summaryrefslogtreecommitdiff
path: root/fs/xfs/xfs_inode_item_recover.c
blob: 2bdba612aa71f6ab1360f1be02e7fd5f240d0ea4 (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
// SPDX-License-Identifier: GPL-2.0
/*
 * Copyright (c) 2000-2006 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_trans.h"
#include "xfs_inode_item.h"
#include "xfs_trace.h"
#include "xfs_trans_priv.h"
#include "xfs_buf_item.h"
#include "xfs_log.h"
#include "xfs_error.h"
#include "xfs_log_priv.h"
#include "xfs_log_recover.h"
#include "xfs_icache.h"
#include "xfs_bmap_btree.h"

STATIC void
xlog_recover_inode_ra_pass2(
	struct xlog                     *log,
	struct xlog_recover_item        *item)
{
	if (item->ri_buf[0].i_len == sizeof(struct xfs_inode_log_format)) {
		struct xfs_inode_log_format	*ilfp = item->ri_buf[0].i_addr;

		xlog_buf_readahead(log, ilfp->ilf_blkno, ilfp->ilf_len,
				   &xfs_inode_buf_ra_ops);
	} else {
		struct xfs_inode_log_format_32	*ilfp = item->ri_buf[0].i_addr;

		xlog_buf_readahead(log, ilfp->ilf_blkno, ilfp->ilf_len,
				   &xfs_inode_buf_ra_ops);
	}
}

/*
 * Inode fork owner changes
 *
 * If we have been told that we have to reparent the inode fork, it's because an
 * extent swap operation on a CRC enabled filesystem has been done and we are
 * replaying it. We need to walk the BMBT of the appropriate fork and change the
 * owners of it.
 *
 * The complexity here is that we don't have an inode context to work with, so
 * after we've replayed the inode we need to instantiate one.  This is where the
 * fun begins.
 *
 * We are in the middle of log recovery, so we can't run transactions. That
 * means we cannot use cache coherent inode instantiation via xfs_iget(), as
 * that will result in the corresponding iput() running the inode through
 * xfs_inactive(). If we've just replayed an inode core that changes the link
 * count to zero (i.e. it's been unlinked), then xfs_inactive() will run
 * transactions (bad!).
 *
 * So, to avoid this, we instantiate an inode directly from the inode core we've
 * just recovered. We have the buffer still locked, and all we really need to
 * instantiate is the inode core and the forks being modified. We can do this
 * manually, then run the inode btree owner change, and then tear down the
 * xfs_inode without having to run any transactions at all.
 *
 * Also, because we don't have a transaction context available here but need to
 * gather all the buffers we modify for writeback so we pass the buffer_list
 * instead for the operation to use.
 */

STATIC int
xfs_recover_inode_owner_change(
	struct xfs_mount	*mp,
	struct xfs_dinode	*dip,
	struct xfs_inode_log_format *in_f,
	struct list_head	*buffer_list)
{
	struct xfs_inode	*ip;
	int			error;

	ASSERT(in_f->ilf_fields & (XFS_ILOG_DOWNER|XFS_ILOG_AOWNER));

	ip = xfs_inode_alloc(mp, in_f->ilf_ino);
	if (!ip)
		return -ENOMEM;

	/* instantiate the inode */
	ASSERT(dip->di_version >= 3);
	xfs_inode_from_disk(ip, dip);

	error = xfs_iformat_fork(ip, dip);
	if (error)
		goto out_free_ip;

	if (!xfs_inode_verify_forks(ip)) {
		error = -EFSCORRUPTED;
		goto out_free_ip;
	}

	if (in_f->ilf_fields & XFS_ILOG_DOWNER) {
		ASSERT(in_f->ilf_fields & XFS_ILOG_DBROOT);
		error = xfs_bmbt_change_owner(NULL, ip, XFS_DATA_FORK,
					      ip->i_ino, buffer_list);
		if (error)
			goto out_free_ip;
	}

