summaryrefslogtreecommitdiff
path: root/io_uring/kbuf.c
blob: d2945c9c812b5c7e81845340d3f28192a7aab623 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
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
125
126
127
128
129
130
131
132
133
134
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
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
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
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
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
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
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
// SPDX-License-Identifier: GPL-2.0
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/namei.h>
#include <linux/poll.h>
#include <linux/vmalloc.h>
#include <linux/io_uring.h>

#include <uapi/linux/io_uring.h>

#include "io_uring.h"
#include "opdef.h"
#include "kbuf.h"
#include "memmap.h"

/* BIDs are addressed by a 16-bit field in a CQE */
#define MAX_BIDS_PER_BGID (1 << 16)

struct kmem_cache *io_buf_cachep;

struct io_provide_buf {
	struct file			*file;
	__u64				addr;
	__u32				len;
	__u32				bgid;
	__u32				nbufs;
	__u16				bid;
};

static inline struct io_buffer_list *io_buffer_get_list(struct io_ring_ctx *ctx,
							unsigned int bgid)
{
	lockdep_assert_held(&ctx->uring_lock);

	return xa_load(&ctx->io_bl_xa, bgid);
}

static int io_buffer_add_list(struct io_ring_ctx *ctx,
			      struct io_buffer_list *bl, unsigned int bgid)
{
	/*
	 * Store buffer group ID and finally mark the list as visible.
	 * The normal lookup doesn't care about the visibility as we're
	 * always under the ->uring_lock, but the RCU lookup from mmap does.
	 */
	bl->bgid = bgid;
	atomic_set(&bl->refs, 1);
	return xa_err(xa_store(&ctx->io_bl_xa, bgid, bl, GFP_KERNEL));
}

bool io_kbuf_recycle_legacy(struct io_kiocb *req, unsigned issue_flags)
{
	struct io_ring_ctx *ctx = req->ctx;
	struct io_buffer_list *bl;
	struct io_buffer *buf;

	io_ring_submit_lock(ctx, issue_flags);

	buf = req->kbuf;
	bl = io_buffer_get_list(ctx, buf->bgid);
	list_add(&buf->list, &bl->buf_list);
	req->flags &= ~REQ_F_BUFFER_SELECTED;
	req->buf_index = buf->bgid;

	io_ring_submit_unlock(ctx, issue_flags);
	return true;
}

void __io_put_kbuf(struct io_kiocb *req, unsigned issue_flags)
{
	/*
	 * We can add this buffer back to two lists:
	 *
	 * 1) The io_buffers_cache list. This one is protected by the
	 *    ctx->uring_lock. If we already hold this lock, add back to this
	 *    list as we can grab it from issue as well.
	 * 2) The io_buffers_comp list. This one is protected by the
	 *    ctx->completion_lock.
	 *
	 * We migrate buffers from the comp_list to the issue cache list
	 * when we need one.
	 */
	if (issue_flags & IO_URING_F_UNLOCKED) {
		struct io_ring_ctx *ctx = req->ctx;

		spin_lock(&ctx->completion_lock);
		__io_put_kbuf_list(req, &ctx->io_buffers_comp);
		spin_unlock(&ctx->completion_lock);
	} else {
		lockdep_assert_held(&req->ctx->uring_lock);

		__io_put_kbuf_list(req, &req->ctx->io_buffers_cache);
	}
}

static void __user *io_provided_buffer_select(struct io_kiocb *req, size_t *len,
					      struct io_buffer_list *bl)
{
	if (!list_empty(&bl->buf_list)) {
		struct io_buffer *kbuf;

		kbuf = list_first_entry(&bl->buf_list, struct io_buffer, list);
		list_del(&kbuf->list);
		if (*len == 0 || *len > kbuf->len)
			*len = kbuf->len;
		if (list_empty(&bl->buf_list))
			req->flags |= REQ_F_BL_EMPTY;
		req->flags |= REQ_F_BUFFER_SELECTED;
		req->kbuf = kbuf;
		req->buf_index = kbuf->bid;
		return u64_to_user_ptr(kbuf->addr);
	}
	return NULL;
}

