Blame view
fs/xfs/xfs_log_recover.c
97.9 KB
0b61f8a40 xfs: convert to S... |
1 |
// SPDX-License-Identifier: GPL-2.0 |
1da177e4c Linux-2.6.12-rc2 |
2 |
/* |
87c199c2a [XFS] Over zealou... |
3 |
* Copyright (c) 2000-2006 Silicon Graphics, Inc. |
7b7187698 [XFS] Update lice... |
4 |
* All Rights Reserved. |
1da177e4c Linux-2.6.12-rc2 |
5 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
6 |
#include "xfs.h" |
a844f4510 [XFS] Remove xfs_... |
7 |
#include "xfs_fs.h" |
70a9883c5 xfs: create a sha... |
8 |
#include "xfs_shared.h" |
239880ef6 xfs: decouple log... |
9 10 11 |
#include "xfs_format.h" #include "xfs_log_format.h" #include "xfs_trans_resv.h" |
a844f4510 [XFS] Remove xfs_... |
12 |
#include "xfs_bit.h" |
a844f4510 [XFS] Remove xfs_... |
13 |
#include "xfs_sb.h" |
1da177e4c Linux-2.6.12-rc2 |
14 |
#include "xfs_mount.h" |
509955823 xfs: log recovery... |
15 |
#include "xfs_defer.h" |
1da177e4c Linux-2.6.12-rc2 |
16 |
#include "xfs_inode.h" |
239880ef6 xfs: decouple log... |
17 |
#include "xfs_trans.h" |
239880ef6 xfs: decouple log... |
18 |
#include "xfs_log.h" |
1da177e4c Linux-2.6.12-rc2 |
19 |
#include "xfs_log_priv.h" |
1da177e4c Linux-2.6.12-rc2 |
20 |
#include "xfs_log_recover.h" |
1da177e4c Linux-2.6.12-rc2 |
21 |
#include "xfs_trans_priv.h" |
a4fbe6ab1 xfs: decouple ino... |
22 23 |
#include "xfs_alloc.h" #include "xfs_ialloc.h" |
0b1b213fc xfs: event tracin... |
24 |
#include "xfs_trace.h" |
33479e054 xfs: remove xfs_i... |
25 |
#include "xfs_icache.h" |
a4fbe6ab1 xfs: decouple ino... |
26 |
#include "xfs_error.h" |
60a4a2225 xfs: update metad... |
27 |
#include "xfs_buf_item.h" |
1da177e4c Linux-2.6.12-rc2 |
28 |
|
fc06c6d06 xfs: separate out... |
29 |
#define BLK_AVG(blk1, blk2) ((blk1+blk2) >> 1) |
9a8d2fdbb xfs: remove xlog_... |
30 31 32 33 34 35 36 37 |
STATIC int xlog_find_zeroed( struct xlog *, xfs_daddr_t *); STATIC int xlog_clear_stale_blocks( struct xlog *, xfs_lsn_t); |
1da177e4c Linux-2.6.12-rc2 |
38 |
#if defined(DEBUG) |
9a8d2fdbb xfs: remove xlog_... |
39 40 41 |
STATIC void xlog_recover_check_summary( struct xlog *); |
1da177e4c Linux-2.6.12-rc2 |
42 43 |
#else #define xlog_recover_check_summary(log) |
1da177e4c Linux-2.6.12-rc2 |
44 |
#endif |
7088c4136 xfs: detect and t... |
45 46 47 |
STATIC int xlog_do_recovery_pass( struct xlog *, xfs_daddr_t, xfs_daddr_t, int, xfs_daddr_t *); |
1da177e4c Linux-2.6.12-rc2 |
48 |
|
1da177e4c Linux-2.6.12-rc2 |
49 50 51 |
/* * Sector aligned buffer routines for buffer create/read/write/access */ |
ff30a6221 xfs: encapsulate ... |
52 |
/* |
99c265950 xfs: more robust ... |
53 54 55 |
* Verify the log-relative block number and length in basic blocks are valid for * an operation involving the given XFS log buffer. Returns true if the fields * are valid, false otherwise. |
ff30a6221 xfs: encapsulate ... |
56 |
*/ |
99c265950 xfs: more robust ... |
57 |
static inline bool |
6e9b3dd80 xfs: stop using b... |
58 |
xlog_verify_bno( |
9a8d2fdbb xfs: remove xlog_... |
59 |
struct xlog *log, |
99c265950 xfs: more robust ... |
60 |
xfs_daddr_t blk_no, |
ff30a6221 xfs: encapsulate ... |
61 62 |
int bbcount) { |
99c265950 xfs: more robust ... |
63 64 65 66 67 |
if (blk_no < 0 || blk_no >= log->l_logBBsize) return false; if (bbcount <= 0 || (blk_no + bbcount) > log->l_logBBsize) return false; return true; |
ff30a6221 xfs: encapsulate ... |
68 |
} |
36adecff5 xfs: nothing spec... |
69 |
/* |
6ad5b3255 xfs: use bios dir... |
70 71 |
* Allocate a buffer to hold log data. The buffer needs to be able to map to * a range of nbblks basic blocks at any valid offset within the log. |
36adecff5 xfs: nothing spec... |
72 |
*/ |
6ad5b3255 xfs: use bios dir... |
73 |
static char * |
6e9b3dd80 xfs: stop using b... |
74 |
xlog_alloc_buffer( |
9a8d2fdbb xfs: remove xlog_... |
75 |
struct xlog *log, |
3228149ce xfs: Check buffer... |
76 |
int nbblks) |
1da177e4c Linux-2.6.12-rc2 |
77 |
{ |
f8f9ee479 xfs: add kmem_all... |
78 |
int align_mask = xfs_buftarg_dma_alignment(log->l_targ); |
99c265950 xfs: more robust ... |
79 80 81 82 |
/* * Pass log block 0 since we don't have an addr yet, buffer will be * verified on read. */ |
a71895c5d xfs: convert open... |
83 |
if (XFS_IS_CORRUPT(log->l_mp, !xlog_verify_bno(log, 0, nbblks))) { |
a0fa2b679 xfs: Convert xlog... |
84 |
xfs_warn(log->l_mp, "Invalid block length (0x%x) for buffer", |
ff30a6221 xfs: encapsulate ... |
85 |
nbblks); |
3228149ce xfs: Check buffer... |
86 87 |
return NULL; } |
1da177e4c Linux-2.6.12-rc2 |
88 |
|
36adecff5 xfs: nothing spec... |
89 |
/* |
6ad5b3255 xfs: use bios dir... |
90 91 92 |
* We do log I/O in units of log sectors (a power-of-2 multiple of the * basic block size), so we round up the requested size to accommodate * the basic blocks required for complete log sectors. |
36adecff5 xfs: nothing spec... |
93 |
* |
6ad5b3255 xfs: use bios dir... |
94 95 96 97 98 99 100 101 |
* In addition, the buffer may be used for a non-sector-aligned block * offset, in which case an I/O of the requested size could extend * beyond the end of the buffer. If the requested size is only 1 basic * block it will never straddle a sector boundary, so this won't be an * issue. Nor will this be a problem if the log I/O is done in basic * blocks (sector size 1). But otherwise we extend the buffer by one * extra log sector to ensure there's space to accommodate this * possibility. |
36adecff5 xfs: nothing spec... |
102 |
*/ |
69ce58f08 xfs: record log s... |
103 104 105 |
if (nbblks > 1 && log->l_sectBBsize > 1) nbblks += log->l_sectBBsize; nbblks = round_up(nbblks, log->l_sectBBsize); |
3219e8cf0 xfs: assure zeroe... |
106 |
return kmem_alloc_io(BBTOB(nbblks), align_mask, KM_MAYFAIL | KM_ZERO); |
1da177e4c Linux-2.6.12-rc2 |
107 |
} |
48389ef17 xfs: kill off l_s... |
108 109 110 111 |
/* * Return the address of the start of the given block number's data * in a log buffer. The buffer covers a log sector-aligned region. */ |
18ffb8c3f xfs: return an of... |
112 |
static inline unsigned int |
076e6acb8 xfs: cleanup xlog... |
113 |
xlog_align( |
9a8d2fdbb xfs: remove xlog_... |
114 |
struct xlog *log, |
18ffb8c3f xfs: return an of... |
115 |
xfs_daddr_t blk_no) |
076e6acb8 xfs: cleanup xlog... |
116 |
{ |
18ffb8c3f xfs: return an of... |
117 |
return BBTOB(blk_no & ((xfs_daddr_t)log->l_sectBBsize - 1)); |
076e6acb8 xfs: cleanup xlog... |
118 |
} |
6ad5b3255 xfs: use bios dir... |
119 120 121 122 123 124 125 |
static int xlog_do_io( struct xlog *log, xfs_daddr_t blk_no, unsigned int nbblks, char *data, unsigned int op) |
1da177e4c Linux-2.6.12-rc2 |
126 |
{ |
6ad5b3255 xfs: use bios dir... |
127 |
int error; |
1da177e4c Linux-2.6.12-rc2 |
128 |
|
a71895c5d xfs: convert open... |
129 |
if (XFS_IS_CORRUPT(log->l_mp, !xlog_verify_bno(log, blk_no, nbblks))) { |
99c265950 xfs: more robust ... |
130 131 132 |
xfs_warn(log->l_mp, "Invalid log block/length (0x%llx, 0x%x) for buffer", blk_no, nbblks); |
2451337dd xfs: global error... |
133 |
return -EFSCORRUPTED; |
3228149ce xfs: Check buffer... |
134 |
} |
69ce58f08 xfs: record log s... |
135 136 |
blk_no = round_down(blk_no, log->l_sectBBsize); nbblks = round_up(nbblks, log->l_sectBBsize); |
1da177e4c Linux-2.6.12-rc2 |
137 |
ASSERT(nbblks > 0); |
1da177e4c Linux-2.6.12-rc2 |
138 |
|
6ad5b3255 xfs: use bios dir... |
139 140 141 142 143 144 145 146 |
error = xfs_rw_bdev(log->l_targ->bt_bdev, log->l_logBBstart + blk_no, BBTOB(nbblks), data, op); if (error && !XFS_FORCED_SHUTDOWN(log->l_mp)) { xfs_alert(log->l_mp, "log recovery %s I/O error at daddr 0x%llx len %d error %d", op == REQ_OP_WRITE ? "write" : "read", blk_no, nbblks, error); } |
1da177e4c Linux-2.6.12-rc2 |
147 148 |
return error; } |
076e6acb8 xfs: cleanup xlog... |
149 |
STATIC int |
6ad5b3255 xfs: use bios dir... |
150 |
xlog_bread_noalign( |
9a8d2fdbb xfs: remove xlog_... |
151 |
struct xlog *log, |
076e6acb8 xfs: cleanup xlog... |
152 153 |
xfs_daddr_t blk_no, int nbblks, |
6ad5b3255 xfs: use bios dir... |
154 |
char *data) |
076e6acb8 xfs: cleanup xlog... |
155 |
{ |
6ad5b3255 xfs: use bios dir... |
156 |
return xlog_do_io(log, blk_no, nbblks, data, REQ_OP_READ); |
076e6acb8 xfs: cleanup xlog... |
157 |
} |
44396476a xfs: reset buffer... |
158 |
STATIC int |
6ad5b3255 xfs: use bios dir... |
159 |
xlog_bread( |
9a8d2fdbb xfs: remove xlog_... |
160 |
struct xlog *log, |
6ad5b3255 xfs: use bios dir... |
161 162 163 164 |
xfs_daddr_t blk_no, int nbblks, char *data, char **offset) |
44396476a xfs: reset buffer... |
165 |
{ |
6ad5b3255 xfs: use bios dir... |
166 |
int error; |
44396476a xfs: reset buffer... |
167 |
|
6ad5b3255 xfs: use bios dir... |
168 169 170 171 |
error = xlog_do_io(log, blk_no, nbblks, data, REQ_OP_READ); if (!error) *offset = data + xlog_align(log, blk_no); return error; |
44396476a xfs: reset buffer... |
172 |
} |
ba0f32d46 [XFS] mark variou... |
173 |
STATIC int |
1da177e4c Linux-2.6.12-rc2 |
174 |
xlog_bwrite( |
9a8d2fdbb xfs: remove xlog_... |
175 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
176 177 |
xfs_daddr_t blk_no, int nbblks, |
6ad5b3255 xfs: use bios dir... |
178 |
char *data) |
1da177e4c Linux-2.6.12-rc2 |
179 |
{ |
6ad5b3255 xfs: use bios dir... |
180 |
return xlog_do_io(log, blk_no, nbblks, data, REQ_OP_WRITE); |
1da177e4c Linux-2.6.12-rc2 |
181 |
} |
1da177e4c Linux-2.6.12-rc2 |
182 183 184 185 186 187 188 189 190 |
#ifdef DEBUG /* * dump debug superblock and log record information */ STATIC void xlog_header_check_dump( xfs_mount_t *mp, xlog_rec_header_t *head) { |
08e96e1a3 xfs: remove newli... |
191 |
xfs_debug(mp, "%s: SB : uuid = %pU, fmt = %d", |
03daa57cd fs/xfs/xfs_log_re... |
192 |
__func__, &mp->m_sb.sb_uuid, XLOG_FMT); |
08e96e1a3 xfs: remove newli... |
193 |
xfs_debug(mp, " log : uuid = %pU, fmt = %d", |
03daa57cd fs/xfs/xfs_log_re... |
194 |
&head->h_fs_uuid, be32_to_cpu(head->h_fmt)); |
1da177e4c Linux-2.6.12-rc2 |
195 196 197 198 199 200 201 202 203 204 205 206 207 |
} #else #define xlog_header_check_dump(mp, head) #endif /* * check log record header for recovery */ STATIC int xlog_header_check_recover( xfs_mount_t *mp, xlog_rec_header_t *head) { |
69ef921b5 xfs: byteswap con... |
208 |
ASSERT(head->h_magicno == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)); |
1da177e4c Linux-2.6.12-rc2 |
209 210 211 212 213 214 |
/* * IRIX doesn't write the h_fmt field and leaves it zeroed * (XLOG_FMT_UNKNOWN). This stops us from trying to recover * a dirty log created in IRIX. */ |
a71895c5d xfs: convert open... |
215 |
if (XFS_IS_CORRUPT(mp, head->h_fmt != cpu_to_be32(XLOG_FMT))) { |
a0fa2b679 xfs: Convert xlog... |
216 217 |
xfs_warn(mp, "dirty log written in incompatible format - can't recover"); |
1da177e4c Linux-2.6.12-rc2 |
218 |
xlog_header_check_dump(mp, head); |
2451337dd xfs: global error... |
219 |
return -EFSCORRUPTED; |
a71895c5d xfs: convert open... |
220 221 222 |
} if (XFS_IS_CORRUPT(mp, !uuid_equal(&mp->m_sb.sb_uuid, &head->h_fs_uuid))) { |
a0fa2b679 xfs: Convert xlog... |
223 224 |
xfs_warn(mp, "dirty log entry has mismatched uuid - can't recover"); |
1da177e4c Linux-2.6.12-rc2 |
225 |
xlog_header_check_dump(mp, head); |
2451337dd xfs: global error... |
226 |
return -EFSCORRUPTED; |
1da177e4c Linux-2.6.12-rc2 |
227 228 229 230 231 232 233 234 235 236 237 238 |
} return 0; } /* * read the head block of the log and check the header */ STATIC int xlog_header_check_mount( xfs_mount_t *mp, xlog_rec_header_t *head) { |
69ef921b5 xfs: byteswap con... |
239 |
ASSERT(head->h_magicno == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)); |
1da177e4c Linux-2.6.12-rc2 |
240 |
|
d905fdaaa xfs: use the comm... |
241 |
if (uuid_is_null(&head->h_fs_uuid)) { |
1da177e4c Linux-2.6.12-rc2 |
242 243 |
/* * IRIX doesn't write the h_fs_uuid or h_fmt fields. If |
d905fdaaa xfs: use the comm... |
244 |
* h_fs_uuid is null, we assume this log was last mounted |
1da177e4c Linux-2.6.12-rc2 |
245 246 |
* by IRIX and continue. */ |
d905fdaaa xfs: use the comm... |
247 |
xfs_warn(mp, "null uuid in log - IRIX style log"); |
a71895c5d xfs: convert open... |
248 249 |
} else if (XFS_IS_CORRUPT(mp, !uuid_equal(&mp->m_sb.sb_uuid, &head->h_fs_uuid))) { |
a0fa2b679 xfs: Convert xlog... |
250 |
xfs_warn(mp, "log has mismatched uuid - can't recover"); |
1da177e4c Linux-2.6.12-rc2 |
251 |
xlog_header_check_dump(mp, head); |
2451337dd xfs: global error... |
252 |
return -EFSCORRUPTED; |
1da177e4c Linux-2.6.12-rc2 |
253 254 255 |
} return 0; } |
1da177e4c Linux-2.6.12-rc2 |
256 257 258 259 260 261 |
/* * This routine finds (to an approximation) the first block in the physical * log which contains the given cycle. It uses a binary search algorithm. * Note that the algorithm can not be perfect because the disk will not * necessarily be perfect. */ |
a8272ce0c [XFS] Fix up spar... |
262 |
STATIC int |
1da177e4c Linux-2.6.12-rc2 |
263 |
xlog_find_cycle_start( |
9a8d2fdbb xfs: remove xlog_... |
264 |
struct xlog *log, |
6e9b3dd80 xfs: stop using b... |
265 |
char *buffer, |
1da177e4c Linux-2.6.12-rc2 |
266 267 268 269 |
xfs_daddr_t first_blk, xfs_daddr_t *last_blk, uint cycle) { |
b2a922cd6 xfs: remove xfs_c... |
270 |
char *offset; |
1da177e4c Linux-2.6.12-rc2 |
271 |
xfs_daddr_t mid_blk; |
e3bb2e30d xfs: avoid repeat... |
272 |
xfs_daddr_t end_blk; |
1da177e4c Linux-2.6.12-rc2 |
273 274 |
uint mid_cycle; int error; |
e3bb2e30d xfs: avoid repeat... |
275 276 277 |
end_blk = *last_blk; mid_blk = BLK_AVG(first_blk, end_blk); while (mid_blk != first_blk && mid_blk != end_blk) { |
6e9b3dd80 xfs: stop using b... |
278 |
error = xlog_bread(log, mid_blk, 1, buffer, &offset); |
076e6acb8 xfs: cleanup xlog... |
279 |
if (error) |
1da177e4c Linux-2.6.12-rc2 |
280 |
return error; |
03bea6fe6 [XFS] clean up so... |
281 |
mid_cycle = xlog_get_cycle(offset); |
e3bb2e30d xfs: avoid repeat... |
282 283 284 285 286 |
if (mid_cycle == cycle) end_blk = mid_blk; /* last_half_cycle == mid_cycle */ else first_blk = mid_blk; /* first_half_cycle == mid_cycle */ mid_blk = BLK_AVG(first_blk, end_blk); |
1da177e4c Linux-2.6.12-rc2 |
287 |
} |
e3bb2e30d xfs: avoid repeat... |
288 289 290 291 |
ASSERT((mid_blk == first_blk && mid_blk+1 == end_blk) || (mid_blk == end_blk && mid_blk-1 == first_blk)); *last_blk = end_blk; |
1da177e4c Linux-2.6.12-rc2 |
292 293 294 295 296 |
return 0; } /* |
3f943d853 xfs: minor odds a... |
297 298 299 300 301 302 |
* Check that a range of blocks does not contain stop_on_cycle_no. * Fill in *new_blk with the block offset where such a block is * found, or with -1 (an invalid block number) if there is no such * block in the range. The scan needs to occur from front to back * and the pointer into the region must be updated since a later * routine will need to perform another test. |
1da177e4c Linux-2.6.12-rc2 |
303 304 305 |
*/ STATIC int xlog_find_verify_cycle( |
9a8d2fdbb xfs: remove xlog_... |
306 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
307 308 309 310 311 312 313 |
xfs_daddr_t start_blk, int nbblks, uint stop_on_cycle_no, xfs_daddr_t *new_blk) { xfs_daddr_t i, j; uint cycle; |
6e9b3dd80 xfs: stop using b... |
314 |
char *buffer; |
1da177e4c Linux-2.6.12-rc2 |
315 |
xfs_daddr_t bufblks; |
b2a922cd6 xfs: remove xfs_c... |
316 |
char *buf = NULL; |
1da177e4c Linux-2.6.12-rc2 |
317 |
int error = 0; |
6881a229f xfs: fix min bufs... |
318 319 320 321 322 323 |
/* * Greedily allocate a buffer big enough to handle the full * range of basic blocks we'll be examining. If that fails, * try a smaller size. We need to be able to read at least * a log sector, or we're out of luck. */ |
1da177e4c Linux-2.6.12-rc2 |
324 |
bufblks = 1 << ffs(nbblks); |
81158e0ce xfs: prevent need... |
325 326 |
while (bufblks > log->l_logBBsize) bufblks >>= 1; |
6e9b3dd80 xfs: stop using b... |
327 |
while (!(buffer = xlog_alloc_buffer(log, bufblks))) { |
1da177e4c Linux-2.6.12-rc2 |
328 |
bufblks >>= 1; |
69ce58f08 xfs: record log s... |
329 |
if (bufblks < log->l_sectBBsize) |
2451337dd xfs: global error... |
330 |
return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
331 332 333 334 335 336 |
} for (i = start_blk; i < start_blk + nbblks; i += bufblks) { int bcount; bcount = min(bufblks, (start_blk + nbblks - i)); |
6e9b3dd80 xfs: stop using b... |
337 |
error = xlog_bread(log, i, bcount, buffer, &buf); |
076e6acb8 xfs: cleanup xlog... |
338 |
if (error) |
1da177e4c Linux-2.6.12-rc2 |
339 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
340 |
for (j = 0; j < bcount; j++) { |
03bea6fe6 [XFS] clean up so... |
341 |
cycle = xlog_get_cycle(buf); |
1da177e4c Linux-2.6.12-rc2 |
342 343 344 345 346 347 348 349 350 351 352 353 |
if (cycle == stop_on_cycle_no) { *new_blk = i+j; goto out; } buf += BBSIZE; } } *new_blk = -1; out: |
6e9b3dd80 xfs: stop using b... |
354 |
kmem_free(buffer); |
1da177e4c Linux-2.6.12-rc2 |
355 356 |
return error; } |
0c771b99d xfs: clean up cal... |
357 358 359 360 361 362 363 364 365 366 367 368 |
static inline int xlog_logrec_hblks(struct xlog *log, struct xlog_rec_header *rh) { if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { int h_size = be32_to_cpu(rh->h_size); if ((be32_to_cpu(rh->h_version) & XLOG_VERSION_2) && h_size > XLOG_HEADER_CYCLE_SIZE) return DIV_ROUND_UP(h_size, XLOG_HEADER_CYCLE_SIZE); } return 1; } |
1da177e4c Linux-2.6.12-rc2 |
369 370 371 372 373 374 375 376 377 378 379 380 381 382 |
/* * Potentially backup over partial log record write. * * In the typical case, last_blk is the number of the block directly after * a good log record. Therefore, we subtract one to get the block number * of the last block in the given buffer. extra_bblks contains the number * of blocks we would have read on a previous read. This happens when the * last log record is split over the end of the physical log. * * extra_bblks is the number of blocks potentially verified on a previous * call to this routine. */ STATIC int xlog_find_verify_log_record( |
9a8d2fdbb xfs: remove xlog_... |
383 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
384 385 386 387 388 |
xfs_daddr_t start_blk, xfs_daddr_t *last_blk, int extra_bblks) { xfs_daddr_t i; |
6e9b3dd80 xfs: stop using b... |
389 |
char *buffer; |
b2a922cd6 xfs: remove xfs_c... |
390 |
char *offset = NULL; |
1da177e4c Linux-2.6.12-rc2 |
391 392 393 394 395 396 397 |
xlog_rec_header_t *head = NULL; int error = 0; int smallmem = 0; int num_blks = *last_blk - start_blk; int xhdrs; ASSERT(start_blk != 0 || *last_blk != start_blk); |
6e9b3dd80 xfs: stop using b... |
398 399 400 401 |
buffer = xlog_alloc_buffer(log, num_blks); if (!buffer) { buffer = xlog_alloc_buffer(log, 1); if (!buffer) |
2451337dd xfs: global error... |
402 |
return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
403 404 |
smallmem = 1; } else { |
6e9b3dd80 xfs: stop using b... |
405 |
error = xlog_bread(log, start_blk, num_blks, buffer, &offset); |
076e6acb8 xfs: cleanup xlog... |
406 |
if (error) |
1da177e4c Linux-2.6.12-rc2 |
407 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
408 409 410 411 412 413 |
