summaryrefslogtreecommitdiff
path: root/fs/nfs/read.c
blob: a6103333b6665695e070c898e9f31642e2551c94 (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
// SPDX-License-Identifier: GPL-2.0-only
/*
 * linux/fs/nfs/read.c
 *
 * Block I/O for NFS
 *
 * Partial copy of Linus' read cache modifications to fs/nfs/file.c
 * modified for async RPC by okir@monad.swb.de
 */

#include <linux/time.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/fcntl.h>
#include <linux/stat.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/task_io_accounting_ops.h>
#include <linux/pagemap.h>
#include <linux/sunrpc/clnt.h>
#include <linux/nfs_fs.h>
#include <linux/nfs_page.h>
#include <linux/module.h>

#include "nfs4_fs.h"
#include "internal.h"
#include "iostat.h"
#include "fscache.h"
#include "pnfs.h"
#include "nfstrace.h"
#include "delegation.h"

#define NFSDBG_FACILITY		NFSDBG_PAGECACHE

const struct nfs_pgio_completion_ops nfs_async_read_completion_ops;
static const struct nfs_rw_ops nfs_rw_read_ops;

static struct kmem_cache *nfs_rdata_cachep;

static struct nfs_pgio_header *nfs_readhdr_alloc(void)
{
	struct nfs_pgio_header *p = kmem_cache_zalloc(nfs_rdata_cachep, GFP_KERNEL);

	if (p)
		p->rw_mode = FMODE_READ;
	return p;
}

static void nfs_readhdr_free(struct nfs_pgio_header *rhdr)
{
	if (rhdr->res.scratch != NULL)
		kfree(rhdr->res.scratch);
	kmem_cache_free(nfs_rdata_cachep, rhdr);
}

static int nfs_return_empty_folio(struct folio *folio)
{
	folio_zero_segment(folio, 0, folio_size(folio));
	folio_mark_uptodate(folio);
	folio_unlock(folio);
	return 0;
}

void nfs_pageio_init_read(struct nfs_pageio_descriptor *pgio,
			      struct inode *inode, bool force_mds,
			      const struct nfs_pgio_completion_ops *compl_ops)
{
	struct nfs_server *server = NFS_SERVER(inode);
	const struct nfs_pageio_ops *pg_ops = &nfs_pgio_rw_ops;

#ifdef CONFIG_NFS_V4_1
	if (server->pnfs_curr_ld && !force_mds)
		pg_ops = server->pnfs_curr_ld->pg_read_ops;
#endif
	nfs_pageio_init(pgio, inode, pg_ops, compl_ops, &nfs_rw_read_ops,
			server->rsize, 0);
}
EXPORT_SYMBOL_GPL(nfs_pageio_init_read);

void nfs_pageio_complete_read(struct nfs_pageio_descriptor *pgio)
{
	struct nfs_pgio_mirror *pgm;
	unsigned long npages;

	nfs_pageio_complete(pgio);

	/* It doesn't make sense to do mirrored reads! */
	WARN_ON_ONCE(pgio->pg_mirror_count != 1);

	pgm = &pgio->pg_mirrors[0];
	NFS_I(pgio->pg_inode)->read_io += pgm->pg_bytes_written;
	npages = (pgm->pg_bytes_written + PAGE_SIZE - 1) >> PAGE_SHIFT;
	nfs_add_stats(pgio->pg_inode, NFSIOS_READPAGES, npages);
}


void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio)
{
	struct nfs_pgio_mirror *mirror;

	if (pgio->pg_ops && pgio->pg_ops->pg_cleanup)
		pgio->pg_ops->pg_cleanup(pgio);

	pgio->pg_ops = &nfs_pgio_rw_ops;

