aboutsummaryrefslogtreecommitdiff
path: root/fs/ceph/export.c
blob: 1d65934c12620efdb805f0757268cc9e27f239e6 (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
// SPDX-License-Identifier: GPL-2.0
#include <linux/ceph/ceph_debug.h>

#include <linux/exportfs.h>
#include <linux/slab.h>
#include <asm/unaligned.h>

#include "super.h"
#include "mds_client.h"

/*
 * Basic fh
 */
struct ceph_nfs_fh {
	u64 ino;
} __attribute__ ((packed));

/*
 * Larger fh that includes parent ino.
 */
struct ceph_nfs_confh {
	u64 ino, parent_ino;
} __attribute__ ((packed));

/*
 * fh for snapped inode
 */
struct ceph_nfs_snapfh {
	u64 ino;
	u64 snapid;
	u64 parent_ino;
	u32 hash;
} __attribute__ ((packed));

static int ceph_encode_snapfh(struct inode *inode, u32 *rawfh, int *max_len,
			      struct inode *parent_inode)
{
	static const int snap_handle_length =
		sizeof(struct ceph_nfs_snapfh) >> 2;
	struct ceph_nfs_snapfh *sfh = (void *)rawfh;
	u64 snapid = ceph_snap(inode);
	int ret;
	bool no_parent = true;

	if (*max_len < snap_handle_length) {
		*max_len = snap_handle_length;
		ret = FILEID_INVALID;
		goto out;
	}

	ret =  -EINVAL;
	if (snapid != CEPH_SNAPDIR) {
		struct inode *dir;
		struct dentry *dentry = d_find_alias(inode);
		if (!dentry)
			goto out;

		rcu_read_lock();
		dir = d_inode_rcu(dentry->d_parent);
		if (ceph_snap(dir) != CEPH_SNAPDIR) {
			sfh->parent_ino = ceph_ino(dir);
			sfh->hash = ceph_dentry_hash(dir, dentry);
			no_parent = false;
		}
		rcu_read_unlock();
		dput(dentry);
	}

	if (no_parent) {
		if (!S_ISDIR(inode->i_mode))
			goto out;
		sfh->parent_ino = sfh->ino;
		sfh->hash = 0;
	}
	sfh->ino = ceph_ino(inode);
	sfh->snapid = snapid;

	*max_len = snap_handle_length;
	ret = FILEID_BTRFS_WITH_PARENT;
out:
	dout("encode_snapfh %llx.%llx ret=%d\n", ceph_vinop(inode), ret);
	return ret;
}

static int ceph_encode_fh(struct inode *inode, u32 *rawfh, int *max_len,
			  struct inode *parent_inode)
{
	static const int handle_length =
		sizeof(struct ceph_nfs_fh) >> 2;
	static const int connected_handle_length =
		sizeof(struct ceph_nfs_confh) >> 2;
	int type;

	if (ceph_snap(inode) != CEPH_NOSNAP)
		return ceph_encode_snapfh(inode, rawfh, max_len, parent_inode);

	if (parent_inode && (*max_len < connected_handle_length)) {
		*max_len = connected_handle_length;
		return FILEID_INVALID;
	} else if (*max_len < handle_length) {
		*max_len = handle_length;
		return FILEID_INVALID;
	}

	if (parent_inode) {
		struct ceph_nfs_confh *cfh = (void *)rawfh;
		dout("encode_fh %llx with parent %llx\n",
		     ceph_ino(inode), ceph_ino(parent_inode));
		cfh->ino = ceph_ino(inode);
		cfh->parent_ino = ceph_ino(parent_inode);
		*max_len = connected_handle_length;
		type = FILEID_INO32_GEN_PARENT;
	} else {
		struct ceph_nfs_fh *fh = (void *)rawfh;
		dout("encode_fh %llx\n", ceph_ino(inode));
		fh->ino = ceph_ino(inode);
		*max_len = handle_length;
		type = FILEID_INO32_GEN;
	}
	return type;
}

static struct inode *__lookup_inode(struct super_block *sb, u64 ino)
{
	struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc;
	struct inode *inode;
	struct ceph_vino vino;
	int err;

	vino.ino = ino;
	vino.snap = CEPH_NOSNAP;

	if (ceph_vino_is_reserved(vino))
		return ERR_PTR(-ESTALE);

	inode = ceph_find_inode(sb, vino);
	if (!inode) {
		struct ceph_mds_request *req;
		int mask;

		req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPINO,
					       USE_ANY_MDS);
		if (IS_ERR(req))
			return ERR_CAST(req);

