Commit | Line | Data |
---|---|---|
08b21fbf | 1 | // SPDX-License-Identifier: GPL-2.0 |
16295bec SK |
2 | /* |
3 | * padata.c - generic interface to process data streams in parallel | |
4 | * | |
bfcdcef8 | 5 | * See Documentation/core-api/padata.rst for more information. |
107f8bda | 6 | * |
16295bec SK |
7 | * Copyright (C) 2008, 2009 secunet Security Networks AG |
8 | * Copyright (C) 2008, 2009 Steffen Klassert <steffen.klassert@secunet.com> | |
9 | * | |
004ed426 DJ |
10 | * Copyright (c) 2020 Oracle and/or its affiliates. |
11 | * Author: Daniel Jordan <daniel.m.jordan@oracle.com> | |
16295bec SK |
12 | */ |
13 | ||
004ed426 | 14 | #include <linux/completion.h> |
9984de1a | 15 | #include <linux/export.h> |
16295bec SK |
16 | #include <linux/cpumask.h> |
17 | #include <linux/err.h> | |
18 | #include <linux/cpu.h> | |
19 | #include <linux/padata.h> | |
20 | #include <linux/mutex.h> | |
21 | #include <linux/sched.h> | |
5a0e3ad6 | 22 | #include <linux/slab.h> |
5e017dc3 | 23 | #include <linux/sysfs.h> |
16295bec SK |
24 | #include <linux/rcupdate.h> |
25 | ||
004ed426 DJ |
26 | #define PADATA_WORK_ONSTACK 1 /* Work's memory is on stack */ |
27 | ||
4611ce22 DJ |
28 | struct padata_work { |
29 | struct work_struct pw_work; | |
30 | struct list_head pw_list; /* padata_free_works linkage */ | |
31 | void *pw_data; | |
32 | }; | |
33 | ||
34 | static DEFINE_SPINLOCK(padata_works_lock); | |
35 | static struct padata_work *padata_works; | |
36 | static LIST_HEAD(padata_free_works); | |
16295bec | 37 | |
004ed426 DJ |
38 | struct padata_mt_job_state { |
39 | spinlock_t lock; | |
40 | struct completion completion; | |
41 | struct padata_mt_job *job; | |
42 | int nworks; | |
43 | int nworks_fini; | |
44 | unsigned long chunk_size; | |
45 | }; | |
46 | ||
07928d9b | 47 | static void padata_free_pd(struct parallel_data *pd); |
004ed426 | 48 | static void __init padata_mt_helper(struct work_struct *work); |
07928d9b | 49 | |
16295bec SK |
50 | static int padata_index_to_cpu(struct parallel_data *pd, int cpu_index) |
51 | { | |
52 | int cpu, target_cpu; | |
53 | ||
e15bacbe | 54 | target_cpu = cpumask_first(pd->cpumask.pcpu); |
16295bec | 55 | for (cpu = 0; cpu < cpu_index; cpu++) |
e15bacbe | 56 | target_cpu = cpumask_next(target_cpu, pd->cpumask.pcpu); |
16295bec SK |
57 | |
58 | return target_cpu; | |
59 | } | |
60 | ||
bfde23ce | 61 | static int padata_cpu_hash(struct parallel_data *pd, unsigned int seq_nr) |
16295bec | 62 | { |
16295bec SK |
63 | /* |
64 | * Hash the sequence numbers to the cpus by taking | |
65 | * seq_nr mod. number of cpus in use. | |
66 | */ | |
bfde23ce | 67 | int cpu_index = seq_nr % cpumask_weight(pd->cpumask.pcpu); |
16295bec SK |
68 | |
69 | return padata_index_to_cpu(pd, cpu_index); | |
70 | } | |
71 | ||
4611ce22 | 72 | static struct padata_work *padata_work_alloc(void) |
16295bec | 73 | { |
4611ce22 | 74 | struct padata_work *pw; |
16295bec | 75 | |
4611ce22 | 76 | lockdep_assert_held(&padata_works_lock); |
16295bec | 77 | |
4611ce22 DJ |
78 | if (list_empty(&padata_free_works)) |
79 | return NULL; /* No more work items allowed to be queued. */ | |
16295bec | 80 | |
4611ce22 DJ |
81 | pw = list_first_entry(&padata_free_works, struct padata_work, pw_list); |
82 | list_del(&pw->pw_list); | |
83 | return pw; | |
84 | } | |
16295bec | 85 | |
4611ce22 | 86 | static void padata_work_init(struct padata_work *pw, work_func_t work_fn, |
004ed426 | 87 | void *data, int flags) |
4611ce22 | 88 | { |
004ed426 DJ |
89 | if (flags & PADATA_WORK_ONSTACK) |
90 | INIT_WORK_ONSTACK(&pw->pw_work, work_fn); | |
91 | else | |
92 | INIT_WORK(&pw->pw_work, work_fn); | |
4611ce22 DJ |
93 | pw->pw_data = data; |
94 | } | |
16295bec | 95 | |
004ed426 DJ |
96 | static int __init padata_work_alloc_mt(int nworks, void *data, |
97 | struct list_head *head) | |
98 | { | |
99 | int i; | |
100 | ||
101 | spin_lock(&padata_works_lock); | |
102 | /* Start at 1 because the current task participates in the job. */ | |
103 | for (i = 1; i < nworks; ++i) { | |
104 | struct padata_work *pw = padata_work_alloc(); | |
105 | ||
106 | if (!pw) | |
107 | break; | |
108 | padata_work_init(pw, padata_mt_helper, data, 0); | |
109 | list_add(&pw->pw_list, head); | |
110 | } | |
111 | spin_unlock(&padata_works_lock); | |
112 | ||
113 | return i; | |
114 | } | |
115 | ||
4611ce22 DJ |
116 | static void padata_work_free(struct padata_work *pw) |
117 | { | |
118 | lockdep_assert_held(&padata_works_lock); | |
119 | list_add(&pw->pw_list, &padata_free_works); | |
120 | } | |
16295bec | 121 | |
004ed426 DJ |
122 | static void __init padata_works_free(struct list_head *works) |
123 | { | |
124 | struct padata_work *cur, *next; | |
125 | ||
126 | if (list_empty(works)) | |
127 | return; | |
128 | ||
129 | spin_lock(&padata_works_lock); | |
130 | list_for_each_entry_safe(cur, next, works, pw_list) { | |
131 | list_del(&cur->pw_list); | |
132 | padata_work_free(cur); | |
133 | } | |
134 | spin_unlock(&padata_works_lock); | |
135 | } | |
136 | ||
4611ce22 DJ |
137 | static void padata_parallel_worker(struct work_struct *parallel_work) |
138 | { | |
139 | struct padata_work *pw = container_of(parallel_work, struct padata_work, | |
140 | pw_work); | |
141 | struct padata_priv *padata = pw->pw_data; | |
16295bec | 142 | |
4611ce22 DJ |
143 | local_bh_disable(); |
