirq %u %s %p" |
8e85b4b55
|
234 |
"with preempt_count %08x," |
f4bc6bb2d
|
235 236 237 238 |
" exited with %08x? ", vec_nr, softirq_to_name[vec_nr], h->action, prev_count, preempt_count()); |
8e85b4b55
|
239 240 |
preempt_count() = prev_count; } |
d6714c22b
|
241 |
rcu_bh_qs(cpu); |
1da177e4c
|
242 243 244 245 |
} h++; pending >>= 1; } while (pending); |
c70f5d661
|
246 |
local_irq_disable(); |
1da177e4c
|
247 248 249 250 251 252 253 |
pending = local_softirq_pending(); if (pending && --max_restart) goto restart; if (pending) wakeup_softirqd(); |
d820ac4c2
|
254 |
lockdep_softirq_exit(); |
829035fd7
|
255 256 |
account_system_vtime(current); |
75e1056f5
|
257 |
__local_bh_enable(SOFTIRQ_OFFSET); |
1da177e4c
|
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 |
} #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); } |
1da177e4c
|
279 |
#endif |
dde4b2b5f
|
280 281 282 283 284 |
/* * Enter an interrupt context. */ void irq_enter(void) { |
6378ddb59
|
285 |
int cpu = smp_processor_id(); |
719254faa
|
286 |
|
64db4cfff
|
287 |
rcu_irq_enter(); |
ee5f80a99
|
288 |
if (idle_cpu(cpu) && !in_interrupt()) { |
d267f87fb
|
289 290 291 292 293 |
/* * Prevent raise_softirq from needlessly waking up ksoftirqd * here, as softirq will be serviced on return from interrupt. */ local_bh_disable(); |
719254faa
|
294 |
tick_check_idle(cpu); |
d267f87fb
|
295 296 297 298 |
_local_bh_enable(); } __irq_enter(); |
dde4b2b5f
|
299 |
} |
1da177e4c
|
300 |
#ifdef __ARCH_IRQ_EXIT_IRQS_DISABLED |
8d32a307e
|
301 302 303 304 |
static inline void invoke_softirq(void) { if (!force_irqthreads) __do_softirq(); |
ec433f0c5
|
305 306 307 |
else { __local_bh_disable((unsigned long)__builtin_return_address(0), SOFTIRQ_OFFSET); |
8d32a307e
|
308 |
wakeup_softirqd(); |
ec433f0c5
|
309 310 |
__local_bh_enable(SOFTIRQ_OFFSET); } |
8d32a307e
|
311 |
} |
1da177e4c
|
312 |
#else |
8d32a307e
|
313 314 315 316 |
static inline void invoke_softirq(void) { if (!force_irqthreads) do_softirq(); |
ec433f0c5
|
317 318 319 |
else { __local_bh_disable((unsigned long)__builtin_return_address(0), SOFTIRQ_OFFSET); |
8d32a307e
|
320 |
wakeup_softirqd(); |
ec433f0c5
|
321 322 |
__local_bh_enable(SOFTIRQ_OFFSET); } |
8d32a307e
|
323 |
} |
1da177e4c
|
324 325 326 327 328 329 330 331 |
#endif /* * Exit an interrupt context. Process softirqs if needed and possible: */ void irq_exit(void) { account_system_vtime(current); |
de30a2b35
|
332 |
trace_hardirq_exit(); |
1da177e4c
|
333 334 335 |
sub_preempt_count(IRQ_EXIT_OFFSET); if (!in_interrupt() && local_softirq_pending()) invoke_softirq(); |
79bf2bb33
|
336 |
|
c5e0cb3dd
|
337 |
rcu_irq_exit(); |
79bf2bb33
|
338 339 |
#ifdef CONFIG_NO_HZ /* Make sure that timer wheel updates are propagated */ |
64db4cfff
|
340 341 |
if (idle_cpu(smp_processor_id()) && !in_interrupt() && !need_resched()) tick_nohz_stop_sched_tick(0); |
79bf2bb33
|
342 |
#endif |
1da177e4c
|
343 344 345 346 347 348 |
preempt_enable_no_resched(); } /* * This function must run with irqs disabled! */ |
7ad5b3a50
|
349 |
inline void raise_softirq_irqoff(unsigned int nr) |
1da177e4c
|
350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 |
{ __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(); } |
7ad5b3a50
|
365 |
void raise_softirq(unsigned int nr) |
1da177e4c
|
366 367 368 369 370 371 372 |
{ unsigned long flags; local_irq_save(flags); raise_softirq_irqoff(nr); local_irq_restore(flags); } |