	/* read path should never have more than one mirror */
	WARN_ON_ONCE(pgio->pg_mirror_count != 1);

	mirror = &pgio->pg_mirrors[0];
	mirror->pg_bsize = NFS_SERVER(pgio->pg_inode)->rsize;
}
EXPORT_SYMBOL_GPL(nfs_pageio_reset_read_mds);

bool nfs_read_alloc_scratch(struct nfs_pgio_header *hdr, size_t size)
{
	WARN_ON(hdr->res.scratch != NULL);
	hdr->res.scratch = kmalloc(size, GFP_KERNEL);
	return hdr->res.scratch != NULL;
}
EXPORT_SYMBOL_GPL(nfs_read_alloc_scratch);

static void nfs_readpage_release(struct nfs_page *req, int error)
{
	struct folio *folio = nfs_page_to_folio(req);

	if (nfs_page_group_sync_on_bit(req, PG_UNLOCKPAGE))
		if (nfs_netfs_folio_unlock(folio))
			folio_unlock(folio);

	nfs_release_request(req);
}

static void nfs_page_group_set_uptodate(struct nfs_page *req)
{
	if (nfs_page_group_sync_on_bit(req, PG_UPTODATE))
		folio_mark_uptodate(nfs_page_to_folio(req));
}

static void nfs_read_completion(struct nfs_pgio_header *hdr)
{
	unsigned long bytes = 0;
	int error;

	if (test_bit(NFS_IOHDR_REDO, &hdr->flags))
		goto out;
	while (!list_empty(&hdr->pages)) {
		struct nfs_page *req = nfs_list_entry(hdr->pages.next);
		struct folio *folio = nfs_page_to_folio(req);
		unsigned long start = req->wb_pgbase;
		unsigned long end = req->wb_pgbase + req->wb_bytes;

		if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) {
			/* note: regions of the page not covered by a
			 * request are zeroed in nfs_read_add_folio
			 */
			if (bytes > hdr->good_bytes) {
				/* nothing in this request was good, so zero
				 * the full extent of the request */
				folio_zero_segment(folio, start, end);

			} else if (hdr->good_bytes - bytes < req->wb_bytes) {
				/* part of this request has good bytes, but
				 * not all. zero the bad bytes */
				start += hdr->good_bytes - bytes;
				WARN_ON(start < req->wb_pgbase);
				folio_zero_segment(folio, start, end);
			}
		}
		error = 0;
		bytes += req->wb_bytes;
		if (test_bit(NFS_IOHDR_ERROR, &hdr->flags)) {
			if (bytes <= hdr->good_bytes)
				nfs_page_group_set_uptodate(req);
			else {
				error = hdr->error;
				xchg(&nfs_req_openctx(req)->error, error);
			}
		} else
			nfs_page_group_set_uptodate(req);
		nfs_list_remove_request(req);
		nfs_readpage_release(req, error);
	}
	nfs_netfs_read_completion(hdr);

out:
	hdr->release(hdr);
}

static void nfs_initiate_read(struct nfs_pgio_header *hdr,
			      struct rpc_message *msg,
			      const struct nfs_rpc_ops *rpc_ops,
			      struct rpc_task_setup *task_setup_data, int how)
{
	rpc_ops->read_setup(hdr, msg);
	nfs_netfs_initiate_read(hdr);
	trace_nfs_initiate_read(hdr);
}

static void
nfs_async_read_error(struct list_head *head, int error)
{
	struct nfs_page	*req;

	while (!list_empty(head)) {
		req = nfs_list_entry(head->next);
		nfs_list_remove_request(req);
		nfs_readpage_release(req, error);
	}
}

const struct nfs_pgio_completion_ops nfs_async_read_completion_ops = {
	.error_cleanup = nfs_async_read_error,
	.completion = nfs_read_completion,
};

