aboutsummaryrefslogtreecommitdiffstats
path: root/fs/nfs/read.c
blob: 8fd6dfbe1bc3b8c248595a85fc5d3f3b86779ee3 (plain) (blame)
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
/*
 * 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/pagemap.h>
#include <linux/sunrpc/clnt.h>
#include <linux/nfs_fs.h>
#include <linux/nfs_page.h>
#include <linux/smp_lock.h>

#include <asm/system.h>

#include "internal.h"
#include "iostat.h"

#define NFSDBG_FACILITY		NFSDBG_PAGECACHE

static int nfs_pagein_multi(struct inode *, struct list_head *, unsigned int, size_t, int);
static int nfs_pagein_one(struct inode *, struct list_head *, unsigned int, size_t, int);
static const struct rpc_call_ops nfs_read_partial_ops;
static const struct rpc_call_ops nfs_read_full_ops;

static struct kmem_cache *nfs_rdata_cachep;
static mempool_t *nfs_rdata_mempool;

#define MIN_POOL_READ	(32)

struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount)
{
	struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, GFP_NOFS);

	if (p) {
		memset(p, 0, sizeof(*p));
		INIT_LIST_HEAD(&p->pages);
		p->npages = pagecount;
		if (pagecount <= ARRAY_SIZE(p->page_array))
			p->pagevec = p->page_array;
		else {
			p->pagevec = kcalloc(pagecount, sizeof(struct page *), GFP_NOFS);
			if (!p->pagevec) {
				mempool_free(p, nfs_rdata_mempool);
				p = NULL;
			}
		}
	}
	return p;
}

static void nfs_readdata_rcu_free(struct rcu_head *head)
{
	struct nfs_read_data *p = container_of(head, struct nfs_read_data, task.u.tk_rcu);
	if (p && (p->pagevec != &p->page_array[0]))
		kfree(p->pagevec);
	mempool_free(p, nfs_rdata_mempool);
}

static void nfs_readdata_free(struct nfs_read_data *rdata)
{
	call_rcu_bh(&rdata->task.u.tk_rcu, nfs_readdata_rcu_free);
}

void nfs_readdata_release(void *data)
{
        nfs_readdata_free(data);
}

static
int nfs_return_empty_page(struct page *page)
{
	zero_user_page(page, 0, PAGE_CACHE_SIZE, KM_USER0);
	SetPageUptodate(page);
	unlock_page(page);
	return 0;
}

static void nfs_readpage_truncate_uninitialised_page(struct nfs_read_data *data)
{
	unsigned int remainder = data->args.count - data->res.count;
	unsigned int base = data->args.pgbase + data->res.count;
	unsigned int pglen;
	struct page **pages;

	if (data->res.eof == 0 || remainder == 0)
		return;
	/*
	 * Note: "remainder" can never be negative, since we check for
	 * 	this in the XDR code.
	 */
	pages = &data->args.pages[base >> PAGE_CACHE_SHIFT];
	base &= ~PAGE_CACHE_MASK;
	pglen = PAGE_CACHE_SIZE - base;
	for (;;) {
		if (remainder <= pglen) {
			zero_user_page(*pages, base, remainder, KM_USER0);
			break;
		}
		zero_user_page(*pages, base, pglen, KM_USER0);
		pages++;
		remainder -= pglen;
		pglen = PAGE_CACHE_SIZE;
		base = 0;
	}
}

static int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode,
		struct page *page)
{
	LIST_HEAD(one_request);
	struct nfs_page	*new;
	unsigned int len;

	len = nfs_page_length(page);
	if (len == 0)
		return nfs_return_empty_page(page);
	new = nfs_create_request(ctx, inode, page, 0, len);
	if (IS_ERR(new)) {
		unlock_page(page);
		return PTR_ERR(new);
	}
	if (len < PAGE_CACHE_SIZE)
		zero_user_page(page, len, PAGE_CACHE_SIZE - len, KM_USER0);

	nfs_list_add_request(new, &one_request);
	if (NFS_SERVER(inode)->rsize < PAGE_CACHE_SIZE)
		nfs_pagein_multi(inode, &one_request, 1, len, 0);
	else
		nfs_pagein_one(inode, &one_request, 1, len, 0);
	return 0;
}

static void nfs_readpage_release(struct nfs_page *req)
{
	unlock_page(req->wb_page);

	dprintk("NFS: read done (%s/%Ld %d@%Ld)\n",
			req->wb_context->path.dentry->d_inode->i_sb->s_id,
			(long long)NFS_FILEID(req->wb_context->path.dentry->d_inode),
			req->wb_bytes,
			(long long)req_offset(req));
	nfs_clear_request(req);
	nfs_release_request(req);
}