offset += ((num_blks - 1) << BBSHIFT); } for (i = (*last_blk) - 1; i >= 0; i--) { if (i < start_blk) { /* valid log record not found */ |
a0fa2b679 xfs: Convert xlog... |
414 415 |
xfs_warn(log->l_mp, "Log inconsistent (didn't find previous header)"); |
1da177e4c Linux-2.6.12-rc2 |
416 |
ASSERT(0); |
895e196fb xfs: convert EIO ... |
417 |
error = -EFSCORRUPTED; |
1da177e4c Linux-2.6.12-rc2 |
418 419 420 421 |
goto out; } if (smallmem) { |
6e9b3dd80 xfs: stop using b... |
422 |
error = xlog_bread(log, i, 1, buffer, &offset); |
076e6acb8 xfs: cleanup xlog... |
423 |
if (error) |
1da177e4c Linux-2.6.12-rc2 |
424 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
425 426 427 |
} head = (xlog_rec_header_t *)offset; |
69ef921b5 xfs: byteswap con... |
428 |
if (head->h_magicno == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) |
1da177e4c Linux-2.6.12-rc2 |
429 430 431 432 433 434 435 436 437 438 439 440 |
break; if (!smallmem) offset -= BBSIZE; } /* * We hit the beginning of the physical log & still no header. Return * to caller. If caller can handle a return of -1, then this routine * will be called again for the end of the physical log. */ if (i == -1) { |
2451337dd xfs: global error... |
441 |
error = 1; |
1da177e4c Linux-2.6.12-rc2 |
442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 |
goto out; } /* * We have the final block of the good log (the first block * of the log record _before_ the head. So we check the uuid. */ if ((error = xlog_header_check_mount(log->l_mp, head))) goto out; /* * We may have found a log record header before we expected one. * last_blk will be the 1st block # with a given cycle #. We may end * up reading an entire log record. In this case, we don't want to * reset last_blk. Only when last_blk points in the middle of a log * record do we update last_blk. */ |
0c771b99d xfs: clean up cal... |
459 |
xhdrs = xlog_logrec_hblks(log, head); |
1da177e4c Linux-2.6.12-rc2 |
460 |
|
b53e675dc [XFS] xlog_rec_he... |
461 462 |
if (*last_blk - i + extra_bblks != BTOBB(be32_to_cpu(head->h_len)) + xhdrs) |
1da177e4c Linux-2.6.12-rc2 |
463 464 465 |
*last_blk = i; out: |
6e9b3dd80 xfs: stop using b... |
466 |
kmem_free(buffer); |
1da177e4c Linux-2.6.12-rc2 |
467 468 469 470 471 |
return error; } /* * Head is defined to be the point of the log where the next log write |
0a94da24b xfs: fix the comm... |
472 |
* could go. This means that incomplete LR writes at the end are |
1da177e4c Linux-2.6.12-rc2 |
473 474 475 476 477 478 479 480 481 482 |
* eliminated when calculating the head. We aren't guaranteed that previous * LR have complete transactions. We only know that a cycle number of * current cycle number -1 won't be present in the log if we start writing * from our current block number. * * last_blk contains the block number of the first block with a given * cycle number. * * Return: zero if normal, non-zero if error. */ |
ba0f32d46 [XFS] mark variou... |
483 |
STATIC int |
1da177e4c Linux-2.6.12-rc2 |
484 |
xlog_find_head( |
9a8d2fdbb xfs: remove xlog_... |
485 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
486 487 |
xfs_daddr_t *return_head_blk) { |
6e9b3dd80 xfs: stop using b... |
488 |
char *buffer; |
b2a922cd6 xfs: remove xfs_c... |
489 |
char *offset; |
1da177e4c Linux-2.6.12-rc2 |
490 491 492 493 494 495 496 |
xfs_daddr_t new_blk, first_blk, start_blk, last_blk, head_blk; int num_scan_bblks; uint first_half_cycle, last_half_cycle; uint stop_on_cycle; int error, log_bbnum = log->l_logBBsize; /* Is the end of the log device zeroed? */ |
2451337dd xfs: global error... |
497 498 499 500 501 502 |
error = xlog_find_zeroed(log, &first_blk); if (error < 0) { xfs_warn(log->l_mp, "empty log check failed"); return error; } if (error == 1) { |
1da177e4c Linux-2.6.12-rc2 |
503 504 505 506 507 508 509 510 |
*return_head_blk = first_blk; /* Is the whole lot zeroed? */ if (!first_blk) { /* Linux XFS shouldn't generate totally zeroed logs - * mkfs etc write a dummy unmount record to a fresh * log so we can store the uuid in there */ |
a0fa2b679 xfs: Convert xlog... |
511 |
xfs_warn(log->l_mp, "totally zeroed log"); |
1da177e4c Linux-2.6.12-rc2 |
512 513 514 |
} return 0; |
1da177e4c Linux-2.6.12-rc2 |
515 516 517 |
} first_blk = 0; /* get cycle # of 1st block */ |
6e9b3dd80 xfs: stop using b... |
518 519 |
buffer = xlog_alloc_buffer(log, 1); if (!buffer) |
2451337dd xfs: global error... |
520 |
return -ENOMEM; |
076e6acb8 xfs: cleanup xlog... |
521 |
|
6e9b3dd80 xfs: stop using b... |
522 |
error = xlog_bread(log, 0, 1, buffer, &offset); |
076e6acb8 xfs: cleanup xlog... |
523 |
if (error) |
6e9b3dd80 xfs: stop using b... |
524 |
goto out_free_buffer; |
076e6acb8 xfs: cleanup xlog... |
525 |
|
03bea6fe6 [XFS] clean up so... |
526 |
first_half_cycle = xlog_get_cycle(offset); |
1da177e4c Linux-2.6.12-rc2 |
527 528 |
last_blk = head_blk = log_bbnum - 1; /* get cycle # of last block */ |
6e9b3dd80 xfs: stop using b... |
529 |
error = xlog_bread(log, last_blk, 1, buffer, &offset); |
076e6acb8 xfs: cleanup xlog... |
530 |
if (error) |
6e9b3dd80 xfs: stop using b... |
531 |
goto out_free_buffer; |
076e6acb8 xfs: cleanup xlog... |
532 |
|
03bea6fe6 [XFS] clean up so... |
533 |
last_half_cycle = xlog_get_cycle(offset); |
1da177e4c Linux-2.6.12-rc2 |
534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 |
ASSERT(last_half_cycle != 0); /* * If the 1st half cycle number is equal to the last half cycle number, * then the entire log is stamped with the same cycle number. In this * case, head_blk can't be set to zero (which makes sense). The below * math doesn't work out properly with head_blk equal to zero. Instead, * we set it to log_bbnum which is an invalid block number, but this * value makes the math correct. If head_blk doesn't changed through * all the tests below, *head_blk is set to zero at the very end rather * than log_bbnum. In a sense, log_bbnum and zero are the same block * in a circular file. */ if (first_half_cycle == last_half_cycle) { /* * In this case we believe that the entire log should have * cycle number last_half_cycle. We need to scan backwards * from the end verifying that there are no holes still * containing last_half_cycle - 1. If we find such a hole, * then the start of that hole will be the new head. The * simple case looks like * x | x ... | x - 1 | x * Another case that fits this picture would be * x | x + 1 | x ... | x |
c41564b5a [XFS] We really s... |
558 |
* In this case the head really is somewhere at the end of the |
1da177e4c Linux-2.6.12-rc2 |
559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 |
* log, as one of the latest writes at the beginning was * incomplete. * One more case is * x | x + 1 | x ... | x - 1 | x * This is really the combination of the above two cases, and * the head has to end up at the start of the x-1 hole at the * end of the log. * * In the 256k log case, we will read from the beginning to the * end of the log and search for cycle numbers equal to x-1. * We don't worry about the x+1 blocks that we encounter, * because we know that they cannot be the head since the log * started with x. */ head_blk = log_bbnum; stop_on_cycle = last_half_cycle - 1; } else { /* * In this case we want to find the first block with cycle * number matching last_half_cycle. We expect the log to be * some variation on |
3f943d853 xfs: minor odds a... |
580 |
* x + 1 ... | x ... | x |
1da177e4c Linux-2.6.12-rc2 |
581 582 583 584 585 586 587 588 589 |
* The first block with cycle number x (last_half_cycle) will * be where the new head belongs. First we do a binary search * for the first occurrence of last_half_cycle. The binary * search may not be totally accurate, so then we scan back * from there looking for occurrences of last_half_cycle before * us. If that backwards scan wraps around the beginning of * the log, then we look for occurrences of last_half_cycle - 1 * at the end of the log. The cases we're looking for look * like |
3f943d853 xfs: minor odds a... |
590 591 592 |
* v binary search stopped here * x + 1 ... | x | x + 1 | x ... | x * ^ but we want to locate this spot |
1da177e4c Linux-2.6.12-rc2 |
593 |
* or |
1da177e4c Linux-2.6.12-rc2 |
594 |
* <---------> less than scan distance |
3f943d853 xfs: minor odds a... |
595 596 |
* x + 1 ... | x ... | x - 1 | x * ^ we want to locate this spot |
1da177e4c Linux-2.6.12-rc2 |
597 598 |
*/ stop_on_cycle = last_half_cycle; |
6e9b3dd80 xfs: stop using b... |
599 600 601 602 |
error = xlog_find_cycle_start(log, buffer, first_blk, &head_blk, last_half_cycle); if (error) goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
603 604 605 606 607 608 609 610 611 |
} /* * Now validate the answer. Scan back some number of maximum possible * blocks and make sure each one has the expected cycle number. The * maximum is determined by the total possible amount of buffering * in the in-core log. The following number can be made tighter if * we actually look at the block size of the filesystem. */ |
9f2a45058 xfs: fix log bloc... |
612 |
num_scan_bblks = min_t(int, log_bbnum, XLOG_TOTAL_REC_SHIFT(log)); |
1da177e4c Linux-2.6.12-rc2 |
613 614 615 616 617 618 619 620 621 |
if (head_blk >= num_scan_bblks) { /* * We are guaranteed that the entire check can be performed * in one buffer. */ start_blk = head_blk - num_scan_bblks; if ((error = xlog_find_verify_cycle(log, start_blk, num_scan_bblks, stop_on_cycle, &new_blk))) |
6e9b3dd80 xfs: stop using b... |
622 |
goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 |
if (new_blk != -1) head_blk = new_blk; } else { /* need to read 2 parts of log */ /* * We are going to scan backwards in the log in two parts. * First we scan the physical end of the log. In this part * of the log, we are looking for blocks with cycle number * last_half_cycle - 1. * If we find one, then we know that the log starts there, as * we've found a hole that didn't get written in going around * the end of the physical log. The simple case for this is * x + 1 ... | x ... | x - 1 | x * <---------> less than scan distance * If all of the blocks at the end of the log have cycle number * last_half_cycle, then we check the blocks at the start of * the log looking for occurrences of last_half_cycle. If we * find one, then our current estimate for the location of the * first occurrence of last_half_cycle is wrong and we move * back to the hole we've found. This case looks like * x + 1 ... | x | x + 1 | x ... * ^ binary search stopped here * Another case we need to handle that only occurs in 256k * logs is * x + 1 ... | x ... | x+1 | x ... * ^ binary search stops here * In a 256k log, the scan at the end of the log will see the * x + 1 blocks. We need to skip past those since that is * certainly not the head of the log. By searching for * last_half_cycle-1 we accomplish that. */ |
1da177e4c Linux-2.6.12-rc2 |
653 |
ASSERT(head_blk <= INT_MAX && |
3f943d853 xfs: minor odds a... |
654 655 |
(xfs_daddr_t) num_scan_bblks >= head_blk); start_blk = log_bbnum - (num_scan_bblks - head_blk); |
1da177e4c Linux-2.6.12-rc2 |
656 657 658 |
if ((error = xlog_find_verify_cycle(log, start_blk, num_scan_bblks - (int)head_blk, (stop_on_cycle - 1), &new_blk))) |
6e9b3dd80 xfs: stop using b... |
659 |
goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
660 661 |
if (new_blk != -1) { head_blk = new_blk; |
9db127edb xfs: change a few... |
662 |
goto validate_head; |
1da177e4c Linux-2.6.12-rc2 |
663 664 665 666 667 668 669 670 671 672 673 674 |
} /* * Scan beginning of log now. The last part of the physical * log is good. This scan needs to verify that it doesn't find * the last_half_cycle. */ start_blk = 0; ASSERT(head_blk <= INT_MAX); if ((error = xlog_find_verify_cycle(log, start_blk, (int)head_blk, stop_on_cycle, &new_blk))) |
6e9b3dd80 xfs: stop using b... |
675 |
goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
676 677 678 |
if (new_blk != -1) head_blk = new_blk; } |
9db127edb xfs: change a few... |
679 |
validate_head: |
1da177e4c Linux-2.6.12-rc2 |
680 681 682 683 684 685 686 687 688 |
/* * Now we need to make sure head_blk is not pointing to a block in * the middle of a log record. */ num_scan_bblks = XLOG_REC_SHIFT(log); if (head_blk >= num_scan_bblks) { start_blk = head_blk - num_scan_bblks; /* don't read head_blk */ /* start ptr at last block ptr before head_blk */ |
2451337dd xfs: global error... |
689 690 691 692 |
error = xlog_find_verify_log_record(log, start_blk, &head_blk, 0); if (error == 1) error = -EIO; if (error) |
6e9b3dd80 xfs: stop using b... |
693 |
goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
694 695 696 |
} else { start_blk = 0; ASSERT(head_blk <= INT_MAX); |
2451337dd xfs: global error... |
697 698 |
error = xlog_find_verify_log_record(log, start_blk, &head_blk, 0); if (error < 0) |
6e9b3dd80 xfs: stop using b... |
699 |
goto out_free_buffer; |
2451337dd xfs: global error... |
700 |
if (error == 1) { |
1da177e4c Linux-2.6.12-rc2 |
701 |
/* We hit the beginning of the log during our search */ |
3f943d853 xfs: minor odds a... |
702 |
start_blk = log_bbnum - (num_scan_bblks - head_blk); |
1da177e4c Linux-2.6.12-rc2 |
703 704 705 706 |
new_blk = log_bbnum; ASSERT(start_blk <= INT_MAX && (xfs_daddr_t) log_bbnum-start_blk >= 0); ASSERT(head_blk <= INT_MAX); |
2451337dd xfs: global error... |
707 708 709 710 711 |
error = xlog_find_verify_log_record(log, start_blk, &new_blk, (int)head_blk); if (error == 1) error = -EIO; if (error) |
6e9b3dd80 xfs: stop using b... |
712 |
goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
713 714 715 |
if (new_blk != log_bbnum) head_blk = new_blk; } else if (error) |
6e9b3dd80 xfs: stop using b... |
716 |
goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
717 |
} |
6e9b3dd80 xfs: stop using b... |
718 |
kmem_free(buffer); |
1da177e4c Linux-2.6.12-rc2 |
719 720 721 722 723 724 725 726 727 728 729 |
if (head_blk == log_bbnum) *return_head_blk = 0; else *return_head_blk = head_blk; /* * When returning here, we have a good block number. Bad block * means that during a previous crash, we didn't have a clean break * from cycle number N to cycle number N-1. In this case, we need * to find the first block with cycle number N-1. */ return 0; |
6e9b3dd80 xfs: stop using b... |
730 731 |
out_free_buffer: kmem_free(buffer); |
1da177e4c Linux-2.6.12-rc2 |
732 |
if (error) |
a0fa2b679 xfs: Convert xlog... |
733 |
xfs_warn(log->l_mp, "failed to find log head"); |
1da177e4c Linux-2.6.12-rc2 |
734 735 736 737 |
return error; } /* |
eed6b462f xfs: refactor log... |
738 739 740 741 742 743 744 745 746 747 748 749 750 |
* Seek backwards in the log for log record headers. * * Given a starting log block, walk backwards until we find the provided number * of records or hit the provided tail block. The return value is the number of * records encountered or a negative error code. The log block and buffer * pointer of the last record seen are returned in rblk and rhead respectively. */ STATIC int xlog_rseek_logrec_hdr( struct xlog *log, xfs_daddr_t head_blk, xfs_daddr_t tail_blk, int count, |
6e9b3dd80 xfs: stop using b... |
751 |
char *buffer, |
eed6b462f xfs: refactor log... |
752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 |
xfs_daddr_t *rblk, struct xlog_rec_header **rhead, bool *wrapped) { int i; int error; int found = 0; char *offset = NULL; xfs_daddr_t end_blk; *wrapped = false; /* * Walk backwards from the head block until we hit the tail or the first * block in the log. */ end_blk = head_blk > tail_blk ? tail_blk : 0; for (i = (int) head_blk - 1; i >= end_blk; i--) { |
6e9b3dd80 xfs: stop using b... |
770 |
error = xlog_bread(log, i, 1, buffer, &offset); |
eed6b462f xfs: refactor log... |
771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 |
if (error) goto out_error; if (*(__be32 *) offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) { *rblk = i; *rhead = (struct xlog_rec_header *) offset; if (++found == count) break; } } /* * If we haven't hit the tail block or the log record header count, * start looking again from the end of the physical log. Note that * callers can pass head == tail if the tail is not yet known. */ if (tail_blk >= head_blk && found != count) { for (i = log->l_logBBsize - 1; i >= (int) tail_blk; i--) { |
6e9b3dd80 xfs: stop using b... |
789 |
error = xlog_bread(log, i, 1, buffer, &offset); |
eed6b462f xfs: refactor log... |
790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 |
if (error) goto out_error; if (*(__be32 *)offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) { *wrapped = true; *rblk = i; *rhead = (struct xlog_rec_header *) offset; if (++found == count) break; } } } return found; out_error: return error; } /* |
7088c4136 xfs: detect and t... |
811 812 813 814 815 816 817 818 819 820 821 822 823 824 |
* Seek forward in the log for log record headers. * * Given head and tail blocks, walk forward from the tail block until we find * the provided number of records or hit the head block. The return value is the * number of records encountered or a negative error code. The log block and * buffer pointer of the last record seen are returned in rblk and rhead * respectively. */ STATIC int xlog_seek_logrec_hdr( struct xlog *log, xfs_daddr_t head_blk, xfs_daddr_t tail_blk, int count, |
6e9b3dd80 xfs: stop using b... |
825 |
char *buffer, |
7088c4136 xfs: detect and t... |
826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 |
xfs_daddr_t *rblk, struct xlog_rec_header **rhead, bool *wrapped) { int i; int error; int found = 0; char *offset = NULL; xfs_daddr_t end_blk; *wrapped = false; /* * Walk forward from the tail block until we hit the head or the last * block in the log. */ end_blk = head_blk > tail_blk ? head_blk : log->l_logBBsize - 1; for (i = (int) tail_blk; i <= end_blk; i++) { |
6e9b3dd80 xfs: stop using b... |
844 |
error = xlog_bread(log, i, 1, buffer, &offset); |
7088c4136 xfs: detect and t... |
845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 |
if (error) goto out_error; if (*(__be32 *) offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) { *rblk = i; *rhead = (struct xlog_rec_header *) offset; if (++found == count) break; } } /* * If we haven't hit the head block or the log record header count, * start looking again from the start of the physical log. */ if (tail_blk > head_blk && found != count) { for (i = 0; i < (int) head_blk; i++) { |
6e9b3dd80 xfs: stop using b... |
862 |
error = xlog_bread(log, i, 1, buffer, &offset); |
7088c4136 xfs: detect and t... |
863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 |
if (error) goto out_error; if (*(__be32 *)offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) { *wrapped = true; *rblk = i; *rhead = (struct xlog_rec_header *) offset; if (++found == count) break; } } } return found; out_error: return error; } /* |
4a4f66eac xfs: fix log reco... |
884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 |