static int io_provided_buffers_select(struct io_kiocb *req, size_t *len,
				      struct io_buffer_list *bl,
				      struct iovec *iov)
{
	void __user *buf;

	buf = io_provided_buffer_select(req, len, bl);
	if (unlikely(!buf))
		return -ENOBUFS;

	iov[0].iov_base = buf;
	iov[0].iov_len = *len;
	return 0;
}

static struct io_uring_buf *io_ring_head_to_buf(struct io_uring_buf_ring *br,
						__u16 head, __u16 mask)
{
	return &br->bufs[head & mask];
}

static void __user *io_ring_buffer_select(struct io_kiocb *req, size_t *len,
					  struct io_buffer_list *bl,
					  unsigned int issue_flags)
{
	struct io_uring_buf_ring *br = bl->buf_ring;
	__u16 tail, head = bl->head;
	struct io_uring_buf *buf;

	tail = smp_load_acquire(&br->tail);
	if (unlikely(tail == head))
		return NULL;

	if (head + 1 == tail)
		req->flags |= REQ_F_BL_EMPTY;

	buf = io_ring_head_to_buf(br, head, bl->mask);
	if (*len == 0 || *len > buf->len)
		*len = buf->len;
	req->flags |= REQ_F_BUFFER_RING | REQ_F_BUFFERS_COMMIT;
	req->buf_list = bl;
	req->buf_index = buf->bid;

	if (issue_flags & IO_URING_F_UNLOCKED || !io_file_can_poll(req)) {
		/*
		 * If we came in unlocked, we have no choice but to consume the
		 * buffer here, otherwise nothing ensures that the buffer won't
		 * get used by others. This does mean it'll be pinned until the
		 * IO completes, coming in unlocked means we're being called from
		 * io-wq context and there may be further retries in async hybrid
		 * mode. For the locked case, the caller must call commit when
		 * the transfer completes (or if we get -EAGAIN and must poll of
		 * retry).
		 */
		req->flags &= ~REQ_F_BUFFERS_COMMIT;
		req->buf_list = NULL;
		bl->head++;
	}
	return u64_to_user_ptr(buf->addr);
}

void __user *io_buffer_select(struct io_kiocb *req, size_t *len,
			      unsigned int issue_flags)
{
	struct io_ring_ctx *ctx = req->ctx;
	struct io_buffer_list *bl;
	void __user *ret = NULL;

	io_ring_submit_lock(req->ctx, issue_flags);

	bl = io_buffer_get_list(ctx, req->buf_index);
	if (likely(bl)) {
		if (bl->is_buf_ring)
			ret = io_ring_buffer_select(req, len, bl, issue_flags);
		else
			ret = io_provided_buffer_select(req, len, bl);
	}
	io_ring_submit_unlock(req->ctx, issue_flags);
	return ret;
}

/* cap it at a reasonable 256, will be one page even for 4K */
#define PEEK_MAX_IMPORT		256

static int io_ring_buffers_peek(struct io_kiocb *req, struct buf_sel_arg *arg,
				struct io_buffer_list *bl)
{
	struct io_uring_buf_ring *br = bl->buf_ring;
	struct iovec *iov = arg->iovs;
	int nr_iovs = arg->nr_iovs;
	__u16 nr_avail, tail, head;
	struct io_uring_buf *buf;

	tail = smp_load_acquire(&br->tail);
	head = bl->head;
	nr_avail = min_t(__u16, tail - head, UIO_MAXIOV);
	if (unlikely(!nr_avail))
		return -ENOBUFS;

	buf = io_ring_head_to_buf(br, head, bl->mask);
	if (arg->max_len) {
		int needed;

		needed = (arg->max_len + buf->len - 1) / buf->len;
		needed = min(needed, PEEK_MAX_IMPORT);
		if (nr_avail > needed)
			nr_avail = needed;
	}