144 | padata->parallel(padata); | |
145 | spin_lock(&padata_works_lock); | |
146 | padata_work_free(pw); | |
147 | spin_unlock(&padata_works_lock); | |
16295bec SK |
148 | local_bh_enable(); |
149 | } | |
150 | ||
0198ffd1 | 151 | /** |
16295bec SK |
152 | * padata_do_parallel - padata parallelization function |
153 | * | |
bbefa1dd | 154 | * @ps: padatashell |
16295bec | 155 | * @padata: object to be parallelized |
e6ce0e08 DJ |
156 | * @cb_cpu: pointer to the CPU that the serialization callback function should |
157 | * run on. If it's not in the serial cpumask of @pinst | |
158 | * (i.e. cpumask.cbcpu), this function selects a fallback CPU and if | |
159 | * none found, returns -EINVAL. | |
16295bec SK |
160 | * |
161 | * The parallelization callback function will run with BHs off. | |
162 | * Note: Every object which is parallelized by padata_do_parallel | |
163 | * must be seen by padata_do_serial. | |
bfcdcef8 DJ |
164 | * |
165 | * Return: 0 on success or else negative error code. | |
16295bec | 166 | */ |
bbefa1dd | 167 | int padata_do_parallel(struct padata_shell *ps, |
e6ce0e08 | 168 | struct padata_priv *padata, int *cb_cpu) |
16295bec | 169 | { |
bbefa1dd | 170 | struct padata_instance *pinst = ps->pinst; |
4611ce22 | 171 | int i, cpu, cpu_index, err; |
16295bec | 172 | struct parallel_data *pd; |
4611ce22 | 173 | struct padata_work *pw; |
16295bec SK |
174 | |
175 | rcu_read_lock_bh(); | |
176 | ||
bbefa1dd | 177 | pd = rcu_dereference_bh(ps->pd); |
16295bec | 178 | |
83f619f3 | 179 | err = -EINVAL; |
7424713b | 180 | if (!(pinst->flags & PADATA_INIT) || pinst->flags & PADATA_INVALID) |
16295bec SK |
181 | goto out; |
182 | ||
e6ce0e08 DJ |
183 | if (!cpumask_test_cpu(*cb_cpu, pd->cpumask.cbcpu)) { |
184 | if (!cpumask_weight(pd->cpumask.cbcpu)) | |
185 | goto out; | |
186 | ||
187 | /* Select an alternate fallback CPU and notify the caller. */ | |
188 | cpu_index = *cb_cpu % cpumask_weight(pd->cpumask.cbcpu); | |
189 | ||
190 | cpu = cpumask_first(pd->cpumask.cbcpu); | |
191 | for (i = 0; i < cpu_index; i++) | |
192 | cpu = cpumask_next(cpu, pd->cpumask.cbcpu); | |
193 | ||
194 | *cb_cpu = cpu; | |
195 | } | |
16295bec SK |
196 | |
197 | err = -EBUSY; | |
198 | if ((pinst->flags & PADATA_RESET)) | |
199 | goto out; | |
200 | ||
d5ee8e75 | 201 | refcount_inc(&pd->refcnt); |
16295bec | 202 | padata->pd = pd; |
e6ce0e08 | 203 | padata->cb_cpu = *cb_cpu; |
16295bec | 204 | |
4611ce22 DJ |
205 | spin_lock(&padata_works_lock); |
206 | padata->seq_nr = ++pd->seq_nr; | |
207 | pw = padata_work_alloc(); | |
208 | spin_unlock(&padata_works_lock); | |
1b0df11f DJ |
209 | |
210 | rcu_read_unlock_bh(); | |
211 | ||
4611ce22 | 212 | if (pw) { |
004ed426 | 213 | padata_work_init(pw, padata_parallel_worker, padata, 0); |
4611ce22 DJ |
214 | queue_work(pinst->parallel_wq, &pw->pw_work); |
215 | } else { | |
216 | /* Maximum works limit exceeded, run in the current task. */ | |
217 | padata->parallel(padata); | |
218 | } | |
16295bec | 219 | |
4611ce22 | 220 | return 0; |
16295bec SK |
221 | out: |
222 | rcu_read_unlock_bh(); | |
223 | ||
224 | return err; | |
225 | } | |
226 | EXPORT_SYMBOL(padata_do_parallel); | |
227 | ||
0198ffd1 | 228 | /* |
bfde23ce | 229 | * padata_find_next - Find the next object that needs serialization. |
0198ffd1 | 230 | * |
bfcdcef8 DJ |
231 | * Return: |
232 | * * A pointer to the control struct of the next object that needs | |
233 | * serialization, if present in one of the percpu reorder queues. | |
234 | * * NULL, if the next object that needs serialization will | |
235 | * be parallel processed by another cpu and is not yet present in | |
236 | * the cpu's reorder queue. | |
0198ffd1 | 237 | */ |
bfde23ce DJ |
238 | static struct padata_priv *padata_find_next(struct parallel_data *pd, |
239 | bool remove_object) | |
16295bec | 240 | { |
16295bec SK |
241 | struct padata_priv *padata; |
242 | struct padata_list *reorder; | |
6fc4dbcf | 243 | int cpu = pd->cpu; |
16295bec | 244 | |
f601c725 | 245 | reorder = per_cpu_ptr(pd->reorder_list, cpu); |
16295bec | 246 | |
de5540d0 | 247 | spin_lock(&reorder->lock); |
bfde23ce DJ |
248 | if (list_empty(&reorder->list)) { |
249 | spin_unlock(&reorder->lock); | |
250 | return NULL; | |
251 | } | |
16295bec | 252 | |
bfde23ce | 253 | padata = list_entry(reorder->list.next, struct padata_priv, list); |
16295bec | 254 | |
bfde23ce DJ |
255 | /* |
256 | * Checks the rare case where two or more parallel jobs have hashed to | |
257 | * the same CPU and one of the later ones finishes first. | |
258 | */ | |
259 | if (padata->seq_nr != pd->processed) { | |
de5540d0 | 260 | spin_unlock(&reorder->lock); |
bfde23ce | 261 | return NULL; |
16295bec SK |
262 | } |
263 | ||
bfde23ce DJ |
264 | if (remove_object) { |
265 | list_del_init(&padata->list); | |
bfde23ce DJ |
266 | ++pd->processed; |
267 | pd->cpu = cpumask_next_wrap(cpu, pd->cpumask.pcpu, -1, false); | |
16295bec SK |
268 | } |
269 | ||
bfde23ce | 270 | spin_unlock(&reorder->lock); |
16295bec SK |
271 | return padata; |
272 | } | |
273 | ||
274 | static void padata_reorder(struct parallel_data *pd) | |
275 | { | |
bbefa1dd | 276 | struct padata_instance *pinst = pd->ps->pinst; |
3047817b | 277 | int cb_cpu; |
16295bec | 278 | struct padata_priv *padata; |
e15bacbe | 279 | struct padata_serial_queue *squeue; |