/*
 * Set up the NFS read request struct
 */
static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
		const struct rpc_call_ops *call_ops,
		unsigned int count, unsigned int offset)
{
	struct inode *inode = req->wb_context->path.dentry->d_inode;
	int swap_flags = IS_SWAPFILE(inode) ? NFS_RPC_SWAPFLAGS : 0;
	struct rpc_task *task;
	struct rpc_message msg = {
		.rpc_argp = &data->args,
		.rpc_resp = &data->res,
		.rpc_cred = req->wb_context->cred,
	};
	struct rpc_task_setup task_setup_data = {
		.task = &data->task,
		.rpc_client = NFS_CLIENT(inode),
		.rpc_message = &msg,
		.callback_ops = call_ops,
		.callback_data = data,
		.flags = RPC_TASK_ASYNC | swap_flags,
	};

	data->req	  = req;
	data->inode	  = inode;
	data->cred	  = msg.rpc_cred;

	data->args.fh     = NFS_FH(inode);
	data->args.offset = req_offset(req) + offset;
	data->args.pgbase = req->wb_pgbase + offset;
	data->args.pages  = data->pagevec;
	data->args.count  = count;
	data->args.context = req->wb_context;

	data->res.fattr   = &data->fattr;
	data->res.count   = count;
	data->res.eof     = 0;
	nfs_fattr_init(&data->fattr);

	/* Set up the initial task struct. */
	NFS_PROTO(inode)->read_setup(data, &msg);

	dprintk("NFS: %5u initiated read call (req %s/%Ld, %u bytes @ offset %Lu)\n",
			data->task.tk_pid,
			inode->i_sb->s_id,
			(long long)NFS_FILEID(inode),
			count,
			(unsigned long long)data->args.offset);

	task = rpc_run_task(&task_setup_data);
	if (!IS_ERR(task))
		rpc_put_task(task);
}

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

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

/*
 * Generate multiple requests to fill a single page.
 *
 * We optimize to reduce the number of read operations on the wire.  If we
 * detect that we're reading a page, or an area of a page, that is past the
 * end of file, we do not generate NFS read operations but just clear the
 * parts of the page that would have come back zero from the server anyway.
 *
 * We rely on the cached value of i_size to make this determination; another
 * client can fill pages on the server past our cached end-of-file, but we
 * won't see the new data until our attribute cache is updated.  This is more
 * or less conventional NFS client behavior.
 */
static int nfs_pagein_multi(struct inode *inode, struct list_head *head, unsigned int npages, size_t count, int flags)
{
	struct nfs_page *req = nfs_list_entry(head->next);
	struct page *page = req->wb_page;
	struct nfs_read_data *data;
	size_t rsize = NFS_SERVER(inode)->rsize, nbytes;
	unsigned int offset;
	int requests = 0;
	LIST_HEAD(list);

	nfs_list_remove_request(req);

	nbytes = count;
	do {
		size_t len = min(nbytes,rsize);

		data = nfs_readdata_alloc(1);
		if (!data)
			goto out_bad;
		INIT_LIST_HEAD(&data->pages);
		list_add(&data->pages, &list);
		requests++;
		nbytes -= len;
	} while(nbytes != 0);
	atomic_set(&req->wb_complete, requests);

	ClearPageError(page);
	offset = 0;
	nbytes = count;
	do {
		data = list_entry(list.next, struct nfs_read_data, pages);
		list_del_init(&data->pages);

		data->pagevec[0] = page;

		if (nbytes < rsize)
			rsize = nbytes;
		nfs_read_rpcsetup(req, data, &nfs_read_partial_ops,
				  rsize, offset);
		offset += rsize;
		nbytes -= rsize;
	} while (nbytes != 0);

	return 0;

out_bad:
	while (!list_empty(&list)) {
		data = list_entry(list.next, struct nfs_read_data, pages);
		list_del(&data->pages);
		nfs_readdata_free(data);
	}
	SetPageError(page);
	nfs_readpage_release(req);
	return -ENOMEM;
}

static int nfs_pagein_one(struct inode *inode, struct list_head *head, unsigned int npages, size_t count, int flags)
{
	struct nfs_page		*req;
	struct page		**pages;
	struct nfs_read_data	*data;

	data = nfs_readdata_alloc(npages);
	if (!data)
		goto out_bad;

