aio.c 45.6 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 *	An async IO implementation for Linux
 *	Written by Benjamin LaHaise <bcrl@kvack.org>
 *
 *	Implements an efficient asynchronous io interface.
 *
 *	Copyright 2000, 2001, 2002 Red Hat, Inc.  All Rights Reserved.
 *
 *	See ../COPYING for licensing terms.
 */
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/time.h>
#include <linux/aio_abi.h>
#include <linux/module.h>
#include <linux/syscalls.h>
18
#include <linux/uio.h>
L
Linus Torvalds 已提交
19 20 21 22 23 24 25 26 27 28 29 30 31 32

#define DEBUG 0

#include <linux/sched.h>
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/mm.h>
#include <linux/mman.h>
#include <linux/slab.h>
#include <linux/timer.h>
#include <linux/aio.h>
#include <linux/highmem.h>
#include <linux/workqueue.h>
#include <linux/security.h>
33
#include <linux/eventfd.h>
L
Linus Torvalds 已提交
34 35 36 37 38 39 40 41 42 43 44 45

#include <asm/kmap_types.h>
#include <asm/uaccess.h>
#include <asm/mmu_context.h>

#if DEBUG > 1
#define dprintk		printk
#else
#define dprintk(x...)	do { ; } while (0)
#endif

/*------ sysctl variables----*/
46 47 48
static DEFINE_SPINLOCK(aio_nr_lock);
unsigned long aio_nr;		/* current system wide number of aio requests */
unsigned long aio_max_nr = 0x10000; /* system wide maximum number of aio requests */
L
Linus Torvalds 已提交
49 50
/*----end sysctl variables---*/

51 52
static struct kmem_cache	*kiocb_cachep;
static struct kmem_cache	*kioctx_cachep;
L
Linus Torvalds 已提交
53 54 55 56

static struct workqueue_struct *aio_wq;

/* Used for rare fput completion. */
57 58
static void aio_fput_routine(struct work_struct *);
static DECLARE_WORK(fput_work, aio_fput_routine);
L
Linus Torvalds 已提交
59 60

static DEFINE_SPINLOCK(fput_lock);
61
static LIST_HEAD(fput_head);
L
Linus Torvalds 已提交
62

63
static void aio_kick_handler(struct work_struct *);
64
static void aio_queue_work(struct kioctx *);
L
Linus Torvalds 已提交
65 66 67 68 69 70 71

/* aio_setup
 *	Creates the slab caches used by the aio routines, panic on
 *	failure as this is done early during the boot sequence.
 */
static int __init aio_setup(void)
{
72 73
	kiocb_cachep = KMEM_CACHE(kiocb, SLAB_HWCACHE_ALIGN|SLAB_PANIC);
	kioctx_cachep = KMEM_CACHE(kioctx,SLAB_HWCACHE_ALIGN|SLAB_PANIC);
L
Linus Torvalds 已提交
74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124

	aio_wq = create_workqueue("aio");

	pr_debug("aio_setup: sizeof(struct page) = %d\n", (int)sizeof(struct page));

	return 0;
}

static void aio_free_ring(struct kioctx *ctx)
{
	struct aio_ring_info *info = &ctx->ring_info;
	long i;

	for (i=0; i<info->nr_pages; i++)
		put_page(info->ring_pages[i]);

	if (info->mmap_size) {
		down_write(&ctx->mm->mmap_sem);
		do_munmap(ctx->mm, info->mmap_base, info->mmap_size);
		up_write(&ctx->mm->mmap_sem);
	}

	if (info->ring_pages && info->ring_pages != info->internal_pages)
		kfree(info->ring_pages);
	info->ring_pages = NULL;
	info->nr = 0;
}

static int aio_setup_ring(struct kioctx *ctx)
{
	struct aio_ring *ring;
	struct aio_ring_info *info = &ctx->ring_info;
	unsigned nr_events = ctx->max_reqs;
	unsigned long size;
	int nr_pages;

	/* Compensate for the ring buffer's head/tail overlap entry */
	nr_events += 2;	/* 1 is required, 2 for good luck */

	size = sizeof(struct aio_ring);
	size += sizeof(struct io_event) * nr_events;
	nr_pages = (size + PAGE_SIZE-1) >> PAGE_SHIFT;

	if (nr_pages < 0)
		return -EINVAL;

	nr_events = (PAGE_SIZE * nr_pages - sizeof(struct aio_ring)) / sizeof(struct io_event);

	info->nr = 0;
	info->ring_pages = info->internal_pages;
	if (nr_pages > AIO_RING_PAGES) {
125
		info->ring_pages = kcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL);
L
Linus Torvalds 已提交
126 127 128 129 130 131 132 133
		if (!info->ring_pages)
			return -ENOMEM;
	}

	info->mmap_size = nr_pages * PAGE_SIZE;
	dprintk("attempting mmap of %lu bytes\n", info->mmap_size);
	down_write(&ctx->mm->mmap_sem);
	info->mmap_base = do_mmap(NULL, 0, info->mmap_size, 
134
				  PROT_READ|PROT_WRITE, MAP_ANONYMOUS|MAP_PRIVATE,
L
Linus Torvalds 已提交
135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
				  0);
	if (IS_ERR((void *)info->mmap_base)) {
		up_write(&ctx->mm->mmap_sem);
		info->mmap_size = 0;
		aio_free_ring(ctx);
		return -EAGAIN;
	}

	dprintk("mmap address: 0x%08lx\n", info->mmap_base);
	info->nr_pages = get_user_pages(current, ctx->mm,
					info->mmap_base, nr_pages, 
					1, 0, info->ring_pages, NULL);
	up_write(&ctx->mm->mmap_sem);

	if (unlikely(info->nr_pages != nr_pages)) {
		aio_free_ring(ctx);
		return -EAGAIN;
	}

	ctx->user_id = info->mmap_base;

	info->nr = nr_events;		/* trusted copy */

	ring = kmap_atomic(info->ring_pages[0], KM_USER0);
	ring->nr = nr_events;	/* user copy */
	ring->id = ctx->user_id;
	ring->head = ring->tail = 0;
	ring->magic = AIO_RING_MAGIC;
	ring->compat_features = AIO_RING_COMPAT_FEATURES;
	ring->incompat_features = AIO_RING_INCOMPAT_FEATURES;
	ring->header_length = sizeof(struct aio_ring);
	kunmap_atomic(ring, KM_USER0);

	return 0;
}


/* aio_ring_event: returns a pointer to the event at the given index from
 * kmap_atomic(, km).  Release the pointer with put_aio_ring_event();
 */
#define AIO_EVENTS_PER_PAGE	(PAGE_SIZE / sizeof(struct io_event))
#define AIO_EVENTS_FIRST_PAGE	((PAGE_SIZE - sizeof(struct aio_ring)) / sizeof(struct io_event))
#define AIO_EVENTS_OFFSET	(AIO_EVENTS_PER_PAGE - AIO_EVENTS_FIRST_PAGE)

#define aio_ring_event(info, nr, km) ({					\
	unsigned pos = (nr) + AIO_EVENTS_OFFSET;			\
	struct io_event *__event;					\
	__event = kmap_atomic(						\
			(info)->ring_pages[pos / AIO_EVENTS_PER_PAGE], km); \
	__event += pos % AIO_EVENTS_PER_PAGE;				\
	__event;							\
})

#define put_aio_ring_event(event, km) do {	\
	struct io_event *__event = (event);	\
	(void)__event;				\
	kunmap_atomic((void *)((unsigned long)__event & PAGE_MASK), km); \
} while(0)

/* ioctx_alloc
 *	Allocates and initializes an ioctx.  Returns an ERR_PTR if it failed.
 */
static struct kioctx *ioctx_alloc(unsigned nr_events)
{
	struct mm_struct *mm;
	struct kioctx *ctx;

	/* Prevent overflows */
	if ((nr_events > (0x10000000U / sizeof(struct io_event))) ||
	    (nr_events > (0x10000000U / sizeof(struct kiocb)))) {
		pr_debug("ENOMEM: nr_events too high\n");
		return ERR_PTR(-EINVAL);
	}

209
	if ((unsigned long)nr_events > aio_max_nr)
L
Linus Torvalds 已提交
210 211
		return ERR_PTR(-EAGAIN);