	/*
	 * only alloc a bigger array if we know we have data to map, eg not
	 * a speculative peek operation.
	 */
	if (arg->mode & KBUF_MODE_EXPAND && nr_avail > nr_iovs && arg->max_len) {
		iov = kmalloc_array(nr_avail, sizeof(struct iovec), GFP_KERNEL);
		if (unlikely(!iov))
			return -ENOMEM;
		if (arg->mode & KBUF_MODE_FREE)
			kfree(arg->iovs);
		arg->iovs = iov;
		nr_iovs = nr_avail;
	} else if (nr_avail < nr_iovs) {
		nr_iovs = nr_avail;
	}

	/* set it to max, if not set, so we can use it unconditionally */
	if (!arg->max_len)
		arg->max_len = INT_MAX;

	req->buf_index = buf->bid;
	do {
		/* truncate end piece, if needed */
		if (buf->len > arg->max_len)
			buf->len = arg->max_len;

		iov->iov_base = u64_to_user_ptr(buf->addr);
		iov->iov_len = buf->len;
		iov++;

		arg->out_len += buf->len;
		arg->max_len -= buf->len;
		if (!arg->max_len)
			break;

		buf = io_ring_head_to_buf(br, ++head, bl->mask);
	} while (--nr_iovs);

	if (head == tail)
		req->flags |= REQ_F_BL_EMPTY;

	req->flags |= REQ_F_BUFFER_RING;
	req->buf_list = bl;
	return iov - arg->iovs;
}

int io_buffers_select(struct io_kiocb *req, struct buf_sel_arg *arg,
		      unsigned int issue_flags)
{
	struct io_ring_ctx *ctx = req->ctx;
	struct io_buffer_list *bl;
	int ret = -ENOENT;

	io_ring_submit_lock(ctx, issue_flags);
	bl = io_buffer_get_list(ctx, req->buf_index);
	if (unlikely(!bl))
		goto out_unlock;

	if (bl->is_buf_ring) {
		ret = io_ring_buffers_peek(req, arg, bl);
		/*
		 * Don't recycle these buffers if we need to go through poll.
		 * Nobody else can use them anyway, and holding on to provided
		 * buffers for a send/write operation would happen on the app
		 * side anyway with normal buffers. Besides, we already
		 * committed them, they cannot be put back in the queue.
		 */
		if (ret > 0) {
			req->flags |= REQ_F_BL_NO_RECYCLE;
			req->buf_list->head += ret;
		}
	} else {
		ret = io_provided_buffers_select(req, &arg->out_len, bl, arg->iovs);
	}
out_unlock:
	io_ring_submit_unlock(ctx, issue_flags);
	return ret;
}

int io_buffers_peek(struct io_kiocb *req, struct buf_sel_arg *arg)
{
	struct io_ring_ctx *ctx = req->ctx;
	struct io_buffer_list *bl;
	int ret;

	lockdep_assert_held(&ctx->uring_lock);

	bl = io_buffer_get_list(ctx, req->buf_index);
	if (unlikely(!bl))
		return -ENOENT;

	if (bl->is_buf_ring) {
		ret = io_ring_buffers_peek(req, arg, bl);
		if (ret > 0)
			req->flags |= REQ_F_BUFFERS_COMMIT;
		return ret;
	}

	/* don't support multiple buffer selections for legacy */
	return io_provided_buffers_select(req, &arg->max_len, bl, arg->iovs);
}

static int __io_remove_buffers(struct io_ring_ctx *ctx,
			       struct io_buffer_list *bl, unsigned nbufs)
{
	unsigned i = 0;

	/* shouldn't happen */
	if (!nbufs)
		return 0;

	if (bl->is_buf_ring) {
		i = bl->buf_ring->tail - bl->head;
		if (bl->buf_nr_pages) {
			int j;

			if (!bl->is_mmap) {
				for (j = 0; j < bl->buf_nr_pages; j++)
					unpin_user_page(bl->buf_pages[j]);
			}
			io_pages_unmap(bl->buf_ring, &bl->buf_pages,
					&bl->buf_nr_pages, bl->is_mmap);
			bl->is_mmap = 0;
		}
		/* make sure it's seen as empty */
		INIT_LIST_HEAD(&bl->buf_list);
		bl->is_buf_ring = 0;
		return i;
	}

