]>
Commit | Line | Data |
---|---|---|
1da177e4 | 1 | /* |
7b718769 NS |
2 | * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. |
3 | * All Rights Reserved. | |
1da177e4 | 4 | * |
7b718769 NS |
5 | * This program is free software; you can redistribute it and/or |
6 | * modify it under the terms of the GNU General Public License as | |
1da177e4 LT |
7 | * published by the Free Software Foundation. |
8 | * | |
7b718769 NS |
9 | * This program is distributed in the hope that it would be useful, |
10 | * but WITHOUT ANY WARRANTY; without even the implied warranty of | |
11 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
12 | * GNU General Public License for more details. | |
1da177e4 | 13 | * |
7b718769 NS |
14 | * You should have received a copy of the GNU General Public License |
15 | * along with this program; if not, write the Free Software Foundation, | |
16 | * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA | |
1da177e4 | 17 | */ |
1da177e4 | 18 | #include "xfs.h" |
a844f451 | 19 | #include "xfs_fs.h" |
70a9883c | 20 | #include "xfs_shared.h" |
a4fbe6ab | 21 | #include "xfs_format.h" |
239880ef DC |
22 | #include "xfs_log_format.h" |
23 | #include "xfs_trans_resv.h" | |
a844f451 | 24 | #include "xfs_bit.h" |
1da177e4 | 25 | #include "xfs_mount.h" |
3ab78df2 | 26 | #include "xfs_defer.h" |
1da177e4 | 27 | #include "xfs_inode.h" |
239880ef | 28 | #include "xfs_trans.h" |
38bb7423 | 29 | #include "xfs_inode_item.h" |
ee1a47ab | 30 | #include "xfs_buf_item.h" |
a844f451 | 31 | #include "xfs_btree.h" |
1da177e4 | 32 | #include "xfs_error.h" |
0b1b213f | 33 | #include "xfs_trace.h" |
ee1a47ab | 34 | #include "xfs_cksum.h" |
cf11da9c | 35 | #include "xfs_alloc.h" |
a45086e2 | 36 | #include "xfs_log.h" |
1da177e4 LT |
37 | |
38 | /* | |
39 | * Cursor allocation zone. | |
40 | */ | |
41 | kmem_zone_t *xfs_btree_cur_zone; | |
42 | ||
43 | /* | |
44 | * Btree magic numbers. | |
45 | */ | |
ee1a47ab | 46 | static const __uint32_t xfs_magics[2][XFS_BTNUM_MAX] = { |
b8704944 | 47 | { XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, 0, XFS_BMAP_MAGIC, XFS_IBT_MAGIC, |
46eeb521 | 48 | XFS_FIBT_MAGIC, 0 }, |
b8704944 | 49 | { XFS_ABTB_CRC_MAGIC, XFS_ABTC_CRC_MAGIC, XFS_RMAP_CRC_MAGIC, |
46eeb521 DW |
50 | XFS_BMAP_CRC_MAGIC, XFS_IBT_CRC_MAGIC, XFS_FIBT_CRC_MAGIC, |
51 | XFS_REFC_CRC_MAGIC } | |
1da177e4 | 52 | }; |
af7d20fd ES |
53 | |
54 | __uint32_t | |
55 | xfs_btree_magic( | |
56 | int crc, | |
57 | xfs_btnum_t btnum) | |
58 | { | |
59 | __uint32_t magic = xfs_magics[crc][btnum]; | |
60 | ||
61 | /* Ensure we asked for crc for crc-only magics. */ | |
62 | ASSERT(magic != 0); | |
63 | return magic; | |
64 | } | |
1da177e4 | 65 | |
7cc95a82 | 66 | STATIC int /* error (0 or EFSCORRUPTED) */ |
a23f6ef8 CH |
67 | xfs_btree_check_lblock( |
68 | struct xfs_btree_cur *cur, /* btree cursor */ | |
7cc95a82 | 69 | struct xfs_btree_block *block, /* btree long form block pointer */ |
a23f6ef8 CH |
70 | int level, /* level of the btree block */ |
71 | struct xfs_buf *bp) /* buffer for block, if any */ | |
72 | { | |
ee1a47ab | 73 | int lblock_ok = 1; /* block passes checks */ |
a23f6ef8 | 74 | struct xfs_mount *mp; /* file system mount point */ |
af7d20fd ES |
75 | xfs_btnum_t btnum = cur->bc_btnum; |
76 | int crc; | |
a23f6ef8 CH |
77 | |
78 | mp = cur->bc_mp; | |
af7d20fd | 79 | crc = xfs_sb_version_hascrc(&mp->m_sb); |
ee1a47ab | 80 | |
af7d20fd | 81 | if (crc) { |
ee1a47ab | 82 | lblock_ok = lblock_ok && |
ce748eaa ES |
83 | uuid_equal(&block->bb_u.l.bb_uuid, |
84 | &mp->m_sb.sb_meta_uuid) && | |