212
	ctx = kmem_cache_zalloc(kioctx_cachep, GFP_KERNEL);
L
Linus Torvalds 已提交
213 214 215 216 217 218 219 220 221 222 223 224 225 226
	if (!ctx)
		return ERR_PTR(-ENOMEM);

	ctx->max_reqs = nr_events;
	mm = ctx->mm = current->mm;
	atomic_inc(&mm->mm_count);

	atomic_set(&ctx->users, 1);
	spin_lock_init(&ctx->ctx_lock);
	spin_lock_init(&ctx->ring_info.ring_lock);
	init_waitqueue_head(&ctx->wait);

	INIT_LIST_HEAD(&ctx->active_reqs);
	INIT_LIST_HEAD(&ctx->run_list);
227
	INIT_DELAYED_WORK(&ctx->wq, aio_kick_handler);
L
Linus Torvalds 已提交
228 229 230 231 232

	if (aio_setup_ring(ctx) < 0)
		goto out_freectx;

	/* limit the number of system wide aios */
233 234 235 236 237 238 239 240
	spin_lock(&aio_nr_lock);
	if (aio_nr + ctx->max_reqs > aio_max_nr ||
	    aio_nr + ctx->max_reqs < aio_nr)
		ctx->max_reqs = 0;
	else
		aio_nr += ctx->max_reqs;
	spin_unlock(&aio_nr_lock);
	if (ctx->max_reqs == 0)
L
Linus Torvalds 已提交
241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292
		goto out_cleanup;

	/* now link into global list.  kludge.  FIXME */
	write_lock(&mm->ioctx_list_lock);
	ctx->next = mm->ioctx_list;
	mm->ioctx_list = ctx;
	write_unlock(&mm->ioctx_list_lock);

	dprintk("aio: allocated ioctx %p[%ld]: mm=%p mask=0x%x\n",
		ctx, ctx->user_id, current->mm, ctx->ring_info.nr);
	return ctx;

out_cleanup:
	__put_ioctx(ctx);
	return ERR_PTR(-EAGAIN);

out_freectx:
	mmdrop(mm);
	kmem_cache_free(kioctx_cachep, ctx);
	ctx = ERR_PTR(-ENOMEM);

	dprintk("aio: error allocating ioctx %p\n", ctx);
	return ctx;
}

/* aio_cancel_all
 *	Cancels all outstanding aio requests on an aio context.  Used 
 *	when the processes owning a context have all exited to encourage 
 *	the rapid destruction of the kioctx.
 */
static void aio_cancel_all(struct kioctx *ctx)
{
	int (*cancel)(struct kiocb *, struct io_event *);
	struct io_event res;
	spin_lock_irq(&ctx->ctx_lock);
	ctx->dead = 1;
	while (!list_empty(&ctx->active_reqs)) {
		struct list_head *pos = ctx->active_reqs.next;
		struct kiocb *iocb = list_kiocb(pos);
		list_del_init(&iocb->ki_list);
		cancel = iocb->ki_cancel;
		kiocbSetCancelled(iocb);
		if (cancel) {
			iocb->ki_users++;
			spin_unlock_irq(&ctx->ctx_lock);
			cancel(iocb, &res);
			spin_lock_irq(&ctx->ctx_lock);
		}
	}
	spin_unlock_irq(&ctx->ctx_lock);
}

293
static void wait_for_all_aios(struct kioctx *ctx)
L
Linus Torvalds 已提交
294 295 296 297
{
	struct task_struct *tsk = current;
	DECLARE_WAITQUEUE(wait, tsk);

298
	spin_lock_irq(&ctx->ctx_lock);
L
Linus Torvalds 已提交
299
	if (!ctx->reqs_active)
300
		goto out;
L
Linus Torvalds 已提交
301 302 303 304

	add_wait_queue(&ctx->wait, &wait);
	set_task_state(tsk, TASK_UNINTERRUPTIBLE);
	while (ctx->reqs_active) {
305
		spin_unlock_irq(&ctx->ctx_lock);
J
Jeff Moyer 已提交
306
		io_schedule();
L
Linus Torvalds 已提交
307
		set_task_state(tsk, TASK_UNINTERRUPTIBLE);
308
		spin_lock_irq(&ctx->ctx_lock);
L
Linus Torvalds 已提交
309 310 311
	}
	__set_task_state(tsk, TASK_RUNNING);
	remove_wait_queue(&ctx->wait, &wait);
312 313 314

out:
	spin_unlock_irq(&ctx->ctx_lock);
L
Linus Torvalds 已提交
315 316 317 318 319
}

/* wait_on_sync_kiocb:
 *	Waits on the given sync kiocb to complete.
 */
320
ssize_t wait_on_sync_kiocb(struct kiocb *iocb)
L
Linus Torvalds 已提交
321 322 323 324 325
{
	while (iocb->ki_users) {
		set_current_state(TASK_UNINTERRUPTIBLE);
		if (!iocb->ki_users)
			break;
J
Jeff Moyer 已提交
326
		io_schedule();
L
Linus Torvalds 已提交
327 328 329 330 331 332 333 334 335 336 337 338
	}
	__set_current_state(TASK_RUNNING);
	return iocb->ki_user_data;
}

/* exit_aio: called when the last user of mm goes away.  At this point, 
 * there is no way for any new requests to be submited or any of the 
 * io_* syscalls to be called on the context.  However, there may be 
 * outstanding requests which hold references to the context; as they 
 * go away, they will call put_ioctx and release any pinned memory
 * associated with the request (held via struct page * references).
 */
339
void exit_aio(struct mm_struct *mm)
L
Linus Torvalds 已提交
340 341 342 343 344 345 346 347 348 349
{
	struct kioctx *ctx = mm->ioctx_list;
	mm->ioctx_list = NULL;
	while (ctx) {
		struct kioctx *next = ctx->next;
		ctx->next = NULL;
		aio_cancel_all(ctx);

		wait_for_all_aios(ctx);
		/*
A
Andrew Morton 已提交
350
		 * Ensure we don't leave the ctx on the aio_wq
L
Linus Torvalds 已提交
351
		 */
352
		cancel_work_sync(&ctx->wq.work);
L
Linus Torvalds 已提交
353 354 355 356 357 358 359 360 361 362 363 364 365 366 367

		if (1 != atomic_read(&ctx->users))
			printk(KERN_DEBUG
				"exit_aio:ioctx still alive: %d %d %d\n",
				atomic_read(&ctx->users), ctx->dead,
				ctx->reqs_active);
		put_ioctx(ctx);
		ctx = next;
	}
}

/* __put_ioctx
 *	Called when the last user of an aio context has gone away,
 *	and the struct needs to be freed.
 */
368
void __put_ioctx(struct kioctx *ctx)
L
Linus Torvalds 已提交
369 370 371
{
	unsigned nr_events = ctx->max_reqs;

E
Eric Sesterhenn 已提交
372
	BUG_ON(ctx->reqs_active);
L
Linus Torvalds 已提交
373 374

	cancel_delayed_work(&ctx->wq);
375
	cancel_work_sync(&ctx->wq.work);
L
Linus Torvalds 已提交
376 377 378 379 380 381
	aio_free_ring(ctx);
	mmdrop(ctx->mm);
	ctx->mm = NULL;
	pr_debug("__put_ioctx: freeing %p\n", ctx);
	kmem_cache_free(kioctx_cachep, ctx);

382 383 384 385 386 387
	if (nr_events) {
		spin_lock(&aio_nr_lock);
		BUG_ON(aio_nr - nr_events > aio_nr);
		aio_nr -= nr_events;
		spin_unlock(&aio_nr_lock);
	}
L
Linus Torvalds 已提交
388 389 390 391 392 393 394 395 396 397 398 399
}

/* aio_get_req
 *	Allocate a slot for an aio request.  Increments the users count
 * of the kioctx so that the kioctx stays around until all requests are
 * complete.  Returns NULL if no requests are free.
 *
 * Returns with kiocb->users set to 2.  The io submit code path holds
 * an extra reference while submitting the i/o.
 * This prevents races between the aio code path referencing the
 * req (after submitting it) and aio_complete() freeing the req.
 */
400
static struct kiocb *__aio_get_req(struct kioctx *ctx)
L
Linus Torvalds 已提交
401 402 403 404 405 406 407 408 409
{
	struct kiocb *req = NULL;
	struct aio_ring *ring;
	int okay = 0;

	req = kmem_cache_alloc(kiocb_cachep, GFP_KERNEL);
	if (unlikely(!req))
		return NULL;

