summaryrefslogtreecommitdiff
path: root/fs/xfs/xfs_itable.c
blob: c08c79d9e311d24f5af78079bbb2f3396c335492 (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
// SPDX-License-Identifier: GPL-2.0
/*
 * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc.
 * All Rights Reserved.
 */
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_shared.h"
#include "xfs_format.h"
#include "xfs_log_format.h"
#include "xfs_trans_resv.h"
#include "xfs_mount.h"
#include "xfs_inode.h"
#include "xfs_btree.h"
#include "xfs_ialloc.h"
#include "xfs_ialloc_btree.h"
#include "xfs_iwalk.h"
#include "xfs_itable.h"
#include "xfs_error.h"
#include "xfs_icache.h"
#include "xfs_health.h"
#include "xfs_trans.h"

/*
 * Bulk Stat
 * =========
 *
 * Use the inode walking functions to fill out struct xfs_bulkstat for every
 * allocated inode, then pass the stat information to some externally provided
 * iteration function.
 */

struct xfs_bstat_chunk {
	bulkstat_one_fmt_pf	formatter;
	struct xfs_ibulk	*breq;
	struct xfs_bulkstat	*buf;
};

/*
 * Fill out the bulkstat info for a single inode and report it somewhere.
 *
 * bc->breq->lastino is effectively the inode cursor as we walk through the
 * filesystem.  Therefore, we update it any time we need to move the cursor
 * forward, regardless of whether or not we're sending any bstat information
 * back to userspace.  If the inode is internal metadata or, has been freed
 * out from under us, we just simply keep going.
 *
 * However, if any other type of error happens we want to stop right where we
 * are so that userspace will call back with exact number of the bad inode and
 * we can send back an error code.
 *
 * Note that if the formatter tells us there's no space left in the buffer we
 * move the cursor forward and abort the walk.
 */
STATIC int
xfs_bulkstat_one_int(
	struct xfs_mount	*mp,
	struct user_namespace	*mnt_userns,
	struct xfs_trans	*tp,
	xfs_ino_t		ino,
	struct xfs_bstat_chunk	*bc)
{
	struct user_namespace	*sb_userns = mp->m_super->s_user_ns;
	struct xfs_inode	*ip;		/* incore inode pointer */
	struct inode		*inode;
	struct xfs_bulkstat	*buf = bc->buf;
	int			error = -EINVAL;

	if (xfs_internal_inum(mp, ino))
		goto out_advance;

	error = xfs_iget(mp, tp, ino,
			 (XFS_IGET_DONTCACHE | XFS_IGET_UNTRUSTED),
			 XFS_ILOCK_SHARED, &ip);
	if (error == -ENOENT || error == -EINVAL)
		goto out_advance;
	if (error)
		goto out;

	ASSERT(ip != NULL);
	ASSERT(ip->i_imap.im_blkno != 0);
	inode = VFS_I(ip);

	/* xfs_iget returns the following without needing
	 * further change.
	 */
	buf->bs_projectid = ip->i_projid;
	buf->bs_ino = ino;
	buf->bs_uid = from_kuid(sb_userns, i_uid_into_mnt(mnt_userns, inode));
	buf->bs_gid = from_kgid(sb_userns, i_gid_into_mnt(mnt_userns, inode));
	buf->bs_size = ip->i_disk_size;

	buf->bs_nlink = inode->i_nlink;
	buf->bs_atime = inode->i_atime.tv_sec;
	buf->bs_atime_nsec = inode->i_atime.tv_nsec;
	buf->bs_mtime = inode->i_mtime.tv_sec;
	buf->bs_mtime_nsec = inode->i_mtime.tv_nsec;
	buf->bs_ctime = inode->i_ctime.tv_sec;
	buf->bs_ctime_nsec = inode->i_ctime.tv_nsec;
	buf->bs_gen = inode->i_generation;
	buf->bs_mode = inode->i_mode;