ee1a47ab CH |
85 | block->bb_u.l.bb_blkno == cpu_to_be64( |
86 | bp ? bp->b_bn : XFS_BUF_DADDR_NULL); | |
87 | } | |
88 | ||
89 | lblock_ok = lblock_ok && | |
af7d20fd | 90 | be32_to_cpu(block->bb_magic) == xfs_btree_magic(crc, btnum) && |
a23f6ef8 CH |
91 | be16_to_cpu(block->bb_level) == level && |
92 | be16_to_cpu(block->bb_numrecs) <= | |
ce5e42db | 93 | cur->bc_ops->get_maxrecs(cur, level) && |
7cc95a82 | 94 | block->bb_u.l.bb_leftsib && |
d5cf09ba | 95 | (block->bb_u.l.bb_leftsib == cpu_to_be64(NULLFSBLOCK) || |
7cc95a82 | 96 | XFS_FSB_SANITY_CHECK(mp, |
ee1a47ab | 97 | be64_to_cpu(block->bb_u.l.bb_leftsib))) && |
7cc95a82 | 98 | block->bb_u.l.bb_rightsib && |
d5cf09ba | 99 | (block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK) || |
7cc95a82 | 100 | XFS_FSB_SANITY_CHECK(mp, |
ee1a47ab CH |
101 | be64_to_cpu(block->bb_u.l.bb_rightsib))); |
102 | ||
a23f6ef8 CH |
103 | if (unlikely(XFS_TEST_ERROR(!lblock_ok, mp, |
104 | XFS_ERRTAG_BTREE_CHECK_LBLOCK, | |
105 | XFS_RANDOM_BTREE_CHECK_LBLOCK))) { | |
106 | if (bp) | |
0b1b213f | 107 | trace_xfs_btree_corrupt(bp, _RET_IP_); |
ee1a47ab | 108 | XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp); |
2451337d | 109 | return -EFSCORRUPTED; |
a23f6ef8 CH |
110 | } |
111 | return 0; | |
112 | } | |
113 | ||
3cc7524c | 114 | STATIC int /* error (0 or EFSCORRUPTED) */ |
1da177e4 | 115 | xfs_btree_check_sblock( |
a23f6ef8 | 116 | struct xfs_btree_cur *cur, /* btree cursor */ |
7cc95a82 | 117 | struct xfs_btree_block *block, /* btree short form block pointer */ |
1da177e4 | 118 | int level, /* level of the btree block */ |
a23f6ef8 | 119 | struct xfs_buf *bp) /* buffer containing block */ |
1da177e4 | 120 | { |
ee1a47ab | 121 | struct xfs_mount *mp; /* file system mount point */ |
a23f6ef8 CH |
122 | struct xfs_buf *agbp; /* buffer for ag. freespace struct */ |
123 | struct xfs_agf *agf; /* ag. freespace structure */ | |
1da177e4 | 124 | xfs_agblock_t agflen; /* native ag. freespace length */ |
ee1a47ab | 125 | int sblock_ok = 1; /* block passes checks */ |
af7d20fd ES |
126 | xfs_btnum_t btnum = cur->bc_btnum; |
127 | int crc; | |
1da177e4 | 128 | |
ee1a47ab | 129 | mp = cur->bc_mp; |
af7d20fd | 130 | crc = xfs_sb_version_hascrc(&mp->m_sb); |
1da177e4 LT |
131 | agbp = cur->bc_private.a.agbp; |
132 | agf = XFS_BUF_TO_AGF(agbp); | |
16259e7d | 133 | agflen = be32_to_cpu(agf->agf_length); |
ee1a47ab | 134 | |
af7d20fd | 135 | if (crc) { |
ee1a47ab | 136 | sblock_ok = sblock_ok && |
ce748eaa ES |
137 | uuid_equal(&block->bb_u.s.bb_uuid, |
138 | &mp->m_sb.sb_meta_uuid) && | |
ee1a47ab CH |
139 | block->bb_u.s.bb_blkno == cpu_to_be64( |
140 | bp ? bp->b_bn : XFS_BUF_DADDR_NULL); | |
141 | } | |
142 | ||
143 | sblock_ok = sblock_ok && | |
af7d20fd | 144 | be32_to_cpu(block->bb_magic) == xfs_btree_magic(crc, btnum) && |
16259e7d CH |
145 | be16_to_cpu(block->bb_level) == level && |
146 | be16_to_cpu(block->bb_numrecs) <= | |
ce5e42db | 147 | cur->bc_ops->get_maxrecs(cur, level) && |
69ef921b | 148 | (block->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK) || |
7cc95a82 CH |
149 | be32_to_cpu(block->bb_u.s.bb_leftsib) < agflen) && |
150 | block->bb_u.s.bb_leftsib && | |
69ef921b | 151 | (block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK) || |
7cc95a82 CH |
152 | be32_to_cpu(block->bb_u.s.bb_rightsib) < agflen) && |
153 | block->bb_u.s.bb_rightsib; | |
ee1a47ab CH |
154 | |
155 | if (unlikely(XFS_TEST_ERROR(!sblock_ok, mp, | |
1da177e4 LT |
156 | XFS_ERRTAG_BTREE_CHECK_SBLOCK, |
157 | XFS_RANDOM_BTREE_CHECK_SBLOCK))) { | |