* Calculate distance from head to tail (i.e., unused space in the log). */ static inline int xlog_tail_distance( struct xlog *log, xfs_daddr_t head_blk, xfs_daddr_t tail_blk) { if (head_blk < tail_blk) return tail_blk - head_blk; return tail_blk + (log->l_logBBsize - head_blk); } /* * Verify the log tail. This is particularly important when torn or incomplete * writes have been detected near the front of the log and the head has been * walked back accordingly. * * We also have to handle the case where the tail was pinned and the head * blocked behind the tail right before a crash. If the tail had been pushed * immediately prior to the crash and the subsequent checkpoint was only * partially written, it's possible it overwrote the last referenced tail in the * log with garbage. This is not a coherency problem because the tail must have * been pushed before it can be overwritten, but appears as log corruption to * recovery because we have no way to know the tail was updated if the * subsequent checkpoint didn't write successfully. |
7088c4136 xfs: detect and t... |
911 |
* |
4a4f66eac xfs: fix log reco... |
912 913 914 915 |
* Therefore, CRC check the log from tail to head. If a failure occurs and the * offending record is within max iclog bufs from the head, walk the tail * forward and retry until a valid tail is found or corruption is detected out * of the range of a possible overwrite. |
7088c4136 xfs: detect and t... |
916 917 918 919 920 |
*/ STATIC int xlog_verify_tail( struct xlog *log, xfs_daddr_t head_blk, |
4a4f66eac xfs: fix log reco... |
921 922 |
xfs_daddr_t *tail_blk, int hsize) |
7088c4136 xfs: detect and t... |
923 924 |
{ struct xlog_rec_header *thead; |
6e9b3dd80 xfs: stop using b... |
925 |
char *buffer; |
7088c4136 xfs: detect and t... |
926 |
xfs_daddr_t first_bad; |
7088c4136 xfs: detect and t... |
927 928 |
int error = 0; bool wrapped; |
4a4f66eac xfs: fix log reco... |
929 930 |
xfs_daddr_t tmp_tail; xfs_daddr_t orig_tail = *tail_blk; |
7088c4136 xfs: detect and t... |
931 |
|
6e9b3dd80 xfs: stop using b... |
932 933 |
buffer = xlog_alloc_buffer(log, 1); if (!buffer) |
7088c4136 xfs: detect and t... |
934 935 936 |
return -ENOMEM; /* |
4a4f66eac xfs: fix log reco... |
937 938 |
* Make sure the tail points to a record (returns positive count on * success). |
7088c4136 xfs: detect and t... |
939 |
*/ |
6e9b3dd80 xfs: stop using b... |
940 |
error = xlog_seek_logrec_hdr(log, head_blk, *tail_blk, 1, buffer, |
4a4f66eac xfs: fix log reco... |
941 942 |
&tmp_tail, &thead, &wrapped); if (error < 0) |
7088c4136 xfs: detect and t... |
943 |
goto out; |
4a4f66eac xfs: fix log reco... |
944 945 |
if (*tail_blk != tmp_tail) *tail_blk = tmp_tail; |
7088c4136 xfs: detect and t... |
946 947 |
/* |
4a4f66eac xfs: fix log reco... |
948 949 950 951 952 |
* Run a CRC check from the tail to the head. We can't just check * MAX_ICLOGS records past the tail because the tail may point to stale * blocks cleared during the search for the head/tail. These blocks are * overwritten with zero-length records and thus record count is not a * reliable indicator of the iclog state before a crash. |
7088c4136 xfs: detect and t... |
953 |
*/ |
4a4f66eac xfs: fix log reco... |
954 955 |
first_bad = 0; error = xlog_do_recovery_pass(log, head_blk, *tail_blk, |
7088c4136 xfs: detect and t... |
956 |
XLOG_RECOVER_CRCPASS, &first_bad); |
a4c9b34d6 xfs: handle -EFSC... |
957 |
while ((error == -EFSBADCRC || error == -EFSCORRUPTED) && first_bad) { |
4a4f66eac xfs: fix log reco... |
958 959 960 961 962 963 964 965 966 |
int tail_distance; /* * Is corruption within range of the head? If so, retry from * the next record. Otherwise return an error. */ tail_distance = xlog_tail_distance(log, head_blk, first_bad); if (tail_distance > BTOBB(XLOG_MAX_ICLOGS * hsize)) break; |
7088c4136 xfs: detect and t... |
967 |
|
4a4f66eac xfs: fix log reco... |
968 |
/* skip to the next record; returns positive count on success */ |
6e9b3dd80 xfs: stop using b... |
969 970 |
error = xlog_seek_logrec_hdr(log, head_blk, first_bad, 2, buffer, &tmp_tail, &thead, &wrapped); |
4a4f66eac xfs: fix log reco... |
971 972 973 974 975 976 977 978 979 980 981 982 983 |
if (error < 0) goto out; *tail_blk = tmp_tail; first_bad = 0; error = xlog_do_recovery_pass(log, head_blk, *tail_blk, XLOG_RECOVER_CRCPASS, &first_bad); } if (!error && *tail_blk != orig_tail) xfs_warn(log->l_mp, "Tail block (0x%llx) overwrite detected. Updated to 0x%llx", orig_tail, *tail_blk); |
7088c4136 xfs: detect and t... |
984 |
out: |
6e9b3dd80 xfs: stop using b... |
985 |
kmem_free(buffer); |
7088c4136 xfs: detect and t... |
986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 |
return error; } /* * Detect and trim torn writes from the head of the log. * * Storage without sector atomicity guarantees can result in torn writes in the * log in the event of a crash. Our only means to detect this scenario is via * CRC verification. While we can't always be certain that CRC verification * failure is due to a torn write vs. an unrelated corruption, we do know that * only a certain number (XLOG_MAX_ICLOGS) of log records can be written out at * one time. Therefore, CRC verify up to XLOG_MAX_ICLOGS records at the head of * the log and treat failures in this range as torn writes as a matter of * policy. In the event of CRC failure, the head is walked back to the last good * record in the log and the tail is updated from that record and verified. */ STATIC int xlog_verify_head( struct xlog *log, xfs_daddr_t *head_blk, /* in/out: unverified head */ xfs_daddr_t *tail_blk, /* out: tail block */ |
6e9b3dd80 xfs: stop using b... |
1007 |
char *buffer, |
7088c4136 xfs: detect and t... |
1008 1009 1010 1011 1012 |
xfs_daddr_t *rhead_blk, /* start blk of last record */ struct xlog_rec_header **rhead, /* ptr to last record */ bool *wrapped) /* last rec. wraps phys. log */ { struct xlog_rec_header *tmp_rhead; |
6e9b3dd80 xfs: stop using b... |
1013 |
char *tmp_buffer; |
7088c4136 xfs: detect and t... |
1014 1015 1016 1017 1018 1019 1020 |
xfs_daddr_t first_bad; xfs_daddr_t tmp_rhead_blk; int found; int error; bool tmp_wrapped; /* |
82ff6cc26 xfs: separate log... |
1021 1022 1023 |
* Check the head of the log for torn writes. Search backwards from the * head until we hit the tail or the maximum number of log record I/Os * that could have been in flight at one time. Use a temporary buffer so |
6e9b3dd80 xfs: stop using b... |
1024 |
* we don't trash the rhead/buffer pointers from the caller. |
7088c4136 xfs: detect and t... |
1025 |
*/ |
6e9b3dd80 xfs: stop using b... |
1026 1027 |
tmp_buffer = xlog_alloc_buffer(log, 1); if (!tmp_buffer) |
7088c4136 xfs: detect and t... |
1028 1029 |
return -ENOMEM; error = xlog_rseek_logrec_hdr(log, *head_blk, *tail_blk, |
6e9b3dd80 xfs: stop using b... |
1030 1031 1032 |
XLOG_MAX_ICLOGS, tmp_buffer, &tmp_rhead_blk, &tmp_rhead, &tmp_wrapped); kmem_free(tmp_buffer); |
7088c4136 xfs: detect and t... |
1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 |
if (error < 0) return error; /* * Now run a CRC verification pass over the records starting at the * block found above to the current head. If a CRC failure occurs, the * log block of the first bad record is saved in first_bad. */ error = xlog_do_recovery_pass(log, *head_blk, tmp_rhead_blk, XLOG_RECOVER_CRCPASS, &first_bad); |
a4c9b34d6 xfs: handle -EFSC... |
1043 |
if ((error == -EFSBADCRC || error == -EFSCORRUPTED) && first_bad) { |
7088c4136 xfs: detect and t... |
1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 |
/* * We've hit a potential torn write. Reset the error and warn * about it. */ error = 0; xfs_warn(log->l_mp, "Torn write (CRC failure) detected at log block 0x%llx. Truncating head block from 0x%llx.", first_bad, *head_blk); /* * Get the header block and buffer pointer for the last good * record before the bad record. * * Note that xlog_find_tail() clears the blocks at the new head * (i.e., the records with invalid CRC) if the cycle number |
b63da6c8d xfs: delete dupli... |
1059 |
* matches the current cycle. |
7088c4136 xfs: detect and t... |
1060 |
*/ |
6e9b3dd80 xfs: stop using b... |
1061 1062 |
found = xlog_rseek_logrec_hdr(log, first_bad, *tail_blk, 1, buffer, rhead_blk, rhead, wrapped); |
7088c4136 xfs: detect and t... |
1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 |
if (found < 0) return found; if (found == 0) /* XXX: right thing to do here? */ return -EIO; /* * Reset the head block to the starting block of the first bad * log record and set the tail block based on the last good * record. * * Bail out if the updated head/tail match as this indicates * possible corruption outside of the acceptable * (XLOG_MAX_ICLOGS) range. This is a job for xfs_repair... */ *head_blk = first_bad; *tail_blk = BLOCK_LSN(be64_to_cpu((*rhead)->h_tail_lsn)); if (*head_blk == *tail_blk) { ASSERT(0); return 0; } |
7088c4136 xfs: detect and t... |
1083 |
} |
5297ac1f6 xfs: always verif... |
1084 1085 |
if (error) return error; |
7088c4136 xfs: detect and t... |
1086 |
|
4a4f66eac xfs: fix log reco... |
1087 1088 |
return xlog_verify_tail(log, *head_blk, tail_blk, be32_to_cpu((*rhead)->h_size)); |
7088c4136 xfs: detect and t... |
1089 1090 1091 |
} /* |
0703a8e1c xfs: replace do_m... |
1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 |
* We need to make sure we handle log wrapping properly, so we can't use the * calculated logbno directly. Make sure it wraps to the correct bno inside the * log. * * The log is limited to 32 bit sizes, so we use the appropriate modulus * operation here and cast it back to a 64 bit daddr on return. */ static inline xfs_daddr_t xlog_wrap_logbno( struct xlog *log, xfs_daddr_t bno) { int mod; div_s64_rem(bno, log->l_logBBsize, &mod); return mod; } /* |
65b99a08b xfs: refactor unm... |
1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 |
* Check whether the head of the log points to an unmount record. In other * words, determine whether the log is clean. If so, update the in-core state * appropriately. */ static int xlog_check_unmount_rec( struct xlog *log, xfs_daddr_t *head_blk, xfs_daddr_t *tail_blk, struct xlog_rec_header *rhead, xfs_daddr_t rhead_blk, |
6e9b3dd80 xfs: stop using b... |
1122 |
char *buffer, |
65b99a08b xfs: refactor unm... |
1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 |
bool *clean) { struct xlog_op_header *op_head; xfs_daddr_t umount_data_blk; xfs_daddr_t after_umount_blk; int hblks; int error; char *offset; *clean = false; /* * Look for unmount record. If we find it, then we know there was a * clean unmount. Since 'i' could be the last block in the physical * log, we convert to a log block before comparing to the head_blk. * * Save the current tail lsn to use to pass to xlog_clear_stale_blocks() * below. We won't want to clear the unmount record if there is one, so * we pass the lsn of the unmount record rather than the block after it. */ |
0c771b99d xfs: clean up cal... |
1143 |
hblks = xlog_logrec_hblks(log, rhead); |
0703a8e1c xfs: replace do_m... |
1144 1145 |
after_umount_blk = xlog_wrap_logbno(log, rhead_blk + hblks + BTOBB(be32_to_cpu(rhead->h_len))); |
65b99a08b xfs: refactor unm... |
1146 1147 |
if (*head_blk == after_umount_blk && be32_to_cpu(rhead->h_num_logops) == 1) { |
0703a8e1c xfs: replace do_m... |
1148 |
umount_data_blk = xlog_wrap_logbno(log, rhead_blk + hblks); |
6e9b3dd80 xfs: stop using b... |
1149 |
error = xlog_bread(log, umount_data_blk, 1, buffer, &offset); |
65b99a08b xfs: refactor unm... |
1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 |
if (error) return error; op_head = (struct xlog_op_header *)offset; if (op_head->oh_flags & XLOG_UNMOUNT_TRANS) { /* * Set tail and last sync so that newly written log * records will point recovery to after the current * unmount record. */ xlog_assign_atomic_lsn(&log->l_tail_lsn, log->l_curr_cycle, after_umount_blk); xlog_assign_atomic_lsn(&log->l_last_sync_lsn, log->l_curr_cycle, after_umount_blk); *tail_blk = after_umount_blk; *clean = true; } } return 0; } |
717bc0ebc xfs: refactor in-... |
1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 |
static void xlog_set_state( struct xlog *log, xfs_daddr_t head_blk, struct xlog_rec_header *rhead, xfs_daddr_t rhead_blk, bool bump_cycle) { /* * Reset log values according to the state of the log when we * crashed. In the case where head_blk == 0, we bump curr_cycle * one because the next write starts a new cycle rather than * continuing the cycle of the last good log record. At this * point we have guaranteed that all partial log records have been * accounted for. Therefore, we know that the last good log record * written was complete and ended exactly on the end boundary * of the physical log. */ log->l_prev_block = rhead_blk; log->l_curr_block = (int)head_blk; log->l_curr_cycle = be32_to_cpu(rhead->h_cycle); if (bump_cycle) log->l_curr_cycle++; atomic64_set(&log->l_tail_lsn, be64_to_cpu(rhead->h_tail_lsn)); atomic64_set(&log->l_last_sync_lsn, be64_to_cpu(rhead->h_lsn)); xlog_assign_grant_head(&log->l_reserve_head.grant, log->l_curr_cycle, BBTOB(log->l_curr_block)); xlog_assign_grant_head(&log->l_write_head.grant, log->l_curr_cycle, BBTOB(log->l_curr_block)); } |
65b99a08b xfs: refactor unm... |
1202 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 |
* Find the sync block number or the tail of the log. * * This will be the block number of the last record to have its * associated buffers synced to disk. Every log record header has * a sync lsn embedded in it. LSNs hold block numbers, so it is easy * to get a sync block number. The only concern is to figure out which * log record header to believe. * * The following algorithm uses the log record header with the largest * lsn. The entire log record does not need to be valid. We only care * that the header is valid. * * We could speed up search by using current head_blk buffer, but it is not * available. */ |
5d77c0dc0 xfs: make several... |
1218 |
STATIC int |
1da177e4c Linux-2.6.12-rc2 |
1219 |
xlog_find_tail( |
9a8d2fdbb xfs: remove xlog_... |
1220 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
1221 |
xfs_daddr_t *head_blk, |
65be60541 [XFS] remove unus... |
1222 |
xfs_daddr_t *tail_blk) |
1da177e4c Linux-2.6.12-rc2 |
1223 1224 |
{ xlog_rec_header_t *rhead; |
b2a922cd6 xfs: remove xfs_c... |
1225 |
char *offset = NULL; |
6e9b3dd80 xfs: stop using b... |
1226 |
char *buffer; |
7088c4136 xfs: detect and t... |
1227 |
int error; |
7088c4136 xfs: detect and t... |
1228 |
xfs_daddr_t rhead_blk; |
1da177e4c Linux-2.6.12-rc2 |
1229 |
xfs_lsn_t tail_lsn; |
eed6b462f xfs: refactor log... |
1230 |
bool wrapped = false; |
65b99a08b xfs: refactor unm... |
1231 |
bool clean = false; |
1da177e4c Linux-2.6.12-rc2 |
1232 1233 1234 1235 1236 1237 |
/* * Find previous log record */ if ((error = xlog_find_head(log, head_blk))) return error; |
82ff6cc26 xfs: separate log... |
1238 |
ASSERT(*head_blk < INT_MAX); |
1da177e4c Linux-2.6.12-rc2 |
1239 |
|
6e9b3dd80 xfs: stop using b... |
1240 1241 |
buffer = xlog_alloc_buffer(log, 1); if (!buffer) |
2451337dd xfs: global error... |
1242 |
return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
1243 |
if (*head_blk == 0) { /* special case */ |
6e9b3dd80 xfs: stop using b... |
1244 |
error = xlog_bread(log, 0, 1, buffer, &offset); |
076e6acb8 xfs: cleanup xlog... |
1245 |
if (error) |
9db127edb xfs: change a few... |
1246 |
goto done; |
076e6acb8 xfs: cleanup xlog... |
1247 |
|
03bea6fe6 [XFS] clean up so... |
1248 |
if (xlog_get_cycle(offset) == 0) { |
1da177e4c Linux-2.6.12-rc2 |
1249 1250 |
*tail_blk = 0; /* leave all other log inited values alone */ |
9db127edb xfs: change a few... |
1251 |
goto done; |
1da177e4c Linux-2.6.12-rc2 |
1252 1253 1254 1255 |
} } /* |
82ff6cc26 xfs: separate log... |
1256 1257 1258 |
* Search backwards through the log looking for the log record header * block. This wraps all the way back around to the head so something is * seriously wrong if we can't find it. |
1da177e4c Linux-2.6.12-rc2 |
1259 |
*/ |
6e9b3dd80 xfs: stop using b... |
1260 |
error = xlog_rseek_logrec_hdr(log, *head_blk, *head_blk, 1, buffer, |
82ff6cc26 xfs: separate log... |
1261 1262 |
&rhead_blk, &rhead, &wrapped); if (error < 0) |
050552cbe xfs: fix some mem... |
1263 |
goto done; |
82ff6cc26 xfs: separate log... |
1264 1265 |
if (!error) { xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__); |
050552cbe xfs: fix some mem... |
1266 1267 |
error = -EFSCORRUPTED; goto done; |
82ff6cc26 xfs: separate log... |
1268 1269 |
} *tail_blk = BLOCK_LSN(be64_to_cpu(rhead->h_tail_lsn)); |
1da177e4c Linux-2.6.12-rc2 |
1270 1271 |
/* |
717bc0ebc xfs: refactor in-... |
1272 |
* Set the log state based on the current head record. |
1da177e4c Linux-2.6.12-rc2 |
1273 |
*/ |
717bc0ebc xfs: refactor in-... |
1274 |
xlog_set_state(log, *head_blk, rhead, rhead_blk, wrapped); |
65b99a08b xfs: refactor unm... |
1275 |
tail_lsn = atomic64_read(&log->l_tail_lsn); |
1da177e4c Linux-2.6.12-rc2 |
1276 1277 |
/* |
65b99a08b xfs: refactor unm... |
1278 1279 |
* Look for an unmount record at the head of the log. This sets the log * state to determine whether recovery is necessary. |
1da177e4c Linux-2.6.12-rc2 |
1280 |
*/ |
65b99a08b xfs: refactor unm... |
1281 |
error = xlog_check_unmount_rec(log, head_blk, tail_blk, rhead, |
6e9b3dd80 xfs: stop using b... |
1282 |
rhead_blk, buffer, &clean); |
65b99a08b xfs: refactor unm... |
1283 1284 |
if (error) goto done; |
1da177e4c Linux-2.6.12-rc2 |
1285 1286 |
/* |
7f6aff3a2 xfs: only run tor... |
1287 1288 1289 1290 |
* Verify the log head if the log is not clean (e.g., we have anything * but an unmount record at the head). This uses CRC verification to * detect and trim torn writes. If discovered, CRC failures are * considered torn writes and the log head is trimmed accordingly. |
1da177e4c Linux-2.6.12-rc2 |
1291 |
* |
7f6aff3a2 xfs: only run tor... |
1292 1293 1294 |
* Note that we can only run CRC verification when the log is dirty * because there's no guarantee that the log data behind an unmount * record is compatible with the current architecture. |
1da177e4c Linux-2.6.12-rc2 |
1295 |
*/ |
7f6aff3a2 xfs: only run tor... |
1296 1297 |
if (!clean) { xfs_daddr_t orig_head = *head_blk; |
1da177e4c Linux-2.6.12-rc2 |
1298 |
|
6e9b3dd80 xfs: stop using b... |
1299 |
error = xlog_verify_head(log, head_blk, tail_blk, buffer, |
7f6aff3a2 xfs: only run tor... |
1300 |
&rhead_blk, &rhead, &wrapped); |
076e6acb8 xfs: cleanup xlog... |
1301 |
if (error) |
9db127edb xfs: change a few... |
1302 |
goto done; |
076e6acb8 xfs: cleanup xlog... |
1303 |
|
7f6aff3a2 xfs: only run tor... |
1304 1305 1306 1307 1308 1309 |
/* update in-core state again if the head changed */ if (*head_blk != orig_head) { xlog_set_state(log, *head_blk, rhead, rhead_blk, wrapped); tail_lsn = atomic64_read(&log->l_tail_lsn); error = xlog_check_unmount_rec(log, head_blk, tail_blk, |
6e9b3dd80 xfs: stop using b... |
1310 |
rhead, rhead_blk, buffer, |
7f6aff3a2 xfs: only run tor... |
1311 1312 1313 |