f601c725 | 280 | struct padata_list *reorder; |
16295bec | 281 | |
0198ffd1 SK |
282 | /* |
283 | * We need to ensure that only one cpu can work on dequeueing of | |
284 | * the reorder queue the time. Calculating in which percpu reorder | |
285 | * queue the next object will arrive takes some time. A spinlock | |
286 | * would be highly contended. Also it is not clear in which order | |
287 | * the objects arrive to the reorder queues. So a cpu could wait to | |
288 | * get the lock just to notice that there is nothing to do at the | |
289 | * moment. Therefore we use a trylock and let the holder of the lock | |
290 | * care for all the objects enqueued during the holdtime of the lock. | |
291 | */ | |
16295bec | 292 | if (!spin_trylock_bh(&pd->lock)) |
d46a5ac7 | 293 | return; |
16295bec SK |
294 | |
295 | while (1) { | |
bfde23ce | 296 | padata = padata_find_next(pd, true); |
16295bec | 297 | |
0198ffd1 | 298 | /* |
69b34844 JD |
299 | * If the next object that needs serialization is parallel |
300 | * processed by another cpu and is still on it's way to the | |
301 | * cpu's reorder queue, nothing to do for now. | |
0198ffd1 | 302 | */ |
bfde23ce | 303 | if (!padata) |
16295bec SK |
304 | break; |
305 | ||
3047817b SK |
306 | cb_cpu = padata->cb_cpu; |
307 | squeue = per_cpu_ptr(pd->squeue, cb_cpu); | |
16295bec | 308 | |
e15bacbe DK |
309 | spin_lock(&squeue->serial.lock); |
310 | list_add_tail(&padata->list, &squeue->serial.list); | |
311 | spin_unlock(&squeue->serial.lock); | |
16295bec | 312 | |
45d153c0 | 313 | queue_work_on(cb_cpu, pinst->serial_wq, &squeue->work); |
16295bec SK |
314 | } |
315 | ||
316 | spin_unlock_bh(&pd->lock); | |
317 | ||
0198ffd1 SK |
318 | /* |
319 | * The next object that needs serialization might have arrived to | |
6fc4dbcf | 320 | * the reorder queues in the meantime. |
cf144f81 | 321 | * |
6fc4dbcf HX |
322 | * Ensure reorder queue is read after pd->lock is dropped so we see |
323 | * new objects from another task in padata_do_serial. Pairs with | |
e04ec0de | 324 | * smp_mb in padata_do_serial. |
0198ffd1 | 325 | */ |
cf144f81 | 326 | smp_mb(); |
16295bec | 327 | |
f601c725 DJ |
328 | reorder = per_cpu_ptr(pd->reorder_list, pd->cpu); |
329 | if (!list_empty(&reorder->list) && padata_find_next(pd, false)) | |
45d153c0 | 330 | queue_work(pinst->serial_wq, &pd->reorder_work); |
16295bec SK |
331 | } |
332 | ||
cf5868c8 MK |
333 | static void invoke_padata_reorder(struct work_struct *work) |
334 | { | |
cf5868c8 MK |
335 | struct parallel_data *pd; |
336 | ||
337 | local_bh_disable(); | |
6fc4dbcf | 338 | pd = container_of(work, struct parallel_data, reorder_work); |
cf5868c8 MK |
339 | padata_reorder(pd); |
340 | local_bh_enable(); | |
341 | } | |
342 | ||
e15bacbe | 343 | static void padata_serial_worker(struct work_struct *serial_work) |
16295bec | 344 | { |
e15bacbe | 345 | struct padata_serial_queue *squeue; |
16295bec SK |
346 | struct parallel_data *pd; |
347 | LIST_HEAD(local_list); | |
07928d9b | 348 | int cnt; |
16295bec SK |
349 | |
350 | local_bh_disable(); | |
e15bacbe DK |
351 | squeue = container_of(serial_work, struct padata_serial_queue, work); |
352 | pd = squeue->pd; | |
16295bec | 353 | |
e15bacbe DK |
354 | spin_lock(&squeue->serial.lock); |
355 | list_replace_init(&squeue->serial.list, &local_list); | |
356 | spin_unlock(&squeue->serial.lock); | |
16295bec | 357 | |
07928d9b HX |
358 | cnt = 0; |
359 | ||
16295bec SK |
360 | while (!list_empty(&local_list)) { |
361 | struct padata_priv *padata; | |
362 | ||
363 | padata = list_entry(local_list.next, | |
364 | struct padata_priv, list); | |
365 | ||
366 | list_del_init(&padata->list); | |
367 | ||
368 | padata->serial(padata); | |
07928d9b | 369 | cnt++; |
16295bec SK |
370 | } |
371 | local_bh_enable(); | |
07928d9b | 372 | |
d5ee8e75 | 373 | if (refcount_sub_and_test(cnt, &pd->refcnt)) |
07928d9b | 374 | padata_free_pd(pd); |
16295bec SK |
375 | } |
376 | ||
0198ffd1 | 377 | /** |
16295bec SK |
378 | * padata_do_serial - padata serialization function |
379 | * | |
380 | * @padata: object to be serialized. | |
381 | * | |
382 | * padata_do_serial must be called for every parallelized object. | |
383 | * The serialization callback function will run with BHs off. | |
384 | */ | |
385 | void padata_do_serial(struct padata_priv *padata) | |
386 | { | |
065cf577 | 387 | struct parallel_data *pd = padata->pd; |
4611ce22 | 388 | int hashed_cpu = padata_cpu_hash(pd, padata->seq_nr); |
f601c725 | 389 | struct padata_list *reorder = per_cpu_ptr(pd->reorder_list, hashed_cpu); |
bfde23ce | 390 | struct padata_priv *cur; |
16295bec | 391 | |
f601c725 | 392 | spin_lock(&reorder->lock); |
bfde23ce | 393 | /* Sort in ascending order of sequence number. */ |
f601c725 | 394 | list_for_each_entry_reverse(cur, &reorder->list, list) |
bfde23ce DJ |
395 | if (cur->seq_nr < padata->seq_nr) |
396 | break; | |
397 | list_add(&padata->list, &cur->list); | |
f601c725 | 398 | spin_unlock(&reorder->lock); |
16295bec | 399 | |
cf144f81 | 400 | /* |
6fc4dbcf | 401 | * Ensure the addition to the reorder list is ordered correctly |
cf144f81 DJ |
402 | * with the trylock of pd->lock in padata_reorder. Pairs with smp_mb |
403 | * in padata_reorder. | |
404 | */ | |
e04ec0de | 405 | smp_mb(); |
cf144f81 | 406 | |
