aboutsummaryrefslogtreecommitdiffstats
path: root/fs/ceph/cache.c
blob: 834f9f3723fbe3980e8a9705e8c1d0f86936eb40 (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
/*
 * Ceph cache definitions.
 *
 *  Copyright (C) 2013 by Adfin Solutions, Inc. All Rights Reserved.
 *  Written by Milosz Tanski (milosz@adfin.com)
 *
 *  This program is free software; you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License version 2
 *  as published by the Free Software Foundation.
 *
 *  This program is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License
 *  along with this program; if not, write to:
 *  Free Software Foundation
 *  51 Franklin Street, Fifth Floor
 *  Boston, MA  02111-1301  USA
 *
 */

#include "super.h"
#include "cache.h"

struct ceph_aux_inode {
	struct timespec	mtime;
	loff_t          size;
};

struct fscache_netfs ceph_cache_netfs = {
	.name		= "ceph",
	.version	= 0,
};

static uint16_t ceph_fscache_session_get_key(const void *cookie_netfs_data,
					     void *buffer, uint16_t maxbuf)
{
	const struct ceph_fs_client* fsc = cookie_netfs_data;
	uint16_t klen;

	klen = sizeof(fsc->client->fsid);
	if (klen > maxbuf)
		return 0;

	memcpy(buffer, &fsc->client->fsid, klen);
	return klen;
}

static const struct fscache_cookie_def ceph_fscache_fsid_object_def = {
	.name		= "CEPH.fsid",
	.type		= FSCACHE_COOKIE_TYPE_INDEX,
	.get_key	= ceph_fscache_session_get_key,
};

int ceph_fscache_register(void)
{
	return fscache_register_netfs(&ceph_cache_netfs);
}

void ceph_fscache_unregister(void)
{
	fscache_unregister_netfs(&ceph_cache_netfs);
}

int ceph_fscache_register_fs(struct ceph_fs_client* fsc)
{
	fsc->fscache = fscache_acquire_cookie(ceph_cache_netfs.primary_index,
					      &ceph_fscache_fsid_object_def,
					      fsc, true);

	if (fsc->fscache == NULL) {
		pr_err("Unable to resgister fsid: %p fscache cookie", fsc);
		return 0;
	}

	fsc->revalidate_wq = alloc_workqueue("ceph-revalidate", 0, 1);
	if (fsc->revalidate_wq == NULL)
		return -ENOMEM;

	return 0;
}

static uint16_t ceph_fscache_inode_get_key(const void *cookie_netfs_data,
					   void *buffer, uint16_t maxbuf)
{
	const struct ceph_inode_info* ci = cookie_netfs_data;
	uint16_t klen;

	/* use ceph virtual inode (id + snaphot) */
	klen = sizeof(ci->i_vino);
	if (klen > maxbuf)
		return 0;

	memcpy(buffer, &ci->i_vino, klen);
	return klen;
}

static uint16_t ceph_fscache_inode_get_aux(const void *cookie_netfs_data,
					   void *buffer, uint16_t bufmax)
{
	struct ceph_aux_inode aux;
	const struct ceph_inode_info* ci = cookie_netfs_data;
	const struct inode* inode = &ci->vfs_inode;

	memset(&aux, 0, sizeof(aux));
	aux.mtime = inode->i_mtime;
	aux.size = inode->i_size;

	memcpy(buffer, &aux, sizeof(aux));

	return sizeof(aux);
}

static void ceph_fscache_inode_get_attr(const void *cookie_netfs_data,
					uint64_t *size)
{
	const struct ceph_inode_info* ci = cookie_netfs_data;
	const struct inode* inode = &ci->vfs_inode;