158 | if (bp) | |
0b1b213f | 159 | trace_xfs_btree_corrupt(bp, _RET_IP_); |
ee1a47ab | 160 | XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp); |
2451337d | 161 | return -EFSCORRUPTED; |
1da177e4 LT |
162 | } |
163 | return 0; | |
164 | } | |
165 | ||
166 | /* | |
a23f6ef8 CH |
167 | * Debug routine: check that block header is ok. |
168 | */ | |
169 | int | |
170 | xfs_btree_check_block( | |
171 | struct xfs_btree_cur *cur, /* btree cursor */ | |
172 | struct xfs_btree_block *block, /* generic btree block pointer */ | |
173 | int level, /* level of the btree block */ | |
174 | struct xfs_buf *bp) /* buffer containing block, if any */ | |
175 | { | |
7cc95a82 CH |
176 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) |
177 | return xfs_btree_check_lblock(cur, block, level, bp); | |
178 | else | |
179 | return xfs_btree_check_sblock(cur, block, level, bp); | |
a23f6ef8 CH |
180 | } |
181 | ||
182 | /* | |
183 | * Check that (long) pointer is ok. | |
184 | */ | |
185 | int /* error (0 or EFSCORRUPTED) */ | |
186 | xfs_btree_check_lptr( | |
187 | struct xfs_btree_cur *cur, /* btree cursor */ | |
d5cf09ba | 188 | xfs_fsblock_t bno, /* btree block disk address */ |
a23f6ef8 CH |
189 | int level) /* btree block level */ |
190 | { | |
5fb5aeee | 191 | XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, |
a23f6ef8 | 192 | level > 0 && |
d5cf09ba | 193 | bno != NULLFSBLOCK && |
a23f6ef8 CH |
194 | XFS_FSB_SANITY_CHECK(cur->bc_mp, bno)); |
195 | return 0; | |
196 | } | |
197 | ||
24ee0e49 | 198 | #ifdef DEBUG |
a23f6ef8 CH |
199 | /* |
200 | * Check that (short) pointer is ok. | |
1da177e4 | 201 | */ |
3cc7524c | 202 | STATIC int /* error (0 or EFSCORRUPTED) */ |
1da177e4 | 203 | xfs_btree_check_sptr( |
a23f6ef8 CH |
204 | struct xfs_btree_cur *cur, /* btree cursor */ |
205 | xfs_agblock_t bno, /* btree block disk address */ | |
206 | int level) /* btree block level */ | |
1da177e4 | 207 | { |
a23f6ef8 | 208 | xfs_agblock_t agblocks = cur->bc_mp->m_sb.sb_agblocks; |
1da177e4 | 209 | |
5fb5aeee | 210 | XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, |
1da177e4 | 211 | level > 0 && |
a23f6ef8 CH |
212 | bno != NULLAGBLOCK && |
213 | bno != 0 && | |
214 | bno < agblocks); | |
1da177e4 LT |
215 | return 0; |
216 | } | |
217 | ||
a23f6ef8 CH |
218 | /* |
219 | * Check that block ptr is ok. | |
220 | */ | |
3cc7524c | 221 | STATIC int /* error (0 or EFSCORRUPTED) */ |
a23f6ef8 CH |
222 | xfs_btree_check_ptr( |
223 | struct xfs_btree_cur *cur, /* btree cursor */ | |
224 | union xfs_btree_ptr *ptr, /* btree block disk address */ | |
225 | int index, /* offset from ptr to check */ | |
226 | int level) /* btree block level */ | |
227 | { | |
228 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) { | |
229 | return xfs_btree_check_lptr(cur, | |
230 | be64_to_cpu((&ptr->l)[index]), level); | |
231 | } else { | |
232 | return xfs_btree_check_sptr(cur, | |
233 | be32_to_cpu((&ptr->s)[index]), level); | |
234 | } | |
235 | } | |
24ee0e49 | 236 | #endif |
a23f6ef8 | 237 | |
ee1a47ab CH |
238 | /* |
239 | * Calculate CRC on the whole btree block and stuff it into the | |
240 | * long-form btree header. | |
241 | * | |
242 | * Prior to calculting the CRC, pull the LSN out of the buffer log item and put | |
fef4ded8 | 243 | * it into the buffer so recovery knows what the last modification was that made |
ee1a47ab CH |
244 | * it to disk. |
245 | */ | |
246 | void | |
247 | xfs_btree_lblock_calc_crc( | |
248 | struct xfs_buf *bp) | |
249 | { | |
250 | struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); | |