	buf->bs_xflags = xfs_ip2xflags(ip);
	buf->bs_extsize_blks = ip->i_extsize;
	buf->bs_extents = xfs_ifork_nextents(&ip->i_df);
	xfs_bulkstat_health(ip, buf);
	buf->bs_aextents = xfs_ifork_nextents(ip->i_afp);
	buf->bs_forkoff = XFS_IFORK_BOFF(ip);
	buf->bs_version = XFS_BULKSTAT_VERSION_V5;

	if (xfs_has_v3inodes(mp)) {
		buf->bs_btime = ip->i_crtime.tv_sec;
		buf->bs_btime_nsec = ip->i_crtime.tv_nsec;
		if (ip->i_diflags2 & XFS_DIFLAG2_COWEXTSIZE)
			buf->bs_cowextsize_blks = ip->i_cowextsize;
	}

	switch (ip->i_df.if_format) {
	case XFS_DINODE_FMT_DEV:
		buf->bs_rdev = sysv_encode_dev(inode->i_rdev);
		buf->bs_blksize = BLKDEV_IOSIZE;
		buf->bs_blocks = 0;
		break;
	case XFS_DINODE_FMT_LOCAL:
		buf->bs_rdev = 0;
		buf->bs_blksize = mp->m_sb.sb_blocksize;
		buf->bs_blocks = 0;
		break;
	case XFS_DINODE_FMT_EXTENTS:
	case XFS_DINODE_FMT_BTREE:
		buf->bs_rdev = 0;
		buf->bs_blksize = mp->m_sb.sb_blocksize;
		buf->bs_blocks = ip->i_nblocks + ip->i_delayed_blks;
		break;
	}
	xfs_iunlock(ip, XFS_ILOCK_SHARED);
	xfs_irele(ip);

	error = bc->formatter(bc->breq, buf);
	if (error == -ECANCELED)
		goto out_advance;
	if (error)
		goto out;

out_advance:
	/*
	 * Advance the cursor to the inode that comes after the one we just
	 * looked at.  We want the caller to move along if the bulkstat
	 * information was copied successfully; if we tried to grab the inode
	 * but it's no longer allocated; or if it's internal metadata.
	 */
	bc->breq->startino = ino + 1;
out:
	return error;
}

/* Bulkstat a single inode. */
int
xfs_bulkstat_one(
	struct xfs_ibulk	*breq,
	bulkstat_one_fmt_pf	formatter)
{
	struct xfs_bstat_chunk	bc = {
		.formatter	= formatter,
		.breq		= breq,
	};
	struct xfs_trans	*tp;
	int			error;

	if (breq->mnt_userns != &init_user_ns) {
		xfs_warn_ratelimited(breq->mp,
			"bulkstat not supported inside of idmapped mounts.");
		return -EINVAL;
	}

	ASSERT(breq->icount == 1);

	bc.buf = kmem_zalloc(sizeof(struct xfs_bulkstat),
			KM_MAYFAIL);
	if (!bc.buf)
		return -ENOMEM;

	/*
	 * Grab an empty transaction so that we can use its recursive buffer
	 * locking abilities to detect cycles in the inobt without deadlocking.
	 */
	error = xfs_trans_alloc_empty(breq->mp, &tp);
	if (error)
		goto out;

	error = xfs_bulkstat_one_int(breq->mp, breq->mnt_userns, tp,
			breq->startino, &bc);
	xfs_trans_cancel(tp);
out:
	kmem_free(bc.buf);

	/*
	 * If we reported one inode to userspace then we abort because we hit
	 * the end of the buffer.  Don't leak that back to userspace.
	 */
	if (error == -ECANCELED)
		error = 0;

	return error;
}

static int
xfs_bulkstat_iwalk(
	struct xfs_mount	*mp,
	struct xfs_trans	*tp,
	xfs_ino_t		ino,
	void			*data)
{
	struct xfs_bstat_chunk	*bc = data;
	int			error;

	error = xfs_bulkstat_one_int(mp, bc->breq->mnt_userns, tp, ino, data);
	/* bulkstat just skips over missing inodes */
	if (error == -ENOENT || error == -EINVAL)
		return 0;
	return error;
}