962cf36c5
|
373 |
void open_softirq(int nr, void (*action)(struct softirq_action *)) |
1da177e4c
|
374 |
{ |
1da177e4c
|
375 376 |
softirq_vec[nr].action = action; } |
9ba5f005c
|
377 378 379 |
/* * Tasklets */ |
1da177e4c
|
380 381 |
struct tasklet_head { |
48f20a9a9
|
382 383 |
struct tasklet_struct *head; struct tasklet_struct **tail; |
1da177e4c
|
384 |
}; |
4620b49f7
|
385 386 |
static DEFINE_PER_CPU(struct tasklet_head, tasklet_vec); static DEFINE_PER_CPU(struct tasklet_head, tasklet_hi_vec); |
1da177e4c
|
387 |
|
7ad5b3a50
|
388 |
void __tasklet_schedule(struct tasklet_struct *t) |
1da177e4c
|
389 390 391 392 |
{ unsigned long flags; local_irq_save(flags); |
48f20a9a9
|
393 |
t->next = NULL; |
909ea9646
|
394 395 |
*__this_cpu_read(tasklet_vec.tail) = t; __this_cpu_write(tasklet_vec.tail, &(t->next)); |
1da177e4c
|
396 397 398 399 400 |
raise_softirq_irqoff(TASKLET_SOFTIRQ); local_irq_restore(flags); } EXPORT_SYMBOL(__tasklet_schedule); |
7ad5b3a50
|
401 |
void __tasklet_hi_schedule(struct tasklet_struct *t) |
1da177e4c
|
402 403 404 405 |
{ unsigned long flags; local_irq_save(flags); |
48f20a9a9
|
406 |
t->next = NULL; |
909ea9646
|
407 408 |
*__this_cpu_read(tasklet_hi_vec.tail) = t; __this_cpu_write(tasklet_hi_vec.tail, &(t->next)); |
1da177e4c
|
409 410 411 412 413 |
raise_softirq_irqoff(HI_SOFTIRQ); local_irq_restore(flags); } EXPORT_SYMBOL(__tasklet_hi_schedule); |
7c692cbad
|
414 415 416 |
void __tasklet_hi_schedule_first(struct tasklet_struct *t) { BUG_ON(!irqs_disabled()); |
909ea9646
|
417 418 |
t->next = __this_cpu_read(tasklet_hi_vec.head); __this_cpu_write(tasklet_hi_vec.head, t); |
7c692cbad
|
419 420 421 422 |
__raise_softirq_irqoff(HI_SOFTIRQ); } EXPORT_SYMBOL(__tasklet_hi_schedule_first); |
1da177e4c
|
423 424 425 426 427 |
static void tasklet_action(struct softirq_action *a) { struct tasklet_struct *list; local_irq_disable(); |
909ea9646
|
428 429 430 |
list = __this_cpu_read(tasklet_vec.head); __this_cpu_write(tasklet_vec.head, NULL); __this_cpu_write(tasklet_vec.tail, &__get_cpu_var(tasklet_vec).head); |
1da177e4c
|
431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 |
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(); |
48f20a9a9
|
450 |
t->next = NULL; |
909ea9646
|
451 452 |
*__this_cpu_read(tasklet_vec.tail) = t; __this_cpu_write(tasklet_vec.tail, &(t->next)); |
1da177e4c
|
453 454 455 456 457 458 459 460 461 462 |
__raise_softirq_irqoff(TASKLET_SOFTIRQ); local_irq_enable(); } } static void tasklet_hi_action(struct softirq_action *a) { struct tasklet_struct *list; local_irq_disable(); |
909ea9646
|
463 464 465 |
list = __this_cpu_read(tasklet_hi_vec.head); __this_cpu_write(tasklet_hi_vec.head, NULL); __this_cpu_write(tasklet_hi_vec.tail, &__get_cpu_var(tasklet_hi_vec).head); |
1da177e4c
|
466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 |
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(); |
48f20a9a9
|
485 |
t->next = NULL; |
909ea9646
|
486 487 |
*__this_cpu_read(tasklet_hi_vec.tail) = t; __this_cpu_write(tasklet_hi_vec.tail, &(t->next)); |
1da177e4c
|
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 |
__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 "); while (test_and_set_bit(TASKLET_STATE_SCHED, &t->state)) { |
79d381c9f
|
513 |
do { |
1da177e4c
|
514 |
yield(); |
79d381c9f
|
515 |
} while (test_bit(TASKLET_STATE_SCHED, &t->state)); |
1da177e4c
|
516 517 518 519 520 521 |