251 | struct xfs_buf_log_item *bip = bp->b_fspriv; | |
252 | ||
253 | if (!xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb)) | |
254 | return; | |
255 | if (bip) | |
256 | block->bb_u.l.bb_lsn = cpu_to_be64(bip->bli_item.li_lsn); | |
f1dbcd7e | 257 | xfs_buf_update_cksum(bp, XFS_BTREE_LBLOCK_CRC_OFF); |
ee1a47ab CH |
258 | } |
259 | ||
260 | bool | |
261 | xfs_btree_lblock_verify_crc( | |
262 | struct xfs_buf *bp) | |
263 | { | |
a45086e2 BF |
264 | struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); |
265 | struct xfs_mount *mp = bp->b_target->bt_mount; | |
266 | ||
267 | if (xfs_sb_version_hascrc(&mp->m_sb)) { | |
268 | if (!xfs_log_check_lsn(mp, be64_to_cpu(block->bb_u.l.bb_lsn))) | |
269 | return false; | |
51582170 | 270 | return xfs_buf_verify_cksum(bp, XFS_BTREE_LBLOCK_CRC_OFF); |
a45086e2 | 271 | } |
51582170 | 272 | |
ee1a47ab CH |
273 | return true; |
274 | } | |
275 | ||
276 | /* | |
277 | * Calculate CRC on the whole btree block and stuff it into the | |
278 | * short-form btree header. | |
279 | * | |
280 | * Prior to calculting the CRC, pull the LSN out of the buffer log item and put | |
fef4ded8 | 281 | * it into the buffer so recovery knows what the last modification was that made |
ee1a47ab CH |
282 | * it to disk. |
283 | */ | |
284 | void | |
285 | xfs_btree_sblock_calc_crc( | |
286 | struct xfs_buf *bp) | |
287 | { | |
288 | struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); | |
289 | struct xfs_buf_log_item *bip = bp->b_fspriv; | |
290 | ||
291 | if (!xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb)) | |
292 | return; | |
293 | if (bip) | |
294 | block->bb_u.s.bb_lsn = cpu_to_be64(bip->bli_item.li_lsn); | |
f1dbcd7e | 295 | xfs_buf_update_cksum(bp, XFS_BTREE_SBLOCK_CRC_OFF); |
ee1a47ab CH |
296 | } |
297 | ||
298 | bool | |
299 | xfs_btree_sblock_verify_crc( | |
300 | struct xfs_buf *bp) | |
301 | { | |
a45086e2 BF |
302 | struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); |
303 | struct xfs_mount *mp = bp->b_target->bt_mount; | |
304 | ||
305 | if (xfs_sb_version_hascrc(&mp->m_sb)) { | |
306 | if (!xfs_log_check_lsn(mp, be64_to_cpu(block->bb_u.s.bb_lsn))) | |
307 | return false; | |
51582170 | 308 | return xfs_buf_verify_cksum(bp, XFS_BTREE_SBLOCK_CRC_OFF); |
a45086e2 | 309 | } |
51582170 | 310 | |
ee1a47ab CH |
311 | return true; |
312 | } | |
313 | ||
c46ee8ad CH |
314 | static int |
315 | xfs_btree_free_block( | |
316 | struct xfs_btree_cur *cur, | |
317 | struct xfs_buf *bp) | |
318 | { | |
319 | int error; | |
320 | ||
321 | error = cur->bc_ops->free_block(cur, bp); | |
edfd9dd5 CH |
322 | if (!error) { |
323 | xfs_trans_binval(cur->bc_tp, bp); | |
c46ee8ad | 324 | XFS_BTREE_STATS_INC(cur, free); |
edfd9dd5 | 325 | } |
c46ee8ad CH |
326 | return error; |
327 | } | |
328 | ||
1da177e4 LT |
329 | /* |
330 | * Delete the btree cursor. | |
331 | */ | |
332 | void | |
333 | xfs_btree_del_cursor( | |
334 | xfs_btree_cur_t *cur, /* btree cursor */ | |
335 | int error) /* del because of error */ | |
336 | { | |
337 | int i; /* btree level */ | |
338 | ||
339 | /* | |
340 | * Clear the buffer pointers, and release the buffers. | |
341 | * If we're doing this in the face of an error, we | |
342 | * need to make sure to inspect all of the entries | |
343 | * in the bc_bufs array for buffers to be unlocked. | |
344 | * This is because some of the btree code works from | |
345 | * level n down to 0, and if we get an error along | |
346 | * the way we won't have initialized all the entries | |
347 | * down to 0. | |
348 | */ | |