	if (in_f->ilf_fields & XFS_ILOG_AOWNER) {
		ASSERT(in_f->ilf_fields & XFS_ILOG_ABROOT);
		error = xfs_bmbt_change_owner(NULL, ip, XFS_ATTR_FORK,
					      ip->i_ino, buffer_list);
		if (error)
			goto out_free_ip;
	}

out_free_ip:
	xfs_inode_free(ip);
	return error;
}

STATIC int
xlog_recover_inode_commit_pass2(
	struct xlog			*log,
	struct list_head		*buffer_list,
	struct xlog_recover_item	*item,
	xfs_lsn_t			current_lsn)
{
	struct xfs_inode_log_format	*in_f;
	struct xfs_mount		*mp = log->l_mp;
	struct xfs_buf			*bp;
	struct xfs_dinode		*dip;
	int				len;
	char				*src;
	char				*dest;
	int				error;
	int				attr_index;
	uint				fields;
	struct xfs_log_dinode		*ldip;
	uint				isize;
	int				need_free = 0;

	if (item->ri_buf[0].i_len == sizeof(struct xfs_inode_log_format)) {
		in_f = item->ri_buf[0].i_addr;
	} else {
		in_f = kmem_alloc(sizeof(struct xfs_inode_log_format), 0);
		need_free = 1;
		error = xfs_inode_item_format_convert(&item->ri_buf[0], in_f);
		if (error)
			goto error;
	}

	/*
	 * Inode buffers can be freed, look out for it,
	 * and do not replay the inode.
	 */
	if (xlog_is_buffer_cancelled(log, in_f->ilf_blkno, in_f->ilf_len)) {
		error = 0;
		trace_xfs_log_recover_inode_cancel(log, in_f);
		goto error;
	}
	trace_xfs_log_recover_inode_recover(log, in_f);

	error = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len,
			0, &bp, &xfs_inode_buf_ops);
	if (error)
		goto error;
	ASSERT(in_f->ilf_fields & XFS_ILOG_CORE);
	dip = xfs_buf_offset(bp, in_f->ilf_boffset);

	/*
	 * Make sure the place we're flushing out to really looks
	 * like an inode!
	 */
	if (XFS_IS_CORRUPT(mp, !xfs_verify_magic16(bp, dip->di_magic))) {
		xfs_alert(mp,
	"%s: Bad inode magic number, dip = "PTR_FMT", dino bp = "PTR_FMT", ino = %Ld",
			__func__, dip, bp, in_f->ilf_ino);
		error = -EFSCORRUPTED;
		goto out_release;
	}
	ldip = item->ri_buf[1].i_addr;
	if (XFS_IS_CORRUPT(mp, ldip->di_magic != XFS_DINODE_MAGIC)) {
		xfs_alert(mp,
			"%s: Bad inode log record, rec ptr "PTR_FMT", ino %Ld",
			__func__, item, in_f->ilf_ino);
		error = -EFSCORRUPTED;
		goto out_release;
	}

	/*
	 * If the inode has an LSN in it, recover the inode only if it's less
	 * than the lsn of the transaction we are replaying. Note: we still
	 * need to replay an owner change even though the inode is more recent
	 * than the transaction as there is no guarantee that all the btree
	 * blocks are more recent than this transaction, too.
	 */
	if (dip->di_version >= 3) {
		xfs_lsn_t	lsn = be64_to_cpu(dip->di_lsn);

		if (lsn && lsn != -1 && XFS_LSN_CMP(lsn, current_lsn) >= 0) {
			trace_xfs_log_recover_inode_skip(log, in_f);
			error = 0;
			goto out_owner_change;
		}
	}