	/* protects io_buffers_cache */
	lockdep_assert_held(&ctx->uring_lock);

	while (!list_empty(&bl->buf_list)) {
		struct io_buffer *nxt;

		nxt = list_first_entry(&bl->buf_list, struct io_buffer, list);
		list_move(&nxt->list, &ctx->io_buffers_cache);
		if (++i == nbufs)
			return i;
		cond_resched();
	}

	return i;
}

void io_put_bl(struct io_ring_ctx *ctx, struct io_buffer_list *bl)
{
	if (atomic_dec_and_test(&bl->refs)) {
		__io_remove_buffers(ctx, bl, -1U);
		kfree_rcu(bl, rcu);
	}
}

void io_destroy_buffers(struct io_ring_ctx *ctx)
{
	struct io_buffer_list *bl;
	struct list_head *item, *tmp;
	struct io_buffer *buf;
	unsigned long index;

	xa_for_each(&ctx->io_bl_xa, index, bl) {
		xa_erase(&ctx->io_bl_xa, bl->bgid);
		io_put_bl(ctx, bl);
	}

	/*
	 * Move deferred locked entries to cache before pruning
	 */
	spin_lock(&ctx->completion_lock);
	if (!list_empty(&ctx->io_buffers_comp))
		list_splice_init(&ctx->io_buffers_comp, &ctx->io_buffers_cache);
	spin_unlock(&ctx->completion_lock);

	list_for_each_safe(item, tmp, &ctx->io_buffers_cache) {
		buf = list_entry(item, struct io_buffer, list);
		kmem_cache_free(io_buf_cachep, buf);
	}
}

int io_remove_buffers_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
	struct io_provide_buf *p = io_kiocb_to_cmd(req, struct io_provide_buf);
	u64 tmp;

	if (sqe->rw_flags || sqe->addr || sqe->len || sqe->off ||
	    sqe->splice_fd_in)
		return -EINVAL;

	tmp = READ_ONCE(sqe->fd);
	if (!tmp || tmp > MAX_BIDS_PER_BGID)
		return -EINVAL;

	memset(p, 0, sizeof(*p));
	p->nbufs = tmp;
	p->bgid = READ_ONCE(sqe->buf_group);
	return 0;
}

int io_remove_buffers(struct io_kiocb *req, unsigned int issue_flags)
{
	struct io_provide_buf *p = io_kiocb_to_cmd(req, struct io_provide_buf);
	struct io_ring_ctx *ctx = req->ctx;
	struct io_buffer_list *bl;
	int ret = 0;

	io_ring_submit_lock(ctx, issue_flags);

	ret = -ENOENT;
	bl = io_buffer_get_list(ctx, p->bgid);
	if (bl) {
		ret = -EINVAL;
		/* can't use provide/remove buffers command on mapped buffers */
		if (!bl->is_buf_ring)
			ret = __io_remove_buffers(ctx, bl, p->nbufs);
	}
	io_ring_submit_unlock(ctx, issue_flags);
	if (ret < 0)
		req_set_fail(req);
	io_req_set_res(req, ret, 0);
	return IOU_OK;
}

int io_provide_buffers_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
	unsigned long size, tmp_check;
	struct io_provide_buf *p = io_kiocb_to_cmd(req, struct io_provide_buf);
	u64 tmp;

	if (sqe->rw_flags || sqe->splice_fd_in)
		return -EINVAL;

	tmp = READ_ONCE(sqe->fd);
	if (!tmp || tmp > MAX_BIDS_PER_BGID)
		return -E2BIG;
	p->nbufs = tmp;
	p->addr = READ_ONCE(sqe->addr);
	p->len = READ_ONCE(sqe->len);

	if (check_mul_overflow((unsigned long)p->len, (unsigned long)p->nbufs,
				&size))
		return -EOVERFLOW;
	if (check_add_overflow((unsigned long)p->addr, size, &tmp_check))
		return -EOVERFLOW;

	size = (unsigned long)p->len * p->nbufs;
	if (!access_ok(u64_to_user_ptr(p->addr), size))
		return -EFAULT;

	p->bgid = READ_ONCE(sqe->buf_group);
	tmp = READ_ONCE(sqe->off);
	if (tmp > USHRT_MAX)
		return -E2BIG;
	if (tmp + p->nbufs > MAX_BIDS_PER_BGID)
		return -EINVAL;
	p->bid = tmp;
	return 0;
}