6fc4dbcf | 407 | padata_reorder(pd); |
16295bec SK |
408 | } |
409 | EXPORT_SYMBOL(padata_do_serial); | |
410 | ||
bbefa1dd | 411 | static int padata_setup_cpumasks(struct padata_instance *pinst) |
16295bec | 412 | { |
bfde23ce | 413 | struct workqueue_attrs *attrs; |
bbefa1dd HX |
414 | int err; |
415 | ||
416 | attrs = alloc_workqueue_attrs(); | |
417 | if (!attrs) | |
418 | return -ENOMEM; | |
419 | ||
420 | /* Restrict parallel_wq workers to pd->cpumask.pcpu. */ | |
421 | cpumask_copy(attrs->cpumask, pinst->cpumask.pcpu); | |
422 | err = apply_workqueue_attrs(pinst->parallel_wq, attrs); | |
423 | free_workqueue_attrs(attrs); | |
424 | ||
425 | return err; | |
426 | } | |
427 | ||
004ed426 DJ |
428 | static void __init padata_mt_helper(struct work_struct *w) |
429 | { | |
430 | struct padata_work *pw = container_of(w, struct padata_work, pw_work); | |
431 | struct padata_mt_job_state *ps = pw->pw_data; | |
432 | struct padata_mt_job *job = ps->job; | |
433 | bool done; | |
434 | ||
435 | spin_lock(&ps->lock); | |
436 | ||
437 | while (job->size > 0) { | |
438 | unsigned long start, size, end; | |
439 | ||
440 | start = job->start; | |
441 | /* So end is chunk size aligned if enough work remains. */ | |
442 | size = roundup(start + 1, ps->chunk_size) - start; | |
443 | size = min(size, job->size); | |
444 | end = start + size; | |
445 | ||
446 | job->start = end; | |
447 | job->size -= size; | |
448 | ||
449 | spin_unlock(&ps->lock); | |
450 | job->thread_fn(start, end, job->fn_arg); | |
451 | spin_lock(&ps->lock); | |
452 | } | |
453 | ||
454 | ++ps->nworks_fini; | |
455 | done = (ps->nworks_fini == ps->nworks); | |
456 | spin_unlock(&ps->lock); | |
457 | ||
458 | if (done) | |
459 | complete(&ps->completion); | |
460 | } | |
461 | ||
462 | /** | |
463 | * padata_do_multithreaded - run a multithreaded job | |
464 | * @job: Description of the job. | |
465 | * | |
466 | * See the definition of struct padata_mt_job for more details. | |
467 | */ | |
468 | void __init padata_do_multithreaded(struct padata_mt_job *job) | |
469 | { | |
470 | /* In case threads finish at different times. */ | |
471 | static const unsigned long load_balance_factor = 4; | |
472 | struct padata_work my_work, *pw; | |
473 | struct padata_mt_job_state ps; | |
474 | LIST_HEAD(works); | |
475 | int nworks; | |
476 | ||
477 | if (job->size == 0) | |
478 | return; | |
479 | ||
480 | /* Ensure at least one thread when size < min_chunk. */ | |
481 | nworks = max(job->size / job->min_chunk, 1ul); | |
482 | nworks = min(nworks, job->max_threads); | |
483 | ||
484 | if (nworks == 1) { | |
485 | /* Single thread, no coordination needed, cut to the chase. */ | |
486 | job->thread_fn(job->start, job->start + job->size, job->fn_arg); | |
487 | return; | |
488 | } | |
489 | ||
490 | spin_lock_init(&ps.lock); | |
491 | init_completion(&ps.completion); | |
492 | ps.job = job; | |
493 | ps.nworks = padata_work_alloc_mt(nworks, &ps, &works); | |
494 | ps.nworks_fini = 0; | |
495 | ||
496 | /* | |
497 | * Chunk size is the amount of work a helper does per call to the | |
498 | * thread function. Load balance large jobs between threads by | |
499 | * increasing the number of chunks, guarantee at least the minimum | |
500 | * chunk size from the caller, and honor the caller's alignment. | |
501 | */ | |
502 | ps.chunk_size = job->size / (ps.nworks * load_balance_factor); | |
503 | ps.chunk_size = max(ps.chunk_size, job->min_chunk); | |
504 | ps.chunk_size = roundup(ps.chunk_size, job->align); | |
505 | ||
506 | list_for_each_entry(pw, &works, pw_list) | |
507 | queue_work(system_unbound_wq, &pw->pw_work); | |
508 | ||
509 | /* Use the current thread, which saves starting a workqueue worker. */ | |
510 | padata_work_init(&my_work, padata_mt_helper, &ps, PADATA_WORK_ONSTACK); | |
511 | padata_mt_helper(&my_work.pw_work); | |
512 | ||
513 | /* Wait for all the helpers to finish. */ | |
514 | wait_for_completion(&ps.completion); | |
515 | ||
516 | destroy_work_on_stack(&my_work.pw_work); | |
517 | padata_works_free(&works); | |
518 | } | |
519 | ||
e15bacbe DK |
520 | static void __padata_list_init(struct padata_list *pd_list) |
521 | { | |
522 | INIT_LIST_HEAD(&pd_list->list); | |
523 | spin_lock_init(&pd_list->lock); | |
524 | } | |
16295bec | 525 | |
e15bacbe DK |
526 | /* Initialize all percpu queues used by serial workers */ |
527 | static void padata_init_squeues(struct parallel_data *pd) | |
528 | { | |
529 | int cpu; | |
530 | struct padata_serial_queue *squeue; | |
7b389b2c | 531 | |
e15bacbe DK |
532 | for_each_cpu(cpu, pd->cpumask.cbcpu) { |
533 | squeue = per_cpu_ptr(pd->squeue, cpu); | |
534 | squeue->pd = pd; | |
535 | __padata_list_init(&squeue->serial); | |
536 | INIT_WORK(&squeue->work, padata_serial_worker); | |
537 | } | |
538 | } | |
16295bec | 539 | |
f601c725 DJ |
540 | /* Initialize per-CPU reorder lists */ |
541 | static void padata_init_reorder_list(struct parallel_data *pd) | |
e15bacbe | 542 | { |
c51636a3 | 543 | int cpu; |
f601c725 | 544 | struct padata_list *list; |
16295bec | 545 | |
c51636a3 | 546 | for_each_cpu(cpu, pd->cpumask.pcpu) { |
f601c725 DJ |
547 | list = per_cpu_ptr(pd->reorder_list, cpu); |
548 | __padata_list_init(list); | |
16295bec | 549 | } |
e15bacbe | 550 | } |
16295bec | 551 | |