349 | for (i = 0; i < cur->bc_nlevels; i++) { | |
350 | if (cur->bc_bufs[i]) | |
c0e59e1a | 351 | xfs_trans_brelse(cur->bc_tp, cur->bc_bufs[i]); |
1da177e4 LT |
352 | else if (!error) |
353 | break; | |
354 | } | |
355 | /* | |
356 | * Can't free a bmap cursor without having dealt with the | |
357 | * allocated indirect blocks' accounting. | |
358 | */ | |
359 | ASSERT(cur->bc_btnum != XFS_BTNUM_BMAP || | |
360 | cur->bc_private.b.allocated == 0); | |
361 | /* | |
362 | * Free the cursor. | |
363 | */ | |
364 | kmem_zone_free(xfs_btree_cur_zone, cur); | |
365 | } | |
366 | ||
367 | /* | |
368 | * Duplicate the btree cursor. | |
369 | * Allocate a new one, copy the record, re-get the buffers. | |
370 | */ | |
371 | int /* error */ | |
372 | xfs_btree_dup_cursor( | |
373 | xfs_btree_cur_t *cur, /* input cursor */ | |
374 | xfs_btree_cur_t **ncur) /* output cursor */ | |
375 | { | |
376 | xfs_buf_t *bp; /* btree block's buffer pointer */ | |
377 | int error; /* error return value */ | |
378 | int i; /* level number of btree block */ | |
379 | xfs_mount_t *mp; /* mount structure for filesystem */ | |
380 | xfs_btree_cur_t *new; /* new cursor value */ | |
381 | xfs_trans_t *tp; /* transaction pointer, can be NULL */ | |
382 | ||
383 | tp = cur->bc_tp; | |
384 | mp = cur->bc_mp; | |
561f7d17 | 385 | |
1da177e4 LT |
386 | /* |
387 | * Allocate a new cursor like the old one. | |
388 | */ | |
561f7d17 CH |
389 | new = cur->bc_ops->dup_cursor(cur); |
390 | ||
1da177e4 LT |
391 | /* |
392 | * Copy the record currently in the cursor. | |
393 | */ | |
394 | new->bc_rec = cur->bc_rec; | |
561f7d17 | 395 | |
1da177e4 LT |
396 | /* |
397 | * For each level current, re-get the buffer and copy the ptr value. | |
398 | */ | |
399 | for (i = 0; i < new->bc_nlevels; i++) { | |
400 | new->bc_ptrs[i] = cur->bc_ptrs[i]; | |
401 | new->bc_ra[i] = cur->bc_ra[i]; | |
c3f8fc73 DC |
402 | bp = cur->bc_bufs[i]; |
403 | if (bp) { | |
404 | error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, | |
405 | XFS_BUF_ADDR(bp), mp->m_bsize, | |
3d3e6f64 | 406 | 0, &bp, |
1813dd64 | 407 | cur->bc_ops->buf_ops); |
c3f8fc73 | 408 | if (error) { |
1da177e4 LT |
409 | xfs_btree_del_cursor(new, error); |
410 | *ncur = NULL; | |
411 | return error; | |
412 | } | |
ee1a47ab CH |
413 | } |
414 | new->bc_bufs[i] = bp; | |
1da177e4 | 415 | } |
1da177e4 LT |
416 | *ncur = new; |
417 | return 0; | |
418 | } | |
419 | ||
65f1eaea CH |
420 | /* |
421 | * XFS btree block layout and addressing: | |
422 | * | |
423 | * There are two types of blocks in the btree: leaf and non-leaf blocks. | |
424 | * | |
425 | * The leaf record start with a header then followed by records containing | |
426 | * the values. A non-leaf block also starts with the same header, and | |
427 | * then first contains lookup keys followed by an equal number of pointers | |
428 | * to the btree blocks at the previous level. | |
429 | * | |
430 | * +--------+-------+-------+-------+-------+-------+-------+ | |
431 | * Leaf: | header | rec 1 | rec 2 | rec 3 | rec 4 | rec 5 | rec N | | |
432 | * +--------+-------+-------+-------+-------+-------+-------+ | |
433 | * | |
434 | * +--------+-------+-------+-------+-------+-------+-------+ | |
435 | * Non-Leaf: | header | key 1 | key 2 | key N | ptr 1 | ptr 2 | ptr N | | |
436 | * +--------+-------+-------+-------+-------+-------+-------+ | |
437 | * | |
438 | * The header is called struct xfs_btree_block for reasons better left unknown | |
439 | * and comes in different versions for short (32bit) and long (64bit) block | |