#define IO_BUFFER_ALLOC_BATCH 64

static int io_refill_buffer_cache(struct io_ring_ctx *ctx)
{
	struct io_buffer *bufs[IO_BUFFER_ALLOC_BATCH];
	int allocated;

	/*
	 * Completions that don't happen inline (eg not under uring_lock) will
	 * add to ->io_buffers_comp. If we don't have any free buffers, check
	 * the completion list and splice those entries first.
	 */
	if (!list_empty_careful(&ctx->io_buffers_comp)) {
		spin_lock(&ctx->completion_lock);
		if (!list_empty(&ctx->io_buffers_comp)) {
			list_splice_init(&ctx->io_buffers_comp,
						&ctx->io_buffers_cache);
			spin_unlock(&ctx->completion_lock);
			return 0;
		}
		spin_unlock(&ctx->completion_lock);
	}

	/*
	 * No free buffers and no completion entries either. Allocate a new
	 * batch of buffer entries and add those to our freelist.
	 */

	allocated = kmem_cache_alloc_bulk(io_buf_cachep, GFP_KERNEL_ACCOUNT,
					  ARRAY_SIZE(bufs), (void **) bufs);
	if (unlikely(!allocated)) {
		/*
		 * Bulk alloc is all-or-nothing. If we fail to get a batch,
		 * retry single alloc to be on the safe side.
		 */
		bufs[0] = kmem_cache_alloc(io_buf_cachep, GFP_KERNEL);
		if (!bufs[0])
			return -ENOMEM;
		allocated = 1;
	}

	while (allocated)
		list_add_tail(&bufs[--allocated]->list, &ctx->io_buffers_cache);

	return 0;
}

static int io_add_buffers(struct io_ring_ctx *ctx, struct io_provide_buf *pbuf,
			  struct io_buffer_list *bl)
{
	struct io_buffer *buf;
	u64 addr = pbuf->addr;
	int i, bid = pbuf->bid;

	for (i = 0; i < pbuf->nbufs; i++) {
		if (list_empty(&ctx->io_buffers_cache) &&
		    io_refill_buffer_cache(ctx))
			break;
		buf = list_first_entry(&ctx->io_buffers_cache, struct io_buffer,
					list);
		list_move_tail(&buf->list, &bl->buf_list);
		buf->addr = addr;
		buf->len = min_t(__u32, pbuf->len, MAX_RW_COUNT);
		buf->bid = bid;
		buf->bgid = pbuf->bgid;
		addr += pbuf->len;
		bid++;
		cond_resched();
	}

	return i ? 0 : -ENOMEM;
}

int io_provide_buffers(struct io_kiocb *req, unsigned int issue_flags)
{
	struct io_provide_buf *p = io_kiocb_to_cmd(req, struct io_provide_buf);
	struct io_ring_ctx *ctx = req->ctx;
	struct io_buffer_list *bl;
	int ret = 0;

	io_ring_submit_lock(ctx, issue_flags);

	bl = io_buffer_get_list(ctx, p->bgid);
	if (unlikely(!bl)) {
		bl = kzalloc(sizeof(*bl), GFP_KERNEL_ACCOUNT);
		if (!bl) {
			ret = -ENOMEM;
			goto err;
		}
		INIT_LIST_HEAD(&bl->buf_list);
		ret = io_buffer_add_list(ctx, bl, p->bgid);
		if (ret) {
			/*
			 * Doesn't need rcu free as it was never visible, but
			 * let's keep it consistent throughout.
			 */
			kfree_rcu(bl, rcu);
			goto err;
		}
	}
	/* can't add buffers via this command for a mapped buffer ring */
	if (bl->is_buf_ring) {
		ret = -EINVAL;
		goto err;
	}