/*
 * This is the callback from RPC telling us whether a reply was
 * received or some error occurred (timeout or socket shutdown).
 */
static int nfs_readpage_done(struct rpc_task *task,
			     struct nfs_pgio_header *hdr,
			     struct inode *inode)
{
	int status = NFS_PROTO(inode)->read_done(task, hdr);
	if (status != 0)
		return status;

	nfs_add_stats(inode, NFSIOS_SERVERREADBYTES, hdr->res.count);
	trace_nfs_readpage_done(task, hdr);

	if (task->tk_status == -ESTALE) {
		nfs_set_inode_stale(inode);
		nfs_mark_for_revalidate(inode);
	}
	return 0;
}

static void nfs_readpage_retry(struct rpc_task *task,
			       struct nfs_pgio_header *hdr)
{
	struct nfs_pgio_args *argp = &hdr->args;
	struct nfs_pgio_res  *resp = &hdr->res;

	/* This is a short read! */
	nfs_inc_stats(hdr->inode, NFSIOS_SHORTREAD);
	trace_nfs_readpage_short(task, hdr);

	/* Has the server at least made some progress? */
	if (resp->count == 0) {
		nfs_set_pgio_error(hdr, -EIO, argp->offset);
		return;
	}

	/* For non rpc-based layout drivers, retry-through-MDS */
	if (!task->tk_ops) {
		hdr->pnfs_error = -EAGAIN;
		return;
	}

	/* Yes, so retry the read at the end of the hdr */
	hdr->mds_offset += resp->count;
	argp->offset += resp->count;
	argp->pgbase += resp->count;
	argp->count -= resp->count;
	resp->count = 0;
	resp->eof = 0;
	rpc_restart_call_prepare(task);
}

static void nfs_readpage_result(struct rpc_task *task,
				struct nfs_pgio_header *hdr)
{
	if (hdr->res.eof) {
		loff_t pos = hdr->args.offset + hdr->res.count;
		unsigned int new = pos - hdr->io_start;

		if (hdr->good_bytes > new) {
			hdr->good_bytes = new;
			set_bit(NFS_IOHDR_EOF, &hdr->flags);
			clear_bit(NFS_IOHDR_ERROR, &hdr->flags);
		}
	} else if (hdr->res.count < hdr->args.count)
		nfs_readpage_retry(task, hdr);
}

int nfs_read_add_folio(struct nfs_pageio_descriptor *pgio,
		       struct nfs_open_context *ctx,
		       struct folio *folio)
{
	struct inode *inode = folio->mapping->host;
	struct nfs_server *server = NFS_SERVER(inode);
	size_t fsize = folio_size(folio);
	unsigned int rsize = server->rsize;
	struct nfs_page *new;
	unsigned int len, aligned_len;
	int error;

	len = nfs_folio_length(folio);
	if (len == 0)
		return nfs_return_empty_folio(folio);

	aligned_len = min_t(unsigned int, ALIGN(len, rsize), fsize);

	new = nfs_page_create_from_folio(ctx, folio, 0, aligned_len);
	if (IS_ERR(new)) {
		error = PTR_ERR(new);
		if (nfs_netfs_folio_unlock(folio))
			folio_unlock(folio);
		goto out;
	}

	if (len < fsize)
		folio_zero_segment(folio, len, fsize);
	if (!nfs_pageio_add_request(pgio, new)) {
		nfs_list_remove_request(new);
		error = pgio->pg_error;
		nfs_readpage_release(new, error);
		goto out;
	}
	return 0;
out:
	return error;
}

/*
 * Actually read a folio over the wire.
 */
static int nfs_do_read_folio(struct file *file, struct folio *folio)
{
	struct inode *inode = file_inode(file);
	struct nfs_pageio_descriptor pgio;
	struct nfs_open_context *ctx;
	int ret;

	ctx = get_nfs_open_context(nfs_file_open_context(file));

	xchg(&ctx->error, 0);
	nfs_pageio_init_read(&pgio, inode, false,
			     &nfs_async_read_completion_ops);

	ret = nfs_read_add_folio(&pgio, ctx, folio);
	if (ret)
		goto out_put;

	nfs_pageio_complete_read(&pgio);
	nfs_update_delegated_atime(inode);
	if (pgio.pg_error < 0) {
		ret = pgio.pg_error;
		goto out_put;
	}