440 | * pointers. The record and key structures are defined by the btree instances | |
441 | * and opaque to the btree core. The block pointers are simple disk endian | |
442 | * integers, available in a short (32bit) and long (64bit) variant. | |
443 | * | |
444 | * The helpers below calculate the offset of a given record, key or pointer | |
445 | * into a btree block (xfs_btree_*_offset) or return a pointer to the given | |
446 | * record, key or pointer (xfs_btree_*_addr). Note that all addressing | |
447 | * inside the btree block is done using indices starting at one, not zero! | |
2c813ad6 DW |
448 | * |
449 | * If XFS_BTREE_OVERLAPPING is set, then this btree supports keys containing | |
450 | * overlapping intervals. In such a tree, records are still sorted lowest to | |
451 | * highest and indexed by the smallest key value that refers to the record. | |
452 | * However, nodes are different: each pointer has two associated keys -- one | |
453 | * indexing the lowest key available in the block(s) below (the same behavior | |
454 | * as the key in a regular btree) and another indexing the highest key | |
455 | * available in the block(s) below. Because records are /not/ sorted by the | |
456 | * highest key, all leaf block updates require us to compute the highest key | |
457 | * that matches any record in the leaf and to recursively update the high keys | |
458 | * in the nodes going further up in the tree, if necessary. Nodes look like | |
459 | * this: | |
460 | * | |
461 | * +--------+-----+-----+-----+-----+-----+-------+-------+-----+ | |
462 | * Non-Leaf: | header | lo1 | hi1 | lo2 | hi2 | ... | ptr 1 | ptr 2 | ... | | |
463 | * +--------+-----+-----+-----+-----+-----+-------+-------+-----+ | |
464 | * | |
465 | * To perform an interval query on an overlapped tree, perform the usual | |
466 | * depth-first search and use the low and high keys to decide if we can skip | |
467 | * that particular node. If a leaf node is reached, return the records that | |
468 | * intersect the interval. Note that an interval query may return numerous | |
469 | * entries. For a non-overlapped tree, simply search for the record associated | |
470 | * with the lowest key and iterate forward until a non-matching record is | |
471 | * found. Section 14.3 ("Interval Trees") of _Introduction to Algorithms_ by | |
472 | * Cormen, Leiserson, Rivest, and Stein (2nd or 3rd ed. only) discuss this in | |
473 | * more detail. | |
474 | * | |
475 | * Why do we care about overlapping intervals? Let's say you have a bunch of | |
476 | * reverse mapping records on a reflink filesystem: | |
477 | * | |
478 | * 1: +- file A startblock B offset C length D -----------+ | |
479 | * 2: +- file E startblock F offset G length H --------------+ | |
480 | * 3: +- file I startblock F offset J length K --+ | |
481 | * 4: +- file L... --+ | |
482 | * | |
483 | * Now say we want to map block (B+D) into file A at offset (C+D). Ideally, | |
484 | * we'd simply increment the length of record 1. But how do we find the record | |
485 | * that ends at (B+D-1) (i.e. record 1)? A LE lookup of (B+D-1) would return | |
486 | * record 3 because the keys are ordered first by startblock. An interval | |
487 | * query would return records 1 and 2 because they both overlap (B+D-1), and | |
488 | * from that we can pick out record 1 as the appropriate left neighbor. | |
489 | * | |
490 | * In the non-overlapped case you can do a LE lookup and decrement the cursor | |
491 | * because a record's interval must end before the next record. | |
65f1eaea CH |