		mask = CEPH_STAT_CAP_INODE;
		if (ceph_security_xattr_wanted(d_inode(sb->s_root)))
			mask |= CEPH_CAP_XATTR_SHARED;
		req->r_args.lookupino.mask = cpu_to_le32(mask);

		req->r_ino1 = vino;
		req->r_num_caps = 1;
		err = ceph_mdsc_do_request(mdsc, NULL, req);
		inode = req->r_target_inode;
		if (inode)
			ihold(inode);
		ceph_mdsc_put_request(req);
		if (!inode)
			return err < 0 ? ERR_PTR(err) : ERR_PTR(-ESTALE);
	}
	return inode;
}

struct inode *ceph_lookup_inode(struct super_block *sb, u64 ino)
{
	struct inode *inode = __lookup_inode(sb, ino);
	if (IS_ERR(inode))
		return inode;
	if (inode->i_nlink == 0) {
		iput(inode);
		return ERR_PTR(-ESTALE);
	}
	return inode;
}

static struct dentry *__fh_to_dentry(struct super_block *sb, u64 ino)
{
	struct inode *inode = __lookup_inode(sb, ino);
	int err;

	if (IS_ERR(inode))
		return ERR_CAST(inode);
	/* We need LINK caps to reliably check i_nlink */
	err = ceph_do_getattr(inode, CEPH_CAP_LINK_SHARED, false);
	if (err) {
		iput(inode);
		return ERR_PTR(err);
	}
	/* -ESTALE if inode as been unlinked and no file is open */
	if ((inode->i_nlink == 0) && (atomic_read(&inode->i_count) == 1)) {
		iput(inode);
		return ERR_PTR(-ESTALE);
	}
	return d_obtain_alias(inode);
}

static struct dentry *__snapfh_to_dentry(struct super_block *sb,
					  struct ceph_nfs_snapfh *sfh,
					  bool want_parent)
{
	struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc;
	struct ceph_mds_request *req;
	struct inode *inode;
	struct ceph_vino vino;
	int mask;
	int err;
	bool unlinked = false;

	if (want_parent) {
		vino.ino = sfh->parent_ino;
		if (sfh->snapid == CEPH_SNAPDIR)
			vino.snap = CEPH_NOSNAP;
		else if (sfh->ino == sfh->parent_ino)
			vino.snap = CEPH_SNAPDIR;
		else
			vino.snap = sfh->snapid;
	} else {
		vino.ino = sfh->ino;
		vino.snap = sfh->snapid;
	}

	if (ceph_vino_is_reserved(vino))
		return ERR_PTR(-ESTALE);

	inode = ceph_find_inode(sb, vino);
	if (inode)
		return d_obtain_alias(inode);

	req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPINO,
				       USE_ANY_MDS);
	if (IS_ERR(req))
		return ERR_CAST(req);

	mask = CEPH_STAT_CAP_INODE;
	if (ceph_security_xattr_wanted(d_inode(sb->s_root)))
		mask |= CEPH_CAP_XATTR_SHARED;
	req->r_args.lookupino.mask = cpu_to_le32(mask);
	if (vino.snap < CEPH_NOSNAP) {
		req->r_args.lookupino.snapid = cpu_to_le64(vino.snap);
		if (!want_parent && sfh->ino != sfh->parent_ino) {
			req->r_args.lookupino.parent =
					cpu_to_le64(sfh->parent_ino);
			req->r_args.lookupino.hash =
					cpu_to_le32(sfh->hash);
		}
	}

	req->r_ino1 = vino;
	req->r_num_caps = 1;
	err = ceph_mdsc_do_request(mdsc, NULL, req);
	inode = req->r_target_inode;
	if (inode) {
		if (vino.snap == CEPH_SNAPDIR) {
			if (inode->i_nlink == 0)
				unlinked = true;
			inode = ceph_get_snapdir(inode);
		} else if (ceph_snap(inode) == vino.snap) {
			ihold(inode);
		} else {
			/* mds does not support lookup snapped inode */
			inode = ERR_PTR(-EOPNOTSUPP);
		}
	} else {
		inode = ERR_PTR(-ESTALE);
	}
	ceph_mdsc_put_request(req);

	if (want_parent) {
		dout("snapfh_to_parent %llx.%llx\n err=%d\n",
		     vino.ino, vino.snap, err);
	} else {
		dout("snapfh_to_dentry %llx.%llx parent %llx hash %x err=%d",
		      vino.ino, vino.snap, sfh->parent_ino, sfh->hash, err);
	}
	if (IS_ERR(inode))
		return ERR_CAST(inode);
	/* see comments in ceph_get_parent() */
	return unlinked ? d_obtain_root(inode) : d_obtain_alias(inode);
}