&clean); if (error) goto done; |
1da177e4c Linux-2.6.12-rc2 |
1314 1315 1316 1317 |
} } /* |
65b99a08b xfs: refactor unm... |
1318 1319 1320 1321 1322 1323 |
* Note that the unmount was clean. If the unmount was not clean, we * need to know this to rebuild the superblock counters from the perag * headers if we have a filesystem using non-persistent counters. */ if (clean) log->l_mp->m_flags |= XFS_MOUNT_WAS_CLEAN; |
1da177e4c Linux-2.6.12-rc2 |
1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 |
/* * Make sure that there are no blocks in front of the head * with the same cycle number as the head. This can happen * because we allow multiple outstanding log writes concurrently, * and the later writes might make it out before earlier ones. * * We use the lsn from before modifying it so that we'll never * overwrite the unmount record after a clean unmount. * * Do this only if we are going to recover the filesystem * * NOTE: This used to say "if (!readonly)" * However on Linux, we can & do recover a read-only filesystem. * We only skip recovery if NORECOVERY is specified on mount, * in which case we would not be here. * * But... if the -device- itself is readonly, just skip this. * We can't recover this device anyway, so it won't matter. */ |
2d15d2c0e xfs: make use of ... |
1344 |
if (!xfs_readonly_buftarg(log->l_targ)) |
1da177e4c Linux-2.6.12-rc2 |
1345 |
error = xlog_clear_stale_blocks(log, tail_lsn); |
1da177e4c Linux-2.6.12-rc2 |
1346 |
|
9db127edb xfs: change a few... |
1347 |
done: |
6e9b3dd80 xfs: stop using b... |
1348 |
kmem_free(buffer); |
1da177e4c Linux-2.6.12-rc2 |
1349 1350 |
if (error) |
a0fa2b679 xfs: Convert xlog... |
1351 |
xfs_warn(log->l_mp, "failed to locate log tail"); |
1da177e4c Linux-2.6.12-rc2 |
1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 |
return error; } /* * Is the log zeroed at all? * * The last binary search should be changed to perform an X block read * once X becomes small enough. You can then search linearly through * the X blocks. This will cut down on the number of reads we need to do. * * If the log is partially zeroed, this routine will pass back the blkno * of the first block with cycle number 0. It won't have a complete LR * preceding it. * * Return: * 0 => the log is completely written to |
2451337dd xfs: global error... |
1368 1369 |
* 1 => use *blk_no as the first block of the log * <0 => error has occurred |
1da177e4c Linux-2.6.12-rc2 |
1370 |
*/ |
a8272ce0c [XFS] Fix up spar... |
1371 |
STATIC int |
1da177e4c Linux-2.6.12-rc2 |
1372 |
xlog_find_zeroed( |
9a8d2fdbb xfs: remove xlog_... |
1373 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
1374 1375 |
xfs_daddr_t *blk_no) { |
6e9b3dd80 xfs: stop using b... |
1376 |
char *buffer; |
b2a922cd6 xfs: remove xfs_c... |
1377 |
char *offset; |
1da177e4c Linux-2.6.12-rc2 |
1378 1379 1380 1381 |
uint first_cycle, last_cycle; xfs_daddr_t new_blk, last_blk, start_blk; xfs_daddr_t num_scan_bblks; int error, log_bbnum = log->l_logBBsize; |
6fdf8ccc0 [XFS] Rework code... |
1382 |
*blk_no = 0; |
1da177e4c Linux-2.6.12-rc2 |
1383 |
/* check totally zeroed log */ |
6e9b3dd80 xfs: stop using b... |
1384 1385 |
buffer = xlog_alloc_buffer(log, 1); if (!buffer) |
2451337dd xfs: global error... |
1386 |
return -ENOMEM; |
6e9b3dd80 xfs: stop using b... |
1387 |
error = xlog_bread(log, 0, 1, buffer, &offset); |
076e6acb8 xfs: cleanup xlog... |
1388 |
if (error) |
6e9b3dd80 xfs: stop using b... |
1389 |
goto out_free_buffer; |
076e6acb8 xfs: cleanup xlog... |
1390 |
|
03bea6fe6 [XFS] clean up so... |
1391 |
first_cycle = xlog_get_cycle(offset); |
1da177e4c Linux-2.6.12-rc2 |
1392 1393 |
if (first_cycle == 0) { /* completely zeroed log */ *blk_no = 0; |
6e9b3dd80 xfs: stop using b... |
1394 |
kmem_free(buffer); |
2451337dd xfs: global error... |
1395 |
return 1; |
1da177e4c Linux-2.6.12-rc2 |
1396 1397 1398 |
} /* check partially zeroed log */ |
6e9b3dd80 xfs: stop using b... |
1399 |
error = xlog_bread(log, log_bbnum-1, 1, buffer, &offset); |
076e6acb8 xfs: cleanup xlog... |
1400 |
if (error) |
6e9b3dd80 xfs: stop using b... |
1401 |
goto out_free_buffer; |
076e6acb8 xfs: cleanup xlog... |
1402 |
|
03bea6fe6 [XFS] clean up so... |
1403 |
last_cycle = xlog_get_cycle(offset); |
1da177e4c Linux-2.6.12-rc2 |
1404 |
if (last_cycle != 0) { /* log completely written to */ |
6e9b3dd80 xfs: stop using b... |
1405 |
kmem_free(buffer); |
1da177e4c Linux-2.6.12-rc2 |
1406 |
return 0; |
1da177e4c Linux-2.6.12-rc2 |
1407 1408 1409 1410 |
} /* we have a partially zeroed log */ last_blk = log_bbnum-1; |
6e9b3dd80 xfs: stop using b... |
1411 1412 1413 |
error = xlog_find_cycle_start(log, buffer, 0, &last_blk, 0); if (error) goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 |
/* * Validate the answer. Because there is no way to guarantee that * the entire log is made up of log records which are the same size, * we scan over the defined maximum blocks. At this point, the maximum * is not chosen to mean anything special. XXXmiken */ num_scan_bblks = XLOG_TOTAL_REC_SHIFT(log); ASSERT(num_scan_bblks <= INT_MAX); if (last_blk < num_scan_bblks) num_scan_bblks = last_blk; start_blk = last_blk - num_scan_bblks; /* * We search for any instances of cycle number 0 that occur before * our current estimate of the head. What we're trying to detect is * 1 ... | 0 | 1 | 0... * ^ binary search ends here */ if ((error = xlog_find_verify_cycle(log, start_blk, (int)num_scan_bblks, 0, &new_blk))) |
6e9b3dd80 xfs: stop using b... |
1436 |
goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
1437 1438 1439 1440 1441 1442 1443 |
if (new_blk != -1) last_blk = new_blk; /* * Potentially backup over partial log record write. We don't need * to search the end of the log because we know it is zero. */ |
2451337dd xfs: global error... |
1444 1445 1446 1447 |
error = xlog_find_verify_log_record(log, start_blk, &last_blk, 0); if (error == 1) error = -EIO; if (error) |
6e9b3dd80 xfs: stop using b... |
1448 |
goto out_free_buffer; |
1da177e4c Linux-2.6.12-rc2 |
1449 1450 |
*blk_no = last_blk; |
6e9b3dd80 xfs: stop using b... |
1451 1452 |
out_free_buffer: kmem_free(buffer); |
1da177e4c Linux-2.6.12-rc2 |
1453 1454 |
if (error) return error; |
2451337dd xfs: global error... |
1455 |
return 1; |
1da177e4c Linux-2.6.12-rc2 |
1456 1457 1458 1459 1460 1461 1462 1463 1464 |
} /* * These are simple subroutines used by xlog_clear_stale_blocks() below * to initialize a buffer full of empty log record headers and write * them into the log. */ STATIC void xlog_add_record( |
9a8d2fdbb xfs: remove xlog_... |
1465 |
struct xlog *log, |
b2a922cd6 xfs: remove xfs_c... |
1466 |
char *buf, |
1da177e4c Linux-2.6.12-rc2 |
1467 1468 1469 1470 1471 1472 1473 1474 |
int cycle, int block, int tail_cycle, int tail_block) { xlog_rec_header_t *recp = (xlog_rec_header_t *)buf; memset(buf, 0, BBSIZE); |
b53e675dc [XFS] xlog_rec_he... |
1475 1476 1477 |
recp->h_magicno = cpu_to_be32(XLOG_HEADER_MAGIC_NUM); recp->h_cycle = cpu_to_be32(cycle); recp->h_version = cpu_to_be32( |
621187099 [XFS] remove shou... |
1478 |
xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? 2 : 1); |
b53e675dc [XFS] xlog_rec_he... |
1479 1480 1481 |
recp->h_lsn = cpu_to_be64(xlog_assign_lsn(cycle, block)); recp->h_tail_lsn = cpu_to_be64(xlog_assign_lsn(tail_cycle, tail_block)); recp->h_fmt = cpu_to_be32(XLOG_FMT); |
1da177e4c Linux-2.6.12-rc2 |
1482 1483 1484 1485 1486 |
memcpy(&recp->h_fs_uuid, &log->l_mp->m_sb.sb_uuid, sizeof(uuid_t)); } STATIC int xlog_write_log_records( |
9a8d2fdbb xfs: remove xlog_... |
1487 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
1488 1489 1490 1491 1492 1493 |
int cycle, int start_block, int blocks, int tail_cycle, int tail_block) { |
b2a922cd6 xfs: remove xfs_c... |
1494 |
char *offset; |
6e9b3dd80 xfs: stop using b... |
1495 |
char *buffer; |
1da177e4c Linux-2.6.12-rc2 |
1496 |
int balign, ealign; |
69ce58f08 xfs: record log s... |
1497 |
int sectbb = log->l_sectBBsize; |
1da177e4c Linux-2.6.12-rc2 |
1498 1499 1500 1501 |
int end_block = start_block + blocks; int bufblks; int error = 0; int i, j = 0; |
6881a229f xfs: fix min bufs... |
1502 1503 1504 1505 1506 1507 |
/* * Greedily allocate a buffer big enough to handle the full * range of basic blocks to be written. If that fails, try * a smaller size. We need to be able to write at least a * log sector, or we're out of luck. */ |
1da177e4c Linux-2.6.12-rc2 |
1508 |
bufblks = 1 << ffs(blocks); |
81158e0ce xfs: prevent need... |
1509 1510 |
while (bufblks > log->l_logBBsize) bufblks >>= 1; |
6e9b3dd80 xfs: stop using b... |
1511 |
while (!(buffer = xlog_alloc_buffer(log, bufblks))) { |
1da177e4c Linux-2.6.12-rc2 |
1512 |
bufblks >>= 1; |
69ce58f08 xfs: record log s... |
1513 |
if (bufblks < sectbb) |
2451337dd xfs: global error... |
1514 |
return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
1515 1516 1517 1518 1519 1520 |
} /* We may need to do a read at the start to fill in part of * the buffer in the starting sector not covered by the first * write below. */ |
5c17f5339 xfs: kill XLOG_SE... |
1521 |
balign = round_down(start_block, sectbb); |
1da177e4c Linux-2.6.12-rc2 |
1522 |
if (balign != start_block) { |
6e9b3dd80 xfs: stop using b... |
1523 |
error = xlog_bread_noalign(log, start_block, 1, buffer); |
076e6acb8 xfs: cleanup xlog... |
1524 |
if (error) |
6e9b3dd80 xfs: stop using b... |
1525 |
goto out_free_buffer; |
076e6acb8 xfs: cleanup xlog... |
1526 |
|
1da177e4c Linux-2.6.12-rc2 |
1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 |
j = start_block - balign; } for (i = start_block; i < end_block; i += bufblks) { int bcount, endcount; bcount = min(bufblks, end_block - start_block); endcount = bcount - j; /* We may need to do a read at the end to fill in part of * the buffer in the final sector not covered by the write. * If this is the same sector as the above read, skip it. */ |
5c17f5339 xfs: kill XLOG_SE... |
1540 |
ealign = round_down(end_block, sectbb); |
1da177e4c Linux-2.6.12-rc2 |
1541 |
if (j == 0 && (start_block + endcount > ealign)) { |
6ad5b3255 xfs: use bios dir... |
1542 |
error = xlog_bread_noalign(log, ealign, sectbb, |
6e9b3dd80 xfs: stop using b... |
1543 |
buffer + BBTOB(ealign - start_block)); |
076e6acb8 xfs: cleanup xlog... |
1544 1545 |
if (error) break; |
1da177e4c Linux-2.6.12-rc2 |
1546 |
} |
6e9b3dd80 xfs: stop using b... |
1547 |
offset = buffer + xlog_align(log, start_block); |
1da177e4c Linux-2.6.12-rc2 |
1548 1549 1550 1551 1552 |
for (; j < endcount; j++) { xlog_add_record(log, offset, cycle, i+j, tail_cycle, tail_block); offset += BBSIZE; } |
6e9b3dd80 xfs: stop using b... |
1553 |
error = xlog_bwrite(log, start_block, endcount, buffer); |
1da177e4c Linux-2.6.12-rc2 |
1554 1555 1556 1557 1558 |
if (error) break; start_block += endcount; j = 0; } |
076e6acb8 xfs: cleanup xlog... |
1559 |
|
6e9b3dd80 xfs: stop using b... |
1560 1561 |
out_free_buffer: kmem_free(buffer); |
1da177e4c Linux-2.6.12-rc2 |
1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 |
return error; } /* * This routine is called to blow away any incomplete log writes out * in front of the log head. We do this so that we won't become confused * if we come up, write only a little bit more, and then crash again. * If we leave the partial log records out there, this situation could * cause us to think those partial writes are valid blocks since they * have the current cycle number. We get rid of them by overwriting them * with empty log records with the old cycle number rather than the * current one. * * The tail lsn is passed in rather than taken from * the log so that we will not write over the unmount record after a * clean unmount in a 512 block log. Doing so would leave the log without * any valid log records in it until a new one was written. If we crashed * during that time we would not be able to recover. */ STATIC int xlog_clear_stale_blocks( |
9a8d2fdbb xfs: remove xlog_... |
1583 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 |
xfs_lsn_t tail_lsn) { int tail_cycle, head_cycle; int tail_block, head_block; int tail_distance, max_distance; int distance; int error; tail_cycle = CYCLE_LSN(tail_lsn); tail_block = BLOCK_LSN(tail_lsn); head_cycle = log->l_curr_cycle; head_block = log->l_curr_block; /* * Figure out the distance between the new head of the log * and the tail. We want to write over any blocks beyond the * head that we may have written just before the crash, but * we don't want to overwrite the tail of the log. */ if (head_cycle == tail_cycle) { /* * The tail is behind the head in the physical log, * so the distance from the head to the tail is the * distance from the head to the end of the log plus * the distance from the beginning of the log to the * tail. */ |
a71895c5d xfs: convert open... |
1611 1612 1613 |
if (XFS_IS_CORRUPT(log->l_mp, head_block < tail_block || head_block >= log->l_logBBsize)) |
2451337dd xfs: global error... |
1614 |
return -EFSCORRUPTED; |
1da177e4c Linux-2.6.12-rc2 |
1615 1616 1617 1618 1619 1620 1621 |
tail_distance = tail_block + (log->l_logBBsize - head_block); } else { /* * The head is behind the tail in the physical log, * so the distance from the head to the tail is just * the tail block minus the head block. */ |
a71895c5d xfs: convert open... |
1622 1623 1624 |
if (XFS_IS_CORRUPT(log->l_mp, head_block >= tail_block || head_cycle != tail_cycle + 1)) |
2451337dd xfs: global error... |
1625 |
return -EFSCORRUPTED; |
1da177e4c Linux-2.6.12-rc2 |
1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 |
tail_distance = tail_block - head_block; } /* * If the head is right up against the tail, we can't clear * anything. */ if (tail_distance <= 0) { ASSERT(tail_distance == 0); return 0; } max_distance = XLOG_TOTAL_REC_SHIFT(log); /* * Take the smaller of the maximum amount of outstanding I/O * we could have and the distance to the tail to clear out. * We take the smaller so that we don't overwrite the tail and * we don't waste all day writing from the head to the tail * for no reason. */ |
9bb54cb56 xfs: clean up MIN... |
1646 |
max_distance = min(max_distance, tail_distance); |
1da177e4c Linux-2.6.12-rc2 |
1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 |
if ((head_block + max_distance) <= log->l_logBBsize) { /* * We can stomp all the blocks we need to without * wrapping around the end of the log. Just do it * in a single write. Use the cycle number of the * current cycle minus one so that the log will look like: * n ... | n - 1 ... */ error = xlog_write_log_records(log, (head_cycle - 1), head_block, max_distance, tail_cycle, tail_block); if (error) return error; } else { /* * We need to wrap around the end of the physical log in * order to clear all the blocks. Do it in two separate * I/Os. The first write should be from the head to the * end of the physical log, and it should use the current * cycle number minus one just like above. */ distance = log->l_logBBsize - head_block; error = xlog_write_log_records(log, (head_cycle - 1), head_block, distance, tail_cycle, tail_block); if (error) return error; /* * Now write the blocks at the start of the physical log. * This writes the remainder of the blocks we want to clear. * It uses the current cycle number since we're now on the * same cycle as the head so that we get: * n ... n ... | n - 1 ... * ^^^^^ blocks we're writing */ distance = max_distance - (log->l_logBBsize - head_block); error = xlog_write_log_records(log, head_cycle, 0, distance, tail_cycle, tail_block); if (error) return error; } return 0; } |
154c733a3 xfs: refactor rel... |
1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 |
/* * Release the recovered intent item in the AIL that matches the given intent * type and intent id. */ void xlog_recover_release_intent( struct xlog *log, unsigned short intent_type, uint64_t intent_id) { struct xfs_ail_cursor cur; struct xfs_log_item *lip; struct xfs_ail *ailp = log->l_ailp; spin_lock(&ailp->ail_lock); for (lip = xfs_trans_ail_cursor_first(ailp, &cur, 0); lip != NULL; lip = xfs_trans_ail_cursor_next(ailp, &cur)) { if (lip->li_type != intent_type) continue; if (!lip->li_ops->iop_match(lip, intent_id)) continue; spin_unlock(&ailp->ail_lock); lip->li_ops->iop_release(lip); spin_lock(&ailp->ail_lock); break; } xfs_trans_ail_cursor_done(&cur); spin_unlock(&ailp->ail_lock); } |
1da177e4c Linux-2.6.12-rc2 |
1725 1726 1727 1728 1729 1730 |
/****************************************************************************** * * Log recover routines * ****************************************************************************** */ |
86ffa471d xfs: refactor log... |
1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 |
static const struct xlog_recover_item_ops *xlog_recover_item_ops[] = { &xlog_buf_item_ops, &xlog_inode_item_ops, &xlog_dquot_item_ops, &xlog_quotaoff_item_ops, &xlog_icreate_item_ops, &xlog_efi_item_ops, &xlog_efd_item_ops, &xlog_rui_item_ops, &xlog_rud_item_ops, &xlog_cui_item_ops, &xlog_cud_item_ops, &xlog_bui_item_ops, &xlog_bud_item_ops, }; static const struct xlog_recover_item_ops * xlog_find_item_ops( struct xlog_recover_item *item) { unsigned int i; for (i = 0; i < ARRAY_SIZE(xlog_recover_item_ops); i++) if (ITEM_TYPE(item) == xlog_recover_item_ops[i]->item_type) return xlog_recover_item_ops[i]; return NULL; } |
1da177e4c Linux-2.6.12-rc2 |
1759 |
|
f0a769538 xfs: Use list_hea... |
1760 |
/* |
a775ad778 xfs: fix log reco... |
1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 |
* Sort the log items in the transaction. * * The ordering constraints are defined by the inode allocation and unlink * behaviour. The rules are: * * 1. Every item is only logged once in a given transaction. Hence it * represents the last logged state of the item. Hence ordering is * dependent on the order in which operations need to be performed so * required initial conditions are always met. * * 2. Cancelled buffers are recorded in pass 1 in a separate table and * there's nothing to replay from them so we can simply cull them * from the transaction. However, we can't do that until after we've * replayed all the other items because they may be dependent on the * cancelled buffer and replaying the cancelled buffer can remove it * form the cancelled buffer table. Hence they have tobe done last. * * 3. Inode allocation buffers must be replayed before inode items that |
28c8e41af xfs: Inode create... |
1779 1780 1781 1782 |
* read the buffer and replay changes into it. For filesystems using the * ICREATE transactions, this means XFS_LI_ICREATE objects need to get * treated the same as inode allocation buffers as they create and * initialise the buffers directly. |
a775ad778 xfs: fix log reco... |
1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 |