	*size = inode->i_size;
}

static enum fscache_checkaux ceph_fscache_inode_check_aux(
	void *cookie_netfs_data, const void *data, uint16_t dlen)
{
	struct ceph_aux_inode aux;
	struct ceph_inode_info* ci = cookie_netfs_data;
	struct inode* inode = &ci->vfs_inode;

	if (dlen != sizeof(aux))
		return FSCACHE_CHECKAUX_OBSOLETE;

	memset(&aux, 0, sizeof(aux));
	aux.mtime = inode->i_mtime;
	aux.size = inode->i_size;

	if (memcmp(data, &aux, sizeof(aux)) != 0)
		return FSCACHE_CHECKAUX_OBSOLETE;

	dout("ceph inode 0x%p cached okay", ci);
	return FSCACHE_CHECKAUX_OKAY;
}

static void ceph_fscache_inode_now_uncached(void* cookie_netfs_data)
{
	struct ceph_inode_info* ci = cookie_netfs_data;
	struct pagevec pvec;
	pgoff_t first;
	int loop, nr_pages;

	pagevec_init(&pvec, 0);
	first = 0;

	dout("ceph inode 0x%p now uncached", ci);

	while (1) {
		nr_pages = pagevec_lookup(&pvec, ci->vfs_inode.i_mapping, first,
					  PAGEVEC_SIZE - pagevec_count(&pvec));

		if (!nr_pages)
			break;

		for (loop = 0; loop < nr_pages; loop++)
			ClearPageFsCache(pvec.pages[loop]);

		first = pvec.pages[nr_pages - 1]->index + 1;

		pvec.nr = nr_pages;
		pagevec_release(&pvec);
		cond_resched();
	}
}

static const struct fscache_cookie_def ceph_fscache_inode_object_def = {
	.name		= "CEPH.inode",
	.type		= FSCACHE_COOKIE_TYPE_DATAFILE,
	.get_key	= ceph_fscache_inode_get_key,
	.get_attr	= ceph_fscache_inode_get_attr,
	.get_aux	= ceph_fscache_inode_get_aux,
	.check_aux	= ceph_fscache_inode_check_aux,
	.now_uncached	= ceph_fscache_inode_now_uncached,
};

void ceph_fscache_register_inode_cookie(struct ceph_fs_client* fsc,
					struct ceph_inode_info* ci)
{
	struct inode* inode = &ci->vfs_inode;

	/* No caching for filesystem */
	if (fsc->fscache == NULL)
		return;

	/* Only cache for regular files that are read only */
	if ((ci->vfs_inode.i_mode & S_IFREG) == 0)
		return;

	/* Avoid multiple racing open requests */
	mutex_lock(&inode->i_mutex);

	if (ci->fscache)
		goto done;

	ci->fscache = fscache_acquire_cookie(fsc->fscache,
					     &ceph_fscache_inode_object_def,
					     ci, true);
	fscache_check_consistency(ci->fscache);
done:
	mutex_unlock(&inode->i_mutex);

}

void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci)
{
	struct fscache_cookie* cookie;

	if ((cookie = ci->fscache) == NULL)
		return;

	ci->fscache = NULL;

	fscache_uncache_all_inode_pages(cookie, &ci->vfs_inode);
	fscache_relinquish_cookie(cookie, 0);
}

static void ceph_vfs_readpage_complete(struct page *page, void *data, int error)
{
	if (!error)
		SetPageUptodate(page);
}

static void ceph_vfs_readpage_complete_unlock(struct page *page, void *data, int error)
{
	if (!error)
		SetPageUptodate(page);

	unlock_page(page);
}

static inline int cache_valid(struct ceph_inode_info *ci)
{
	return ((ceph_caps_issued(ci) & CEPH_CAP_FILE_CACHE) &&
		(ci->i_fscache_gen == ci->i_rdcache_gen));
}