/*
 * Check the incoming lastino parameter.
 *
 * We allow any inode value that could map to physical space inside the
 * filesystem because if there are no inodes there, bulkstat moves on to the
 * next chunk.  In other words, the magic agino value of zero takes us to the
 * first chunk in the AG, and an agino value past the end of the AG takes us to
 * the first chunk in the next AG.
 *
 * Therefore we can end early if the requested inode is beyond the end of the
 * filesystem or doesn't map properly.
 */
static inline bool
xfs_bulkstat_already_done(
	struct xfs_mount	*mp,
	xfs_ino_t		startino)
{
	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, startino);
	xfs_agino_t		agino = XFS_INO_TO_AGINO(mp, startino);

	return agno >= mp->m_sb.sb_agcount ||
	       startino != XFS_AGINO_TO_INO(mp, agno, agino);
}

/* Return stat information in bulk (by-inode) for the filesystem. */
int
xfs_bulkstat(
	struct xfs_ibulk	*breq,
	bulkstat_one_fmt_pf	formatter)
{
	struct xfs_bstat_chunk	bc = {
		.formatter	= formatter,
		.breq		= breq,
	};
	struct xfs_trans	*tp;
	int			error;

	if (breq->mnt_userns != &init_user_ns) {
		xfs_warn_ratelimited(breq->mp,
			"bulkstat not supported inside of idmapped mounts.");
		return -EINVAL;
	}
	if (xfs_bulkstat_already_done(breq->mp, breq->startino))
		return 0;

	bc.buf = kmem_zalloc(sizeof(struct xfs_bulkstat),
			KM_MAYFAIL);
	if (!bc.buf)
		return -ENOMEM;

	/*
	 * Grab an empty transaction so that we can use its recursive buffer
	 * locking abilities to detect cycles in the inobt without deadlocking.
	 */
	error = xfs_trans_alloc_empty(breq->mp, &tp);
	if (error)
		goto out;

	error = xfs_iwalk(breq->mp, tp, breq->startino, breq->flags,
			xfs_bulkstat_iwalk, breq->icount, &bc);
	xfs_trans_cancel(tp);
out:
	kmem_free(bc.buf);

	/*
	 * We found some inodes, so clear the error status and return them.
	 * The lastino pointer will point directly at the inode that triggered
	 * any error that occurred, so on the next call the error will be
	 * triggered again and propagated to userspace as there will be no
	 * formatted inodes in the buffer.
	 */
	if (breq->ocount > 0)
		error = 0;

	return error;
}

/* Convert bulkstat (v5) to bstat (v1). */
void
xfs_bulkstat_to_bstat(
	struct xfs_mount		*mp,
	struct xfs_bstat		*bs1,
	const struct xfs_bulkstat	*bstat)
{
	/* memset is needed here because of padding holes in the structure. */
	memset(bs1, 0, sizeof(struct xfs_bstat));
	bs1->bs_ino = bstat->bs_ino;
	bs1->bs_mode = bstat->bs_mode;
	bs1->bs_nlink = bstat->bs_nlink;
	bs1->bs_uid = bstat->bs_uid;
	bs1->bs_gid = bstat->bs_gid;
	bs1->bs_rdev = bstat->bs_rdev;
	bs1->bs_blksize = bstat->bs_blksize;
	bs1->bs_size = bstat->bs_size;
	bs1->bs_atime.tv_sec = bstat->bs_atime;
	bs1->bs_mtime.tv_sec = bstat->bs_mtime;
	bs1->bs_ctime.tv_sec = bstat->bs_ctime;
	bs1->bs_atime.tv_nsec = bstat->bs_atime_nsec;
	bs1->bs_mtime.tv_nsec = bstat->bs_mtime_nsec;
	bs1->bs_ctime.tv_nsec = bstat->bs_ctime_nsec;
	bs1->bs_blocks = bstat->bs_blocks;
	bs1->bs_xflags = bstat->bs_xflags;
	bs1->bs_extsize = XFS_FSB_TO_B(mp, bstat->bs_extsize_blks);
	bs1->bs_extents = bstat->bs_extents;
	bs1->bs_gen = bstat->bs_gen;
	bs1->bs_projid_lo = bstat->bs_projectid & 0xFFFF;
	bs1->bs_forkoff = bstat->bs_forkoff;
	bs1->bs_projid_hi = bstat->bs_projectid >> 16;
	bs1->bs_sick = bstat->bs_sick;
	bs1->bs_checked = bstat->bs_checked;
	bs1->bs_cowextsize = XFS_FSB_TO_B(mp, bstat->bs_cowextsize_blks);
	bs1->bs_dmevmask = 0;
	bs1->bs_dmstate = 0;
	bs1->bs_aextents = bstat->bs_aextents;
}