* * 4. Inode unlink buffers must be replayed after inode items are replayed. * This ensures that inodes are completely flushed to the inode buffer * in a "free" state before we remove the unlinked inode list pointer. * * Hence the ordering needs to be inode allocation buffers first, inode items * second, inode unlink buffers third and cancelled buffers last. * * But there's a problem with that - we can't tell an inode allocation buffer * apart from a regular buffer, so we can't separate them. We can, however, * tell an inode unlink buffer from the others, and so we can separate them out * from all the other buffers and move them to last. * * Hence, 4 lists, in order from head to tail: |
28c8e41af xfs: Inode create... |
1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 |
* - buffer_list for all buffers except cancelled/inode unlink buffers * - item_list for all non-buffer items * - inode_buffer_list for inode unlink buffers * - cancel_list for the cancelled buffers * * Note that we add objects to the tail of the lists so that first-to-last * ordering is preserved within the lists. Adding objects to the head of the * list means when we traverse from the head we walk them in last-to-first * order. For cancelled buffers and inode unlink buffers this doesn't matter, * but for all other items there may be specific ordering that we need to * preserve. |
f0a769538 xfs: Use list_hea... |
1808 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
1809 1810 |
STATIC int xlog_recover_reorder_trans( |
ad223e603 xfs: rename log s... |
1811 1812 |
struct xlog *log, struct xlog_recover *trans, |
9abbc539b xfs: add log item... |
1813 |
int pass) |
1da177e4c Linux-2.6.12-rc2 |
1814 |
{ |
35f4521fd xfs: convert xfs_... |
1815 |
struct xlog_recover_item *item, *n; |
2a84108fe xfs: free the lis... |
1816 |
int error = 0; |
f0a769538 xfs: Use list_hea... |
1817 |
LIST_HEAD(sort_list); |
a775ad778 xfs: fix log reco... |
1818 1819 1820 |
LIST_HEAD(cancel_list); LIST_HEAD(buffer_list); LIST_HEAD(inode_buffer_list); |
5ce70b770 xfs: rename inode... |
1821 |
LIST_HEAD(item_list); |
f0a769538 xfs: Use list_hea... |
1822 1823 1824 |
list_splice_init(&trans->r_itemq, &sort_list); list_for_each_entry_safe(item, n, &sort_list, ri_list) { |
86ffa471d xfs: refactor log... |
1825 |
enum xlog_recover_reorder fate = XLOG_REORDER_ITEM_LIST; |
1da177e4c Linux-2.6.12-rc2 |
1826 |
|
86ffa471d xfs: refactor log... |
1827 1828 |
item->ri_ops = xlog_find_item_ops(item); if (!item->ri_ops) { |
a0fa2b679 xfs: Convert xlog... |
1829 |
xfs_warn(log->l_mp, |
0d2d35a33 xfs: report unrec... |
1830 1831 |
"%s: unrecognized type of log operation (%d)", __func__, ITEM_TYPE(item)); |
1da177e4c Linux-2.6.12-rc2 |
1832 |
ASSERT(0); |
2a84108fe xfs: free the lis... |
1833 1834 1835 1836 1837 1838 |
/* * return the remaining items back to the transaction * item list so they can be freed in caller. */ if (!list_empty(&sort_list)) list_splice_init(&sort_list, &trans->r_itemq); |
86ffa471d xfs: refactor log... |
1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 |
error = -EFSCORRUPTED; break; } if (item->ri_ops->reorder) fate = item->ri_ops->reorder(item); switch (fate) { case XLOG_REORDER_BUFFER_LIST: list_move_tail(&item->ri_list, &buffer_list); break; case XLOG_REORDER_CANCEL_LIST: trace_xfs_log_recover_item_reorder_head(log, trans, item, pass); list_move(&item->ri_list, &cancel_list); break; case XLOG_REORDER_INODE_BUFFER_LIST: list_move(&item->ri_list, &inode_buffer_list); break; case XLOG_REORDER_ITEM_LIST: trace_xfs_log_recover_item_reorder_tail(log, trans, item, pass); list_move_tail(&item->ri_list, &item_list); break; |
1da177e4c Linux-2.6.12-rc2 |
1863 |
} |
f0a769538 xfs: Use list_hea... |
1864 |
} |
86ffa471d xfs: refactor log... |
1865 |
|
f0a769538 xfs: Use list_hea... |
1866 |
ASSERT(list_empty(&sort_list)); |
a775ad778 xfs: fix log reco... |
1867 1868 |
if (!list_empty(&buffer_list)) list_splice(&buffer_list, &trans->r_itemq); |
5ce70b770 xfs: rename inode... |
1869 1870 |
if (!list_empty(&item_list)) list_splice_tail(&item_list, &trans->r_itemq); |
a775ad778 xfs: fix log reco... |
1871 1872 1873 1874 |
if (!list_empty(&inode_buffer_list)) list_splice_tail(&inode_buffer_list, &trans->r_itemq); if (!list_empty(&cancel_list)) list_splice_tail(&cancel_list, &trans->r_itemq); |
2a84108fe xfs: free the lis... |
1875 |
return error; |
1da177e4c Linux-2.6.12-rc2 |
1876 |
} |
8ea5682d0 xfs: refactor log... |
1877 |
void |
7d4894b4c xfs: factor out a... |
1878 1879 1880 1881 1882 1883 1884 1885 1886 |
xlog_buf_readahead( struct xlog *log, xfs_daddr_t blkno, uint len, const struct xfs_buf_ops *ops) { if (!xlog_is_buffer_cancelled(log, blkno, len)) xfs_buf_readahead(log->l_mp->m_ddev_targp, blkno, len, ops); } |
1da177e4c Linux-2.6.12-rc2 |
1887 |
STATIC int |
00574da19 xfs: introduce ob... |
1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 |
xlog_recover_items_pass2( struct xlog *log, struct xlog_recover *trans, struct list_head *buffer_list, struct list_head *item_list) { struct xlog_recover_item *item; int error = 0; list_for_each_entry(item, item_list, ri_list) { |
2565a11b2 xfs: remove log r... |
1898 1899 1900 1901 1902 1903 |
trace_xfs_log_recover_item_recover(log, trans, item, XLOG_RECOVER_PASS2); if (item->ri_ops->commit_pass2) error = item->ri_ops->commit_pass2(log, buffer_list, item, trans->r_lsn); |
00574da19 xfs: introduce ob... |
1904 1905 1906 1907 1908 1909 |
if (error) return error; } return error; } |
d04509486 xfs: refactor xlo... |
1910 1911 1912 1913 1914 1915 1916 |
/* * Perform the transaction. * * If the transaction modifies a buffer or inode, do it now. Otherwise, * EFIs and EFDs get queued up by adding entries into the AIL for them. */ STATIC int |
1da177e4c Linux-2.6.12-rc2 |
1917 |
xlog_recover_commit_trans( |
ad223e603 xfs: rename log s... |
1918 |
struct xlog *log, |
d04509486 xfs: refactor xlo... |
1919 |
struct xlog_recover *trans, |
12818d24d xfs: rework log r... |
1920 1921 |
int pass, struct list_head *buffer_list) |
1da177e4c Linux-2.6.12-rc2 |
1922 |
{ |
00574da19 xfs: introduce ob... |
1923 |
int error = 0; |
00574da19 xfs: introduce ob... |
1924 1925 1926 |
int items_queued = 0; struct xlog_recover_item *item; struct xlog_recover_item *next; |
00574da19 xfs: introduce ob... |
1927 1928 1929 1930 |
LIST_HEAD (ra_list); LIST_HEAD (done_list); #define XLOG_RECOVER_COMMIT_QUEUE_MAX 100 |
1da177e4c Linux-2.6.12-rc2 |
1931 |
|
39775431f xfs: free uncommi... |
1932 |
hlist_del_init(&trans->r_list); |
d04509486 xfs: refactor xlo... |
1933 1934 1935 |
error = xlog_recover_reorder_trans(log, trans, pass); if (error) |
1da177e4c Linux-2.6.12-rc2 |
1936 |
return error; |
d04509486 xfs: refactor xlo... |
1937 |
|
00574da19 xfs: introduce ob... |
1938 |
list_for_each_entry_safe(item, next, &trans->r_itemq, ri_list) { |
3304a4fab xfs: refactor log... |
1939 |
trace_xfs_log_recover_item_recover(log, trans, item, pass); |
43ff2122e xfs: on-stack del... |
1940 1941 |
switch (pass) { case XLOG_RECOVER_PASS1: |
3304a4fab xfs: refactor log... |
1942 1943 |
if (item->ri_ops->commit_pass1) error = item->ri_ops->commit_pass1(log, item); |
43ff2122e xfs: on-stack del... |
1944 1945 |
break; case XLOG_RECOVER_PASS2: |
8ea5682d0 xfs: refactor log... |
1946 1947 |
if (item->ri_ops->ra_pass2) item->ri_ops->ra_pass2(log, item); |
00574da19 xfs: introduce ob... |
1948 1949 1950 1951 |
list_move_tail(&item->ri_list, &ra_list); items_queued++; if (items_queued >= XLOG_RECOVER_COMMIT_QUEUE_MAX) { error = xlog_recover_items_pass2(log, trans, |
12818d24d xfs: rework log r... |
1952 |
buffer_list, &ra_list); |
00574da19 xfs: introduce ob... |
1953 1954 1955 |
list_splice_tail_init(&ra_list, &done_list); items_queued = 0; } |
43ff2122e xfs: on-stack del... |
1956 1957 1958 1959 |
break; default: ASSERT(0); } |
d04509486 xfs: refactor xlo... |
1960 |
if (error) |
43ff2122e xfs: on-stack del... |
1961 |
goto out; |
d04509486 xfs: refactor xlo... |
1962 |
} |
00574da19 xfs: introduce ob... |
1963 1964 1965 1966 |
out: if (!list_empty(&ra_list)) { if (!error) error = xlog_recover_items_pass2(log, trans, |
12818d24d xfs: rework log r... |
1967 |
buffer_list, &ra_list); |
00574da19 xfs: introduce ob... |
1968 1969 1970 1971 1972 |
list_splice_tail_init(&ra_list, &done_list); } if (!list_empty(&done_list)) list_splice_init(&done_list, &trans->r_itemq); |
12818d24d xfs: rework log r... |
1973 |
return error; |
1da177e4c Linux-2.6.12-rc2 |
1974 |
} |
765606698 xfs: reorganise t... |
1975 1976 1977 1978 |
STATIC void xlog_recover_add_item( struct list_head *head) { |
35f4521fd xfs: convert xfs_... |
1979 |
struct xlog_recover_item *item; |
765606698 xfs: reorganise t... |
1980 |
|
35f4521fd xfs: convert xfs_... |
1981 |
item = kmem_zalloc(sizeof(struct xlog_recover_item), 0); |
765606698 xfs: reorganise t... |
1982 1983 1984 |
INIT_LIST_HEAD(&item->ri_list); list_add_tail(&item->ri_list, head); } |
1da177e4c Linux-2.6.12-rc2 |
1985 |
STATIC int |
765606698 xfs: reorganise t... |
1986 1987 1988 |
xlog_recover_add_to_cont_trans( struct xlog *log, struct xlog_recover *trans, |
b2a922cd6 xfs: remove xfs_c... |
1989 |
char *dp, |
765606698 xfs: reorganise t... |
1990 |
int len) |
1da177e4c Linux-2.6.12-rc2 |
1991 |
{ |
35f4521fd xfs: convert xfs_... |
1992 |
struct xlog_recover_item *item; |
b2a922cd6 xfs: remove xfs_c... |
1993 |
char *ptr, *old_ptr; |
765606698 xfs: reorganise t... |
1994 |
int old_len; |
89cebc847 xfs: validate tra... |
1995 1996 1997 1998 |
/* * If the transaction is empty, the header was split across this and the * previous record. Copy the rest of the header. */ |
765606698 xfs: reorganise t... |
1999 |
if (list_empty(&trans->r_itemq)) { |
848ccfc8f xfs: fix log reco... |
2000 |
ASSERT(len <= sizeof(struct xfs_trans_header)); |
89cebc847 xfs: validate tra... |
2001 2002 |
if (len > sizeof(struct xfs_trans_header)) { xfs_warn(log->l_mp, "%s: bad header length", __func__); |
895e196fb xfs: convert EIO ... |
2003 |
return -EFSCORRUPTED; |
89cebc847 xfs: validate tra... |
2004 |
} |
765606698 xfs: reorganise t... |
2005 |
xlog_recover_add_item(&trans->r_itemq); |
b2a922cd6 xfs: remove xfs_c... |
2006 |
ptr = (char *)&trans->r_theader + |
89cebc847 xfs: validate tra... |
2007 |
sizeof(struct xfs_trans_header) - len; |
765606698 xfs: reorganise t... |
2008 2009 2010 |
memcpy(ptr, dp, len); return 0; } |
89cebc847 xfs: validate tra... |
2011 |
|
765606698 xfs: reorganise t... |
2012 |
/* take the tail entry */ |
35f4521fd xfs: convert xfs_... |
2013 2014 |
item = list_entry(trans->r_itemq.prev, struct xlog_recover_item, ri_list); |
765606698 xfs: reorganise t... |
2015 2016 2017 |
old_ptr = item->ri_buf[item->ri_cnt-1].i_addr; old_len = item->ri_buf[item->ri_cnt-1].i_len; |
771915c4f xfs: remove kmem_... |
2018 |
ptr = krealloc(old_ptr, len + old_len, GFP_KERNEL | __GFP_NOFAIL); |
765606698 xfs: reorganise t... |
2019 2020 2021 2022 |
memcpy(&ptr[old_len], dp, len); item->ri_buf[item->ri_cnt-1].i_len += len; item->ri_buf[item->ri_cnt-1].i_addr = ptr; trace_xfs_log_recover_item_add_cont(log, trans, item, 0); |
1da177e4c Linux-2.6.12-rc2 |
2023 2024 2025 2026 |
return 0; } /* |
765606698 xfs: reorganise t... |
2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 |
* The next region to add is the start of a new region. It could be * a whole region or it could be the first part of a new region. Because * of this, the assumption here is that the type and size fields of all * format structures fit into the first 32 bits of the structure. * * This works because all regions must be 32 bit aligned. Therefore, we * either have both fields or we have neither field. In the case we have * neither field, the data part of the region is zero length. We only have * a log_op_header and can throw away the header since a new one will appear * later. If we have at least 4 bytes, then we can determine how many regions * will appear in the current log item. */ STATIC int xlog_recover_add_to_trans( struct xlog *log, struct xlog_recover *trans, |
b2a922cd6 xfs: remove xfs_c... |
2043 |
char *dp, |
765606698 xfs: reorganise t... |
2044 2045 |
int len) { |
06b113212 xfs: remove inode... |
2046 |
struct xfs_inode_log_format *in_f; /* any will do */ |
35f4521fd xfs: convert xfs_... |
2047 |
struct xlog_recover_item *item; |
b2a922cd6 xfs: remove xfs_c... |
2048 |
char *ptr; |
765606698 xfs: reorganise t... |
2049 2050 2051 2052 2053 2054 2055 2056 2057 |
if (!len) return 0; if (list_empty(&trans->r_itemq)) { /* we need to catch log corruptions here */ if (*(uint *)dp != XFS_TRANS_HEADER_MAGIC) { xfs_warn(log->l_mp, "%s: bad header magic number", __func__); ASSERT(0); |
895e196fb xfs: convert EIO ... |
2058 |
return -EFSCORRUPTED; |
765606698 xfs: reorganise t... |
2059 |
} |
89cebc847 xfs: validate tra... |
2060 2061 2062 2063 |
if (len > sizeof(struct xfs_trans_header)) { xfs_warn(log->l_mp, "%s: bad header length", __func__); ASSERT(0); |
895e196fb xfs: convert EIO ... |
2064 |
return -EFSCORRUPTED; |
89cebc847 xfs: validate tra... |
2065 2066 2067 2068 2069 2070 2071 2072 |
} /* * The transaction header can be arbitrarily split across op * records. If we don't have the whole thing here, copy what we * do have and handle the rest in the next record. */ if (len == sizeof(struct xfs_trans_header)) |
765606698 xfs: reorganise t... |
2073 2074 2075 2076 |
xlog_recover_add_item(&trans->r_itemq); memcpy(&trans->r_theader, dp, len); return 0; } |
707e0ddaf fs: xfs: Remove K... |
2077 |
ptr = kmem_alloc(len, 0); |
765606698 xfs: reorganise t... |
2078 |
memcpy(ptr, dp, len); |
06b113212 xfs: remove inode... |
2079 |
in_f = (struct xfs_inode_log_format *)ptr; |
765606698 xfs: reorganise t... |
2080 2081 |
/* take the tail entry */ |
35f4521fd xfs: convert xfs_... |
2082 2083 |
item = list_entry(trans->r_itemq.prev, struct xlog_recover_item, ri_list); |
765606698 xfs: reorganise t... |
2084 2085 2086 2087 2088 |
if (item->ri_total != 0 && item->ri_total == item->ri_cnt) { /* tail item is in use, get a new one */ xlog_recover_add_item(&trans->r_itemq); item = list_entry(trans->r_itemq.prev, |
35f4521fd xfs: convert xfs_... |
2089 |
struct xlog_recover_item, ri_list); |
765606698 xfs: reorganise t... |
2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 |
} if (item->ri_total == 0) { /* first region to be added */ if (in_f->ilf_size == 0 || in_f->ilf_size > XLOG_MAX_REGIONS_IN_ITEM) { xfs_warn(log->l_mp, "bad number of regions (%d) in inode log format", in_f->ilf_size); ASSERT(0); kmem_free(ptr); |
895e196fb xfs: convert EIO ... |
2100 |
return -EFSCORRUPTED; |
765606698 xfs: reorganise t... |
2101 2102 2103 2104 2105 |
} item->ri_total = in_f->ilf_size; item->ri_buf = kmem_zalloc(item->ri_total * sizeof(xfs_log_iovec_t), |
707e0ddaf fs: xfs: Remove K... |
2106 |
0); |
765606698 xfs: reorganise t... |
2107 |
} |
d6abecb82 xfs: range check ... |
2108 2109 2110 2111 2112 2113 2114 2115 2116 |
if (item->ri_total <= item->ri_cnt) { xfs_warn(log->l_mp, "log item region count (%d) overflowed size (%d)", item->ri_cnt, item->ri_total); ASSERT(0); kmem_free(ptr); return -EFSCORRUPTED; } |
765606698 xfs: reorganise t... |
2117 2118 2119 2120 2121 2122 2123 |
/* Description region is ri_buf[0] */ item->ri_buf[item->ri_cnt].i_addr = ptr; item->ri_buf[item->ri_cnt].i_len = len; item->ri_cnt++; trace_xfs_log_recover_item_add(log, trans, item, 0); return 0; } |
b818cca19 xfs: refactor rec... |
2124 |
|
765606698 xfs: reorganise t... |
2125 2126 2127 2128 2129 2130 2131 2132 2133 |
/* * Free up any resources allocated by the transaction * * Remember that EFIs, EFDs, and IUNLINKs are handled later. */ STATIC void xlog_recover_free_trans( struct xlog_recover *trans) { |
35f4521fd xfs: convert xfs_... |
2134 |
struct xlog_recover_item *item, *n; |
765606698 xfs: reorganise t... |
2135 |
int i; |
39775431f xfs: free uncommi... |
2136 |
hlist_del_init(&trans->r_list); |
765606698 xfs: reorganise t... |
2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 |
list_for_each_entry_safe(item, n, &trans->r_itemq, ri_list) { /* Free the regions in the item. */ list_del(&item->ri_list); for (i = 0; i < item->ri_cnt; i++) kmem_free(item->ri_buf[i].i_addr); /* Free the item itself */ kmem_free(item->ri_buf); kmem_free(item); } /* Free the transaction recover structure */ kmem_free(trans); } |
e9131e50f xfs: recovery of ... |
2149 2150 2151 |
/* * On error or completion, trans is freed. */ |
1da177e4c Linux-2.6.12-rc2 |
2152 |
STATIC int |
eeb116881 xfs: refactor xlo... |
2153 2154 2155 |
xlog_recovery_process_trans( struct xlog *log, struct xlog_recover *trans, |
b2a922cd6 xfs: remove xfs_c... |
2156 |
char *dp, |
eeb116881 xfs: refactor xlo... |
2157 2158 |
unsigned int len, unsigned int flags, |
12818d24d xfs: rework log r... |
2159 2160 |
int pass, struct list_head *buffer_list) |
1da177e4c Linux-2.6.12-rc2 |
2161 |
{ |
e9131e50f xfs: recovery of ... |
2162 2163 |
int error = 0; bool freeit = false; |
eeb116881 xfs: refactor xlo... |
2164 2165 2166 2167 2168 |
/* mask off ophdr transaction container flags */ flags &= ~XLOG_END_TRANS; if (flags & XLOG_WAS_CONT_TRANS) flags &= ~XLOG_CONTINUE_TRANS; |
88b863db9 xfs: fix double f... |
2169 2170 2171 2172 |
/* * Callees must not free the trans structure. We'll decide if we need to * free it or not based on the operation being done and it's result. */ |
eeb116881 xfs: refactor xlo... |
2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 |
switch (flags) { /* expected flag values */ case 0: case XLOG_CONTINUE_TRANS: error = xlog_recover_add_to_trans(log, trans, dp, len); break; case XLOG_WAS_CONT_TRANS: error = xlog_recover_add_to_cont_trans(log, trans, dp, len); break; case XLOG_COMMIT_TRANS: |
12818d24d xfs: rework log r... |
2183 2184 |
error = xlog_recover_commit_trans(log, trans, pass, buffer_list); |
88b863db9 xfs: fix double f... |
2185 2186 |
/* success or fail, we are now done with this transaction. */ freeit = true; |
eeb116881 xfs: refactor xlo... |
2187 2188 2189 2190 |
break; /* unexpected flag values */ case XLOG_UNMOUNT_TRANS: |
e9131e50f xfs: recovery of ... |
2191 |
/* just skip trans */ |
eeb116881 xfs: refactor xlo... |
2192 |
xfs_warn(log->l_mp, "%s: Unmount LR", __func__); |
e9131e50f xfs: recovery of ... |
2193 |
freeit = true; |
eeb116881 xfs: refactor xlo... |
2194 2195 |
break; case XLOG_START_TRANS: |
eeb116881 xfs: refactor xlo... |
2196 2197 2198 |
default: xfs_warn(log->l_mp, "%s: bad flag 0x%x", __func__, flags); ASSERT(0); |
895e196fb xfs: convert EIO ... |
2199 |
error = -EFSCORRUPTED; |
eeb116881 xfs: refactor xlo... |
2200 2201 |
break; } |
e9131e50f xfs: recovery of ... |
2202 2203 |
if (error || freeit) xlog_recover_free_trans(trans); |
eeb116881 xfs: refactor xlo... |
2204 2205 |
return error; } |
b818cca19 xfs: refactor rec... |
2206 2207 2208 2209 2210 2211 2212 |
/* * Lookup the transaction recovery structure associated with the ID in the * current ophdr. If the transaction doesn't exist and the start flag is set in * the ophdr, then allocate a new transaction for future ID matches to find. * Either way, return what we found during the lookup - an existing transaction * or nothing. */ |