492 | */ |
493 | ||
494 | /* | |
495 | * Return size of the btree block header for this btree instance. | |
496 | */ | |
497 | static inline size_t xfs_btree_block_len(struct xfs_btree_cur *cur) | |
498 | { | |
ee1a47ab CH |
499 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) { |
500 | if (cur->bc_flags & XFS_BTREE_CRC_BLOCKS) | |
501 | return XFS_BTREE_LBLOCK_CRC_LEN; | |
502 | return XFS_BTREE_LBLOCK_LEN; | |
503 | } | |
504 | if (cur->bc_flags & XFS_BTREE_CRC_BLOCKS) | |
505 | return XFS_BTREE_SBLOCK_CRC_LEN; | |
506 | return XFS_BTREE_SBLOCK_LEN; | |
65f1eaea CH |
507 | } |
508 | ||
509 | /* | |
510 | * Return size of btree block pointers for this btree instance. | |
511 | */ | |
512 | static inline size_t xfs_btree_ptr_len(struct xfs_btree_cur *cur) | |
513 | { | |
514 | return (cur->bc_flags & XFS_BTREE_LONG_PTRS) ? | |
515 | sizeof(__be64) : sizeof(__be32); | |
516 | } | |
517 | ||
518 | /* | |
519 | * Calculate offset of the n-th record in a btree block. | |
520 | */ | |
521 | STATIC size_t | |
522 | xfs_btree_rec_offset( | |
523 | struct xfs_btree_cur *cur, | |
524 | int n) | |
525 | { | |
526 | return xfs_btree_block_len(cur) + | |
527 | (n - 1) * cur->bc_ops->rec_len; | |
528 | } | |
529 | ||
530 | /* | |
531 | * Calculate offset of the n-th key in a btree block. | |
532 | */ | |
533 | STATIC size_t | |
534 | xfs_btree_key_offset( | |
535 | struct xfs_btree_cur *cur, | |
536 | int n) | |
537 | { | |
538 | return xfs_btree_block_len(cur) + | |
539 | (n - 1) * cur->bc_ops->key_len; | |
540 | } | |
541 | ||
2c813ad6 DW |
542 | /* |
543 | * Calculate offset of the n-th high key in a btree block. | |
544 | */ | |
545 | STATIC size_t | |
546 | xfs_btree_high_key_offset( | |
547 | struct xfs_btree_cur *cur, | |
548 | int n) | |
549 | { | |
550 | return xfs_btree_block_len(cur) + | |
551 | (n - 1) * cur->bc_ops->key_len + (cur->bc_ops->key_len / 2); | |
552 | } | |
553 | ||
65f1eaea CH |
554 | /* |
555 | * Calculate offset of the n-th block pointer in a btree block. | |
556 | */ | |
557 | STATIC size_t | |
558 | xfs_btree_ptr_offset( | |
559 | struct xfs_btree_cur *cur, | |
560 | int n, | |
561 | int level) | |
562 | { | |
563 | return xfs_btree_block_len(cur) + | |
564 | cur->bc_ops->get_maxrecs(cur, level) * cur->bc_ops->key_len + | |
565 | (n - 1) * xfs_btree_ptr_len(cur); | |
566 | } | |
567 | ||
568 | /* | |
569 | * Return a pointer to the n-th record in the btree block. | |
570 | */ | |
571 | STATIC union xfs_btree_rec * | |
572 | xfs_btree_rec_addr( | |
573 | struct xfs_btree_cur *cur, | |
574 | int n, | |
575 | struct xfs_btree_block *block) | |
576 | { | |
577 | return (union xfs_btree_rec *) | |
578 | ((char *)block + xfs_btree_rec_offset(cur, n)); | |
579 | } | |
580 | ||
581 | /* | |
582 | * Return a pointer to the n-th key in the btree block. | |
583 | */ | |
584 | STATIC union xfs_btree_key * | |
585 | xfs_btree_key_addr( | |
586 | struct xfs_btree_cur *cur, | |
587 | int n, | |
588 | struct xfs_btree_block *block) | |
589 | { | |
590 | return (union xfs_btree_key *) | |
591 | ((char *)block + xfs_btree_key_offset(cur, n)); | |
592 | } | |
593 | ||
2c813ad6 DW |
594 | /* |
595 | * Return a pointer to the n-th high key in the btree block. | |
596 | */ | |
597 | STATIC union xfs_btree_key * | |
598 | xfs_btree_high_key_addr( | |
599 | struct xfs_btree_cur *cur, | |
600 | int n, | |
601 | struct xfs_btree_block *block) | |
602 | { | |
603 | return (union xfs_btree_key *) | |
604 | ((char *)block + xfs_btree_high_key_offset(cur, n)); | |
605 | } | |
606 | ||
65f1eaea CH |