e15bacbe | 552 | /* Allocate and initialize the internal cpumask dependend resources. */ |
bbefa1dd | 553 | static struct parallel_data *padata_alloc_pd(struct padata_shell *ps) |
e15bacbe | 554 | { |
bbefa1dd | 555 | struct padata_instance *pinst = ps->pinst; |
e15bacbe | 556 | struct parallel_data *pd; |
16295bec | 557 | |
e15bacbe DK |
558 | pd = kzalloc(sizeof(struct parallel_data), GFP_KERNEL); |
559 | if (!pd) | |
560 | goto err; | |
16295bec | 561 | |
f601c725 DJ |
562 | pd->reorder_list = alloc_percpu(struct padata_list); |
563 | if (!pd->reorder_list) | |
e15bacbe DK |
564 | goto err_free_pd; |
565 | ||
566 | pd->squeue = alloc_percpu(struct padata_serial_queue); | |
567 | if (!pd->squeue) | |
f601c725 | 568 | goto err_free_reorder_list; |
bfde23ce | 569 | |
bbefa1dd | 570 | pd->ps = ps; |
cec00e6e DJ |
571 | |
572 | if (!alloc_cpumask_var(&pd->cpumask.pcpu, GFP_KERNEL)) | |
e15bacbe | 573 | goto err_free_squeue; |
cec00e6e DJ |
574 | if (!alloc_cpumask_var(&pd->cpumask.cbcpu, GFP_KERNEL)) |
575 | goto err_free_pcpu; | |
576 | ||
d69e037b DJ |
577 | cpumask_and(pd->cpumask.pcpu, pinst->cpumask.pcpu, cpu_online_mask); |
578 | cpumask_and(pd->cpumask.cbcpu, pinst->cpumask.cbcpu, cpu_online_mask); | |
16295bec | 579 | |
f601c725 | 580 | padata_init_reorder_list(pd); |
e15bacbe | 581 | padata_init_squeues(pd); |
4611ce22 | 582 | pd->seq_nr = -1; |
d5ee8e75 | 583 | refcount_set(&pd->refcnt, 1); |
16295bec | 584 | spin_lock_init(&pd->lock); |
ec9c7d19 | 585 | pd->cpu = cpumask_first(pd->cpumask.pcpu); |
6fc4dbcf | 586 | INIT_WORK(&pd->reorder_work, invoke_padata_reorder); |
16295bec SK |
587 | |
588 | return pd; | |
589 | ||
cec00e6e DJ |
590 | err_free_pcpu: |
591 | free_cpumask_var(pd->cpumask.pcpu); | |
e15bacbe DK |
592 | err_free_squeue: |
593 | free_percpu(pd->squeue); | |
f601c725 DJ |
594 | err_free_reorder_list: |
595 | free_percpu(pd->reorder_list); | |
16295bec SK |
596 | err_free_pd: |
597 | kfree(pd); | |
598 | err: | |
599 | return NULL; | |
600 | } | |
601 | ||
602 | static void padata_free_pd(struct parallel_data *pd) | |
603 | { | |
e15bacbe DK |
604 | free_cpumask_var(pd->cpumask.pcpu); |
605 | free_cpumask_var(pd->cpumask.cbcpu); | |
f601c725 | 606 | free_percpu(pd->reorder_list); |
e15bacbe | 607 | free_percpu(pd->squeue); |
16295bec SK |
608 | kfree(pd); |
609 | } | |
610 | ||
4c879170 SK |
611 | static void __padata_start(struct padata_instance *pinst) |
612 | { | |
613 | pinst->flags |= PADATA_INIT; | |
614 | } | |
615 | ||
ee836555 SK |
616 | static void __padata_stop(struct padata_instance *pinst) |
617 | { | |
618 | if (!(pinst->flags & PADATA_INIT)) | |
619 | return; | |
620 | ||
621 | pinst->flags &= ~PADATA_INIT; | |
622 | ||
623 | synchronize_rcu(); | |
ee836555 SK |
624 | } |
625 | ||
25985edc | 626 | /* Replace the internal control structure with a new one. */ |
bbefa1dd | 627 | static int padata_replace_one(struct padata_shell *ps) |
16295bec | 628 | { |
bbefa1dd | 629 | struct parallel_data *pd_new; |
16295bec | 630 | |
bbefa1dd HX |
631 | pd_new = padata_alloc_pd(ps); |
632 | if (!pd_new) | |
633 | return -ENOMEM; | |
16295bec | 634 | |
bbefa1dd HX |
635 | ps->opd = rcu_dereference_protected(ps->pd, 1); |
636 | rcu_assign_pointer(ps->pd, pd_new); | |
16295bec | 637 | |
bbefa1dd HX |
638 | return 0; |
639 | } | |
640 | ||
894c9ef9 | 641 | static int padata_replace(struct padata_instance *pinst) |
bbefa1dd | 642 | { |
bbefa1dd | 643 | struct padata_shell *ps; |
41ccdbfd | 644 | int err = 0; |
bbefa1dd HX |
645 | |
646 | pinst->flags |= PADATA_RESET; | |
16295bec | 647 | |
bbefa1dd HX |
648 | list_for_each_entry(ps, &pinst->pslist, list) { |
649 | err = padata_replace_one(ps); | |
650 | if (err) | |
651 | break; | |
652 | } | |
653 | ||
654 | synchronize_rcu(); | |
655 | ||
656 | list_for_each_entry_continue_reverse(ps, &pinst->pslist, list) | |
d5ee8e75 | 657 | if (refcount_dec_and_test(&ps->opd->refcnt)) |
bbefa1dd | 658 | padata_free_pd(ps->opd); |
16295bec SK |
659 | |
660 | pinst->flags &= ~PADATA_RESET; | |
bbefa1dd HX |
661 | |
662 | return err; | |
16295bec SK |
663 | } |
664 | ||
33e54450 SK |
665 | /* If cpumask contains no active cpu, we mark the instance as invalid. */ |
666 | static bool padata_validate_cpumask(struct padata_instance *pinst, | |
667 | const struct cpumask *cpumask) | |
668 | { | |
13614e0f | 669 | if (!cpumask_intersects(cpumask, cpu_online_mask)) { |
33e54450 SK |
670 | pinst->flags |= PADATA_INVALID; |
671 | return false; | |
672 | } | |
673 | ||
674 | pinst->flags &= ~PADATA_INVALID; | |
675 | return true; | |
676 | } | |
677 | ||
65ff577e SK |
678 | static int __padata_set_cpumasks(struct padata_instance *pinst, |
679 | cpumask_var_t pcpumask, | |
680 | cpumask_var_t cbcpumask) | |
681 | { | |
682 | int valid; | |
bbefa1dd | 683 | int err; |
65ff577e SK |
684 | |
685 | valid = padata_validate_cpumask(pinst, pcpumask); | |
686 | if (!valid) { | |
687 | __padata_stop(pinst); | |
688 | goto out_replace; | |
689 | } | |
690 | ||
691 | valid = padata_validate_cpumask(pinst, cbcpumask); | |
692 | if (!valid) | |
693 | __padata_stop(pinst); | |
694 | ||
695 | out_replace: | |
65ff577e SK |
696 | cpumask_copy(pinst->cpumask.pcpu, pcpumask); |
697 | cpumask_copy(pinst->cpumask.cbcpu, cbcpumask); | |
698 | ||
894c9ef9 | 699 | err = padata_setup_cpumasks(pinst) ?: padata_replace(pinst); |
65ff577e SK |