Z
Zach Brown 已提交
410
	req->ki_flags = 0;
L
Linus Torvalds 已提交
411 412 413 414 415 416 417
	req->ki_users = 2;
	req->ki_key = 0;
	req->ki_ctx = ctx;
	req->ki_cancel = NULL;
	req->ki_retry = NULL;
	req->ki_dtor = NULL;
	req->private = NULL;
B
Badari Pulavarty 已提交
418
	req->ki_iovec = NULL;
L
Linus Torvalds 已提交
419
	INIT_LIST_HEAD(&req->ki_run_list);
420
	req->ki_eventfd = ERR_PTR(-EINVAL);
L
Linus Torvalds 已提交
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

	/* Check if the completion queue has enough free space to
	 * accept an event from this io.
	 */
	spin_lock_irq(&ctx->ctx_lock);
	ring = kmap_atomic(ctx->ring_info.ring_pages[0], KM_USER0);
	if (ctx->reqs_active < aio_ring_avail(&ctx->ring_info, ring)) {
		list_add(&req->ki_list, &ctx->active_reqs);
		ctx->reqs_active++;
		okay = 1;
	}
	kunmap_atomic(ring, KM_USER0);
	spin_unlock_irq(&ctx->ctx_lock);

	if (!okay) {
		kmem_cache_free(kiocb_cachep, req);
		req = NULL;
	}

	return req;
}

static inline struct kiocb *aio_get_req(struct kioctx *ctx)
{
	struct kiocb *req;
	/* Handle a potential starvation case -- should be exceedingly rare as 
	 * requests will be stuck on fput_head only if the aio_fput_routine is 
	 * delayed and the requests were the last user of the struct file.
	 */
	req = __aio_get_req(ctx);
	if (unlikely(NULL == req)) {
		aio_fput_routine(NULL);
		req = __aio_get_req(ctx);
	}
	return req;
}

static inline void really_put_req(struct kioctx *ctx, struct kiocb *req)
{
460 461
	assert_spin_locked(&ctx->ctx_lock);

462 463
	if (!IS_ERR(req->ki_eventfd))
		fput(req->ki_eventfd);
L
Linus Torvalds 已提交
464 465
	if (req->ki_dtor)
		req->ki_dtor(req);
B
Badari Pulavarty 已提交
466 467
	if (req->ki_iovec != &req->ki_inline_vec)
		kfree(req->ki_iovec);
L
Linus Torvalds 已提交
468 469 470 471 472 473 474
	kmem_cache_free(kiocb_cachep, req);
	ctx->reqs_active--;

	if (unlikely(!ctx->reqs_active && ctx->dead))
		wake_up(&ctx->wait);
}

475
static void aio_fput_routine(struct work_struct *data)
L
Linus Torvalds 已提交
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
{
	spin_lock_irq(&fput_lock);
	while (likely(!list_empty(&fput_head))) {
		struct kiocb *req = list_kiocb(fput_head.next);
		struct kioctx *ctx = req->ki_ctx;

		list_del(&req->ki_list);
		spin_unlock_irq(&fput_lock);

		/* Complete the fput */
		__fput(req->ki_filp);

		/* Link the iocb into the context's free list */
		spin_lock_irq(&ctx->ctx_lock);
		really_put_req(ctx, req);
		spin_unlock_irq(&ctx->ctx_lock);

		put_ioctx(ctx);
		spin_lock_irq(&fput_lock);
	}
	spin_unlock_irq(&fput_lock);
}

/* __aio_put_req
 *	Returns true if this put was the last user of the request.
 */
static int __aio_put_req(struct kioctx *ctx, struct kiocb *req)
{
	dprintk(KERN_DEBUG "aio_put(%p): f_count=%d\n",
		req, atomic_read(&req->ki_filp->f_count));

507 508
	assert_spin_locked(&ctx->ctx_lock);

L
Linus Torvalds 已提交
509
	req->ki_users --;
E
Eric Sesterhenn 已提交
510
	BUG_ON(req->ki_users < 0);
L
Linus Torvalds 已提交
511 512 513 514 515 516 517 518 519
	if (likely(req->ki_users))
		return 0;
	list_del(&req->ki_list);		/* remove from active_reqs */
	req->ki_cancel = NULL;
	req->ki_retry = NULL;

	/* Must be done under the lock to serialise against cancellation.
	 * Call this aio_fput as it duplicates fput via the fput_work.
	 */
520
	if (unlikely(atomic_dec_and_test(&req->ki_filp->f_count))) {
L
Linus Torvalds 已提交
521 522 523 524 525 526 527 528 529 530 531 532 533 534
		get_ioctx(ctx);
		spin_lock(&fput_lock);
		list_add(&req->ki_list, &fput_head);
		spin_unlock(&fput_lock);
		queue_work(aio_wq, &fput_work);
	} else
		really_put_req(ctx, req);
	return 1;
}

/* aio_put_req
 *	Returns true if this put was the last user of the kiocb,
 *	false if the request is still in use.
 */
535
int aio_put_req(struct kiocb *req)
L
Linus Torvalds 已提交
536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585
{
	struct kioctx *ctx = req->ki_ctx;
	int ret;
	spin_lock_irq(&ctx->ctx_lock);
	ret = __aio_put_req(ctx, req);
	spin_unlock_irq(&ctx->ctx_lock);
	return ret;
}

/*	Lookup an ioctx id.  ioctx_list is lockless for reads.
 *	FIXME: this is O(n) and is only suitable for development.
 */
struct kioctx *lookup_ioctx(unsigned long ctx_id)
{
	struct kioctx *ioctx;
	struct mm_struct *mm;

	mm = current->mm;
	read_lock(&mm->ioctx_list_lock);
	for (ioctx = mm->ioctx_list; ioctx; ioctx = ioctx->next)
		if (likely(ioctx->user_id == ctx_id && !ioctx->dead)) {
			get_ioctx(ioctx);
			break;
		}
	read_unlock(&mm->ioctx_list_lock);

	return ioctx;
}

/*
 * use_mm
 *	Makes the calling kernel thread take on the specified
 *	mm context.
 *	Called by the retry thread execute retries within the
 *	iocb issuer's mm context, so that copy_from/to_user
 *	operations work seamlessly for aio.
 *	(Note: this routine is intended to be called only
 *	from a kernel thread context)
 */
static void use_mm(struct mm_struct *mm)
{
	struct mm_struct *active_mm;
	struct task_struct *tsk = current;

	task_lock(tsk);
	tsk->flags |= PF_BORROWED_MM;
	active_mm = tsk->active_mm;
	atomic_inc(&mm->mm_count);
	tsk->mm = mm;
	tsk->active_mm = mm;
586 587 588 589
	/*
	 * Note that on UML this *requires* PF_BORROWED_MM to be set, otherwise
	 * it won't work. Update it accordingly if you change it here
	 */
590
	switch_mm(active_mm, mm, tsk);
L
Linus Torvalds 已提交
591 592 593 594 595 596 597 598 599 600 601 602 603
	task_unlock(tsk);

	mmdrop(active_mm);
}

/*
 * unuse_mm
 *	Reverses the effect of use_mm, i.e. releases the
 *	specified mm context which was earlier taken on
 *	by the calling kernel thread
 *	(Note: this routine is intended to be called only
 *	from a kernel thread context)
 */
604
static void unuse_mm(struct mm_struct *mm)
L
Linus Torvalds 已提交
605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628
{
	struct task_struct *tsk = current;

	task_lock(tsk);
	tsk->flags &= ~PF_BORROWED_MM;
	tsk->mm = NULL;
	/* active_mm is still 'mm' */
	enter_lazy_tlb(mm, tsk);
	task_unlock(tsk);
}

/*
 * Queue up a kiocb to be retried. Assumes that the kiocb
 * has already been marked as kicked, and places it on
 * the retry run list for the corresponding ioctx, if it
 * isn't already queued. Returns 1 if it actually queued
 * the kiocb (to tell the caller to activate the work
 * queue to process it), or 0, if it found that it was
 * already queued.
 */
static inline int __queue_kicked_iocb(struct kiocb *iocb)
{
	struct kioctx *ctx = iocb->ki_ctx;

629 630
	assert_spin_locked(&ctx->ctx_lock);

L
Linus Torvalds 已提交
631 632 633 634 635 636 637 638 639 640 641 642 643
	if (list_empty(&iocb->ki_run_list)) {
		list_add_tail(&iocb->ki_run_list,
			&ctx->run_list);
		return 1;
	}
	return 0;
}