struct xfs_inumbers_chunk {
	inumbers_fmt_pf		formatter;
	struct xfs_ibulk	*breq;
};

/*
 * INUMBERS
 * ========
 * This is how we export inode btree records to userspace, so that XFS tools
 * can figure out where inodes are allocated.
 */

/*
 * Format the inode group structure and report it somewhere.
 *
 * Similar to xfs_bulkstat_one_int, lastino is the inode cursor as we walk
 * through the filesystem so we move it forward unless there was a runtime
 * error.  If the formatter tells us the buffer is now full we also move the
 * cursor forward and abort the walk.
 */
STATIC int
xfs_inumbers_walk(
	struct xfs_mount	*mp,
	struct xfs_trans	*tp,
	xfs_agnumber_t		agno,
	const struct xfs_inobt_rec_incore *irec,
	void			*data)
{
	struct xfs_inumbers	inogrp = {
		.xi_startino	= XFS_AGINO_TO_INO(mp, agno, irec->ir_startino),
		.xi_alloccount	= irec->ir_count - irec->ir_freecount,
		.xi_allocmask	= ~irec->ir_free,
		.xi_version	= XFS_INUMBERS_VERSION_V5,
	};
	struct xfs_inumbers_chunk *ic = data;
	int			error;

	error = ic->formatter(ic->breq, &inogrp);
	if (error && error != -ECANCELED)
		return error;

	ic->breq->startino = XFS_AGINO_TO_INO(mp, agno, irec->ir_startino) +
			XFS_INODES_PER_CHUNK;
	return error;
}

/*
 * Return inode number table for the filesystem.
 */
int
xfs_inumbers(
	struct xfs_ibulk	*breq,
	inumbers_fmt_pf		formatter)
{
	struct xfs_inumbers_chunk ic = {
		.formatter	= formatter,
		.breq		= breq,
	};
	struct xfs_trans	*tp;
	int			error = 0;

	if (xfs_bulkstat_already_done(breq->mp, breq->startino))
		return 0;

	/*
	 * Grab an empty transaction so that we can use its recursive buffer
	 * locking abilities to detect cycles in the inobt without deadlocking.
	 */
	error = xfs_trans_alloc_empty(breq->mp, &tp);
	if (error)
		goto out;

	error = xfs_inobt_walk(breq->mp, tp, breq->startino, breq->flags,
			xfs_inumbers_walk, breq->icount, &ic);
	xfs_trans_cancel(tp);
out:

	/*
	 * We found some inode groups, so clear the error status and return
	 * them.  The lastino pointer will point directly at the inode that
	 * triggered any error that occurred, so on the next call the error
	 * will be triggered again and propagated to userspace as there will be
	 * no formatted inode groups in the buffer.
	 */
	if (breq->ocount > 0)
		error = 0;

	return error;
}

/* Convert an inumbers (v5) struct to a inogrp (v1) struct. */
void
xfs_inumbers_to_inogrp(
	struct xfs_inogrp		*ig1,
	const struct xfs_inumbers	*ig)
{
	/* memset is needed here because of padding holes in the structure. */
	memset(ig1, 0, sizeof(struct xfs_inogrp));
	ig1->xi_startino = ig->xi_startino;
	ig1->xi_alloccount = ig->xi_alloccount;
	ig1->xi_allocmask = ig->xi_allocmask;
}