} tasklet_unlock_wait(t); clear_bit(TASKLET_STATE_SCHED, &t->state); } EXPORT_SYMBOL(tasklet_kill); |
9ba5f005c
|
522 523 524 525 526 |
/* * tasklet_hrtimer */ /* |
b9c303227
|
527 528 529 |
* The trampoline is called when the hrtimer expires. It schedules a tasklet * to run __tasklet_hrtimer_trampoline() which in turn will call the intended * hrtimer callback, but from softirq context. |
9ba5f005c
|
530 531 532 533 534 |
*/ static enum hrtimer_restart __hrtimer_tasklet_trampoline(struct hrtimer *timer) { struct tasklet_hrtimer *ttimer = container_of(timer, struct tasklet_hrtimer, timer); |
b9c303227
|
535 536 |
tasklet_hi_schedule(&ttimer->tasklet); return HRTIMER_NORESTART; |
9ba5f005c
|
537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 |
} /* * 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 |
25985edce
|
556 |
* @function: hrtimer callback function which gets called from softirq context |
9ba5f005c
|
557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 |
* @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 */ |
54514a70a
|
575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 |
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; |
6e2756376
|
610 |
__smp_call_function_single(cpu, cp, 0); |
54514a70a
|
611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 |
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, }; |
1da177e4c
|
694 695 |
void __init softirq_init(void) { |
48f20a9a9
|
696 697 698 |
int cpu; for_each_possible_cpu(cpu) { |
54514a70a
|
699 |
int i; |
48f20a9a9
|
700 701 702 703 |
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; |
54514a70a
|
704 705 |
for (i = 0; i < NR_SOFTIRQS; i++) INIT_LIST_HEAD(&per_cpu(softirq_work_list[i], cpu)); |
48f20a9a9
|
706 |
} |
54514a70a
|
707 |
register_hotcpu_notifier(&remote_softirq_cpu_notifier); |
962cf36c5
|
708 709 |
open_softirq(TASKLET_SOFTIRQ, tasklet_action); open_softirq(HI_SOFTIRQ, tasklet_hi_action); |
1da177e4c
|
710 |
} |
1871e52c7
|
711 |
static int run_ksoftirqd(void * __bind_cpu) |
1da177e4c
|
712 |
{ |
1da177e4c
|
713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 |
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; |
c305d524e
|
731 732 733 734 |
local_irq_disable(); if (local_softirq_pending()) __do_softirq(); local_irq_enable(); |
1da177e4c
|
735 736 737 |
preempt_enable_no_resched(); cond_resched(); preempt_disable(); |
25502a6c1
|
738 |
rcu_note_context_switch((long)__bind_cpu); |
1da177e4c
|
739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 |
} 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. */ |
48f20a9a9
|
779 |
for (i = &per_cpu(tasklet_vec, cpu).head; *i; i = &(*i)->next) { |
1da177e4c
|
780 781 |
if (*i == t) { *i = t->next; |
48f20a9a9
|
782 783 784 |
/* If this was the tail element, move the tail ptr */ if (*i == NULL) per_cpu(tasklet_vec, cpu).tail = i; |
1da177e4c
|
785 786 787 788 789 790 791 792 |
return; } } BUG(); } static void takeover_tasklets(unsigned int cpu) { |
1da177e4c
|
793 794 795 796 |
/* CPU is dead, so no lock needed. */ local_irq_disable(); /* Find end, append list for that CPU. */ |
e5e417232
|
797 |
if (&per_cpu(tasklet_vec, cpu).head != per_cpu(tasklet_vec, cpu).tail) { |
909ea9646
|
798 799 |
*__this_cpu_read(tasklet_vec.tail) = per_cpu(tasklet_vec, cpu).head; this_cpu_write(tasklet_vec.tail, per_cpu(tasklet_vec, cpu).tail); |
e5e417232
|
800 801 802 |