700 | |
701 | if (valid) | |
702 | __padata_start(pinst); | |
703 | ||
bbefa1dd | 704 | return err; |
65ff577e SK |
705 | } |
706 | ||
e15bacbe | 707 | /** |
bfcdcef8 DJ |
708 | * padata_set_cpumask - Sets specified by @cpumask_type cpumask to the value |
709 | * equivalent to @cpumask. | |
16295bec | 710 | * @pinst: padata instance |
e15bacbe DK |
711 | * @cpumask_type: PADATA_CPU_SERIAL or PADATA_CPU_PARALLEL corresponding |
712 | * to parallel and serial cpumasks respectively. | |
16295bec | 713 | * @cpumask: the cpumask to use |
bfcdcef8 DJ |
714 | * |
715 | * Return: 0 on success or negative error code | |
16295bec | 716 | */ |
e15bacbe DK |
717 | int padata_set_cpumask(struct padata_instance *pinst, int cpumask_type, |
718 | cpumask_var_t cpumask) | |
719 | { | |
720 | struct cpumask *serial_mask, *parallel_mask; | |
65ff577e SK |
721 | int err = -EINVAL; |
722 | ||
80771c82 | 723 | cpus_read_lock(); |
38228e88 | 724 | mutex_lock(&pinst->lock); |
6751fb3c | 725 | |
e15bacbe DK |
726 | switch (cpumask_type) { |
727 | case PADATA_CPU_PARALLEL: | |
728 | serial_mask = pinst->cpumask.cbcpu; | |
729 | parallel_mask = cpumask; | |
730 | break; | |
731 | case PADATA_CPU_SERIAL: | |
732 | parallel_mask = pinst->cpumask.pcpu; | |
733 | serial_mask = cpumask; | |
734 | break; | |
735 | default: | |
65ff577e | 736 | goto out; |
16295bec SK |
737 | } |
738 | ||
65ff577e | 739 | err = __padata_set_cpumasks(pinst, parallel_mask, serial_mask); |
16295bec SK |
740 | |
741 | out: | |
742 | mutex_unlock(&pinst->lock); | |
80771c82 | 743 | cpus_read_unlock(); |
16295bec SK |
744 | |
745 | return err; | |
746 | } | |
747 | EXPORT_SYMBOL(padata_set_cpumask); | |
748 | ||
19d795b6 AB |
749 | #ifdef CONFIG_HOTPLUG_CPU |
750 | ||
16295bec SK |
751 | static int __padata_add_cpu(struct padata_instance *pinst, int cpu) |
752 | { | |
bbefa1dd | 753 | int err = 0; |
16295bec | 754 | |
13614e0f | 755 | if (cpumask_test_cpu(cpu, cpu_online_mask)) { |
894c9ef9 | 756 | err = padata_replace(pinst); |
33e54450 | 757 | |
e15bacbe DK |
758 | if (padata_validate_cpumask(pinst, pinst->cpumask.pcpu) && |
759 | padata_validate_cpumask(pinst, pinst->cpumask.cbcpu)) | |
33e54450 | 760 | __padata_start(pinst); |
16295bec SK |
761 | } |
762 | ||
bbefa1dd | 763 | return err; |
16295bec SK |
764 | } |
765 | ||
16295bec SK |
766 | static int __padata_remove_cpu(struct padata_instance *pinst, int cpu) |
767 | { | |
bbefa1dd | 768 | int err = 0; |
16295bec | 769 | |
894c9ef9 | 770 | if (!cpumask_test_cpu(cpu, cpu_online_mask)) { |
e15bacbe | 771 | if (!padata_validate_cpumask(pinst, pinst->cpumask.pcpu) || |
b89661df | 772 | !padata_validate_cpumask(pinst, pinst->cpumask.cbcpu)) |
33e54450 | 773 | __padata_stop(pinst); |
33e54450 | 774 | |
894c9ef9 | 775 | err = padata_replace(pinst); |
16295bec SK |
776 | } |
777 | ||
bbefa1dd | 778 | return err; |
16295bec SK |
779 | } |
780 | ||
e15bacbe DK |
781 | static inline int pinst_has_cpu(struct padata_instance *pinst, int cpu) |
782 | { | |
783 | return cpumask_test_cpu(cpu, pinst->cpumask.pcpu) || | |
784 | cpumask_test_cpu(cpu, pinst->cpumask.cbcpu); | |
785 | } | |
786 | ||
30e92153 | 787 | static int padata_cpu_online(unsigned int cpu, struct hlist_node *node) |
16295bec | 788 | { |
16295bec | 789 | struct padata_instance *pinst; |
30e92153 | 790 | int ret; |
16295bec | 791 | |
3c2214b6 | 792 | pinst = hlist_entry_safe(node, struct padata_instance, cpu_online_node); |
30e92153 SAS |
793 | if (!pinst_has_cpu(pinst, cpu)) |
794 | return 0; | |
16295bec | 795 | |
30e92153 SAS |
796 | mutex_lock(&pinst->lock); |
797 | ret = __padata_add_cpu(pinst, cpu); | |
798 | mutex_unlock(&pinst->lock); | |
799 | return ret; | |
800 | } | |
16295bec | 801 | |
894c9ef9 | 802 | static int padata_cpu_dead(unsigned int cpu, struct hlist_node *node) |
30e92153 SAS |
803 | { |
804 | struct padata_instance *pinst; | |
805 | int ret; | |
806 | ||
3c2214b6 | 807 | pinst = hlist_entry_safe(node, struct padata_instance, cpu_dead_node); |
30e92153 SAS |
808 | if (!pinst_has_cpu(pinst, cpu)) |
809 | return 0; | |
16295bec | 810 | |
30e92153 SAS |
811 | mutex_lock(&pinst->lock); |
812 | ret = __padata_remove_cpu(pinst, cpu); | |
813 | mutex_unlock(&pinst->lock); | |
814 | return ret; | |
16295bec | 815 | } |
30e92153 SAS |
816 | |
817 | static enum cpuhp_state hp_online; | |
e2cb2f1c | 818 | #endif |
16295bec | 819 | |
5e017dc3 DK |
820 | static void __padata_free(struct padata_instance *pinst) |
821 | { | |
822 | #ifdef CONFIG_HOTPLUG_CPU | |
3c2214b6 DJ |
823 | cpuhp_state_remove_instance_nocalls(CPUHP_PADATA_DEAD, |
824 | &pinst->cpu_dead_node); | |
825 | cpuhp_state_remove_instance_nocalls(hp_online, &pinst->cpu_online_node); | |
5e017dc3 DK |
826 | #endif |
827 | ||
bbefa1dd HX |
828 | WARN_ON(!list_empty(&pinst->pslist)); |
829 | ||
5e017dc3 DK |
830 | free_cpumask_var(pinst->cpumask.pcpu); |
831 | free_cpumask_var(pinst->cpumask.cbcpu); | |
45d153c0 DJ |
832 | destroy_workqueue(pinst->serial_wq); |
833 | destroy_workqueue(pinst->parallel_wq); | |
5e017dc3 DK |
834 | kfree(pinst); |
835 | } | |
836 | ||
837 | #define kobj2pinst(_kobj) \ | |
838 | container_of(_kobj, struct padata_instance, kobj) | |
839 | #define attr2pentry(_attr) \ | |
840 | container_of(_attr, struct padata_sysfs_entry, attr) | |
841 | ||