	/*
	 * di_flushiter is only valid for v1/2 inodes. All changes for v3 inodes
	 * are transactional and if ordering is necessary we can determine that
	 * more accurately by the LSN field in the V3 inode core. Don't trust
	 * the inode versions we might be changing them here - use the
	 * superblock flag to determine whether we need to look at di_flushiter
	 * to skip replay when the on disk inode is newer than the log one
	 */
	if (!xfs_sb_version_has_v3inode(&mp->m_sb) &&
	    ldip->di_flushiter < be16_to_cpu(dip->di_flushiter)) {
		/*
		 * Deal with the wrap case, DI_MAX_FLUSH is less
		 * than smaller numbers
		 */
		if (be16_to_cpu(dip->di_flushiter) == DI_MAX_FLUSH &&
		    ldip->di_flushiter < (DI_MAX_FLUSH >> 1)) {
			/* do nothing */
		} else {
			trace_xfs_log_recover_inode_skip(log, in_f);
			error = 0;
			goto out_release;
		}
	}

	/* Take the opportunity to reset the flush iteration count */
	ldip->di_flushiter = 0;

	if (unlikely(S_ISREG(ldip->di_mode))) {
		if ((ldip->di_format != XFS_DINODE_FMT_EXTENTS) &&
		    (ldip->di_format != XFS_DINODE_FMT_BTREE)) {
			XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(3)",
					 XFS_ERRLEVEL_LOW, mp, ldip,
					 sizeof(*ldip));
			xfs_alert(mp,
		"%s: Bad regular inode log record, rec ptr "PTR_FMT", "
		"ino ptr = "PTR_FMT", ino bp = "PTR_FMT", ino %Ld",
				__func__, item, dip, bp, in_f->ilf_ino);
			error = -EFSCORRUPTED;
			goto out_release;
		}
	} else if (unlikely(S_ISDIR(ldip->di_mode))) {
		if ((ldip->di_format != XFS_DINODE_FMT_EXTENTS) &&
		    (ldip->di_format != XFS_DINODE_FMT_BTREE) &&
		    (ldip->di_format != XFS_DINODE_FMT_LOCAL)) {
			XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(4)",
					     XFS_ERRLEVEL_LOW, mp, ldip,
					     sizeof(*ldip));
			xfs_alert(mp,
		"%s: Bad dir inode log record, rec ptr "PTR_FMT", "
		"ino ptr = "PTR_FMT", ino bp = "PTR_FMT", ino %Ld",
				__func__, item, dip, bp, in_f->ilf_ino);
			error = -EFSCORRUPTED;
			goto out_release;
		}
	}
	if (unlikely(ldip->di_nextents + ldip->di_anextents > ldip->di_nblocks)){
		XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(5)",
				     XFS_ERRLEVEL_LOW, mp, ldip,
				     sizeof(*ldip));
		xfs_alert(mp,
	"%s: Bad inode log record, rec ptr "PTR_FMT", dino ptr "PTR_FMT", "
	"dino bp "PTR_FMT", ino %Ld, total extents = %d, nblocks = %Ld",
			__func__, item, dip, bp, in_f->ilf_ino,
			ldip->di_nextents + ldip->di_anextents,
			ldip->di_nblocks);
		error = -EFSCORRUPTED;
		goto out_release;
	}
	if (unlikely(ldip->di_forkoff > mp->m_sb.sb_inodesize)) {
		XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(6)",
				     XFS_ERRLEVEL_LOW, mp, ldip,
				     sizeof(*ldip));
		xfs_alert(mp,
	"%s: Bad inode log record, rec ptr "PTR_FMT", dino ptr "PTR_FMT", "
	"dino bp "PTR_FMT", ino %Ld, forkoff 0x%x", __func__,
			item, dip, bp, in_f->ilf_ino, ldip->di_forkoff);
		error = -EFSCORRUPTED;
		goto out_release;
	}
	isize = xfs_log_dinode_size(mp);
	if (unlikely(item->ri_buf[1].i_len > isize)) {
		XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(7)",
				     XFS_ERRLEVEL_LOW, mp, ldip,
				     sizeof(*ldip));
		xfs_alert(mp,
			"%s: Bad inode log record length %d, rec ptr "PTR_FMT,
			__func__, item->ri_buf[1].i_len, item);
		error = -EFSCORRUPTED;
		goto out_release;
	}