/* aio_run_iocb
 *	This is the core aio execution routine. It is
 *	invoked both for initial i/o submission and
 *	subsequent retries via the aio_kick_handler.
 *	Expects to be invoked with iocb->ki_ctx->lock
A
Andreas Mohr 已提交
644
 *	already held. The lock is released and reacquired
L
Linus Torvalds 已提交
645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711
 *	as needed during processing.
 *
 * Calls the iocb retry method (already setup for the
 * iocb on initial submission) for operation specific
 * handling, but takes care of most of common retry
 * execution details for a given iocb. The retry method
 * needs to be non-blocking as far as possible, to avoid
 * holding up other iocbs waiting to be serviced by the
 * retry kernel thread.
 *
 * The trickier parts in this code have to do with
 * ensuring that only one retry instance is in progress
 * for a given iocb at any time. Providing that guarantee
 * simplifies the coding of individual aio operations as
 * it avoids various potential races.
 */
static ssize_t aio_run_iocb(struct kiocb *iocb)
{
	struct kioctx	*ctx = iocb->ki_ctx;
	ssize_t (*retry)(struct kiocb *);
	ssize_t ret;

	if (!(retry = iocb->ki_retry)) {
		printk("aio_run_iocb: iocb->ki_retry = NULL\n");
		return 0;
	}

	/*
	 * We don't want the next retry iteration for this
	 * operation to start until this one has returned and
	 * updated the iocb state. However, wait_queue functions
	 * can trigger a kick_iocb from interrupt context in the
	 * meantime, indicating that data is available for the next
	 * iteration. We want to remember that and enable the
	 * next retry iteration _after_ we are through with
	 * this one.
	 *
	 * So, in order to be able to register a "kick", but
	 * prevent it from being queued now, we clear the kick
	 * flag, but make the kick code *think* that the iocb is
	 * still on the run list until we are actually done.
	 * When we are done with this iteration, we check if
	 * the iocb was kicked in the meantime and if so, queue
	 * it up afresh.
	 */

	kiocbClearKicked(iocb);

	/*
	 * This is so that aio_complete knows it doesn't need to
	 * pull the iocb off the run list (We can't just call
	 * INIT_LIST_HEAD because we don't want a kick_iocb to
	 * queue this on the run list yet)
	 */
	iocb->ki_run_list.next = iocb->ki_run_list.prev = NULL;
	spin_unlock_irq(&ctx->ctx_lock);

	/* Quit retrying if the i/o has been cancelled */
	if (kiocbIsCancelled(iocb)) {
		ret = -EINTR;
		aio_complete(iocb, ret, 0);
		/* must not access the iocb after this */
		goto out;
	}

	/*
	 * Now we are all set to call the retry method in async
712
	 * context.
L
Linus Torvalds 已提交
713 714 715
	 */
	ret = retry(iocb);

716 717 718
	if (ret != -EIOCBRETRY && ret != -EIOCBQUEUED) {
		BUG_ON(!list_empty(&iocb->ki_wait.task_list));
		aio_complete(iocb, ret, 0);
L
Linus Torvalds 已提交
719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736
	}
out:
	spin_lock_irq(&ctx->ctx_lock);

	if (-EIOCBRETRY == ret) {
		/*
		 * OK, now that we are done with this iteration
		 * and know that there is more left to go,
		 * this is where we let go so that a subsequent
		 * "kick" can start the next iteration
		 */

		/* will make __queue_kicked_iocb succeed from here on */
		INIT_LIST_HEAD(&iocb->ki_run_list);
		/* we must queue the next iteration ourselves, if it
		 * has already been kicked */
		if (kiocbIsKicked(iocb)) {
			__queue_kicked_iocb(iocb);
737 738 739 740 741 742 743 744

			/*
			 * __queue_kicked_iocb will always return 1 here, because
			 * iocb->ki_run_list is empty at this point so it should
			 * be safe to unconditionally queue the context into the
			 * work queue.
			 */
			aio_queue_work(ctx);
L
Linus Torvalds 已提交
745 746 747 748 749 750 751 752 753 754
		}
	}
	return ret;
}

/*
 * __aio_run_iocbs:
 * 	Process all pending retries queued on the ioctx
 * 	run list.
 * Assumes it is operating within the aio issuer's mm
755
 * context.
L
Linus Torvalds 已提交
756 757 758 759
 */
static int __aio_run_iocbs(struct kioctx *ctx)
{
	struct kiocb *iocb;
760
	struct list_head run_list;
L
Linus Torvalds 已提交
761

762 763
	assert_spin_locked(&ctx->ctx_lock);

764
	list_replace_init(&ctx->run_list, &run_list);
L
Linus Torvalds 已提交
765 766 767 768 769 770 771 772 773
	while (!list_empty(&run_list)) {
		iocb = list_entry(run_list.next, struct kiocb,
			ki_run_list);
		list_del(&iocb->ki_run_list);
		/*
		 * Hold an extra reference while retrying i/o.
		 */
		iocb->ki_users++;       /* grab extra reference */
		aio_run_iocb(iocb);
774
		__aio_put_req(ctx, iocb);
L
Linus Torvalds 已提交
775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836
 	}
	if (!list_empty(&ctx->run_list))
		return 1;
	return 0;
}

static void aio_queue_work(struct kioctx * ctx)
{
	unsigned long timeout;
	/*
	 * if someone is waiting, get the work started right
	 * away, otherwise, use a longer delay
	 */
	smp_mb();
	if (waitqueue_active(&ctx->wait))
		timeout = 1;
	else
		timeout = HZ/10;
	queue_delayed_work(aio_wq, &ctx->wq, timeout);
}


/*
 * aio_run_iocbs:
 * 	Process all pending retries queued on the ioctx
 * 	run list.
 * Assumes it is operating within the aio issuer's mm
 * context.
 */
static inline void aio_run_iocbs(struct kioctx *ctx)
{
	int requeue;

	spin_lock_irq(&ctx->ctx_lock);

	requeue = __aio_run_iocbs(ctx);
	spin_unlock_irq(&ctx->ctx_lock);
	if (requeue)
		aio_queue_work(ctx);
}

/*
 * just like aio_run_iocbs, but keeps running them until
 * the list stays empty
 */
static inline void aio_run_all_iocbs(struct kioctx *ctx)
{
	spin_lock_irq(&ctx->ctx_lock);
	while (__aio_run_iocbs(ctx))
		;
	spin_unlock_irq(&ctx->ctx_lock);
}

/*
 * aio_kick_handler:
 * 	Work queue handler triggered to process pending
 * 	retries on an ioctx. Takes on the aio issuer's
 *	mm context before running the iocbs, so that
 *	copy_xxx_user operates on the issuer's address
 *      space.
 * Run on aiod's context.
 */
837
static void aio_kick_handler(struct work_struct *work)
L
Linus Torvalds 已提交
838
{
839
	struct kioctx *ctx = container_of(work, struct kioctx, wq.work);
L
Linus Torvalds 已提交
840
	mm_segment_t oldfs = get_fs();
841
	struct mm_struct *mm;
L
Linus Torvalds 已提交
842 843 844 845 846 847
	int requeue;

	set_fs(USER_DS);
	use_mm(ctx->mm);
	spin_lock_irq(&ctx->ctx_lock);
	requeue =__aio_run_iocbs(ctx);
848
	mm = ctx->mm;
L
Linus Torvalds 已提交
849
	spin_unlock_irq(&ctx->ctx_lock);
850
 	unuse_mm(mm);
L
Linus Torvalds 已提交
851 852 853 854 855
	set_fs(oldfs);
	/*
	 * we're in a worker thread already, don't use queue_delayed_work,
	 */
	if (requeue)
856
		queue_delayed_work(aio_wq, &ctx->wq, 0);
L
Linus Torvalds 已提交
857 858 859 860 861 862 863 864
}


/*
 * Called by kick_iocb to queue the kiocb for retry
 * and if required activate the aio work queue to process
 * it
 */
865
static void try_queue_kicked_iocb(struct kiocb *iocb)
L
Linus Torvalds 已提交
866 867 868 869 870
{
 	struct kioctx	*ctx = iocb->ki_ctx;
	unsigned long flags;
	int run = 0;

871 872 873 874 875 876
	/* We're supposed to be the only path putting the iocb back on the run
	 * list.  If we find that the iocb is *back* on a wait queue already
	 * than retry has happened before we could queue the iocb.  This also
	 * means that the retry could have completed and freed our iocb, no
	 * good. */
	BUG_ON((!list_empty(&iocb->ki_wait.task_list)));
L
Linus Torvalds 已提交
877 878

	spin_lock_irqsave(&ctx->ctx_lock, flags);
879 880 881 882
	/* set this inside the lock so that we can't race with aio_run_iocb()
	 * testing it and putting the iocb on the run list under the lock */
	if (!kiocbTryKick(iocb))
		run = __queue_kicked_iocb(iocb);
L
Linus Torvalds 已提交
883
	spin_unlock_irqrestore(&ctx->ctx_lock, flags);
K
Ken Chen 已提交
884
	if (run)
L
Linus Torvalds 已提交
885 886 887 888 889 890 891 892 893 894
		aio_queue_work(ctx);
}