eeb116881 xfs: refactor xlo... |
2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 |
STATIC struct xlog_recover * xlog_recover_ophdr_to_trans( struct hlist_head rhash[], struct xlog_rec_header *rhead, struct xlog_op_header *ohead) { struct xlog_recover *trans; xlog_tid_t tid; struct hlist_head *rhp; tid = be32_to_cpu(ohead->oh_tid); rhp = &rhash[XLOG_RHASH(tid)]; |
b818cca19 xfs: refactor rec... |
2225 2226 2227 2228 |
hlist_for_each_entry(trans, rhp, r_list) { if (trans->r_log_tid == tid) return trans; } |
eeb116881 xfs: refactor xlo... |
2229 2230 |
/* |
b818cca19 xfs: refactor rec... |
2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 |
* skip over non-start transaction headers - we could be * processing slack space before the next transaction starts */ if (!(ohead->oh_flags & XLOG_START_TRANS)) return NULL; ASSERT(be32_to_cpu(ohead->oh_len) == 0); /* * This is a new transaction so allocate a new recovery container to * hold the recovery ops that will follow. */ |
707e0ddaf fs: xfs: Remove K... |
2243 |
trans = kmem_zalloc(sizeof(struct xlog_recover), 0); |
b818cca19 xfs: refactor rec... |
2244 2245 2246 2247 2248 2249 2250 2251 2252 |
trans->r_log_tid = tid; trans->r_lsn = be64_to_cpu(rhead->h_lsn); INIT_LIST_HEAD(&trans->r_itemq); INIT_HLIST_NODE(&trans->r_list); hlist_add_head(&trans->r_list, rhp); /* * Nothing more to do for this ophdr. Items to be added to this new * transaction will be in subsequent ophdr containers. |
eeb116881 xfs: refactor xlo... |
2253 |
*/ |
eeb116881 xfs: refactor xlo... |
2254 2255 2256 2257 2258 2259 2260 2261 2262 |
return NULL; } STATIC int xlog_recover_process_ophdr( struct xlog *log, struct hlist_head rhash[], struct xlog_rec_header *rhead, struct xlog_op_header *ohead, |
b2a922cd6 xfs: remove xfs_c... |
2263 2264 |
char *dp, char *end, |
12818d24d xfs: rework log r... |
2265 2266 |
int pass, struct list_head *buffer_list) |
eeb116881 xfs: refactor xlo... |
2267 2268 |
{ struct xlog_recover *trans; |
eeb116881 xfs: refactor xlo... |
2269 |
unsigned int len; |
12818d24d xfs: rework log r... |
2270 |
int error; |
eeb116881 xfs: refactor xlo... |
2271 2272 2273 2274 2275 2276 2277 |
/* Do we understand who wrote this op? */ if (ohead->oh_clientid != XFS_TRANSACTION && ohead->oh_clientid != XFS_LOG) { xfs_warn(log->l_mp, "%s: bad clientid 0x%x", __func__, ohead->oh_clientid); ASSERT(0); |
895e196fb xfs: convert EIO ... |
2278 |
return -EFSCORRUPTED; |
eeb116881 xfs: refactor xlo... |
2279 2280 2281 2282 2283 2284 2285 2286 2287 |
} /* * Check the ophdr contains all the data it is supposed to contain. */ len = be32_to_cpu(ohead->oh_len); if (dp + len > end) { xfs_warn(log->l_mp, "%s: bad length 0x%x", __func__, len); WARN_ON(1); |
895e196fb xfs: convert EIO ... |
2288 |
return -EFSCORRUPTED; |
eeb116881 xfs: refactor xlo... |
2289 2290 2291 2292 2293 2294 2295 |
} trans = xlog_recover_ophdr_to_trans(rhash, rhead, ohead); if (!trans) { /* nothing to do, so skip over this ophdr */ return 0; } |
12818d24d xfs: rework log r... |
2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 |
/* * The recovered buffer queue is drained only once we know that all * recovery items for the current LSN have been processed. This is * required because: * * - Buffer write submission updates the metadata LSN of the buffer. * - Log recovery skips items with a metadata LSN >= the current LSN of * the recovery item. * - Separate recovery items against the same metadata buffer can share * a current LSN. I.e., consider that the LSN of a recovery item is * defined as the starting LSN of the first record in which its * transaction appears, that a record can hold multiple transactions, * and/or that a transaction can span multiple records. * * In other words, we are allowed to submit a buffer from log recovery * once per current LSN. Otherwise, we may incorrectly skip recovery * items and cause corruption. * * We don't know up front whether buffers are updated multiple times per * LSN. Therefore, track the current LSN of each commit log record as it * is processed and drain the queue when it changes. Use commit records * because they are ordered correctly by the logging code. */ if (log->l_recovery_lsn != trans->r_lsn && ohead->oh_flags & XLOG_COMMIT_TRANS) { error = xfs_buf_delwri_submit(buffer_list); if (error) return error; log->l_recovery_lsn = trans->r_lsn; } |
e9131e50f xfs: recovery of ... |
2326 |
return xlog_recovery_process_trans(log, trans, dp, len, |
12818d24d xfs: rework log r... |
2327 |
ohead->oh_flags, pass, buffer_list); |
1da177e4c Linux-2.6.12-rc2 |
2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 |
} /* * There are two valid states of the r_state field. 0 indicates that the * transaction structure is in a normal state. We have either seen the * start of the transaction or the last operation we added was not a partial * operation. If the last operation we added to the transaction was a * partial operation, we need to mark r_state with XLOG_WAS_CONT_TRANS. * * NOTE: skip LRs with 0 data length. */ STATIC int xlog_recover_process_data( |
9a8d2fdbb xfs: remove xlog_... |
2341 |
struct xlog *log, |
f0a769538 xfs: Use list_hea... |
2342 |
struct hlist_head rhash[], |
9a8d2fdbb xfs: remove xlog_... |
2343 |
struct xlog_rec_header *rhead, |
b2a922cd6 xfs: remove xfs_c... |
2344 |
char *dp, |
12818d24d xfs: rework log r... |
2345 2346 |
int pass, struct list_head *buffer_list) |
1da177e4c Linux-2.6.12-rc2 |
2347 |
{ |
eeb116881 xfs: refactor xlo... |
2348 |
struct xlog_op_header *ohead; |
b2a922cd6 xfs: remove xfs_c... |
2349 |
char *end; |
1da177e4c Linux-2.6.12-rc2 |
2350 |
int num_logops; |
1da177e4c Linux-2.6.12-rc2 |
2351 |
int error; |
1da177e4c Linux-2.6.12-rc2 |
2352 |
|
eeb116881 xfs: refactor xlo... |
2353 |
end = dp + be32_to_cpu(rhead->h_len); |
b53e675dc [XFS] xlog_rec_he... |
2354 |
num_logops = be32_to_cpu(rhead->h_num_logops); |
1da177e4c Linux-2.6.12-rc2 |
2355 2356 2357 |
/* check the log format matches our own - else we can't recover */ if (xlog_header_check_recover(log->l_mp, rhead)) |
2451337dd xfs: global error... |
2358 |
return -EIO; |
1da177e4c Linux-2.6.12-rc2 |
2359 |
|
5cd9cee98 xfs: log recovery... |
2360 |
trace_xfs_log_recover_record(log, rhead, pass); |
eeb116881 xfs: refactor xlo... |
2361 2362 2363 2364 2365 2366 2367 2368 |
while ((dp < end) && num_logops) { ohead = (struct xlog_op_header *)dp; dp += sizeof(*ohead); ASSERT(dp <= end); /* errors will abort recovery */ error = xlog_recover_process_ophdr(log, rhash, rhead, ohead, |
12818d24d xfs: rework log r... |
2369 |
dp, end, pass, buffer_list); |
eeb116881 xfs: refactor xlo... |
2370 2371 |
if (error) return error; |
67fcb7bfb [XFS] clean up so... |
2372 |
dp += be32_to_cpu(ohead->oh_len); |
1da177e4c Linux-2.6.12-rc2 |
2373 2374 2375 2376 |
num_logops--; } return 0; } |
509955823 xfs: log recovery... |
2377 2378 2379 |
/* Take all the collected deferred ops and finish them in order. */ static int xlog_finish_defer_ops( |
e6fff81e4 xfs: proper repla... |
2380 2381 |
struct xfs_mount *mp, struct list_head *capture_list) |
509955823 xfs: log recovery... |
2382 |
{ |
e6fff81e4 xfs: proper repla... |
2383 |
struct xfs_defer_capture *dfc, *next; |
509955823 xfs: log recovery... |
2384 |
struct xfs_trans *tp; |
ff4ab5e02 xfs: fix an incor... |
2385 |
struct xfs_inode *ip; |
e6fff81e4 xfs: proper repla... |
2386 |
int error = 0; |
509955823 xfs: log recovery... |
2387 |
|
e6fff81e4 xfs: proper repla... |
2388 |
list_for_each_entry_safe(dfc, next, capture_list, dfc_list) { |
929b92f64 xfs: xfs_defer_ca... |
2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 |
struct xfs_trans_res resv; /* * Create a new transaction reservation from the captured * information. Set logcount to 1 to force the new transaction * to regrant every roll so that we can make forward progress * in recovery no matter how full the log might be. */ resv.tr_logres = dfc->dfc_logres; resv.tr_logcount = 1; resv.tr_logflags = XFS_TRANS_PERM_LOG_RES; error = xfs_trans_alloc(mp, &resv, dfc->dfc_blkres, dfc->dfc_rtxres, XFS_TRANS_RESERVE, &tp); |
e6fff81e4 xfs: proper repla... |
2403 2404 |
if (error) return error; |
509955823 xfs: log recovery... |
2405 |
|
e6fff81e4 xfs: proper repla... |
2406 2407 2408 2409 2410 |
/* * Transfer to this new transaction all the dfops we captured * from recovering a single intent item. */ list_del_init(&dfc->dfc_list); |
ff4ab5e02 xfs: fix an incor... |
2411 |
xfs_defer_ops_continue(dfc, tp, &ip); |
e6fff81e4 xfs: proper repla... |
2412 2413 |
error = xfs_trans_commit(tp); |
ff4ab5e02 xfs: fix an incor... |
2414 2415 2416 2417 |
if (ip) { xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_irele(ip); } |
e6fff81e4 xfs: proper repla... |
2418 2419 2420 2421 2422 2423 |
if (error) return error; } ASSERT(list_empty(capture_list)); return 0; |
509955823 xfs: log recovery... |
2424 |
} |
e6fff81e4 xfs: proper repla... |
2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 |
/* Release all the captured defer ops and capture structures in this list. */ static void xlog_abort_defer_ops( struct xfs_mount *mp, struct list_head *capture_list) { struct xfs_defer_capture *dfc; struct xfs_defer_capture *next; list_for_each_entry_safe(dfc, next, capture_list, dfc_list) { list_del_init(&dfc->dfc_list); xfs_defer_ops_release(mp, dfc); } } |
1da177e4c Linux-2.6.12-rc2 |
2439 |
/* |
dc42375d5 xfs: refactor red... |
2440 2441 2442 |
* When this is called, all of the log intent items which did not have * corresponding log done items should be in the AIL. What we do now * is update the data structures associated with each one. |
1da177e4c Linux-2.6.12-rc2 |
2443 |
* |
dc42375d5 xfs: refactor red... |
2444 2445 2446 2447 2448 2449 |
* Since we process the log intent items in normal transactions, they * will be removed at some point after the commit. This prevents us * from just walking down the list processing each one. We'll use a * flag in the intent item to skip those that we've already processed * and use the AIL iteration mechanism's generation count to try to * speed this up at least a bit. |
1da177e4c Linux-2.6.12-rc2 |
2450 |
* |
dc42375d5 xfs: refactor red... |
2451 2452 2453 |
* When we start, we know that the intents are the only things in the * AIL. As we process them, however, other items are added to the * AIL. |
1da177e4c Linux-2.6.12-rc2 |
2454 |
*/ |
3c1e2bbe5 [XFS] Propagate x... |
2455 |
STATIC int |
dc42375d5 xfs: refactor red... |
2456 |
xlog_recover_process_intents( |
f0b2efad1 xfs: don't leave ... |
2457 |
struct xlog *log) |
1da177e4c Linux-2.6.12-rc2 |
2458 |
{ |
e6fff81e4 xfs: proper repla... |
2459 |
LIST_HEAD(capture_list); |
27d8d5fe0 [XFS] Use a curso... |
2460 |
struct xfs_ail_cursor cur; |
509955823 xfs: log recovery... |
2461 |
struct xfs_log_item *lip; |
a9c21c1b9 [XFS] Given the l... |
2462 |
struct xfs_ail *ailp; |
e6fff81e4 xfs: proper repla... |
2463 |
int error = 0; |
7bf7a193a xfs: fix compiler... |
2464 |
#if defined(DEBUG) || defined(XFS_WARN) |
dc42375d5 xfs: refactor red... |
2465 |
xfs_lsn_t last_lsn; |
7bf7a193a xfs: fix compiler... |
2466 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
2467 |
|
a9c21c1b9 [XFS] Given the l... |
2468 |
ailp = log->l_ailp; |
57e809561 xfs: Rename xa_ e... |
2469 |
spin_lock(&ailp->ail_lock); |
7bf7a193a xfs: fix compiler... |
2470 |
#if defined(DEBUG) || defined(XFS_WARN) |
dc42375d5 xfs: refactor red... |
2471 |
last_lsn = xlog_assign_lsn(log->l_curr_cycle, log->l_curr_block); |
7bf7a193a xfs: fix compiler... |
2472 |
#endif |
e6fff81e4 xfs: proper repla... |
2473 2474 2475 |
for (lip = xfs_trans_ail_cursor_first(ailp, &cur, 0); lip != NULL; lip = xfs_trans_ail_cursor_next(ailp, &cur)) { |
1da177e4c Linux-2.6.12-rc2 |
2476 |
/* |
dc42375d5 xfs: refactor red... |
2477 2478 |
* We're done when we see something other than an intent. * There should be no intents left in the AIL now. |
1da177e4c Linux-2.6.12-rc2 |
2479 |
*/ |
dc42375d5 xfs: refactor red... |
2480 |
if (!xlog_item_is_intent(lip)) { |
27d8d5fe0 [XFS] Use a curso... |
2481 |
#ifdef DEBUG |
a9c21c1b9 [XFS] Given the l... |
2482 |
for (; lip; lip = xfs_trans_ail_cursor_next(ailp, &cur)) |
dc42375d5 xfs: refactor red... |
2483 |
ASSERT(!xlog_item_is_intent(lip)); |
27d8d5fe0 [XFS] Use a curso... |
2484 |
#endif |
1da177e4c Linux-2.6.12-rc2 |
2485 2486 2487 2488 |
break; } /* |
dc42375d5 xfs: refactor red... |
2489 2490 2491 |
* We should never see a redo item with a LSN higher than * the last transaction we found in the log at the start * of recovery. |
1da177e4c Linux-2.6.12-rc2 |
2492 |
*/ |
dc42375d5 xfs: refactor red... |
2493 |
ASSERT(XFS_LSN_CMP(last_lsn, lip->li_lsn) >= 0); |
1da177e4c Linux-2.6.12-rc2 |
2494 |
|
509955823 xfs: log recovery... |
2495 2496 |
/* * NOTE: If your intent processing routine can create more |
e6fff81e4 xfs: proper repla... |
2497 2498 |
* deferred ops, you /must/ attach them to the capture list in * the recover routine or else those subsequent intents will be |
509955823 xfs: log recovery... |
2499 2500 |
* replayed in the wrong order! */ |
901219bb2 xfs: remove XFS_L... |
2501 |
spin_unlock(&ailp->ail_lock); |
e6fff81e4 xfs: proper repla... |
2502 |
error = lip->li_ops->iop_recover(lip, &capture_list); |
901219bb2 xfs: remove XFS_L... |
2503 |
spin_lock(&ailp->ail_lock); |
27d8d5fe0 [XFS] Use a curso... |
2504 |
if (error) |
e6fff81e4 xfs: proper repla... |
2505 |
break; |
1da177e4c Linux-2.6.12-rc2 |
2506 |
} |
e6fff81e4 xfs: proper repla... |
2507 |
|
e4a1e29cb xfs: remove unuse... |
2508 |
xfs_trans_ail_cursor_done(&cur); |
57e809561 xfs: Rename xa_ e... |
2509 |
spin_unlock(&ailp->ail_lock); |
e6fff81e4 xfs: proper repla... |
2510 2511 |
if (error) goto err; |
509955823 xfs: log recovery... |
2512 |
|
e6fff81e4 xfs: proper repla... |
2513 2514 2515 2516 2517 2518 2519 |
error = xlog_finish_defer_ops(log->l_mp, &capture_list); if (error) goto err; return 0; err: xlog_abort_defer_ops(log->l_mp, &capture_list); |
3c1e2bbe5 [XFS] Propagate x... |
2520 |
return error; |
1da177e4c Linux-2.6.12-rc2 |
2521 2522 2523 |
} /* |
dc42375d5 xfs: refactor red... |
2524 2525 |
* A cancel occurs when the mount has failed and we're bailing out. * Release all pending log intent items so they don't pin the AIL. |
f0b2efad1 xfs: don't leave ... |
2526 |
*/ |
a7a9250e1 fs: xfs: xfs_log:... |
2527 |
STATIC void |
dc42375d5 xfs: refactor red... |
2528 |
xlog_recover_cancel_intents( |
f0b2efad1 xfs: don't leave ... |
2529 2530 2531 |
struct xlog *log) { struct xfs_log_item *lip; |
f0b2efad1 xfs: don't leave ... |
2532 2533 2534 2535 |
struct xfs_ail_cursor cur; struct xfs_ail *ailp; ailp = log->l_ailp; |
57e809561 xfs: Rename xa_ e... |
2536 |
spin_lock(&ailp->ail_lock); |
f0b2efad1 xfs: don't leave ... |
2537 2538 2539 |
lip = xfs_trans_ail_cursor_first(ailp, &cur, 0); while (lip != NULL) { /* |
dc42375d5 xfs: refactor red... |
2540 2541 |
* We're done when we see something other than an intent. * There should be no intents left in the AIL now. |
f0b2efad1 xfs: don't leave ... |
2542 |
*/ |
dc42375d5 xfs: refactor red... |
2543 |
if (!xlog_item_is_intent(lip)) { |
f0b2efad1 xfs: don't leave ... |
2544 2545 |
#ifdef DEBUG for (; lip; lip = xfs_trans_ail_cursor_next(ailp, &cur)) |
dc42375d5 xfs: refactor red... |
2546 |
ASSERT(!xlog_item_is_intent(lip)); |
f0b2efad1 xfs: don't leave ... |
2547 2548 2549 |
#endif break; } |
9329ba89c xfs: refactor rec... |
2550 2551 2552 |
spin_unlock(&ailp->ail_lock); lip->li_ops->iop_release(lip); spin_lock(&ailp->ail_lock); |
f0b2efad1 xfs: don't leave ... |
2553 2554 2555 2556 |
lip = xfs_trans_ail_cursor_next(ailp, &cur); } xfs_trans_ail_cursor_done(&cur); |
57e809561 xfs: Rename xa_ e... |
2557 |
spin_unlock(&ailp->ail_lock); |
f0b2efad1 xfs: don't leave ... |
2558 2559 2560 |
} /* |
1da177e4c Linux-2.6.12-rc2 |
2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 |
* This routine performs a transaction to null out a bad inode pointer * in an agi unlinked inode hash bucket. */ STATIC void xlog_recover_clear_agi_bucket( xfs_mount_t *mp, xfs_agnumber_t agno, int bucket) { xfs_trans_t *tp; xfs_agi_t *agi; xfs_buf_t *agibp; int offset; int error; |
253f4911f xfs: better xfs_t... |
2575 |
error = xfs_trans_alloc(mp, &M_RES(mp)->tr_clearagi, 0, 0, 0, &tp); |
e5720eec0 [XFS] Propagate e... |
2576 |
if (error) |
253f4911f xfs: better xfs_t... |
2577 |
goto out_error; |
1da177e4c Linux-2.6.12-rc2 |
2578 |
|
5e1be0fb1 [XFS] factor out ... |
2579 2580 |
error = xfs_read_agi(mp, tp, agno, &agibp); if (error) |
e5720eec0 [XFS] Propagate e... |
2581 |
goto out_abort; |
1da177e4c Linux-2.6.12-rc2 |
2582 |
|
370c782b9 xfs: remove XFS_B... |
2583 |
agi = agibp->b_addr; |
16259e7d9 [XFS] Endianess a... |
2584 |
agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO); |
1da177e4c Linux-2.6.12-rc2 |
2585 2586 2587 2588 |
offset = offsetof(xfs_agi_t, agi_unlinked) + (sizeof(xfs_agino_t) * bucket); xfs_trans_log_buf(tp, agibp, offset, (offset + sizeof(xfs_agino_t) - 1)); |
70393313d xfs: saner xfs_tr... |
2589 |
error = xfs_trans_commit(tp); |
e5720eec0 [XFS] Propagate e... |
2590 2591 2592 2593 2594 |
if (error) goto out_error; return; out_abort: |
4906e2154 xfs: remove the f... |
2595 |
xfs_trans_cancel(tp); |
e5720eec0 [XFS] Propagate e... |
2596 |
out_error: |
a0fa2b679 xfs: Convert xlog... |
2597 |
xfs_warn(mp, "%s: failed to clear agi %d. Continuing.", __func__, agno); |
e5720eec0 [XFS] Propagate e... |
2598 |
return; |
1da177e4c Linux-2.6.12-rc2 |
2599 |
} |
23fac50f9 [XFS] split up xl... |
2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 |
STATIC xfs_agino_t xlog_recover_process_one_iunlink( struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agino_t agino, int bucket) { struct xfs_buf *ibp; struct xfs_dinode *dip; struct xfs_inode *ip; xfs_ino_t ino; int error; ino = XFS_AGINO_TO_INO(mp, agno, agino); |
7b6259e7a xfs: remove block... |
2614 |
error = xfs_iget(mp, NULL, ino, 0, 0, &ip); |
23fac50f9 [XFS] split up xl... |
2615 2616 2617 2618 2619 2620 |
if (error) goto fail; /* * Get the on disk inode to find the next inode in the bucket. */ |
c19950799 xfs: remove unuse... |
2621 |
error = xfs_imap_to_bp(mp, NULL, &ip->i_imap, &dip, &ibp, 0); |
23fac50f9 [XFS] split up xl... |
2622 |
if (error) |
0e446673a [XFS] fix error h... |
2623 |
goto fail_iput; |
23fac50f9 [XFS] split up xl... |
2624 |
|
17c12bcd3 xfs: when replayi... |
2625 |
xfs_iflags_clear(ip, XFS_IRECOVERY); |
54d7b5c1d xfs: use vfs inod... |
2626 |
ASSERT(VFS_I(ip)->i_nlink == 0); |
c19b3b05a xfs: mode di_mode... |
2627 |
ASSERT(VFS_I(ip)->i_mode != 0); |
23fac50f9 [XFS] split up xl... |
2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 |
/* setup for the next pass */ agino = be32_to_cpu(dip->di_next_unlinked); xfs_buf_relse(ibp); /* * Prevent any DMAPI event from being sent when the reference on * the inode is dropped. */ ip->i_d.di_dmevmask = 0; |
44a8736bd xfs: clean up IRE... |
2638 |
xfs_irele(ip); |
23fac50f9 [XFS] split up xl... |