	/* recover the log dinode inode into the on disk inode */
	xfs_log_dinode_to_disk(ldip, dip);

	fields = in_f->ilf_fields;
	if (fields & XFS_ILOG_DEV)
		xfs_dinode_put_rdev(dip, in_f->ilf_u.ilfu_rdev);

	if (in_f->ilf_size == 2)
		goto out_owner_change;
	len = item->ri_buf[2].i_len;
	src = item->ri_buf[2].i_addr;
	ASSERT(in_f->ilf_size <= 4);
	ASSERT((in_f->ilf_size == 3) || (fields & XFS_ILOG_AFORK));
	ASSERT(!(fields & XFS_ILOG_DFORK) ||
	       (len == in_f->ilf_dsize));

	switch (fields & XFS_ILOG_DFORK) {
	case XFS_ILOG_DDATA:
	case XFS_ILOG_DEXT:
		memcpy(XFS_DFORK_DPTR(dip), src, len);
		break;

	case XFS_ILOG_DBROOT:
		xfs_bmbt_to_bmdr(mp, (struct xfs_btree_block *)src, len,
				 (struct xfs_bmdr_block *)XFS_DFORK_DPTR(dip),
				 XFS_DFORK_DSIZE(dip, mp));
		break;

	default:
		/*
		 * There are no data fork flags set.
		 */
		ASSERT((fields & XFS_ILOG_DFORK) == 0);
		break;
	}

	/*
	 * If we logged any attribute data, recover it.  There may or
	 * may not have been any other non-core data logged in this
	 * transaction.
	 */
	if (in_f->ilf_fields & XFS_ILOG_AFORK) {
		if (in_f->ilf_fields & XFS_ILOG_DFORK) {
			attr_index = 3;
		} else {
			attr_index = 2;
		}
		len = item->ri_buf[attr_index].i_len;
		src = item->ri_buf[attr_index].i_addr;
		ASSERT(len == in_f->ilf_asize);

		switch (in_f->ilf_fields & XFS_ILOG_AFORK) {
		case XFS_ILOG_ADATA:
		case XFS_ILOG_AEXT:
			dest = XFS_DFORK_APTR(dip);
			ASSERT(len <= XFS_DFORK_ASIZE(dip, mp));
			memcpy(dest, src, len);
			break;

		case XFS_ILOG_ABROOT:
			dest = XFS_DFORK_APTR(dip);
			xfs_bmbt_to_bmdr(mp, (struct xfs_btree_block *)src,
					 len, (struct xfs_bmdr_block *)dest,
					 XFS_DFORK_ASIZE(dip, mp));
			break;

		default:
			xfs_warn(log->l_mp, "%s: Invalid flag", __func__);
			ASSERT(0);
			error = -EFSCORRUPTED;
			goto out_release;
		}
	}

out_owner_change:
	/* Recover the swapext owner change unless inode has been deleted */
	if ((in_f->ilf_fields & (XFS_ILOG_DOWNER|XFS_ILOG_AOWNER)) &&
	    (dip->di_mode != 0))
		error = xfs_recover_inode_owner_change(mp, dip, in_f,
						       buffer_list);
	/* re-generate the checksum. */
	xfs_dinode_calc_crc(log->l_mp, dip);

	ASSERT(bp->b_mount == mp);
	bp->b_iodone = xlog_recover_iodone;
	xfs_buf_delwri_queue(bp, buffer_list);

out_release:
	xfs_buf_relse(bp);
error:
	if (need_free)
		kmem_free(in_f);
	return error;
}

const struct xlog_recover_item_ops xlog_inode_item_ops = {
	.item_type		= XFS_LI_INODE,
	.ra_pass2		= xlog_recover_inode_ra_pass2,
	.commit_pass2		= xlog_recover_inode_commit_pass2,
};