	INIT_LIST_HEAD(&data->pages);
	pages = data->pagevec;
	while (!list_empty(head)) {
		req = nfs_list_entry(head->next);
		nfs_list_remove_request(req);
		nfs_list_add_request(req, &data->pages);
		ClearPageError(req->wb_page);
		*pages++ = req->wb_page;
	}
	req = nfs_list_entry(data->pages.next);

	nfs_read_rpcsetup(req, data, &nfs_read_full_ops, count, 0);
	return 0;
out_bad:
	nfs_async_read_error(head);
	return -ENOMEM;
}

/*
 * This is the callback from RPC telling us whether a reply was
 * received or some error occurred (timeout or socket shutdown).
 */
int nfs_readpage_result(struct rpc_task *task, struct nfs_read_data *data)
{
	int status;

	dprintk("NFS: %s: %5u, (status %d)\n", __FUNCTION__, task->tk_pid,
			task->tk_status);

	status = NFS_PROTO(data->inode)->read_done(task, data);
	if (status != 0)
		return status;

	nfs_add_stats(data->inode, NFSIOS_SERVERREADBYTES, data->res.count);

	if (task->tk_status == -ESTALE) {
		set_bit(NFS_INO_STALE, &NFS_I(data->inode)->flags);
		nfs_mark_for_revalidate(data->inode);
	}
	return 0;
}

static int nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data)
{
	struct nfs_readargs *argp = &data->args;
	struct nfs_readres *resp = &data->res;

	if (resp->eof || resp->count == argp->count)
		return 0;

	/* This is a short read! */
	nfs_inc_stats(data->inode, NFSIOS_SHORTREAD);
	/* Has the server at least made some progress? */
	if (resp->count == 0)
		return 0;

	/* Yes, so retry the read at the end of the data */
	argp->offset += resp->count;
	argp->pgbase += resp->count;
	argp->count -= resp->count;
	rpc_restart_call(task);
	return -EAGAIN;
}

/*
 * Handle a read reply that fills part of a page.
 */
static void nfs_readpage_result_partial(struct rpc_task *task, void *calldata)
{
	struct nfs_read_data *data = calldata;
	struct nfs_page *req = data->req;
	struct page *page = req->wb_page;
 
	if (nfs_readpage_result(task, data) != 0)
		return;

	if (likely(task->tk_status >= 0)) {
		nfs_readpage_truncate_uninitialised_page(data);
		if (nfs_readpage_retry(task, data) != 0)
			return;
	}
	if (unlikely(task->tk_status < 0))
		SetPageError(page);
	if (atomic_dec_and_test(&req->wb_complete)) {
		if (!PageError(page))
			SetPageUptodate(page);
		nfs_readpage_release(req);
	}
}

static const struct rpc_call_ops nfs_read_partial_ops = {
	.rpc_call_done = nfs_readpage_result_partial,
	.rpc_release = nfs_readdata_release,
};

static void nfs_readpage_set_pages_uptodate(struct nfs_read_data *data)
{
	unsigned int count = data->res.count;
	unsigned int base = data->args.pgbase;
	struct page **pages;

	if (data->res.eof)
		count = data->args.count;
	if (unlikely(count == 0))
		return;
	pages = &data->args.pages[base >> PAGE_CACHE_SHIFT];
	base &= ~PAGE_CACHE_MASK;
	count += base;
	for (;count >= PAGE_CACHE_SIZE; count -= PAGE_CACHE_SIZE, pages++)
		SetPageUptodate(*pages);
	if (count == 0)
		return;
	/* Was this a short read? */
	if (data->res.eof || data->res.count == data->args.count)
		SetPageUptodate(*pages);
}

/*
 * This is the callback from RPC telling us whether a reply was
 * received or some error occurred (timeout or socket shutdown).
 */
static void nfs_readpage_result_full(struct rpc_task *task, void *calldata)
{
	struct nfs_read_data *data = calldata;

	if (nfs_readpage_result(task, data) != 0)
		return;
	/*
	 * Note: nfs_readpage_retry may change the values of
	 * data->args. In the multi-page case, we therefore need
	 * to ensure that we call nfs_readpage_set_pages_uptodate()
	 * first.
	 */
	if (likely(task->tk_status >= 0)) {
		nfs_readpage_truncate_uninitialised_page(data);
		nfs_readpage_set_pages_uptodate(data);
		if (nfs_readpage_retry(task, data) != 0)
			return;
	}
	while (!list_empty(&data->pages)) {
		struct nfs_page *req = nfs_list_entry(data->pages.next);

		nfs_list_remove_request(req);
		nfs_readpage_release(req);
	}
}

static const struct rpc_call_ops nfs_read_full_ops = {
	.rpc_call_done = nfs_readpage_result_full,
	.rpc_release = nfs_readdata_release,
};