/*
 * kick_iocb:
 *      Called typically from a wait queue callback context
 *      (aio_wake_function) to trigger a retry of the iocb.
 *      The retry is usually executed by aio workqueue
 *      threads (See aio_kick_handler).
 */
895
void kick_iocb(struct kiocb *iocb)
L
Linus Torvalds 已提交
896 897 898 899 900 901 902 903 904
{
	/* sync iocbs are easy: they can only ever be executing from a 
	 * single context. */
	if (is_sync_kiocb(iocb)) {
		kiocbSetKicked(iocb);
	        wake_up_process(iocb->ki_obj.tsk);
		return;
	}

905
	try_queue_kicked_iocb(iocb);
L
Linus Torvalds 已提交
906 907 908 909 910 911 912 913
}
EXPORT_SYMBOL(kick_iocb);

/* aio_complete
 *	Called when the io request on the given iocb is complete.
 *	Returns true if this is the last user of the request.  The 
 *	only other user of the request can be the cancellation code.
 */
914
int aio_complete(struct kiocb *iocb, long res, long res2)
L
Linus Torvalds 已提交
915 916 917 918 919 920 921 922 923
{
	struct kioctx	*ctx = iocb->ki_ctx;
	struct aio_ring_info	*info;
	struct aio_ring	*ring;
	struct io_event	*event;
	unsigned long	flags;
	unsigned long	tail;
	int		ret;

924 925 926 927 928 929
	/*
	 * Special case handling for sync iocbs:
	 *  - events go directly into the iocb for fast handling
	 *  - the sync task with the iocb in its stack holds the single iocb
	 *    ref, no other paths have a way to get another ref
	 *  - the sync task helpfully left a reference to itself in the iocb
L
Linus Torvalds 已提交
930 931
	 */
	if (is_sync_kiocb(iocb)) {
932
		BUG_ON(iocb->ki_users != 1);
L
Linus Torvalds 已提交
933
		iocb->ki_user_data = res;
934
		iocb->ki_users = 0;
L
Linus Torvalds 已提交
935
		wake_up_process(iocb->ki_obj.tsk);
936
		return 1;
L
Linus Torvalds 已提交
937 938
	}

939 940 941 942 943 944 945 946
	/*
	 * Check if the user asked us to deliver the result through an
	 * eventfd. The eventfd_signal() function is safe to be called
	 * from IRQ context.
	 */
	if (!IS_ERR(iocb->ki_eventfd))
		eventfd_signal(iocb->ki_eventfd, 1);

L
Linus Torvalds 已提交
947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970
	info = &ctx->ring_info;

	/* add a completion event to the ring buffer.
	 * must be done holding ctx->ctx_lock to prevent
	 * other code from messing with the tail
	 * pointer since we might be called from irq
	 * context.
	 */
	spin_lock_irqsave(&ctx->ctx_lock, flags);

	if (iocb->ki_run_list.prev && !list_empty(&iocb->ki_run_list))
		list_del_init(&iocb->ki_run_list);

	/*
	 * cancelled requests don't get events, userland was given one
	 * when the event got cancelled.
	 */
	if (kiocbIsCancelled(iocb))
		goto put_rq;

	ring = kmap_atomic(info->ring_pages[0], KM_IRQ1);

	tail = info->tail;
	event = aio_ring_event(info, tail, KM_IRQ0);
971 972
	if (++tail >= info->nr)
		tail = 0;
L
Linus Torvalds 已提交
973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998

	event->obj = (u64)(unsigned long)iocb->ki_obj.user;
	event->data = iocb->ki_user_data;
	event->res = res;
	event->res2 = res2;

	dprintk("aio_complete: %p[%lu]: %p: %p %Lx %lx %lx\n",
		ctx, tail, iocb, iocb->ki_obj.user, iocb->ki_user_data,
		res, res2);

	/* after flagging the request as done, we
	 * must never even look at it again
	 */
	smp_wmb();	/* make event visible before updating tail */

	info->tail = tail;
	ring->tail = tail;

	put_aio_ring_event(event, KM_IRQ0);
	kunmap_atomic(ring, KM_IRQ1);

	pr_debug("added to ring %p at [%lu]\n", iocb, tail);
put_rq:
	/* everything turned out well, dispose of the aiocb. */
	ret = __aio_put_req(ctx, iocb);

999 1000 1001 1002 1003 1004 1005 1006
	/*
	 * We have to order our ring_info tail store above and test
	 * of the wait list below outside the wait lock.  This is
	 * like in wake_up_bit() where clearing a bit has to be
	 * ordered with the unlocked test.
	 */
	smp_mb();

L
Linus Torvalds 已提交
1007 1008 1009
	if (waitqueue_active(&ctx->wait))
		wake_up(&ctx->wait);

1010
	spin_unlock_irqrestore(&ctx->ctx_lock, flags);
L
Linus Torvalds 已提交
1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170
	return ret;
}

/* aio_read_evt
 *	Pull an event off of the ioctx's event ring.  Returns the number of 
 *	events fetched (0 or 1 ;-)
 *	FIXME: make this use cmpxchg.
 *	TODO: make the ringbuffer user mmap()able (requires FIXME).
 */
static int aio_read_evt(struct kioctx *ioctx, struct io_event *ent)
{
	struct aio_ring_info *info = &ioctx->ring_info;
	struct aio_ring *ring;
	unsigned long head;
	int ret = 0;

	ring = kmap_atomic(info->ring_pages[0], KM_USER0);
	dprintk("in aio_read_evt h%lu t%lu m%lu\n",
		 (unsigned long)ring->head, (unsigned long)ring->tail,
		 (unsigned long)ring->nr);

	if (ring->head == ring->tail)
		goto out;

	spin_lock(&info->ring_lock);

	head = ring->head % info->nr;
	if (head != ring->tail) {
		struct io_event *evp = aio_ring_event(info, head, KM_USER1);
		*ent = *evp;
		head = (head + 1) % info->nr;
		smp_mb(); /* finish reading the event before updatng the head */
		ring->head = head;
		ret = 1;
		put_aio_ring_event(evp, KM_USER1);
	}
	spin_unlock(&info->ring_lock);

out:
	kunmap_atomic(ring, KM_USER0);
	dprintk("leaving aio_read_evt: %d  h%lu t%lu\n", ret,
		 (unsigned long)ring->head, (unsigned long)ring->tail);
	return ret;
}

struct aio_timeout {
	struct timer_list	timer;
	int			timed_out;
	struct task_struct	*p;
};

static void timeout_func(unsigned long data)
{
	struct aio_timeout *to = (struct aio_timeout *)data;

	to->timed_out = 1;
	wake_up_process(to->p);
}

static inline void init_timeout(struct aio_timeout *to)
{
	init_timer(&to->timer);
	to->timer.data = (unsigned long)to;
	to->timer.function = timeout_func;
	to->timed_out = 0;
	to->p = current;
}

static inline void set_timeout(long start_jiffies, struct aio_timeout *to,
			       const struct timespec *ts)
{
	to->timer.expires = start_jiffies + timespec_to_jiffies(ts);
	if (time_after(to->timer.expires, jiffies))
		add_timer(&to->timer);
	else
		to->timed_out = 1;
}

static inline void clear_timeout(struct aio_timeout *to)
{
	del_singleshot_timer_sync(&to->timer);
}

static int read_events(struct kioctx *ctx,
			long min_nr, long nr,
			struct io_event __user *event,
			struct timespec __user *timeout)
{
	long			start_jiffies = jiffies;
	struct task_struct	*tsk = current;
	DECLARE_WAITQUEUE(wait, tsk);
	int			ret;
	int			i = 0;
	struct io_event		ent;
	struct aio_timeout	to;
	int			retry = 0;

	/* needed to zero any padding within an entry (there shouldn't be 
	 * any, but C is fun!
	 */
	memset(&ent, 0, sizeof(ent));
retry:
	ret = 0;
	while (likely(i < nr)) {
		ret = aio_read_evt(ctx, &ent);
		if (unlikely(ret <= 0))
			break;

		dprintk("read event: %Lx %Lx %Lx %Lx\n",
			ent.data, ent.obj, ent.res, ent.res2);