842 | static void padata_sysfs_release(struct kobject *kobj) | |
843 | { | |
844 | struct padata_instance *pinst = kobj2pinst(kobj); | |
845 | __padata_free(pinst); | |
846 | } | |
847 | ||
848 | struct padata_sysfs_entry { | |
849 | struct attribute attr; | |
850 | ssize_t (*show)(struct padata_instance *, struct attribute *, char *); | |
851 | ssize_t (*store)(struct padata_instance *, struct attribute *, | |
852 | const char *, size_t); | |
853 | }; | |
854 | ||
855 | static ssize_t show_cpumask(struct padata_instance *pinst, | |
856 | struct attribute *attr, char *buf) | |
857 | { | |
858 | struct cpumask *cpumask; | |
859 | ssize_t len; | |
860 | ||
861 | mutex_lock(&pinst->lock); | |
862 | if (!strcmp(attr->name, "serial_cpumask")) | |
863 | cpumask = pinst->cpumask.cbcpu; | |
864 | else | |
865 | cpumask = pinst->cpumask.pcpu; | |
866 | ||
4497da6f TH |
867 | len = snprintf(buf, PAGE_SIZE, "%*pb\n", |
868 | nr_cpu_ids, cpumask_bits(cpumask)); | |
5e017dc3 | 869 | mutex_unlock(&pinst->lock); |
4497da6f | 870 | return len < PAGE_SIZE ? len : -EINVAL; |
5e017dc3 DK |
871 | } |
872 | ||
873 | static ssize_t store_cpumask(struct padata_instance *pinst, | |
874 | struct attribute *attr, | |
875 | const char *buf, size_t count) | |
876 | { | |
877 | cpumask_var_t new_cpumask; | |
878 | ssize_t ret; | |
879 | int mask_type; | |
880 | ||
881 | if (!alloc_cpumask_var(&new_cpumask, GFP_KERNEL)) | |
882 | return -ENOMEM; | |
883 | ||
884 | ret = bitmap_parse(buf, count, cpumask_bits(new_cpumask), | |
885 | nr_cpumask_bits); | |
886 | if (ret < 0) | |
887 | goto out; | |
888 | ||
889 | mask_type = !strcmp(attr->name, "serial_cpumask") ? | |
890 | PADATA_CPU_SERIAL : PADATA_CPU_PARALLEL; | |
891 | ret = padata_set_cpumask(pinst, mask_type, new_cpumask); | |
892 | if (!ret) | |
893 | ret = count; | |
894 | ||
895 | out: | |
896 | free_cpumask_var(new_cpumask); | |
897 | return ret; | |
898 | } | |
899 | ||
900 | #define PADATA_ATTR_RW(_name, _show_name, _store_name) \ | |
901 | static struct padata_sysfs_entry _name##_attr = \ | |
902 | __ATTR(_name, 0644, _show_name, _store_name) | |
903 | #define PADATA_ATTR_RO(_name, _show_name) \ | |
904 | static struct padata_sysfs_entry _name##_attr = \ | |
905 | __ATTR(_name, 0400, _show_name, NULL) | |
906 | ||
907 | PADATA_ATTR_RW(serial_cpumask, show_cpumask, store_cpumask); | |
908 | PADATA_ATTR_RW(parallel_cpumask, show_cpumask, store_cpumask); | |
909 | ||
910 | /* | |
911 | * Padata sysfs provides the following objects: | |
912 | * serial_cpumask [RW] - cpumask for serial workers | |
913 | * parallel_cpumask [RW] - cpumask for parallel workers | |
914 | */ | |
915 | static struct attribute *padata_default_attrs[] = { | |
916 | &serial_cpumask_attr.attr, | |
917 | ¶llel_cpumask_attr.attr, | |
918 | NULL, | |
919 | }; | |
2064fbc7 | 920 | ATTRIBUTE_GROUPS(padata_default); |
5e017dc3 DK |
921 | |
922 | static ssize_t padata_sysfs_show(struct kobject *kobj, | |
923 | struct attribute *attr, char *buf) | |
924 | { | |
925 | struct padata_instance *pinst; | |
926 | struct padata_sysfs_entry *pentry; | |
927 | ssize_t ret = -EIO; | |
928 | ||
929 | pinst = kobj2pinst(kobj); | |
930 | pentry = attr2pentry(attr); | |
931 | if (pentry->show) | |
932 | ret = pentry->show(pinst, attr, buf); | |
933 | ||
934 | return ret; | |
935 | } | |
936 | ||
937 | static ssize_t padata_sysfs_store(struct kobject *kobj, struct attribute *attr, | |
938 | const char *buf, size_t count) | |
939 | { | |
940 | struct padata_instance *pinst; | |
941 | struct padata_sysfs_entry *pentry; | |
942 | ssize_t ret = -EIO; | |
943 | ||
944 | pinst = kobj2pinst(kobj); | |
945 | pentry = attr2pentry(attr); | |
946 | if (pentry->show) | |
947 | ret = pentry->store(pinst, attr, buf, count); | |
948 | ||
949 | return ret; | |
950 | } | |
951 | ||
952 | static const struct sysfs_ops padata_sysfs_ops = { | |
953 | .show = padata_sysfs_show, | |
954 | .store = padata_sysfs_store, | |
955 | }; | |
956 | ||
957 | static struct kobj_type padata_attr_type = { | |
958 | .sysfs_ops = &padata_sysfs_ops, | |
2064fbc7 | 959 | .default_groups = padata_default_groups, |
5e017dc3 DK |
960 | .release = padata_sysfs_release, |
961 | }; | |
962 | ||
e15bacbe | 963 | /** |
3f257191 | 964 | * padata_alloc - allocate and initialize a padata instance |
b128a304 | 965 | * @name: used to identify the instance |
bfcdcef8 DJ |
966 | * |
967 | * Return: new instance on success, NULL on error | |
16295bec | 968 | */ |
3f257191 | 969 | struct padata_instance *padata_alloc(const char *name) |
16295bec | 970 | { |
16295bec | 971 | struct padata_instance *pinst; |
16295bec SK |
972 | |
973 | pinst = kzalloc(sizeof(struct padata_instance), GFP_KERNEL); | |
974 | if (!pinst) | |
975 | goto err; | |
976 | ||
bfde23ce DJ |
977 | pinst->parallel_wq = alloc_workqueue("%s_parallel", WQ_UNBOUND, 0, |
978 | name); | |
45d153c0 | 979 | if (!pinst->parallel_wq) |
16295bec | 980 | goto err_free_inst; |
b128a304 | 981 | |
80771c82 | 982 | cpus_read_lock(); |
cc491d8e | 983 | |
45d153c0 DJ |
984 | pinst->serial_wq = alloc_workqueue("%s_serial", WQ_MEM_RECLAIM | |
985 | WQ_CPU_INTENSIVE, 1, name); | |
986 | if (!pinst->serial_wq) | |
cc491d8e | 987 | goto err_put_cpus; |
45d153c0 DJ |
988 | |
989 | if (!alloc_cpumask_var(&pinst->cpumask.pcpu, GFP_KERNEL)) | |
990 | goto err_free_serial_wq; | |