/*
 * Read a page over NFS.
 * We read the page synchronously in the following case:
 *  -	The error flag is set for this page. This happens only when a
 *	previous async read operation failed.
 */
int nfs_readpage(struct file *file, struct page *page)
{
	struct nfs_open_context *ctx;
	struct inode *inode = page->mapping->host;
	int		error;

	dprintk("NFS: nfs_readpage (%p %ld@%lu)\n",
		page, PAGE_CACHE_SIZE, page->index);
	nfs_inc_stats(inode, NFSIOS_VFSREADPAGE);
	nfs_add_stats(inode, NFSIOS_READPAGES, 1);

	/*
	 * Try to flush any pending writes to the file..
	 *
	 * NOTE! Because we own the page lock, there cannot
	 * be any new pending writes generated at this point
	 * for this page (other pages can be written to).
	 */
	error = nfs_wb_page(inode, page);
	if (error)
		goto out_unlock;
	if (PageUptodate(page))
		goto out_unlock;

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

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

	error = nfs_readpage_async(ctx, inode, page);

	put_nfs_open_context(ctx);
	return error;
out_unlock:
	unlock_page(page);
	return error;
}

struct nfs_readdesc {
	struct nfs_pageio_descriptor *pgio;
	struct nfs_open_context *ctx;
};

static int
readpage_async_filler(void *data, struct page *page)
{
	struct nfs_readdesc *desc = (struct nfs_readdesc *)data;
	struct inode *inode = page->mapping->host;
	struct nfs_page *new;
	unsigned int len;
	int error;

	error = nfs_wb_page(inode, page);
	if (error)
		goto out_unlock;
	if (PageUptodate(page))
		goto out_unlock;

	len = nfs_page_length(page);
	if (len == 0)
		return nfs_return_empty_page(page);

	new = nfs_create_request(desc->ctx, inode, page, 0, len);
	if (IS_ERR(new))
		goto out_error;

	if (len < PAGE_CACHE_SIZE)
		zero_user_page(page, len, PAGE_CACHE_SIZE - len, KM_USER0);
	nfs_pageio_add_request(desc->pgio, new);
	return 0;
out_error:
	error = PTR_ERR(new);
	SetPageError(page);
out_unlock:
	unlock_page(page);
	return error;
}

int nfs_readpages(struct file *filp, struct address_space *mapping,
		struct list_head *pages, unsigned nr_pages)
{
	struct nfs_pageio_descriptor pgio;
	struct nfs_readdesc desc = {
		.pgio = &pgio,
	};
	struct inode *inode = mapping->host;
	struct nfs_server *server = NFS_SERVER(inode);
	size_t rsize = server->rsize;
	unsigned long npages;
	int ret = -ESTALE;

	dprintk("NFS: nfs_readpages (%s/%Ld %d)\n",
			inode->i_sb->s_id,
			(long long)NFS_FILEID(inode),
			nr_pages);
	nfs_inc_stats(inode, NFSIOS_VFSREADPAGES);

	if (NFS_STALE(inode))
		goto out;

	if (filp == NULL) {
		desc.ctx = nfs_find_open_context(inode, NULL, FMODE_READ);
		if (desc.ctx == NULL)
			return -EBADF;
	} else
		desc.ctx = get_nfs_open_context(nfs_file_open_context(filp));
	if (rsize < PAGE_CACHE_SIZE)
		nfs_pageio_init(&pgio, inode, nfs_pagein_multi, rsize, 0);
	else
		nfs_pageio_init(&pgio, inode, nfs_pagein_one, rsize, 0);

	ret = read_cache_pages(mapping, pages, readpage_async_filler, &desc);

	nfs_pageio_complete(&pgio);
	npages = (pgio.pg_bytes_written + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
	nfs_add_stats(inode, NFSIOS_READPAGES, npages);
	put_nfs_open_context(desc.ctx);
out:
	return ret;
}

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

	nfs_rdata_mempool = mempool_create_slab_pool(MIN_POOL_READ,
						     nfs_rdata_cachep);
	if (nfs_rdata_mempool == NULL)
		return -ENOMEM;

	return 0;
}

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