		/* Could we split the check in two? */
		ret = -EFAULT;
		if (unlikely(copy_to_user(event, &ent, sizeof(ent)))) {
			dprintk("aio: lost an event due to EFAULT.\n");
			break;
		}
		ret = 0;

		/* Good, event copied to userland, update counts. */
		event ++;
		i ++;
	}

	if (min_nr <= i)
		return i;
	if (ret)
		return ret;

	/* End fast path */

	/* racey check, but it gets redone */
	if (!retry && unlikely(!list_empty(&ctx->run_list))) {
		retry = 1;
		aio_run_all_iocbs(ctx);
		goto retry;
	}

	init_timeout(&to);
	if (timeout) {
		struct timespec	ts;
		ret = -EFAULT;
		if (unlikely(copy_from_user(&ts, timeout, sizeof(ts))))
			goto out;

		set_timeout(start_jiffies, &to, &ts);
	}

	while (likely(i < nr)) {
		add_wait_queue_exclusive(&ctx->wait, &wait);
		do {
			set_task_state(tsk, TASK_INTERRUPTIBLE);
			ret = aio_read_evt(ctx, &ent);
			if (ret)
				break;
			if (min_nr <= i)
				break;
			ret = 0;
			if (to.timed_out)	/* Only check after read evt */
				break;
1171 1172 1173 1174 1175 1176
			/* Try to only show up in io wait if there are ops
			 *  in flight */
			if (ctx->reqs_active)
				io_schedule();
			else
				schedule();
L
Linus Torvalds 已提交
1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259
			if (signal_pending(tsk)) {
				ret = -EINTR;
				break;
			}
			/*ret = aio_read_evt(ctx, &ent);*/
		} while (1) ;

		set_task_state(tsk, TASK_RUNNING);
		remove_wait_queue(&ctx->wait, &wait);

		if (unlikely(ret <= 0))
			break;

		ret = -EFAULT;
		if (unlikely(copy_to_user(event, &ent, sizeof(ent)))) {
			dprintk("aio: lost an event due to EFAULT.\n");
			break;
		}

		/* Good, event copied to userland, update counts. */
		event ++;
		i ++;
	}

	if (timeout)
		clear_timeout(&to);
out:
	return i ? i : ret;
}

/* Take an ioctx and remove it from the list of ioctx's.  Protects 
 * against races with itself via ->dead.
 */
static void io_destroy(struct kioctx *ioctx)
{
	struct mm_struct *mm = current->mm;
	struct kioctx **tmp;
	int was_dead;

	/* delete the entry from the list is someone else hasn't already */
	write_lock(&mm->ioctx_list_lock);
	was_dead = ioctx->dead;
	ioctx->dead = 1;
	for (tmp = &mm->ioctx_list; *tmp && *tmp != ioctx;
	     tmp = &(*tmp)->next)
		;
	if (*tmp)
		*tmp = ioctx->next;
	write_unlock(&mm->ioctx_list_lock);

	dprintk("aio_release(%p)\n", ioctx);
	if (likely(!was_dead))
		put_ioctx(ioctx);	/* twice for the list */

	aio_cancel_all(ioctx);
	wait_for_all_aios(ioctx);
	put_ioctx(ioctx);	/* once for the lookup */
}

/* sys_io_setup:
 *	Create an aio_context capable of receiving at least nr_events.
 *	ctxp must not point to an aio_context that already exists, and
 *	must be initialized to 0 prior to the call.  On successful
 *	creation of the aio_context, *ctxp is filled in with the resulting 
 *	handle.  May fail with -EINVAL if *ctxp is not initialized,
 *	if the specified nr_events exceeds internal limits.  May fail 
 *	with -EAGAIN if the specified nr_events exceeds the user's limit 
 *	of available events.  May fail with -ENOMEM if insufficient kernel
 *	resources are available.  May fail with -EFAULT if an invalid
 *	pointer is passed for ctxp.  Will fail with -ENOSYS if not
 *	implemented.
 */
asmlinkage long sys_io_setup(unsigned nr_events, aio_context_t __user *ctxp)
{
	struct kioctx *ioctx = NULL;
	unsigned long ctx;
	long ret;

	ret = get_user(ctx, ctxp);
	if (unlikely(ret))
		goto out;

	ret = -EINVAL;
1260 1261 1262
	if (unlikely(ctx || nr_events == 0)) {
		pr_debug("EINVAL: io_setup: ctx %lu nr_events %u\n",
		         ctx, nr_events);
L
Linus Torvalds 已提交
1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297
		goto out;
	}

	ioctx = ioctx_alloc(nr_events);
	ret = PTR_ERR(ioctx);
	if (!IS_ERR(ioctx)) {
		ret = put_user(ioctx->user_id, ctxp);
		if (!ret)
			return 0;

		get_ioctx(ioctx); /* io_destroy() expects us to hold a ref */
		io_destroy(ioctx);
	}

out:
	return ret;
}

/* sys_io_destroy:
 *	Destroy the aio_context specified.  May cancel any outstanding 
 *	AIOs and block on completion.  Will fail with -ENOSYS if not
 *	implemented.  May fail with -EFAULT if the context pointed to
 *	is invalid.
 */
asmlinkage long sys_io_destroy(aio_context_t ctx)
{
	struct kioctx *ioctx = lookup_ioctx(ctx);
	if (likely(NULL != ioctx)) {
		io_destroy(ioctx);
		return 0;
	}
	pr_debug("EINVAL: io_destroy: invalid context id\n");
	return -EINVAL;
}

B
Badari Pulavarty 已提交
1298
static void aio_advance_iovec(struct kiocb *iocb, ssize_t ret)
L
Linus Torvalds 已提交
1299
{
B
Badari Pulavarty 已提交
1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312
	struct iovec *iov = &iocb->ki_iovec[iocb->ki_cur_seg];

	BUG_ON(ret <= 0);

	while (iocb->ki_cur_seg < iocb->ki_nr_segs && ret > 0) {
		ssize_t this = min((ssize_t)iov->iov_len, ret);
		iov->iov_base += this;
		iov->iov_len -= this;
		iocb->ki_left -= this;
		ret -= this;
		if (iov->iov_len == 0) {
			iocb->ki_cur_seg++;
			iov++;
1313
		}
B
Badari Pulavarty 已提交
1314
	}
L
Linus Torvalds 已提交
1315

B
Badari Pulavarty 已提交
1316 1317 1318
	/* the caller should not have done more io than what fit in
	 * the remaining iovecs */
	BUG_ON(ret > 0 && iocb->ki_left == 0);
L
Linus Torvalds 已提交
1319 1320
}

B
Badari Pulavarty 已提交
1321
static ssize_t aio_rw_vect_retry(struct kiocb *iocb)
L
Linus Torvalds 已提交
1322 1323
{
	struct file *file = iocb->ki_filp;
B
Badari Pulavarty 已提交
1324 1325 1326 1327
	struct address_space *mapping = file->f_mapping;
	struct inode *inode = mapping->host;
	ssize_t (*rw_op)(struct kiocb *, const struct iovec *,
			 unsigned long, loff_t);
L
Linus Torvalds 已提交
1328
	ssize_t ret = 0;
B
Badari Pulavarty 已提交
1329 1330 1331 1332 1333 1334 1335 1336 1337 1338
	unsigned short opcode;

	if ((iocb->ki_opcode == IOCB_CMD_PREADV) ||
		(iocb->ki_opcode == IOCB_CMD_PREAD)) {
		rw_op = file->f_op->aio_read;
		opcode = IOCB_CMD_PREADV;
	} else {
		rw_op = file->f_op->aio_write;
		opcode = IOCB_CMD_PWRITEV;
	}
L
Linus Torvalds 已提交
1339

1340 1341 1342 1343
	/* This matches the pread()/pwrite() logic */
	if (iocb->ki_pos < 0)
		return -EINVAL;

1344
	do {
B
Badari Pulavarty 已提交
1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355
		ret = rw_op(iocb, &iocb->ki_iovec[iocb->ki_cur_seg],
			    iocb->ki_nr_segs - iocb->ki_cur_seg,
			    iocb->ki_pos);
		if (ret > 0)
			aio_advance_iovec(iocb, ret);

	/* retry all partial writes.  retry partial reads as long as its a
	 * regular file. */
	} while (ret > 0 && iocb->ki_left > 0 &&
		 (opcode == IOCB_CMD_PWRITEV ||
		  (!S_ISFIFO(inode->i_mode) && !S_ISSOCK(inode->i_mode))));
L
Linus Torvalds 已提交
1356