per_cpu(tasklet_vec, cpu).head = NULL; per_cpu(tasklet_vec, cpu).tail = &per_cpu(tasklet_vec, cpu).head; } |
1da177e4c
|
803 |
raise_softirq_irqoff(TASKLET_SOFTIRQ); |
e5e417232
|
804 |
if (&per_cpu(tasklet_hi_vec, cpu).head != per_cpu(tasklet_hi_vec, cpu).tail) { |
909ea9646
|
805 806 |
*__this_cpu_read(tasklet_hi_vec.tail) = per_cpu(tasklet_hi_vec, cpu).head; __this_cpu_write(tasklet_hi_vec.tail, per_cpu(tasklet_hi_vec, cpu).tail); |
e5e417232
|
807 808 809 |
per_cpu(tasklet_hi_vec, cpu).head = NULL; per_cpu(tasklet_hi_vec, cpu).tail = &per_cpu(tasklet_hi_vec, cpu).head; } |
1da177e4c
|
810 811 812 813 814 |
raise_softirq_irqoff(HI_SOFTIRQ); local_irq_enable(); } #endif /* CONFIG_HOTPLUG_CPU */ |
8c78f3075
|
815 |
static int __cpuinit cpu_callback(struct notifier_block *nfb, |
1da177e4c
|
816 817 818 819 820 821 822 823 |
unsigned long action, void *hcpu) { int hotcpu = (unsigned long)hcpu; struct task_struct *p; switch (action) { case CPU_UP_PREPARE: |
8bb784428
|
824 |
case CPU_UP_PREPARE_FROZEN: |
94dcf29a1
|
825 826 827 828 |
p = kthread_create_on_node(run_ksoftirqd, hcpu, cpu_to_node(hotcpu), "ksoftirqd/%d", hotcpu); |
1da177e4c
|
829 830 831 |
if (IS_ERR(p)) { printk("ksoftirqd for %i failed ", hotcpu); |
80b5184cc
|
832 |
return notifier_from_errno(PTR_ERR(p)); |
1da177e4c
|
833 834 835 836 837 |
} kthread_bind(p, hotcpu); per_cpu(ksoftirqd, hotcpu) = p; break; case CPU_ONLINE: |
8bb784428
|
838 |
case CPU_ONLINE_FROZEN: |
1da177e4c
|
839 840 841 842 |
wake_up_process(per_cpu(ksoftirqd, hotcpu)); break; #ifdef CONFIG_HOTPLUG_CPU case CPU_UP_CANCELED: |
8bb784428
|
843 |
case CPU_UP_CANCELED_FROZEN: |
fc75cdfa5
|
844 845 |
if (!per_cpu(ksoftirqd, hotcpu)) break; |
1da177e4c
|
846 |
/* Unbind so it can run. Fall thru. */ |
a4c4af7c8
|
847 |
kthread_bind(per_cpu(ksoftirqd, hotcpu), |
f1fc057c7
|
848 |
cpumask_any(cpu_online_mask)); |
1da177e4c
|
849 |
case CPU_DEAD: |
1c6b4aa94
|
850 |
case CPU_DEAD_FROZEN: { |
c9b5f501e
|
851 |
static const struct sched_param param = { |
fe7de49f9
|
852 853 |
.sched_priority = MAX_RT_PRIO-1 }; |
1c6b4aa94
|
854 |
|
1da177e4c
|
855 856 |
p = per_cpu(ksoftirqd, hotcpu); per_cpu(ksoftirqd, hotcpu) = NULL; |
961ccddd5
|
857 |
sched_setscheduler_nocheck(p, SCHED_FIFO, ¶m); |
1da177e4c
|
858 859 860 |
kthread_stop(p); takeover_tasklets(hotcpu); break; |
1c6b4aa94
|
861 |
} |
1da177e4c
|
862 863 864 865 |
#endif /* CONFIG_HOTPLUG_CPU */ } return NOTIFY_OK; } |
8c78f3075
|
866 |
static struct notifier_block __cpuinitdata cpu_nfb = { |
1da177e4c
|
867 868 |
.notifier_call = cpu_callback }; |
7babe8db9
|
869 |
static __init int spawn_ksoftirqd(void) |
1da177e4c
|
870 871 |
{ void *cpu = (void *)(long)smp_processor_id(); |
07dccf334
|
872 |
int err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu); |
9e506f7ad
|
873 |
BUG_ON(err != NOTIFY_OK); |
1da177e4c
|
874 875 876 877 |
cpu_callback(&cpu_nfb, CPU_ONLINE, cpu); register_cpu_notifier(&cpu_nfb); return 0; } |
7babe8db9
|
878 |
early_initcall(spawn_ksoftirqd); |
78eef01b0
|
879 |
|
43a256322
|
880 881 882 883 884 885 886 887 888 |
/* * [ 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; } |
b683de2b3
|
889 |
#ifdef CONFIG_GENERIC_HARDIRQS |
4a046d175
|
890 891 |
int __init __weak arch_probe_nr_irqs(void) { |
b683de2b3
|
892 |
return NR_IRQS_LEGACY; |
4a046d175
|
893 |
} |
43a256322
|
894 895 896 897 |
int __init __weak arch_early_irq_init(void) { return 0; } |
b683de2b3
|
898 |
#endif |