/*
 * convert regular fh to dentry
 */
static struct dentry *ceph_fh_to_dentry(struct super_block *sb,
					struct fid *fid,
					int fh_len, int fh_type)
{
	struct ceph_nfs_fh *fh = (void *)fid->raw;

	if (fh_type == FILEID_BTRFS_WITH_PARENT) {
		struct ceph_nfs_snapfh *sfh = (void *)fid->raw;
		return __snapfh_to_dentry(sb, sfh, false);
	}

	if (fh_type != FILEID_INO32_GEN  &&
	    fh_type != FILEID_INO32_GEN_PARENT)
		return NULL;
	if (fh_len < sizeof(*fh) / 4)
		return NULL;

	dout("fh_to_dentry %llx\n", fh->ino);
	return __fh_to_dentry(sb, fh->ino);
}

static struct dentry *__get_parent(struct super_block *sb,
				   struct dentry *child, u64 ino)
{
	struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc;
	struct ceph_mds_request *req;
	struct inode *inode;
	int mask;
	int err;

	req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPPARENT,
				       USE_ANY_MDS);
	if (IS_ERR(req))
		return ERR_CAST(req);

	if (child) {
		req->r_inode = d_inode(child);
		ihold(d_inode(child));
	} else {
		req->r_ino1 = (struct ceph_vino) {
			.ino = ino,
			.snap = CEPH_NOSNAP,
		};
	}

	mask = CEPH_STAT_CAP_INODE;
	if (ceph_security_xattr_wanted(d_inode(sb->s_root)))
		mask |= CEPH_CAP_XATTR_SHARED;
	req->r_args.getattr.mask = cpu_to_le32(mask);

	req->r_num_caps = 1;
	err = ceph_mdsc_do_request(mdsc, NULL, req);
	if (err) {
		ceph_mdsc_put_request(req);
		return ERR_PTR(err);
	}

	inode = req->r_target_inode;
	if (inode)
		ihold(inode);
	ceph_mdsc_put_request(req);
	if (!inode)
		return ERR_PTR(-ENOENT);

	return d_obtain_alias(inode);
}

static struct dentry *ceph_get_parent(struct dentry *child)
{
	struct inode *inode = d_inode(child);
	struct dentry *dn;

	if (ceph_snap(inode) != CEPH_NOSNAP) {
		struct inode* dir;
		bool unlinked = false;
		/* do not support non-directory */
		if (!d_is_dir(child)) {
			dn = ERR_PTR(-EINVAL);
			goto out;
		}
		dir = __lookup_inode(inode->i_sb, ceph_ino(inode));
		if (IS_ERR(dir)) {
			dn = ERR_CAST(dir);
			goto out;
		}
		/* There can be multiple paths to access snapped inode.
		 * For simplicity, treat snapdir of head inode as parent */
		if (ceph_snap(inode) != CEPH_SNAPDIR) {
			struct inode *snapdir = ceph_get_snapdir(dir);
			if (dir->i_nlink == 0)
				unlinked = true;
			iput(dir);
			if (IS_ERR(snapdir)) {
				dn = ERR_CAST(snapdir);
				goto out;
			}
			dir = snapdir;
		}
		/* If directory has already been deleted, futher get_parent
		 * will fail. Do not mark snapdir dentry as disconnected,
		 * this prevent exportfs from doing futher get_parent. */
		if (unlinked)
			dn = d_obtain_root(dir);
		else
			dn = d_obtain_alias(dir);
	} else {
		dn = __get_parent(child->d_sb, child, 0);
	}
out:
	dout("get_parent %p ino %llx.%llx err=%ld\n",
	     child, ceph_vinop(inode), (long)PTR_ERR_OR_ZERO(dn));
	return dn;
}

/*
 * convert regular fh to parent
 */
static struct dentry *ceph_fh_to_parent(struct super_block *sb,
					struct fid *fid,
					int fh_len, int fh_type)
{
	struct ceph_nfs_confh *cfh = (void *)fid->raw;
	struct dentry *dentry;

	if (fh_type == FILEID_BTRFS_WITH_PARENT) {
		struct ceph_nfs_snapfh *sfh = (void *)fid->raw;
		return __snapfh_to_dentry(sb, sfh, true);
	}

	if (fh_type != FILEID_INO32_GEN_PARENT)
		return NULL;
	if (fh_len < sizeof(*cfh) / 4)
		return NULL;

	dout("fh_to_parent %llx\n", cfh->parent_ino);
	dentry = __get_parent(sb, NULL, cfh->ino);
	if (unlikely(dentry == ERR_PTR(-ENOENT)))
		dentry = __fh_to_dentry(sb, cfh->parent_ino);
	return dentry;
}