	ret = folio_wait_locked_killable(folio);
	if (!folio_test_uptodate(folio) && !ret)
		ret = xchg(&ctx->error, 0);

out_put:
	put_nfs_open_context(ctx);
	return ret;
}

/*
 * Synchronously read a folio.
 *
 * This is not heavily used as most users to try an asynchronous
 * large read through ->readahead first.
 */
int nfs_read_folio(struct file *file, struct folio *folio)
{
	struct inode *inode = file_inode(file);
	loff_t pos = folio_pos(folio);
	size_t len = folio_size(folio);
	int ret;

	trace_nfs_aop_readpage(inode, pos, len);
	nfs_inc_stats(inode, NFSIOS_VFSREADPAGE);
	task_io_account_read(len);

	/*
	 * Try to flush any pending writes to the file..
	 *
	 * NOTE! Because we own the folio lock, there cannot
	 * be any new pending writes generated at this point
	 * for this folio (other folios can be written to).
	 */
	ret = nfs_wb_folio(inode, folio);
	if (ret)
		goto out_unlock;
	if (folio_test_uptodate(folio))
		goto out_unlock;

	ret = -ESTALE;
	if (NFS_STALE(inode))
		goto out_unlock;

	ret = nfs_netfs_read_folio(file, folio);
	if (ret)
		ret = nfs_do_read_folio(file, folio);
out:
	trace_nfs_aop_readpage_done(inode, pos, len, ret);
	return ret;
out_unlock:
	folio_unlock(folio);
	goto out;
}

void nfs_readahead(struct readahead_control *ractl)
{
	struct nfs_pageio_descriptor pgio;
	struct nfs_open_context *ctx;
	unsigned int nr_pages = readahead_count(ractl);
	struct file *file = ractl->file;
	struct inode *inode = ractl->mapping->host;
	struct folio *folio;
	int ret;

	trace_nfs_aop_readahead(inode, readahead_pos(ractl), nr_pages);
	nfs_inc_stats(inode, NFSIOS_VFSREADPAGES);
	task_io_account_read(readahead_length(ractl));

	ret = -ESTALE;
	if (NFS_STALE(inode))
		goto out;

	ret = nfs_netfs_readahead(ractl);
	if (!ret)
		goto out;

	if (file == NULL) {
		ret = -EBADF;
		ctx = nfs_find_open_context(inode, NULL, FMODE_READ);
		if (ctx == NULL)
			goto out;
	} else
		ctx = get_nfs_open_context(nfs_file_open_context(file));

	nfs_pageio_init_read(&pgio, inode, false,
			     &nfs_async_read_completion_ops);

	while ((folio = readahead_folio(ractl)) != NULL) {
		ret = nfs_read_add_folio(&pgio, ctx, folio);
		if (ret)
			break;
	}

	nfs_pageio_complete_read(&pgio);
	nfs_update_delegated_atime(inode);

	put_nfs_open_context(ctx);
out:
	trace_nfs_aop_readahead_done(inode, nr_pages, ret);
}

int __init nfs_init_readpagecache(void)
{
	nfs_rdata_cachep = kmem_cache_create("nfs_read_data",
					     sizeof(struct nfs_pgio_header),
					     0, SLAB_HWCACHE_ALIGN,
					     NULL);
	if (nfs_rdata_cachep == NULL)
		return -ENOMEM;

	return 0;
}

void nfs_destroy_readpagecache(void)
{
	kmem_cache_destroy(nfs_rdata_cachep);
}

static const struct nfs_rw_ops nfs_rw_read_ops = {
	.rw_alloc_header	= nfs_readhdr_alloc,
	.rw_free_header		= nfs_readhdr_free,
	.rw_done		= nfs_readpage_done,
	.rw_result		= nfs_readpage_result,
	.rw_initiate		= nfs_initiate_read,
};