/* Atempt to read from the fscache,
 *
 * This function is called from the readpage_nounlock context. DO NOT attempt to
 * unlock the page here (or in the callback).
 */
int ceph_readpage_from_fscache(struct inode *inode, struct page *page)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	int ret;

	if (!cache_valid(ci))
		return -ENOBUFS;

	ret = fscache_read_or_alloc_page(ci->fscache, page,
					 ceph_vfs_readpage_complete, NULL,
					 GFP_KERNEL);

	switch (ret) {
		case 0: /* Page found */
			dout("page read submitted\n");
			return 0;
		case -ENOBUFS: /* Pages were not found, and can't be */
		case -ENODATA: /* Pages were not found */
			dout("page/inode not in cache\n");
			return ret;
		default:
			dout("%s: unknown error ret = %i\n", __func__, ret);
			return ret;
	}
}

int ceph_readpages_from_fscache(struct inode *inode,
				  struct address_space *mapping,
				  struct list_head *pages,
				  unsigned *nr_pages)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	int ret;

	if (!cache_valid(ci))
		return -ENOBUFS;

	ret = fscache_read_or_alloc_pages(ci->fscache, mapping, pages, nr_pages,
					  ceph_vfs_readpage_complete_unlock,
					  NULL, mapping_gfp_mask(mapping));

	switch (ret) {
		case 0: /* All pages found */
			dout("all-page read submitted\n");
			return 0;
		case -ENOBUFS: /* Some pages were not found, and can't be */
		case -ENODATA: /* some pages were not found */
			dout("page/inode not in cache\n");
			return ret;
		default:
			dout("%s: unknown error ret = %i\n", __func__, ret);
			return ret;
	}
}

void ceph_readpage_to_fscache(struct inode *inode, struct page *page)
{
	struct ceph_inode_info *ci = ceph_inode(inode);
	int ret;

	if (!PageFsCache(page))
		return;

	if (!cache_valid(ci))
		return;

	ret = fscache_write_page(ci->fscache, page, GFP_KERNEL);
	if (ret)
		 fscache_uncache_page(ci->fscache, page);
}

void ceph_invalidate_fscache_page(struct inode* inode, struct page *page)
{
	struct ceph_inode_info *ci = ceph_inode(inode);

	if (!PageFsCache(page))
		return;

	fscache_wait_on_page_write(ci->fscache, page);
	fscache_uncache_page(ci->fscache, page);
}

void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc)
{
	if (fsc->revalidate_wq)
		destroy_workqueue(fsc->revalidate_wq);

	fscache_relinquish_cookie(fsc->fscache, 0);
	fsc->fscache = NULL;
}

static void ceph_revalidate_work(struct work_struct *work)
{
	int issued;
	u32 orig_gen;
	struct ceph_inode_info *ci = container_of(work, struct ceph_inode_info,
						  i_revalidate_work);
	struct inode *inode = &ci->vfs_inode;

	spin_lock(&ci->i_ceph_lock);
	issued = __ceph_caps_issued(ci, NULL);
	orig_gen = ci->i_rdcache_gen;
	spin_unlock(&ci->i_ceph_lock);

	if (!(issued & CEPH_CAP_FILE_CACHE)) {
		dout("revalidate_work lost cache before validation %p\n",
		     inode);
		goto out;
	}

	if (!fscache_check_consistency(ci->fscache))
		fscache_invalidate(ci->fscache);

	spin_lock(&ci->i_ceph_lock);
	/* Update the new valid generation (backwards sanity check too) */
	if (orig_gen > ci->i_fscache_gen) {
		ci->i_fscache_gen = orig_gen;
	}
	spin_unlock(&ci->i_ceph_lock);

out:
	iput(&ci->vfs_inode);
}

void ceph_queue_revalidate(struct inode *inode)
{
	struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
	struct ceph_inode_info *ci = ceph_inode(inode);

	if (fsc->revalidate_wq == NULL || ci->fscache == NULL)
		return;

	ihold(inode);

	if (queue_work(ceph_sb_to_client(inode->i_sb)->revalidate_wq,
		       &ci->i_revalidate_work)) {
		dout("ceph_queue_revalidate %p\n", inode);
	} else {
		dout("ceph_queue_revalidate %p failed\n)", inode);
		iput(inode);
	}
}

void ceph_fscache_inode_init(struct ceph_inode_info *ci)
{
	ci->fscache = NULL;
	/* The first load is verifed cookie open time */
	ci->i_fscache_gen = 1;
	INIT_WORK(&ci->i_revalidate_work, ceph_revalidate_work);
}