	ret = io_add_buffers(ctx, p, bl);
err:
	io_ring_submit_unlock(ctx, issue_flags);

	if (ret < 0)
		req_set_fail(req);
	io_req_set_res(req, ret, 0);
	return IOU_OK;
}

static int io_pin_pbuf_ring(struct io_uring_buf_reg *reg,
			    struct io_buffer_list *bl)
{
	struct io_uring_buf_ring *br = NULL;
	struct page **pages;
	int nr_pages, ret;

	pages = io_pin_pages(reg->ring_addr,
			     flex_array_size(br, bufs, reg->ring_entries),
			     &nr_pages);
	if (IS_ERR(pages))
		return PTR_ERR(pages);

	br = vmap(pages, nr_pages, VM_MAP, PAGE_KERNEL);
	if (!br) {
		ret = -ENOMEM;
		goto error_unpin;
	}

#ifdef SHM_COLOUR
	/*
	 * On platforms that have specific aliasing requirements, SHM_COLOUR
	 * is set and we must guarantee that the kernel and user side align
	 * nicely. We cannot do that if IOU_PBUF_RING_MMAP isn't set and
	 * the application mmap's the provided ring buffer. Fail the request
	 * if we, by chance, don't end up with aligned addresses. The app
	 * should use IOU_PBUF_RING_MMAP instead, and liburing will handle
	 * this transparently.
	 */
	if ((reg->ring_addr | (unsigned long) br) & (SHM_COLOUR - 1)) {
		ret = -EINVAL;
		goto error_unpin;
	}
#endif
	bl->buf_pages = pages;
	bl->buf_nr_pages = nr_pages;
	bl->buf_ring = br;
	bl->is_buf_ring = 1;
	bl->is_mmap = 0;
	return 0;
error_unpin:
	unpin_user_pages(pages, nr_pages);
	kvfree(pages);
	vunmap(br);
	return ret;
}

static int io_alloc_pbuf_ring(struct io_ring_ctx *ctx,
			      struct io_uring_buf_reg *reg,
			      struct io_buffer_list *bl)
{
	size_t ring_size;

	ring_size = reg->ring_entries * sizeof(struct io_uring_buf_ring);

	bl->buf_ring = io_pages_map(&bl->buf_pages, &bl->buf_nr_pages, ring_size);
	if (!bl->buf_ring)
		return -ENOMEM;

	bl->is_buf_ring = 1;
	bl->is_mmap = 1;
	return 0;
}

int io_register_pbuf_ring(struct io_ring_ctx *ctx, void __user *arg)
{
	struct io_uring_buf_reg reg;
	struct io_buffer_list *bl, *free_bl = NULL;
	int ret;

	lockdep_assert_held(&ctx->uring_lock);

	if (copy_from_user(&reg, arg, sizeof(reg)))
		return -EFAULT;

	if (reg.resv[0] || reg.resv[1] || reg.resv[2])
		return -EINVAL;
	if (reg.flags & ~IOU_PBUF_RING_MMAP)
		return -EINVAL;
	if (!(reg.flags & IOU_PBUF_RING_MMAP)) {
		if (!reg.ring_addr)
			return -EFAULT;
		if (reg.ring_addr & ~PAGE_MASK)
			return -EINVAL;
	} else {
		if (reg.ring_addr)
			return -EINVAL;
	}

	if (!is_power_of_2(reg.ring_entries))
		return -EINVAL;