2639 |
return agino; |
0e446673a [XFS] fix error h... |
2640 |
fail_iput: |
44a8736bd xfs: clean up IRE... |
2641 |
xfs_irele(ip); |
23fac50f9 [XFS] split up xl... |
2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 |
fail: /* * We can't read in the inode this bucket points to, or this inode * is messed up. Just ditch this bucket of inodes. We will lose * some inodes and space, but at least we won't hang. * * Call xlog_recover_clear_agi_bucket() to perform a transaction to * clear the inode pointer in the bucket. */ xlog_recover_clear_agi_bucket(mp, agno, bucket); return NULLAGINO; } |
1da177e4c Linux-2.6.12-rc2 |
2654 |
/* |
8ab39f11d xfs: prevent CIL ... |
2655 |
* Recover AGI unlinked lists |
1da177e4c Linux-2.6.12-rc2 |
2656 |
* |
8ab39f11d xfs: prevent CIL ... |
2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 |
* This is called during recovery to process any inodes which we unlinked but * not freed when the system crashed. These inodes will be on the lists in the * AGI blocks. What we do here is scan all the AGIs and fully truncate and free * any inodes found on the lists. Each inode is removed from the lists when it * has been fully truncated and is freed. The freeing of the inode and its * removal from the list must be atomic. * * If everything we touch in the agi processing loop is already in memory, this * loop can hold the cpu for a long time. It runs without lock contention, * memory allocation contention, the need wait for IO, etc, and so will run * until we either run out of inodes to process, run low on memory or we run out * of log space. * * This behaviour is bad for latency on single CPU and non-preemptible kernels, * and can prevent other filesytem work (such as CIL pushes) from running. This * can lead to deadlocks if the recovery process runs out of log reservation * space. Hence we need to yield the CPU when there is other kernel work * scheduled on this CPU to ensure other scheduled work can run without undue * latency. |
1da177e4c Linux-2.6.12-rc2 |
2676 |
*/ |
d96f8f891 xfs: add more sta... |
2677 |
STATIC void |
1da177e4c Linux-2.6.12-rc2 |
2678 |
xlog_recover_process_iunlinks( |
9a8d2fdbb xfs: remove xlog_... |
2679 |
struct xlog *log) |
1da177e4c Linux-2.6.12-rc2 |
2680 2681 2682 2683 2684 |
{ xfs_mount_t *mp; xfs_agnumber_t agno; xfs_agi_t *agi; xfs_buf_t *agibp; |
1da177e4c Linux-2.6.12-rc2 |
2685 |
xfs_agino_t agino; |
1da177e4c Linux-2.6.12-rc2 |
2686 2687 |
int bucket; int error; |
1da177e4c Linux-2.6.12-rc2 |
2688 2689 |
mp = log->l_mp; |
1da177e4c Linux-2.6.12-rc2 |
2690 2691 2692 2693 |
for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) { /* * Find the agi for this ag. */ |
5e1be0fb1 [XFS] factor out ... |
2694 2695 2696 2697 2698 2699 2700 2701 2702 |
error = xfs_read_agi(mp, NULL, agno, &agibp); if (error) { /* * AGI is b0rked. Don't process it. * * We should probably mark the filesystem as corrupt * after we've recovered all the ag's we can.... */ continue; |
1da177e4c Linux-2.6.12-rc2 |
2703 |
} |
d97d32edc xfs: Fix oops on ... |
2704 2705 2706 2707 2708 2709 2710 2711 2712 |
/* * Unlock the buffer so that it can be acquired in the normal * course of the transaction to truncate and free each inode. * Because we are not racing with anyone else here for the AGI * buffer, we don't even need to hold it locked to read the * initial unlinked bucket entries out of the buffer. We keep * buffer reference though, so that it stays pinned in memory * while we need the buffer. */ |
370c782b9 xfs: remove XFS_B... |
2713 |
agi = agibp->b_addr; |
d97d32edc xfs: Fix oops on ... |
2714 |
xfs_buf_unlock(agibp); |
1da177e4c Linux-2.6.12-rc2 |
2715 2716 |
for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++) { |
16259e7d9 [XFS] Endianess a... |
2717 |
agino = be32_to_cpu(agi->agi_unlinked[bucket]); |
1da177e4c Linux-2.6.12-rc2 |
2718 |
while (agino != NULLAGINO) { |
23fac50f9 [XFS] split up xl... |
2719 2720 |
agino = xlog_recover_process_one_iunlink(mp, agno, agino, bucket); |
8ab39f11d xfs: prevent CIL ... |
2721 |
cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
2722 2723 |
} } |
d97d32edc xfs: Fix oops on ... |
2724 |
xfs_buf_rele(agibp); |
1da177e4c Linux-2.6.12-rc2 |
2725 |
} |
1da177e4c Linux-2.6.12-rc2 |
2726 |
} |
910832697 xfs: change some ... |
2727 |
STATIC void |
1da177e4c Linux-2.6.12-rc2 |
2728 |
xlog_unpack_data( |
9a8d2fdbb xfs: remove xlog_... |
2729 |
struct xlog_rec_header *rhead, |
b2a922cd6 xfs: remove xfs_c... |
2730 |
char *dp, |
9a8d2fdbb xfs: remove xlog_... |
2731 |
struct xlog *log) |
1da177e4c Linux-2.6.12-rc2 |
2732 2733 |
{ int i, j, k; |
1da177e4c Linux-2.6.12-rc2 |
2734 |
|
b53e675dc [XFS] xlog_rec_he... |
2735 |
for (i = 0; i < BTOBB(be32_to_cpu(rhead->h_len)) && |
1da177e4c Linux-2.6.12-rc2 |
2736 |
i < (XLOG_HEADER_CYCLE_SIZE / BBSIZE); i++) { |
b53e675dc [XFS] xlog_rec_he... |
2737 |
*(__be32 *)dp = *(__be32 *)&rhead->h_cycle_data[i]; |
1da177e4c Linux-2.6.12-rc2 |
2738 2739 |
dp += BBSIZE; } |
621187099 [XFS] remove shou... |
2740 |
if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { |
b28708d6a [XFS] sanitize xl... |
2741 |
xlog_in_core_2_t *xhdr = (xlog_in_core_2_t *)rhead; |
b53e675dc [XFS] xlog_rec_he... |
2742 |
for ( ; i < BTOBB(be32_to_cpu(rhead->h_len)); i++) { |
1da177e4c Linux-2.6.12-rc2 |
2743 2744 |
j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); k = i % (XLOG_HEADER_CYCLE_SIZE / BBSIZE); |
b53e675dc [XFS] xlog_rec_he... |
2745 |
*(__be32 *)dp = xhdr[j].hic_xheader.xh_cycle_data[k]; |
1da177e4c Linux-2.6.12-rc2 |
2746 2747 2748 |
dp += BBSIZE; } } |
1da177e4c Linux-2.6.12-rc2 |
2749 |
} |
9d94901f6 xfs: refactor log... |
2750 |
/* |
b94fb2d17 xfs: refactor and... |
2751 |
* CRC check, unpack and process a log record. |
9d94901f6 xfs: refactor log... |
2752 2753 2754 2755 2756 2757 2758 |
*/ STATIC int xlog_recover_process( struct xlog *log, struct hlist_head rhash[], struct xlog_rec_header *rhead, char *dp, |
12818d24d xfs: rework log r... |
2759 2760 |
int pass, struct list_head *buffer_list) |
9d94901f6 xfs: refactor log... |
2761 |
{ |
cae028df5 xfs: optimise CRC... |
2762 |
__le32 old_crc = rhead->h_crc; |
b94fb2d17 xfs: refactor and... |
2763 |
__le32 crc; |
6528250b7 xfs: support a cr... |
2764 |
crc = xlog_cksum(log, rhead, dp, be32_to_cpu(rhead->h_len)); |
b94fb2d17 xfs: refactor and... |
2765 |
/* |
6528250b7 xfs: support a cr... |
2766 2767 |
* Nothing else to do if this is a CRC verification pass. Just return * if this a record with a non-zero crc. Unfortunately, mkfs always |
cae028df5 xfs: optimise CRC... |
2768 |
* sets old_crc to 0 so we must consider this valid even on v5 supers. |
6528250b7 xfs: support a cr... |
2769 2770 2771 2772 |
* Otherwise, return EFSBADCRC on failure so the callers up the stack * know precisely what failed. */ if (pass == XLOG_RECOVER_CRCPASS) { |
cae028df5 xfs: optimise CRC... |
2773 |
if (old_crc && crc != old_crc) |
6528250b7 xfs: support a cr... |
2774 2775 2776 2777 2778 2779 2780 2781 2782 |
return -EFSBADCRC; return 0; } /* * We're in the normal recovery path. Issue a warning if and only if the * CRC in the header is non-zero. This is an advisory warning and the * zero CRC check prevents warnings from being emitted when upgrading * the kernel from one that does not add CRCs by default. |
b94fb2d17 xfs: refactor and... |
2783 |
*/ |
cae028df5 xfs: optimise CRC... |
2784 2785 |
if (crc != old_crc) { if (old_crc || xfs_sb_version_hascrc(&log->l_mp->m_sb)) { |
b94fb2d17 xfs: refactor and... |
2786 2787 |
xfs_alert(log->l_mp, "log record CRC mismatch: found 0x%x, expected 0x%x.", |
cae028df5 xfs: optimise CRC... |
2788 |
le32_to_cpu(old_crc), |
b94fb2d17 xfs: refactor and... |
2789 2790 2791 2792 2793 2794 2795 2796 |
le32_to_cpu(crc)); xfs_hex_dump(dp, 32); } /* * If the filesystem is CRC enabled, this mismatch becomes a * fatal log corruption failure. */ |
a5155b870 xfs: always log c... |
2797 2798 |
if (xfs_sb_version_hascrc(&log->l_mp->m_sb)) { XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, log->l_mp); |
b94fb2d17 xfs: refactor and... |
2799 |
return -EFSCORRUPTED; |
a5155b870 xfs: always log c... |
2800 |
} |
b94fb2d17 xfs: refactor and... |
2801 |
} |
9d94901f6 xfs: refactor log... |
2802 |
|
910832697 xfs: change some ... |
2803 |
xlog_unpack_data(rhead, dp, log); |
9d94901f6 xfs: refactor log... |
2804 |
|
12818d24d xfs: rework log r... |
2805 2806 |
return xlog_recover_process_data(log, rhash, rhead, dp, pass, buffer_list); |
9d94901f6 xfs: refactor log... |
2807 |
} |
1da177e4c Linux-2.6.12-rc2 |
2808 2809 |
STATIC int xlog_valid_rec_header( |
9a8d2fdbb xfs: remove xlog_... |
2810 2811 |
struct xlog *log, struct xlog_rec_header *rhead, |
f692d09e9 xfs: avoid LR buf... |
2812 2813 |
xfs_daddr_t blkno, int bufsize) |
1da177e4c Linux-2.6.12-rc2 |
2814 2815 |
{ int hlen; |
a71895c5d xfs: convert open... |
2816 2817 |
if (XFS_IS_CORRUPT(log->l_mp, rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM))) |
2451337dd xfs: global error... |
2818 |
return -EFSCORRUPTED; |
a71895c5d xfs: convert open... |
2819 2820 2821 2822 |
if (XFS_IS_CORRUPT(log->l_mp, (!rhead->h_version || (be32_to_cpu(rhead->h_version) & (~XLOG_VERSION_OKBITS))))) { |
a0fa2b679 xfs: Convert xlog... |
2823 |
xfs_warn(log->l_mp, "%s: unrecognised log version (%d).", |
34a622b2e [XFS] replace rem... |
2824 |
__func__, be32_to_cpu(rhead->h_version)); |
895e196fb xfs: convert EIO ... |
2825 |
return -EFSCORRUPTED; |
1da177e4c Linux-2.6.12-rc2 |
2826 |
} |
f692d09e9 xfs: avoid LR buf... |
2827 2828 2829 2830 |
/* * LR body must have data (or it wouldn't have been written) * and h_len must not be greater than LR buffer size. */ |
b53e675dc [XFS] xlog_rec_he... |
2831 |
hlen = be32_to_cpu(rhead->h_len); |
f692d09e9 xfs: avoid LR buf... |
2832 |
if (XFS_IS_CORRUPT(log->l_mp, hlen <= 0 || hlen > bufsize)) |
2451337dd xfs: global error... |
2833 |
return -EFSCORRUPTED; |
f692d09e9 xfs: avoid LR buf... |
2834 |
|
a71895c5d xfs: convert open... |
2835 2836 |
if (XFS_IS_CORRUPT(log->l_mp, blkno > log->l_logBBsize || blkno > INT_MAX)) |
2451337dd xfs: global error... |
2837 |
return -EFSCORRUPTED; |
1da177e4c Linux-2.6.12-rc2 |
2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 |
return 0; } /* * Read the log from tail to head and process the log records found. * Handle the two cases where the tail and head are in the same cycle * and where the active portion of the log wraps around the end of * the physical log separately. The pass parameter is passed through * to the routines called to process the data and is not looked at * here. */ STATIC int xlog_do_recovery_pass( |
9a8d2fdbb xfs: remove xlog_... |
2851 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
2852 2853 |
xfs_daddr_t head_blk, xfs_daddr_t tail_blk, |
d7f37692e xfs: return start... |
2854 2855 |
int pass, xfs_daddr_t *first_bad) /* out: first bad log rec */ |
1da177e4c Linux-2.6.12-rc2 |
2856 2857 |
{ xlog_rec_header_t *rhead; |
284f1c2c9 xfs: fix recovery... |
2858 |
xfs_daddr_t blk_no, rblk_no; |
d7f37692e xfs: return start... |
2859 |
xfs_daddr_t rhead_blk; |
b2a922cd6 xfs: remove xfs_c... |
2860 |
char *offset; |
6ad5b3255 xfs: use bios dir... |
2861 |
char *hbp, *dbp; |
a70f9fe52 xfs: detect and h... |
2862 |
int error = 0, h_size, h_len; |
12818d24d xfs: rework log r... |
2863 |
int error2 = 0; |
1da177e4c Linux-2.6.12-rc2 |
2864 2865 |
int bblks, split_bblks; int hblks, split_hblks, wrapped_hblks; |
39775431f xfs: free uncommi... |
2866 |
int i; |
f0a769538 xfs: Use list_hea... |
2867 |
struct hlist_head rhash[XLOG_RHASH_SIZE]; |
12818d24d xfs: rework log r... |
2868 |
LIST_HEAD (buffer_list); |
1da177e4c Linux-2.6.12-rc2 |
2869 2870 |
ASSERT(head_blk != tail_blk); |
a4c9b34d6 xfs: handle -EFSC... |
2871 |
blk_no = rhead_blk = tail_blk; |
1da177e4c Linux-2.6.12-rc2 |
2872 |
|
39775431f xfs: free uncommi... |
2873 2874 |
for (i = 0; i < XLOG_RHASH_SIZE; i++) INIT_HLIST_HEAD(&rhash[i]); |
1da177e4c Linux-2.6.12-rc2 |
2875 2876 2877 2878 |
/* * Read the header of the tail block and get the iclog buffer size from * h_size. Use this to tell how many sectors make up the log header. */ |
621187099 [XFS] remove shou... |
2879 |
if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { |
1da177e4c Linux-2.6.12-rc2 |
2880 2881 2882 2883 2884 |
/* * When using variable length iclogs, read first sector of * iclog header and extract the header size from it. Get a * new hbp that is the correct size. */ |
6e9b3dd80 xfs: stop using b... |
2885 |
hbp = xlog_alloc_buffer(log, 1); |
1da177e4c Linux-2.6.12-rc2 |
2886 |
if (!hbp) |
2451337dd xfs: global error... |
2887 |
return -ENOMEM; |
076e6acb8 xfs: cleanup xlog... |
2888 2889 2890 |
error = xlog_bread(log, tail_blk, 1, hbp, &offset); if (error) |
1da177e4c Linux-2.6.12-rc2 |
2891 |
goto bread_err1; |
076e6acb8 xfs: cleanup xlog... |
2892 |
|
1da177e4c Linux-2.6.12-rc2 |
2893 |
rhead = (xlog_rec_header_t *)offset; |
a70f9fe52 xfs: detect and h... |
2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 |
/* * xfsprogs has a bug where record length is based on lsunit but * h_size (iclog size) is hardcoded to 32k. Now that we * unconditionally CRC verify the unmount record, this means the * log buffer can be too small for the record and cause an * overrun. * * Detect this condition here. Use lsunit for the buffer size as * long as this looks like the mkfs case. Otherwise, return an * error to avoid a buffer overrun. */ |
b53e675dc [XFS] xlog_rec_he... |
2906 |
h_size = be32_to_cpu(rhead->h_size); |
a70f9fe52 xfs: detect and h... |
2907 |
h_len = be32_to_cpu(rhead->h_len); |
f692d09e9 xfs: avoid LR buf... |
2908 2909 2910 |
if (h_len > h_size && h_len <= log->l_mp->m_logbsize && rhead->h_num_logops == cpu_to_be32(1)) { xfs_warn(log->l_mp, |
a70f9fe52 xfs: detect and h... |
2911 |
"invalid iclog size (%d bytes), using lsunit (%d bytes)", |
f692d09e9 xfs: avoid LR buf... |
2912 2913 |
h_size, log->l_mp->m_logbsize); h_size = log->l_mp->m_logbsize; |
a70f9fe52 xfs: detect and h... |
2914 |
} |
f692d09e9 xfs: avoid LR buf... |
2915 2916 2917 |
error = xlog_valid_rec_header(log, rhead, tail_blk, h_size); if (error) goto bread_err1; |
0c771b99d xfs: clean up cal... |
2918 2919 |
hblks = xlog_logrec_hblks(log, rhead); if (hblks != 1) { |
6ad5b3255 xfs: use bios dir... |
2920 |
kmem_free(hbp); |
6e9b3dd80 xfs: stop using b... |
2921 |
hbp = xlog_alloc_buffer(log, hblks); |
1da177e4c Linux-2.6.12-rc2 |
2922 2923 |
} } else { |
69ce58f08 xfs: record log s... |
2924 |
ASSERT(log->l_sectBBsize == 1); |
1da177e4c Linux-2.6.12-rc2 |
2925 |
hblks = 1; |
6e9b3dd80 xfs: stop using b... |
2926 |
hbp = xlog_alloc_buffer(log, 1); |
1da177e4c Linux-2.6.12-rc2 |
2927 2928 2929 2930 |
h_size = XLOG_BIG_RECORD_BSIZE; } if (!hbp) |
2451337dd xfs: global error... |
2931 |
return -ENOMEM; |
6e9b3dd80 xfs: stop using b... |
2932 |
dbp = xlog_alloc_buffer(log, BTOBB(h_size)); |
1da177e4c Linux-2.6.12-rc2 |
2933 |
if (!dbp) { |
6ad5b3255 xfs: use bios dir... |
2934 |
kmem_free(hbp); |
2451337dd xfs: global error... |
2935 |
return -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2936 2937 2938 |
} memset(rhash, 0, sizeof(rhash)); |
970fd3f04 xfs: deduplicate ... |
2939 |
if (tail_blk > head_blk) { |
1da177e4c Linux-2.6.12-rc2 |
2940 2941 2942 |
/* * Perform recovery around the end of the physical log. * When the head is not on the same cycle number as the tail, |
970fd3f04 xfs: deduplicate ... |
2943 |
* we can't do a sequential recovery. |
1da177e4c Linux-2.6.12-rc2 |
2944 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
2945 2946 2947 2948 |
while (blk_no < log->l_logBBsize) { /* * Check for header wrapping around physical end-of-log */ |
6ad5b3255 xfs: use bios dir... |
2949 |
offset = hbp; |
1da177e4c Linux-2.6.12-rc2 |
2950 2951 2952 2953 |
split_hblks = 0; wrapped_hblks = 0; if (blk_no + hblks <= log->l_logBBsize) { /* Read header in one read */ |
076e6acb8 xfs: cleanup xlog... |
2954 2955 |
error = xlog_bread(log, blk_no, hblks, hbp, &offset); |
1da177e4c Linux-2.6.12-rc2 |
2956 2957 |
if (error) goto bread_err2; |
1da177e4c Linux-2.6.12-rc2 |
2958 2959 2960 2961 2962 2963 2964 |
} else { /* This LR is split across physical log end */ if (blk_no != log->l_logBBsize) { /* some data before physical log end */ ASSERT(blk_no <= INT_MAX); split_hblks = log->l_logBBsize - (int)blk_no; ASSERT(split_hblks > 0); |
076e6acb8 xfs: cleanup xlog... |
2965 2966 2967 2968 |
error = xlog_bread(log, blk_no, split_hblks, hbp, &offset); if (error) |
1da177e4c Linux-2.6.12-rc2 |
2969 |
goto bread_err2; |
1da177e4c Linux-2.6.12-rc2 |
2970 |
} |
076e6acb8 xfs: cleanup xlog... |
2971 |
|
1da177e4c Linux-2.6.12-rc2 |
2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 |
/* * Note: this black magic still works with * large sector sizes (non-512) only because: * - we increased the buffer size originally * by 1 sector giving us enough extra space * for the second read; * - the log start is guaranteed to be sector * aligned; * - we read the log end (LR header start) * _first_, then the log start (LR header end) * - order is important. */ |
234f56aca [XFS] Check for e... |
2984 |
wrapped_hblks = hblks - split_hblks; |
6ad5b3255 xfs: use bios dir... |
2985 2986 |
error = xlog_bread_noalign(log, 0, wrapped_hblks, |
44396476a xfs: reset buffer... |
2987 |
offset + BBTOB(split_hblks)); |
1da177e4c Linux-2.6.12-rc2 |
2988 2989 |
if (error) goto bread_err2; |
1da177e4c Linux-2.6.12-rc2 |
2990 2991 2992 |
} rhead = (xlog_rec_header_t *)offset; error = xlog_valid_rec_header(log, rhead, |
f692d09e9 xfs: avoid LR buf... |
2993 |
split_hblks ? blk_no : 0, h_size); |
1da177e4c Linux-2.6.12-rc2 |
2994 2995 |
if (error) goto bread_err2; |
b53e675dc [XFS] xlog_rec_he... |
2996 |
bblks = (int)BTOBB(be32_to_cpu(rhead->h_len)); |
1da177e4c Linux-2.6.12-rc2 |
2997 |
blk_no += hblks; |
284f1c2c9 xfs: fix recovery... |
2998 2999 3000 3001 3002 3003 3004 3005 3006 |
/* * Read the log record data in multiple reads if it * wraps around the end of the log. Note that if the * header already wrapped, blk_no could point past the * end of the log. The record data is contiguous in * that case. */ if (blk_no + bblks <= log->l_logBBsize || blk_no >= log->l_logBBsize) { |
0703a8e1c xfs: replace do_m... |
3007 |
rblk_no = xlog_wrap_logbno(log, blk_no); |
284f1c2c9 xfs: fix recovery... |
3008 |
error = xlog_bread(log, rblk_no, bblks, dbp, |
076e6acb8 xfs: cleanup xlog... |
3009 |
&offset); |
1da177e4c Linux-2.6.12-rc2 |
3010 3011 |
if (error) goto bread_err2; |
1da177e4c Linux-2.6.12-rc2 |
3012 3013 3014 |
} else { /* This log record is split across the * physical end of log */ |
6ad5b3255 xfs: use bios dir... |
3015 |
offset = dbp; |
1da177e4c Linux-2.6.12-rc2 |
3016 3017 3018 3019 3020 3021 3022 3023 3024 |
split_bblks = 0; if (blk_no != log->l_logBBsize) { /* some data is before the physical * end of log */ ASSERT(!wrapped_hblks); ASSERT(blk_no <= INT_MAX); split_bblks = log->l_logBBsize - (int)blk_no; ASSERT(split_bblks > 0); |