B
Badari Pulavarty 已提交
1357 1358
	/* This means we must have transferred all that we could */
	/* No need to retry anymore */
L
Linus Torvalds 已提交
1359 1360 1361
	if ((ret == 0) || (iocb->ki_left == 0))
		ret = iocb->ki_nbytes - iocb->ki_left;

1362 1363 1364 1365 1366 1367 1368
	/* If we managed to write some out we return that, rather than
	 * the eventual error. */
	if (opcode == IOCB_CMD_PWRITEV
	    && ret < 0 && ret != -EIOCBQUEUED && ret != -EIOCBRETRY
	    && iocb->ki_nbytes - iocb->ki_left)
		ret = iocb->ki_nbytes - iocb->ki_left;

L
Linus Torvalds 已提交
1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391
	return ret;
}

static ssize_t aio_fdsync(struct kiocb *iocb)
{
	struct file *file = iocb->ki_filp;
	ssize_t ret = -EINVAL;

	if (file->f_op->aio_fsync)
		ret = file->f_op->aio_fsync(iocb, 1);
	return ret;
}

static ssize_t aio_fsync(struct kiocb *iocb)
{
	struct file *file = iocb->ki_filp;
	ssize_t ret = -EINVAL;

	if (file->f_op->aio_fsync)
		ret = file->f_op->aio_fsync(iocb, 0);
	return ret;
}

B
Badari Pulavarty 已提交
1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422
static ssize_t aio_setup_vectored_rw(int type, struct kiocb *kiocb)
{
	ssize_t ret;

	ret = rw_copy_check_uvector(type, (struct iovec __user *)kiocb->ki_buf,
				    kiocb->ki_nbytes, 1,
				    &kiocb->ki_inline_vec, &kiocb->ki_iovec);
	if (ret < 0)
		goto out;

	kiocb->ki_nr_segs = kiocb->ki_nbytes;
	kiocb->ki_cur_seg = 0;
	/* ki_nbytes/left now reflect bytes instead of segs */
	kiocb->ki_nbytes = ret;
	kiocb->ki_left = ret;

	ret = 0;
out:
	return ret;
}

static ssize_t aio_setup_single_vector(struct kiocb *kiocb)
{
	kiocb->ki_iovec = &kiocb->ki_inline_vec;
	kiocb->ki_iovec->iov_base = kiocb->ki_buf;
	kiocb->ki_iovec->iov_len = kiocb->ki_left;
	kiocb->ki_nr_segs = 1;
	kiocb->ki_cur_seg = 0;
	return 0;
}

L
Linus Torvalds 已提交
1423 1424 1425 1426 1427
/*
 * aio_setup_iocb:
 *	Performs the initial checks and aio retry method
 *	setup for the kiocb at the time of io submission.
 */
1428
static ssize_t aio_setup_iocb(struct kiocb *kiocb)
L
Linus Torvalds 已提交
1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441
{
	struct file *file = kiocb->ki_filp;
	ssize_t ret = 0;

	switch (kiocb->ki_opcode) {
	case IOCB_CMD_PREAD:
		ret = -EBADF;
		if (unlikely(!(file->f_mode & FMODE_READ)))
			break;
		ret = -EFAULT;
		if (unlikely(!access_ok(VERIFY_WRITE, kiocb->ki_buf,
			kiocb->ki_left)))
			break;
1442 1443 1444
		ret = security_file_permission(file, MAY_READ);
		if (unlikely(ret))
			break;
B
Badari Pulavarty 已提交
1445 1446 1447
		ret = aio_setup_single_vector(kiocb);
		if (ret)
			break;
L
Linus Torvalds 已提交
1448 1449
		ret = -EINVAL;
		if (file->f_op->aio_read)
B
Badari Pulavarty 已提交
1450
			kiocb->ki_retry = aio_rw_vect_retry;
L
Linus Torvalds 已提交
1451 1452 1453 1454 1455 1456 1457 1458 1459
		break;
	case IOCB_CMD_PWRITE:
		ret = -EBADF;
		if (unlikely(!(file->f_mode & FMODE_WRITE)))
			break;
		ret = -EFAULT;
		if (unlikely(!access_ok(VERIFY_READ, kiocb->ki_buf,
			kiocb->ki_left)))
			break;
1460 1461 1462
		ret = security_file_permission(file, MAY_WRITE);
		if (unlikely(ret))
			break;
B
Badari Pulavarty 已提交
1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493
		ret = aio_setup_single_vector(kiocb);
		if (ret)
			break;
		ret = -EINVAL;
		if (file->f_op->aio_write)
			kiocb->ki_retry = aio_rw_vect_retry;
		break;
	case IOCB_CMD_PREADV:
		ret = -EBADF;
		if (unlikely(!(file->f_mode & FMODE_READ)))
			break;
		ret = security_file_permission(file, MAY_READ);
		if (unlikely(ret))
			break;
		ret = aio_setup_vectored_rw(READ, kiocb);
		if (ret)
			break;
		ret = -EINVAL;
		if (file->f_op->aio_read)
			kiocb->ki_retry = aio_rw_vect_retry;
		break;
	case IOCB_CMD_PWRITEV:
		ret = -EBADF;
		if (unlikely(!(file->f_mode & FMODE_WRITE)))
			break;
		ret = security_file_permission(file, MAY_WRITE);
		if (unlikely(ret))
			break;
		ret = aio_setup_vectored_rw(WRITE, kiocb);
		if (ret)
			break;
L
Linus Torvalds 已提交
1494 1495
		ret = -EINVAL;
		if (file->f_op->aio_write)
B
Badari Pulavarty 已提交
1496
			kiocb->ki_retry = aio_rw_vect_retry;
L
Linus Torvalds 已提交
1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524
		break;
	case IOCB_CMD_FDSYNC:
		ret = -EINVAL;
		if (file->f_op->aio_fsync)
			kiocb->ki_retry = aio_fdsync;
		break;
	case IOCB_CMD_FSYNC:
		ret = -EINVAL;
		if (file->f_op->aio_fsync)
			kiocb->ki_retry = aio_fsync;
		break;
	default:
		dprintk("EINVAL: io_submit: no operation provided\n");
		ret = -EINVAL;
	}

	if (!kiocb->ki_retry)
		return ret;

	return 0;
}

/*
 * aio_wake_function:
 * 	wait queue callback function for aio notification,
 * 	Simply triggers a retry of the operation via kick_iocb.
 *
 * 	This callback is specified in the wait queue entry in
1525
 *	a kiocb.
L
Linus Torvalds 已提交
1526 1527 1528 1529 1530 1531 1532 1533
 *
 * Note:
 * This routine is executed with the wait queue lock held.
 * Since kick_iocb acquires iocb->ctx->ctx_lock, it nests
 * the ioctx lock inside the wait queue lock. This is safe
 * because this callback isn't used for wait queues which
 * are nested inside ioctx lock (i.e. ctx->wait)
 */
1534 1535
static int aio_wake_function(wait_queue_t *wait, unsigned mode,
			     int sync, void *key)
L
Linus Torvalds 已提交
1536 1537 1538 1539 1540 1541 1542 1543
{
	struct kiocb *iocb = container_of(wait, struct kiocb, ki_wait);

	list_del_init(&wait->task_list);
	kick_iocb(iocb);
	return 1;
}

1544
int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
L
Linus Torvalds 已提交
1545 1546 1547 1548 1549 1550 1551
			 struct iocb *iocb)
{
	struct kiocb *req;
	struct file *file;
	ssize_t ret;

	/* enforce forwards compatibility on users */
1552
	if (unlikely(iocb->aio_reserved1 || iocb->aio_reserved2)) {
L
Linus Torvalds 已提交
1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575
		pr_debug("EINVAL: io_submit: reserve field set\n");
		return -EINVAL;
	}

	/* prevent overflows */
	if (unlikely(
	    (iocb->aio_buf != (unsigned long)iocb->aio_buf) ||
	    (iocb->aio_nbytes != (size_t)iocb->aio_nbytes) ||
	    ((ssize_t)iocb->aio_nbytes < 0)
	   )) {
		pr_debug("EINVAL: io_submit: overflow check\n");
		return -EINVAL;
	}

	file = fget(iocb->aio_fildes);
	if (unlikely(!file))
		return -EBADF;

	req = aio_get_req(ctx);		/* returns with 2 references to req */
	if (unlikely(!req)) {
		fput(file);
		return -EAGAIN;
	}
Y
Yan Zheng 已提交
1576
	req->ki_filp = file;
1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589
	if (iocb->aio_flags & IOCB_FLAG_RESFD) {
		/*
		 * If the IOCB_FLAG_RESFD flag of aio_flags is set, get an
		 * instance of the file* now. The file descriptor must be
		 * an eventfd() fd, and will be signaled for each completed
		 * event using the eventfd_signal() function.
		 */
		req->ki_eventfd = eventfd_fget((int) iocb->aio_resfd);
		if (unlikely(IS_ERR(req->ki_eventfd))) {
			ret = PTR_ERR(req->ki_eventfd);
			goto out_put_req;
		}
	}
L
Linus Torvalds 已提交
1590