	/* cannot disambiguate full vs empty due to head/tail size */
	if (reg.ring_entries >= 65536)
		return -EINVAL;

	bl = io_buffer_get_list(ctx, reg.bgid);
	if (bl) {
		/* if mapped buffer ring OR classic exists, don't allow */
		if (bl->is_buf_ring || !list_empty(&bl->buf_list))
			return -EEXIST;
	} else {
		free_bl = bl = kzalloc(sizeof(*bl), GFP_KERNEL);
		if (!bl)
			return -ENOMEM;
	}

	if (!(reg.flags & IOU_PBUF_RING_MMAP))
		ret = io_pin_pbuf_ring(&reg, bl);
	else
		ret = io_alloc_pbuf_ring(ctx, &reg, bl);

	if (!ret) {
		bl->nr_entries = reg.ring_entries;
		bl->mask = reg.ring_entries - 1;

		io_buffer_add_list(ctx, bl, reg.bgid);
		return 0;
	}

	kfree_rcu(free_bl, rcu);
	return ret;
}

int io_unregister_pbuf_ring(struct io_ring_ctx *ctx, void __user *arg)
{
	struct io_uring_buf_reg reg;
	struct io_buffer_list *bl;

	lockdep_assert_held(&ctx->uring_lock);

	if (copy_from_user(&reg, arg, sizeof(reg)))
		return -EFAULT;
	if (reg.resv[0] || reg.resv[1] || reg.resv[2])
		return -EINVAL;
	if (reg.flags)
		return -EINVAL;

	bl = io_buffer_get_list(ctx, reg.bgid);
	if (!bl)
		return -ENOENT;
	if (!bl->is_buf_ring)
		return -EINVAL;

	xa_erase(&ctx->io_bl_xa, bl->bgid);
	io_put_bl(ctx, bl);
	return 0;
}

int io_register_pbuf_status(struct io_ring_ctx *ctx, void __user *arg)
{
	struct io_uring_buf_status buf_status;
	struct io_buffer_list *bl;
	int i;

	if (copy_from_user(&buf_status, arg, sizeof(buf_status)))
		return -EFAULT;

	for (i = 0; i < ARRAY_SIZE(buf_status.resv); i++)
		if (buf_status.resv[i])
			return -EINVAL;

	bl = io_buffer_get_list(ctx, buf_status.buf_group);
	if (!bl)
		return -ENOENT;
	if (!bl->is_buf_ring)
		return -EINVAL;

	buf_status.head = bl->head;
	if (copy_to_user(arg, &buf_status, sizeof(buf_status)))
		return -EFAULT;

	return 0;
}

struct io_buffer_list *io_pbuf_get_bl(struct io_ring_ctx *ctx,
				      unsigned long bgid)
{
	struct io_buffer_list *bl;
	bool ret;

	/*
	 * We have to be a bit careful here - we're inside mmap and cannot grab
	 * the uring_lock. This means the buffer_list could be simultaneously
	 * going away, if someone is trying to be sneaky. Look it up under rcu
	 * so we know it's not going away, and attempt to grab a reference to
	 * it. If the ref is already zero, then fail the mapping. If successful,
	 * the caller will call io_put_bl() to drop the the reference at at the
	 * end. This may then safely free the buffer_list (and drop the pages)
	 * at that point, vm_insert_pages() would've already grabbed the
	 * necessary vma references.
	 */
	rcu_read_lock();
	bl = xa_load(&ctx->io_bl_xa, bgid);
	/* must be a mmap'able buffer ring and have pages */
	ret = false;
	if (bl && bl->is_mmap)
		ret = atomic_inc_not_zero(&bl->refs);
	rcu_read_unlock();

	if (ret)
		return bl;

	return ERR_PTR(-EINVAL);
}

int io_pbuf_mmap(struct file *file, struct vm_area_struct *vma)
{
	struct io_ring_ctx *ctx = file->private_data;
	loff_t pgoff = vma->vm_pgoff << PAGE_SHIFT;
	struct io_buffer_list *bl;
	int bgid, ret;

	bgid = (pgoff & ~IORING_OFF_MMAP_MASK) >> IORING_OFF_PBUF_SHIFT;
	bl = io_pbuf_get_bl(ctx, bgid);
	if (IS_ERR(bl))
		return PTR_ERR(bl);

	ret = io_uring_mmap_pages(ctx, vma, bl->buf_pages, bl->buf_nr_pages);
	io_put_bl(ctx, bl);
	return ret;
}