static int __get_snap_name(struct dentry *parent, char *name,
			   struct dentry *child)
{
	struct inode *inode = d_inode(child);
	struct inode *dir = d_inode(parent);
	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
	struct ceph_mds_request *req = NULL;
	char *last_name = NULL;
	unsigned next_offset = 2;
	int err = -EINVAL;

	if (ceph_ino(inode) != ceph_ino(dir))
		goto out;
	if (ceph_snap(inode) == CEPH_SNAPDIR) {
		if (ceph_snap(dir) == CEPH_NOSNAP) {
			strcpy(name, fsc->mount_options->snapdir_name);
			err = 0;
		}
		goto out;
	}
	if (ceph_snap(dir) != CEPH_SNAPDIR)
		goto out;

	while (1) {
		struct ceph_mds_reply_info_parsed *rinfo;
		struct ceph_mds_reply_dir_entry *rde;
		int i;

		req = ceph_mdsc_create_request(fsc->mdsc, CEPH_MDS_OP_LSSNAP,
					       USE_AUTH_MDS);
		if (IS_ERR(req)) {
			err = PTR_ERR(req);
			req = NULL;
			goto out;
		}
		err = ceph_alloc_readdir_reply_buffer(req, inode);
		if (err)
			goto out;

		req->r_direct_mode = USE_AUTH_MDS;
		req->r_readdir_offset = next_offset;
		req->r_args.readdir.flags =
				cpu_to_le16(CEPH_READDIR_REPLY_BITFLAGS);
		if (last_name) {
			req->r_path2 = last_name;
			last_name = NULL;
		}

		req->r_inode = dir;
		ihold(dir);
		req->r_dentry = dget(parent);

		inode_lock(dir);
		err = ceph_mdsc_do_request(fsc->mdsc, NULL, req);
		inode_unlock(dir);

		if (err < 0)
			goto out;

		rinfo = &req->r_reply_info;
		for (i = 0; i < rinfo->dir_nr; i++) {
			rde = rinfo->dir_entries + i;
			BUG_ON(!rde->inode.in);
			if (ceph_snap(inode) ==
			    le64_to_cpu(rde->inode.in->snapid)) {
				memcpy(name, rde->name, rde->name_len);
				name[rde->name_len] = '\0';
				err = 0;
				goto out;
			}
		}

		if (rinfo->dir_end)
			break;

		BUG_ON(rinfo->dir_nr <= 0);
		rde = rinfo->dir_entries + (rinfo->dir_nr - 1);
		next_offset += rinfo->dir_nr;
		last_name = kstrndup(rde->name, rde->name_len, GFP_KERNEL);
		if (!last_name) {
			err = -ENOMEM;
			goto out;
		}

		ceph_mdsc_put_request(req);
		req = NULL;
	}
	err = -ENOENT;
out:
	if (req)
		ceph_mdsc_put_request(req);
	kfree(last_name);
	dout("get_snap_name %p ino %llx.%llx err=%d\n",
	     child, ceph_vinop(inode), err);
	return err;
}

static int ceph_get_name(struct dentry *parent, char *name,
			 struct dentry *child)
{
	struct ceph_mds_client *mdsc;
	struct ceph_mds_request *req;
	struct inode *inode = d_inode(child);
	int err;

	if (ceph_snap(inode) != CEPH_NOSNAP)
		return __get_snap_name(parent, name, child);

	mdsc = ceph_inode_to_client(inode)->mdsc;
	req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPNAME,
				       USE_ANY_MDS);
	if (IS_ERR(req))
		return PTR_ERR(req);

	inode_lock(d_inode(parent));

	req->r_inode = inode;
	ihold(inode);
	req->r_ino2 = ceph_vino(d_inode(parent));
	req->r_parent = d_inode(parent);
	ihold(req->r_parent);
	set_bit(CEPH_MDS_R_PARENT_LOCKED, &req->r_req_flags);
	req->r_num_caps = 2;
	err = ceph_mdsc_do_request(mdsc, NULL, req);

	inode_unlock(d_inode(parent));

	if (!err) {
		struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
		memcpy(name, rinfo->dname, rinfo->dname_len);
		name[rinfo->dname_len] = 0;
		dout("get_name %p ino %llx.%llx name %s\n",
		     child, ceph_vinop(inode), name);
	} else {
		dout("get_name %p ino %llx.%llx err %d\n",
		     child, ceph_vinop(inode), err);
	}

	ceph_mdsc_put_request(req);
	return err;
}

const struct export_operations ceph_export_ops = {
	.encode_fh = ceph_encode_fh,
	.fh_to_dentry = ceph_fh_to_dentry,
	.fh_to_parent = ceph_fh_to_parent,
	.get_parent = ceph_get_parent,
	.get_name = ceph_get_name,
};