1591
	ret = put_user(req->ki_key, &user_iocb->aio_key);
L
Linus Torvalds 已提交
1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612
	if (unlikely(ret)) {
		dprintk("EFAULT: aio_key\n");
		goto out_put_req;
	}

	req->ki_obj.user = user_iocb;
	req->ki_user_data = iocb->aio_data;
	req->ki_pos = iocb->aio_offset;

	req->ki_buf = (char __user *)(unsigned long)iocb->aio_buf;
	req->ki_left = req->ki_nbytes = iocb->aio_nbytes;
	req->ki_opcode = iocb->aio_lio_opcode;
	init_waitqueue_func_entry(&req->ki_wait, aio_wake_function);
	INIT_LIST_HEAD(&req->ki_wait.task_list);

	ret = aio_setup_iocb(req);

	if (ret)
		goto out_put_req;

	spin_lock_irq(&ctx->ctx_lock);
1613 1614
	aio_run_iocb(req);
	if (!list_empty(&ctx->run_list)) {
K
Ken Chen 已提交
1615 1616 1617 1618
		/* drain the run list */
		while (__aio_run_iocbs(ctx))
			;
	}
L
Linus Torvalds 已提交
1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689
	spin_unlock_irq(&ctx->ctx_lock);
	aio_put_req(req);	/* drop extra ref to req */
	return 0;

out_put_req:
	aio_put_req(req);	/* drop extra ref to req */
	aio_put_req(req);	/* drop i/o ref to req */
	return ret;
}

/* sys_io_submit:
 *	Queue the nr iocbs pointed to by iocbpp for processing.  Returns
 *	the number of iocbs queued.  May return -EINVAL if the aio_context
 *	specified by ctx_id is invalid, if nr is < 0, if the iocb at
 *	*iocbpp[0] is not properly initialized, if the operation specified
 *	is invalid for the file descriptor in the iocb.  May fail with
 *	-EFAULT if any of the data structures point to invalid data.  May
 *	fail with -EBADF if the file descriptor specified in the first
 *	iocb is invalid.  May fail with -EAGAIN if insufficient resources
 *	are available to queue any iocbs.  Will return 0 if nr is 0.  Will
 *	fail with -ENOSYS if not implemented.
 */
asmlinkage long sys_io_submit(aio_context_t ctx_id, long nr,
			      struct iocb __user * __user *iocbpp)
{
	struct kioctx *ctx;
	long ret = 0;
	int i;

	if (unlikely(nr < 0))
		return -EINVAL;

	if (unlikely(!access_ok(VERIFY_READ, iocbpp, (nr*sizeof(*iocbpp)))))
		return -EFAULT;

	ctx = lookup_ioctx(ctx_id);
	if (unlikely(!ctx)) {
		pr_debug("EINVAL: io_submit: invalid context id\n");
		return -EINVAL;
	}

	/*
	 * AKPM: should this return a partial result if some of the IOs were
	 * successfully submitted?
	 */
	for (i=0; i<nr; i++) {
		struct iocb __user *user_iocb;
		struct iocb tmp;

		if (unlikely(__get_user(user_iocb, iocbpp + i))) {
			ret = -EFAULT;
			break;
		}

		if (unlikely(copy_from_user(&tmp, user_iocb, sizeof(tmp)))) {
			ret = -EFAULT;
			break;
		}

		ret = io_submit_one(ctx, user_iocb, &tmp);
		if (ret)
			break;
	}

	put_ioctx(ctx);
	return i ? i : ret;
}

/* lookup_kiocb
 *	Finds a given iocb for cancellation.
 */
1690 1691
static struct kiocb *lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb,
				  u32 key)
L
Linus Torvalds 已提交
1692 1693
{
	struct list_head *pos;
1694 1695 1696

	assert_spin_locked(&ctx->ctx_lock);

L
Linus Torvalds 已提交
1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758
	/* TODO: use a hash or array, this sucks. */
	list_for_each(pos, &ctx->active_reqs) {
		struct kiocb *kiocb = list_kiocb(pos);
		if (kiocb->ki_obj.user == iocb && kiocb->ki_key == key)
			return kiocb;
	}
	return NULL;
}

/* sys_io_cancel:
 *	Attempts to cancel an iocb previously passed to io_submit.  If
 *	the operation is successfully cancelled, the resulting event is
 *	copied into the memory pointed to by result without being placed
 *	into the completion queue and 0 is returned.  May fail with
 *	-EFAULT if any of the data structures pointed to are invalid.
 *	May fail with -EINVAL if aio_context specified by ctx_id is
 *	invalid.  May fail with -EAGAIN if the iocb specified was not
 *	cancelled.  Will fail with -ENOSYS if not implemented.
 */
asmlinkage long sys_io_cancel(aio_context_t ctx_id, struct iocb __user *iocb,
			      struct io_event __user *result)
{
	int (*cancel)(struct kiocb *iocb, struct io_event *res);
	struct kioctx *ctx;
	struct kiocb *kiocb;
	u32 key;
	int ret;

	ret = get_user(key, &iocb->aio_key);
	if (unlikely(ret))
		return -EFAULT;

	ctx = lookup_ioctx(ctx_id);
	if (unlikely(!ctx))
		return -EINVAL;

	spin_lock_irq(&ctx->ctx_lock);
	ret = -EAGAIN;
	kiocb = lookup_kiocb(ctx, iocb, key);
	if (kiocb && kiocb->ki_cancel) {
		cancel = kiocb->ki_cancel;
		kiocb->ki_users ++;
		kiocbSetCancelled(kiocb);
	} else
		cancel = NULL;
	spin_unlock_irq(&ctx->ctx_lock);

	if (NULL != cancel) {
		struct io_event tmp;
		pr_debug("calling cancel\n");
		memset(&tmp, 0, sizeof(tmp));
		tmp.obj = (u64)(unsigned long)kiocb->ki_obj.user;
		tmp.data = kiocb->ki_user_data;
		ret = cancel(kiocb, &tmp);
		if (!ret) {
			/* Cancellation succeeded -- copy the result
			 * into the user's buffer.
			 */
			if (copy_to_user(result, &tmp, sizeof(tmp)))
				ret = -EFAULT;
		}
	} else
1759
		ret = -EINVAL;
L
Linus Torvalds 已提交
1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792

	put_ioctx(ctx);

	return ret;
}

/* io_getevents:
 *	Attempts to read at least min_nr events and up to nr events from
 *	the completion queue for the aio_context specified by ctx_id.  May
 *	fail with -EINVAL if ctx_id is invalid, if min_nr is out of range,
 *	if nr is out of range, if when is out of range.  May fail with
 *	-EFAULT if any of the memory specified to is invalid.  May return
 *	0 or < min_nr if no events are available and the timeout specified
 *	by when	has elapsed, where when == NULL specifies an infinite
 *	timeout.  Note that the timeout pointed to by when is relative and
 *	will be updated if not NULL and the operation blocks.  Will fail
 *	with -ENOSYS if not implemented.
 */
asmlinkage long sys_io_getevents(aio_context_t ctx_id,
				 long min_nr,
				 long nr,
				 struct io_event __user *events,
				 struct timespec __user *timeout)
{
	struct kioctx *ioctx = lookup_ioctx(ctx_id);
	long ret = -EINVAL;

	if (likely(ioctx)) {
		if (likely(min_nr <= nr && min_nr >= 0 && nr >= 0))
			ret = read_events(ioctx, min_nr, nr, events, timeout);
		put_ioctx(ioctx);
	}

1793
	asmlinkage_protect(5, ret, ctx_id, min_nr, nr, events, timeout);
L
Linus Torvalds 已提交
1794 1795 1796 1797 1798 1799 1800 1801
	return ret;
}

__initcall(aio_setup);

EXPORT_SYMBOL(aio_complete);
EXPORT_SYMBOL(aio_put_req);
EXPORT_SYMBOL(wait_on_sync_kiocb);