e15bacbe DK |
991 | if (!alloc_cpumask_var(&pinst->cpumask.cbcpu, GFP_KERNEL)) { |
992 | free_cpumask_var(pinst->cpumask.pcpu); | |
45d153c0 | 993 | goto err_free_serial_wq; |
33e54450 | 994 | } |
16295bec | 995 | |
bbefa1dd | 996 | INIT_LIST_HEAD(&pinst->pslist); |
16295bec | 997 | |
3f257191 DJ |
998 | cpumask_copy(pinst->cpumask.pcpu, cpu_possible_mask); |
999 | cpumask_copy(pinst->cpumask.cbcpu, cpu_possible_mask); | |
bbefa1dd HX |
1000 | |
1001 | if (padata_setup_cpumasks(pinst)) | |
d69e037b | 1002 | goto err_free_masks; |
16295bec | 1003 | |
bd25b488 | 1004 | __padata_start(pinst); |
16295bec | 1005 | |
5e017dc3 | 1006 | kobject_init(&pinst->kobj, &padata_attr_type); |
16295bec SK |
1007 | mutex_init(&pinst->lock); |
1008 | ||
b8b4a416 | 1009 | #ifdef CONFIG_HOTPLUG_CPU |
3c2214b6 DJ |
1010 | cpuhp_state_add_instance_nocalls_cpuslocked(hp_online, |
1011 | &pinst->cpu_online_node); | |
894c9ef9 | 1012 | cpuhp_state_add_instance_nocalls_cpuslocked(CPUHP_PADATA_DEAD, |
3c2214b6 | 1013 | &pinst->cpu_dead_node); |
b8b4a416 | 1014 | #endif |
cc491d8e | 1015 | |
80771c82 | 1016 | cpus_read_unlock(); |
cc491d8e | 1017 | |
16295bec SK |
1018 | return pinst; |
1019 | ||
e15bacbe DK |
1020 | err_free_masks: |
1021 | free_cpumask_var(pinst->cpumask.pcpu); | |
1022 | free_cpumask_var(pinst->cpumask.cbcpu); | |
45d153c0 DJ |
1023 | err_free_serial_wq: |
1024 | destroy_workqueue(pinst->serial_wq); | |
cc491d8e | 1025 | err_put_cpus: |
80771c82 | 1026 | cpus_read_unlock(); |
45d153c0 | 1027 | destroy_workqueue(pinst->parallel_wq); |
16295bec SK |
1028 | err_free_inst: |
1029 | kfree(pinst); | |
1030 | err: | |
1031 | return NULL; | |
1032 | } | |
3f257191 | 1033 | EXPORT_SYMBOL(padata_alloc); |
9596695e | 1034 | |
0198ffd1 | 1035 | /** |
16295bec SK |
1036 | * padata_free - free a padata instance |
1037 | * | |
bfcdcef8 | 1038 | * @pinst: padata instance to free |
16295bec SK |
1039 | */ |
1040 | void padata_free(struct padata_instance *pinst) | |
1041 | { | |
5e017dc3 | 1042 | kobject_put(&pinst->kobj); |
16295bec SK |
1043 | } |
1044 | EXPORT_SYMBOL(padata_free); | |
30e92153 | 1045 | |
bbefa1dd HX |
1046 | /** |
1047 | * padata_alloc_shell - Allocate and initialize padata shell. | |
1048 | * | |
1049 | * @pinst: Parent padata_instance object. | |
bfcdcef8 DJ |
1050 | * |
1051 | * Return: new shell on success, NULL on error | |
bbefa1dd HX |
1052 | */ |
1053 | struct padata_shell *padata_alloc_shell(struct padata_instance *pinst) | |
1054 | { | |
1055 | struct parallel_data *pd; | |
1056 | struct padata_shell *ps; | |
1057 | ||
1058 | ps = kzalloc(sizeof(*ps), GFP_KERNEL); | |
1059 | if (!ps) | |
1060 | goto out; | |
1061 | ||
1062 | ps->pinst = pinst; | |
1063 | ||
80771c82 | 1064 | cpus_read_lock(); |
bbefa1dd | 1065 | pd = padata_alloc_pd(ps); |
80771c82 | 1066 | cpus_read_unlock(); |
bbefa1dd HX |
1067 | |
1068 | if (!pd) | |
1069 | goto out_free_ps; | |
1070 | ||
1071 | mutex_lock(&pinst->lock); | |
1072 | RCU_INIT_POINTER(ps->pd, pd); | |
1073 | list_add(&ps->list, &pinst->pslist); | |
1074 | mutex_unlock(&pinst->lock); | |
1075 | ||
1076 | return ps; | |
1077 | ||
1078 | out_free_ps: | |
1079 | kfree(ps); | |
1080 | out: | |
1081 | return NULL; | |
1082 | } | |
1083 | EXPORT_SYMBOL(padata_alloc_shell); | |
1084 | ||
1085 | /** | |
1086 | * padata_free_shell - free a padata shell | |
1087 | * | |
1088 | * @ps: padata shell to free | |
1089 | */ | |
1090 | void padata_free_shell(struct padata_shell *ps) | |
1091 | { | |
07b24c7c EB |
1092 | if (!ps) |
1093 | return; | |
bbefa1dd | 1094 | |
07b24c7c | 1095 | mutex_lock(&ps->pinst->lock); |
bbefa1dd HX |
1096 | list_del(&ps->list); |
1097 | padata_free_pd(rcu_dereference_protected(ps->pd, 1)); | |
07b24c7c | 1098 | mutex_unlock(&ps->pinst->lock); |
bbefa1dd HX |
1099 | |
1100 | kfree(ps); | |
1101 | } | |
1102 | EXPORT_SYMBOL(padata_free_shell); | |
1103 | ||
f1b192b1 | 1104 | void __init padata_init(void) |
30e92153 | 1105 | { |
4611ce22 | 1106 | unsigned int i, possible_cpus; |
f1b192b1 | 1107 | #ifdef CONFIG_HOTPLUG_CPU |
30e92153 SAS |
1108 | int ret; |
1109 | ||
1110 | ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, "padata:online", | |
894c9ef9 | 1111 | padata_cpu_online, NULL); |
30e92153 | 1112 | if (ret < 0) |
f1b192b1 | 1113 | goto err; |
30e92153 | 1114 | hp_online = ret; |
894c9ef9 DJ |
1115 | |
1116 | ret = cpuhp_setup_state_multi(CPUHP_PADATA_DEAD, "padata:dead", | |
1117 | NULL, padata_cpu_dead); | |
4611ce22 DJ |
1118 | if (ret < 0) |
1119 | goto remove_online_state; | |
1120 | #endif | |
1121 | ||
1122 | possible_cpus = num_possible_cpus(); | |
1123 | padata_works = kmalloc_array(possible_cpus, sizeof(struct padata_work), | |
1124 | GFP_KERNEL); | |
1125 | if (!padata_works) | |
1126 | goto remove_dead_state; | |
1127 | ||
1128 | for (i = 0; i < possible_cpus; ++i) | |
1129 | list_add(&padata_works[i].pw_list, &padata_free_works); | |
30e92153 | 1130 | |
f1b192b1 | 1131 | return; |
4611ce22 DJ |
1132 | |
1133 | remove_dead_state: | |
1134 | #ifdef CONFIG_HOTPLUG_CPU | |
1135 | cpuhp_remove_multi_state(CPUHP_PADATA_DEAD); | |
1136 | remove_online_state: | |
1137 | cpuhp_remove_multi_state(hp_online); | |
f1b192b1 | 1138 | err: |
30e92153 | 1139 | #endif |
4611ce22 | 1140 | pr_warn("padata: initialization failed\n"); |
f1b192b1 | 1141 | } |