076e6acb8 xfs: cleanup xlog... |
3025 3026 3027 3028 |
error = xlog_bread(log, blk_no, split_bblks, dbp, &offset); if (error) |
1da177e4c Linux-2.6.12-rc2 |
3029 |
goto bread_err2; |
1da177e4c Linux-2.6.12-rc2 |
3030 |
} |
076e6acb8 xfs: cleanup xlog... |
3031 |
|
1da177e4c Linux-2.6.12-rc2 |
3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 |
/* * Note: this black magic still works with * large sector sizes (non-512) only because: * - we increased the buffer size originally * by 1 sector giving us enough extra space * for the second read; * - the log start is guaranteed to be sector * aligned; * - we read the log end (LR header start) * _first_, then the log start (LR header end) * - order is important. */ |
6ad5b3255 xfs: use bios dir... |
3044 3045 |
error = xlog_bread_noalign(log, 0, bblks - split_bblks, |
44396476a xfs: reset buffer... |
3046 |
offset + BBTOB(split_bblks)); |
076e6acb8 xfs: cleanup xlog... |
3047 3048 |
if (error) goto bread_err2; |
1da177e4c Linux-2.6.12-rc2 |
3049 |
} |
0e446be44 xfs: add CRC chec... |
3050 |
|
9d94901f6 xfs: refactor log... |
3051 |
error = xlog_recover_process(log, rhash, rhead, offset, |
12818d24d xfs: rework log r... |
3052 |
pass, &buffer_list); |
0e446be44 xfs: add CRC chec... |
3053 |
if (error) |
1da177e4c Linux-2.6.12-rc2 |
3054 |
goto bread_err2; |
d7f37692e xfs: return start... |
3055 |
|
1da177e4c Linux-2.6.12-rc2 |
3056 |
blk_no += bblks; |
d7f37692e xfs: return start... |
3057 |
rhead_blk = blk_no; |
1da177e4c Linux-2.6.12-rc2 |
3058 3059 3060 3061 |
} ASSERT(blk_no >= log->l_logBBsize); blk_no -= log->l_logBBsize; |
d7f37692e xfs: return start... |
3062 |
rhead_blk = blk_no; |
970fd3f04 xfs: deduplicate ... |
3063 |
} |
1da177e4c Linux-2.6.12-rc2 |
3064 |
|
970fd3f04 xfs: deduplicate ... |
3065 3066 3067 3068 3069 |
/* read first part of physical log */ while (blk_no < head_blk) { error = xlog_bread(log, blk_no, hblks, hbp, &offset); if (error) goto bread_err2; |
076e6acb8 xfs: cleanup xlog... |
3070 |
|
970fd3f04 xfs: deduplicate ... |
3071 |
rhead = (xlog_rec_header_t *)offset; |
f692d09e9 xfs: avoid LR buf... |
3072 |
error = xlog_valid_rec_header(log, rhead, blk_no, h_size); |
970fd3f04 xfs: deduplicate ... |
3073 3074 |
if (error) goto bread_err2; |
076e6acb8 xfs: cleanup xlog... |
3075 |
|
970fd3f04 xfs: deduplicate ... |
3076 3077 3078 3079 3080 3081 |
/* blocks in data section */ bblks = (int)BTOBB(be32_to_cpu(rhead->h_len)); error = xlog_bread(log, blk_no+hblks, bblks, dbp, &offset); if (error) goto bread_err2; |
076e6acb8 xfs: cleanup xlog... |
3082 |
|
12818d24d xfs: rework log r... |
3083 3084 |
error = xlog_recover_process(log, rhash, rhead, offset, pass, &buffer_list); |
970fd3f04 xfs: deduplicate ... |
3085 3086 |
if (error) goto bread_err2; |
d7f37692e xfs: return start... |
3087 |
|
970fd3f04 xfs: deduplicate ... |
3088 |
blk_no += bblks + hblks; |
d7f37692e xfs: return start... |
3089 |
rhead_blk = blk_no; |
1da177e4c Linux-2.6.12-rc2 |
3090 3091 3092 |
} bread_err2: |
6ad5b3255 xfs: use bios dir... |
3093 |
kmem_free(dbp); |
1da177e4c Linux-2.6.12-rc2 |
3094 |
bread_err1: |
6ad5b3255 xfs: use bios dir... |
3095 |
kmem_free(hbp); |
d7f37692e xfs: return start... |
3096 |
|
12818d24d xfs: rework log r... |
3097 3098 3099 3100 3101 3102 |
/* * Submit buffers that have been added from the last record processed, * regardless of error status. */ if (!list_empty(&buffer_list)) error2 = xfs_buf_delwri_submit(&buffer_list); |
d7f37692e xfs: return start... |
3103 3104 |
if (error && first_bad) *first_bad = rhead_blk; |
39775431f xfs: free uncommi... |
3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 |
/* * Transactions are freed at commit time but transactions without commit * records on disk are never committed. Free any that may be left in the * hash table. */ for (i = 0; i < XLOG_RHASH_SIZE; i++) { struct hlist_node *tmp; struct xlog_recover *trans; hlist_for_each_entry_safe(trans, tmp, &rhash[i], r_list) xlog_recover_free_trans(trans); } |
12818d24d xfs: rework log r... |
3117 |
return error ? error : error2; |
1da177e4c Linux-2.6.12-rc2 |
3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 |
} /* * Do the recovery of the log. We actually do this in two phases. * The two passes are necessary in order to implement the function * of cancelling a record written into the log. The first pass * determines those things which have been cancelled, and the * second pass replays log items normally except for those which * have been cancelled. The handling of the replay and cancellations * takes place in the log item type specific routines. * * The table of items which have cancel records in the log is allocated * and freed at this level, since only here do we know when all of * the log recovery has been completed. */ STATIC int xlog_do_log_recovery( |
9a8d2fdbb xfs: remove xlog_... |
3135 |
struct xlog *log, |
1da177e4c Linux-2.6.12-rc2 |
3136 3137 3138 |
xfs_daddr_t head_blk, xfs_daddr_t tail_blk) { |
d5689eaa0 xfs: use struct l... |
3139 |
int error, i; |
1da177e4c Linux-2.6.12-rc2 |
3140 3141 3142 3143 3144 3145 3146 |
ASSERT(head_blk != tail_blk); /* * First do a pass to find all of the cancelled buf log items. * Store them in the buf_cancel_table for use in the second pass. */ |
d5689eaa0 xfs: use struct l... |
3147 3148 |
log->l_buf_cancel_table = kmem_zalloc(XLOG_BC_TABLE_SIZE * sizeof(struct list_head), |
707e0ddaf fs: xfs: Remove K... |
3149 |
0); |
d5689eaa0 xfs: use struct l... |
3150 3151 |
for (i = 0; i < XLOG_BC_TABLE_SIZE; i++) INIT_LIST_HEAD(&log->l_buf_cancel_table[i]); |
1da177e4c Linux-2.6.12-rc2 |
3152 |
error = xlog_do_recovery_pass(log, head_blk, tail_blk, |
d7f37692e xfs: return start... |
3153 |
XLOG_RECOVER_PASS1, NULL); |
1da177e4c Linux-2.6.12-rc2 |
3154 |
if (error != 0) { |
f0e2d93c2 [XFS] Remove unus... |
3155 |
kmem_free(log->l_buf_cancel_table); |
1da177e4c Linux-2.6.12-rc2 |
3156 3157 3158 3159 3160 3161 3162 3163 |
log->l_buf_cancel_table = NULL; return error; } /* * Then do a second pass to actually recover the items in the log. * When it is complete free the table of buf cancel items. */ error = xlog_do_recovery_pass(log, head_blk, tail_blk, |
d7f37692e xfs: return start... |
3164 |
XLOG_RECOVER_PASS2, NULL); |
1da177e4c Linux-2.6.12-rc2 |
3165 |
#ifdef DEBUG |
6d192a9b8 [XFS] inode items... |
3166 |
if (!error) { |
1da177e4c Linux-2.6.12-rc2 |
3167 3168 3169 |
int i; for (i = 0; i < XLOG_BC_TABLE_SIZE; i++) |
d5689eaa0 xfs: use struct l... |
3170 |
ASSERT(list_empty(&log->l_buf_cancel_table[i])); |
1da177e4c Linux-2.6.12-rc2 |
3171 3172 |
} #endif /* DEBUG */ |
f0e2d93c2 [XFS] Remove unus... |
3173 |
kmem_free(log->l_buf_cancel_table); |
1da177e4c Linux-2.6.12-rc2 |
3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 |
log->l_buf_cancel_table = NULL; return error; } /* * Do the actual recovery */ STATIC int xlog_do_recover( |
b3f8e08ca xfs: remove xfs_g... |
3184 3185 3186 |
struct xlog *log, xfs_daddr_t head_blk, xfs_daddr_t tail_blk) |
1da177e4c Linux-2.6.12-rc2 |
3187 |
{ |
b3f8e08ca xfs: remove xfs_g... |
3188 3189 3190 3191 |
struct xfs_mount *mp = log->l_mp; struct xfs_buf *bp = mp->m_sb_bp; struct xfs_sb *sbp = &mp->m_sb; int error; |
1da177e4c Linux-2.6.12-rc2 |
3192 |
|
e67d3d424 xfs: add log reco... |
3193 |
trace_xfs_log_recover(log, head_blk, tail_blk); |
1da177e4c Linux-2.6.12-rc2 |
3194 3195 3196 3197 |
/* * First replay the images in the log. */ error = xlog_do_log_recovery(log, head_blk, tail_blk); |
43ff2122e xfs: on-stack del... |
3198 |
if (error) |
1da177e4c Linux-2.6.12-rc2 |
3199 |
return error; |
1da177e4c Linux-2.6.12-rc2 |
3200 3201 3202 3203 |
/* * If IO errors happened during recovery, bail out. */ |
b3f8e08ca xfs: remove xfs_g... |
3204 |
if (XFS_FORCED_SHUTDOWN(mp)) |
2451337dd xfs: global error... |
3205 |
return -EIO; |
1da177e4c Linux-2.6.12-rc2 |
3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 |
/* * We now update the tail_lsn since much of the recovery has completed * and there may be space available to use. If there were no extent * or iunlinks, we can free up the entire log and set the tail_lsn to * be the last_sync_lsn. This was set in xlog_find_tail to be the * lsn of the last known good LR on disk. If there are extent frees * or iunlinks they will have some entries in the AIL; so we look at * the AIL to determine how to set the tail_lsn. */ |
a798011c8 xfs: reinitialise... |
3216 |
xlog_assign_tail_lsn(mp); |
1da177e4c Linux-2.6.12-rc2 |
3217 3218 |
/* |
b3f8e08ca xfs: remove xfs_g... |
3219 3220 |
* Now that we've finished replaying all buffer and inode updates, * re-read the superblock and reverify it. |
1da177e4c Linux-2.6.12-rc2 |
3221 |
*/ |
b3f8e08ca xfs: remove xfs_g... |
3222 3223 |
xfs_buf_lock(bp); xfs_buf_hold(bp); |
26e328759 xfs: reuse _xfs_b... |
3224 |
error = _xfs_buf_read(bp, XBF_READ); |
d64e31a2f [XFS] Ensure erro... |
3225 |
if (error) { |
a798011c8 xfs: reinitialise... |
3226 |
if (!XFS_FORCED_SHUTDOWN(mp)) { |
cdbcf82b8 xfs: fix xfs_buf_... |
3227 |
xfs_buf_ioerror_alert(bp, __this_address); |
595bff75d xfs: introduce xf... |
3228 3229 |
ASSERT(0); } |
1da177e4c Linux-2.6.12-rc2 |
3230 3231 3232 3233 3234 |
xfs_buf_relse(bp); return error; } /* Convert superblock from on-disk format */ |
3e6e8afd3 xfs: remove XFS_B... |
3235 |
xfs_sb_from_disk(sbp, bp->b_addr); |
1da177e4c Linux-2.6.12-rc2 |
3236 |
xfs_buf_relse(bp); |
a798011c8 xfs: reinitialise... |
3237 3238 3239 3240 3241 3242 3243 |
/* re-initialise in-core superblock and geometry structures */ xfs_reinit_percpu_counters(mp); error = xfs_initialize_perag(mp, sbp->sb_agcount, &mp->m_maxagi); if (error) { xfs_warn(mp, "Failed post-recovery per-ag init: %d", error); return error; } |
525488520 xfs: rmap btree r... |
3244 |
mp->m_alloc_set_aside = xfs_alloc_set_aside(mp); |
5478eead8 [XFS] Re-initiali... |
3245 |
|
1da177e4c Linux-2.6.12-rc2 |
3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 |
xlog_recover_check_summary(log); /* Normal transactions can now occur */ log->l_flags &= ~XLOG_ACTIVE_RECOVERY; return 0; } /* * Perform recovery and re-initialize some log variables in xlog_find_tail. * * Return error or zero. */ int xlog_recover( |
9a8d2fdbb xfs: remove xlog_... |
3260 |
struct xlog *log) |
1da177e4c Linux-2.6.12-rc2 |
3261 3262 3263 3264 3265 |
{ xfs_daddr_t head_blk, tail_blk; int error; /* find the tail of the log */ |
a45086e27 xfs: validate met... |
3266 3267 |
error = xlog_find_tail(log, &head_blk, &tail_blk); if (error) |
1da177e4c Linux-2.6.12-rc2 |
3268 |
return error; |
a45086e27 xfs: validate met... |
3269 3270 3271 3272 3273 3274 3275 3276 |
/* * The superblock was read before the log was available and thus the LSN * could not be verified. Check the superblock LSN against the current * LSN now that it's known. */ if (xfs_sb_version_hascrc(&log->l_mp->m_sb) && !xfs_log_check_lsn(log->l_mp, log->l_mp->m_sb.sb_lsn)) return -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 |
if (tail_blk != head_blk) { /* There used to be a comment here: * * disallow recovery on read-only mounts. note -- mount * checks for ENOSPC and turns it into an intelligent * error message. * ...but this is no longer true. Now, unless you specify * NORECOVERY (in which case this function would never be * called), we just go ahead and recover. We do this all * under the vfs layer, so we can get away with it unless * the device itself is read-only, in which case we fail. */ |
3a02ee182 [XFS] Get rid of ... |
3289 |
if ((error = xfs_dev_is_read_only(log->l_mp, "recovery"))) { |
1da177e4c Linux-2.6.12-rc2 |
3290 3291 |
return error; } |
e721f504c xfs: implement ex... |
3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 |
/* * Version 5 superblock log feature mask validation. We know the * log is dirty so check if there are any unknown log features * in what we need to recover. If there are unknown features * (e.g. unsupported transactions, then simply reject the * attempt at recovery before touching anything. */ if (XFS_SB_VERSION_NUM(&log->l_mp->m_sb) == XFS_SB_VERSION_5 && xfs_sb_has_incompat_log_feature(&log->l_mp->m_sb, XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN)) { xfs_warn(log->l_mp, |
f41febd2e xfs: Use consiste... |
3303 |
"Superblock has unknown incompatible log features (0x%x) enabled.", |
e721f504c xfs: implement ex... |
3304 3305 |
(log->l_mp->m_sb.sb_features_log_incompat & XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN)); |
f41febd2e xfs: Use consiste... |
3306 3307 3308 3309 |
xfs_warn(log->l_mp, "The log can not be fully and/or safely recovered by this kernel."); xfs_warn(log->l_mp, "Please recover the log on a kernel that supports the unknown features."); |
2451337dd xfs: global error... |
3310 |
return -EINVAL; |
e721f504c xfs: implement ex... |
3311 |
} |
2e2271787 xfs: export log_r... |
3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 |
/* * Delay log recovery if the debug hook is set. This is debug * instrumention to coordinate simulation of I/O failures with * log recovery. */ if (xfs_globals.log_recovery_delay) { xfs_notice(log->l_mp, "Delaying log recovery for %d seconds.", xfs_globals.log_recovery_delay); msleep(xfs_globals.log_recovery_delay * 1000); } |
a0fa2b679 xfs: Convert xlog... |
3323 3324 3325 |
xfs_notice(log->l_mp, "Starting recovery (logdev: %s)", log->l_mp->m_logname ? log->l_mp->m_logname : "internal"); |
1da177e4c Linux-2.6.12-rc2 |
3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 |
error = xlog_do_recover(log, head_blk, tail_blk); log->l_flags |= XLOG_RECOVERY_NEEDED; } return error; } /* * In the first part of recovery we replay inodes and buffers and build * up the list of extent free items which need to be processed. Here * we process the extent free items and clean up the on disk unlinked * inode lists. This is separated from the first part of recovery so * that the root and real-time bitmap inodes can be read in from disk in * between the two stages. This is necessary so that we can free space * in the real-time portion of the file system. */ int xlog_recover_finish( |
9a8d2fdbb xfs: remove xlog_... |
3344 |
struct xlog *log) |
1da177e4c Linux-2.6.12-rc2 |
3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 |
{ /* * Now we're ready to do the transactions needed for the * rest of recovery. Start with completing all the extent * free intent records and then process the unlinked inode * lists. At this point, we essentially run in normal mode * except that we're still performing recovery actions * rather than accepting new requests. */ if (log->l_flags & XLOG_RECOVERY_NEEDED) { |
3c1e2bbe5 [XFS] Propagate x... |
3355 |
int error; |
dc42375d5 xfs: refactor red... |
3356 |
error = xlog_recover_process_intents(log); |
3c1e2bbe5 [XFS] Propagate x... |
3357 |
if (error) { |
2e76f188f xfs: cancel inten... |
3358 3359 3360 3361 3362 3363 3364 3365 |
/* * Cancel all the unprocessed intent items now so that * we don't leave them pinned in the AIL. This can * cause the AIL to livelock on the pinned item if * anyone tries to push the AIL (inode reclaim does * this) before we get around to xfs_log_mount_cancel. */ xlog_recover_cancel_intents(log); |
dc42375d5 xfs: refactor red... |
3366 |
xfs_alert(log->l_mp, "Failed to recover intents"); |
3c1e2bbe5 [XFS] Propagate x... |
3367 3368 |
return error; } |
9e88b5d86 xfs: log rmap int... |
3369 |
|
1da177e4c Linux-2.6.12-rc2 |
3370 |
/* |
dc42375d5 xfs: refactor red... |
3371 |
* Sync the log to get all the intents out of the AIL. |
1da177e4c Linux-2.6.12-rc2 |
3372 3373 |
* This isn't absolutely necessary, but it helps in * case the unlink transactions would have problems |
dc42375d5 xfs: refactor red... |
3374 |
* pushing the intents out of the way. |
1da177e4c Linux-2.6.12-rc2 |
3375 |
*/ |
a14a348bf xfs: cleanup up x... |
3376 |
xfs_log_force(log->l_mp, XFS_LOG_SYNC); |
1da177e4c Linux-2.6.12-rc2 |
3377 |
|
4249023a5 [XFS] cleanup xfs... |
3378 |
xlog_recover_process_iunlinks(log); |
1da177e4c Linux-2.6.12-rc2 |
3379 3380 |
xlog_recover_check_summary(log); |
a0fa2b679 xfs: Convert xlog... |
3381 3382 3383 |
xfs_notice(log->l_mp, "Ending recovery (logdev: %s)", log->l_mp->m_logname ? log->l_mp->m_logname : "internal"); |
1da177e4c Linux-2.6.12-rc2 |
3384 3385 |
log->l_flags &= ~XLOG_RECOVERY_NEEDED; } else { |
a0fa2b679 xfs: Convert xlog... |
3386 |
xfs_info(log->l_mp, "Ending clean mount"); |
1da177e4c Linux-2.6.12-rc2 |
3387 3388 3389 |
} return 0; } |
a7a9250e1 fs: xfs: xfs_log:... |
3390 |
void |
f0b2efad1 xfs: don't leave ... |
3391 3392 3393 |
xlog_recover_cancel( struct xlog *log) { |
f0b2efad1 xfs: don't leave ... |
3394 |
if (log->l_flags & XLOG_RECOVERY_NEEDED) |
a7a9250e1 fs: xfs: xfs_log:... |
3395 |
xlog_recover_cancel_intents(log); |
f0b2efad1 xfs: don't leave ... |
3396 |
} |
1da177e4c Linux-2.6.12-rc2 |
3397 3398 3399 3400 3401 3402 |
#if defined(DEBUG) /* * Read all of the agf and agi counters and check that they * are consistent with the superblock counters. */ |
e89fbb5ee xfs: mark xlog_re... |
3403 |
STATIC void |
1da177e4c Linux-2.6.12-rc2 |
3404 |
xlog_recover_check_summary( |
9a8d2fdbb xfs: remove xlog_... |
3405 |
struct xlog *log) |
1da177e4c Linux-2.6.12-rc2 |
3406 3407 |
{ xfs_mount_t *mp; |
1da177e4c Linux-2.6.12-rc2 |
3408 3409 |
xfs_buf_t *agfbp; xfs_buf_t *agibp; |
1da177e4c Linux-2.6.12-rc2 |
3410 |
xfs_agnumber_t agno; |
c8ce540db xfs: remove doubl... |
3411 3412 3413 |
uint64_t freeblks; uint64_t itotal; uint64_t ifree; |
5e1be0fb1 [XFS] factor out ... |
3414 |
int error; |
1da177e4c Linux-2.6.12-rc2 |
3415 3416 3417 3418 3419 3420 3421 |
mp = log->l_mp; freeblks = 0LL; itotal = 0LL; ifree = 0LL; for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) { |
4805621a3 [XFS] factor out ... |
3422 3423 |
error = xfs_read_agf(mp, NULL, agno, 0, &agfbp); if (error) { |
a0fa2b679 xfs: Convert xlog... |
3424 3425 |
xfs_alert(mp, "%s agf read failed agno %d error %d", __func__, agno, error); |
4805621a3 [XFS] factor out ... |
3426 |
} else { |
9798f615a xfs: remove XFS_B... |
3427 |
struct xfs_agf *agfp = agfbp->b_addr; |
4805621a3 [XFS] factor out ... |
3428 3429 3430 |
freeblks += be32_to_cpu(agfp->agf_freeblks) + be32_to_cpu(agfp->agf_flcount); xfs_buf_relse(agfbp); |
1da177e4c Linux-2.6.12-rc2 |
3431 |
} |
1da177e4c Linux-2.6.12-rc2 |
3432 |
|
5e1be0fb1 [XFS] factor out ... |
3433 |
error = xfs_read_agi(mp, NULL, agno, &agibp); |
a0fa2b679 xfs: Convert xlog... |
3434 3435 3436 3437 |
if (error) { xfs_alert(mp, "%s agi read failed agno %d error %d", __func__, agno, error); } else { |
370c782b9 xfs: remove XFS_B... |
3438 |
struct xfs_agi *agi = agibp->b_addr; |
16259e7d9 [XFS] Endianess a... |
3439 |
|
5e1be0fb1 [XFS] factor out ... |
3440 3441 3442 3443 |
itotal += be32_to_cpu(agi->agi_count); ifree += be32_to_cpu(agi->agi_freecount); xfs_buf_relse(agibp); } |
1da177e4c Linux-2.6.12-rc2 |
3444 |
} |
1da177e4c Linux-2.6.12-rc2 |
3445 3446 |
} #endif /* DEBUG */ |