607 | /* |
608 | * Return a pointer to the n-th block pointer in the btree block. | |
609 | */ | |
610 | STATIC union xfs_btree_ptr * | |
611 | xfs_btree_ptr_addr( | |
612 | struct xfs_btree_cur *cur, | |
613 | int n, | |
614 | struct xfs_btree_block *block) | |
615 | { | |
616 | int level = xfs_btree_get_level(block); | |
617 | ||
618 | ASSERT(block->bb_level != 0); | |
619 | ||
620 | return (union xfs_btree_ptr *) | |
621 | ((char *)block + xfs_btree_ptr_offset(cur, n, level)); | |
622 | } | |
623 | ||
8186e517 | 624 | /* |
1cb93863 | 625 | * Get the root block which is stored in the inode. |
8186e517 CH |
626 | * |
627 | * For now this btree implementation assumes the btree root is always | |
628 | * stored in the if_broot field of an inode fork. | |
629 | */ | |
630 | STATIC struct xfs_btree_block * | |
631 | xfs_btree_get_iroot( | |
fbfb24bf | 632 | struct xfs_btree_cur *cur) |
8186e517 | 633 | { |
fbfb24bf | 634 | struct xfs_ifork *ifp; |
8186e517 | 635 | |
fbfb24bf KN |
636 | ifp = XFS_IFORK_PTR(cur->bc_private.b.ip, cur->bc_private.b.whichfork); |
637 | return (struct xfs_btree_block *)ifp->if_broot; | |
8186e517 CH |
638 | } |
639 | ||
1da177e4 LT |
640 | /* |
641 | * Retrieve the block pointer from the cursor at the given level. | |
8186e517 | 642 | * This may be an inode btree root or from a buffer. |
1da177e4 | 643 | */ |
8186e517 | 644 | STATIC struct xfs_btree_block * /* generic btree block pointer */ |
1da177e4 | 645 | xfs_btree_get_block( |
8186e517 | 646 | struct xfs_btree_cur *cur, /* btree cursor */ |
1da177e4 | 647 | int level, /* level in btree */ |
8186e517 | 648 | struct xfs_buf **bpp) /* buffer containing the block */ |
1da177e4 | 649 | { |
8186e517 CH |
650 | if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) && |
651 | (level == cur->bc_nlevels - 1)) { | |
652 | *bpp = NULL; | |
653 | return xfs_btree_get_iroot(cur); | |
1da177e4 | 654 | } |
8186e517 CH |
655 | |
656 | *bpp = cur->bc_bufs[level]; | |
657 | return XFS_BUF_TO_BLOCK(*bpp); | |
1da177e4 LT |
658 | } |
659 | ||
660 | /* | |
661 | * Get a buffer for the block, return it with no data read. | |
662 | * Long-form addressing. | |
663 | */ | |
664 | xfs_buf_t * /* buffer for fsbno */ | |
665 | xfs_btree_get_bufl( | |
666 | xfs_mount_t *mp, /* file system mount point */ | |
667 | xfs_trans_t *tp, /* transaction pointer */ | |
668 | xfs_fsblock_t fsbno, /* file system block number */ | |
669 | uint lock) /* lock flags for get_buf */ | |
670 | { | |
1da177e4 LT |
671 | xfs_daddr_t d; /* real disk block address */ |
672 | ||
673 | ASSERT(fsbno != NULLFSBLOCK); | |
674 | d = XFS_FSB_TO_DADDR(mp, fsbno); | |
36de9556 | 675 | return xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, lock); |
1da177e4 LT |
676 | } |
677 | ||
678 | /* | |
679 | * Get a buffer for the block, return it with no data read. | |
680 | * Short-form addressing. | |
681 | */ | |
682 | xfs_buf_t * /* buffer for agno/agbno */ | |
683 | xfs_btree_get_bufs( | |
684 | xfs_mount_t *mp, /* file system mount point */ | |
685 | xfs_trans_t *tp, /* transaction pointer */ | |
686 | xfs_agnumber_t agno, /* allocation group number */ | |
687 | xfs_agblock_t agbno, /* allocation group block number */ | |
688 | uint lock) /* lock flags for get_buf */ | |
689 | { | |
1da177e4 LT |
690 | xfs_daddr_t d; /* real disk block address */ |
691 | ||
692 | ASSERT(agno != NULLAGNUMBER); | |
693 | ASSERT(agbno != NULLAGBLOCK); | |
694 | d = XFS_AGB_TO_DADDR(mp, agno, agbno); | |
36de9556 | 695 | return xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, lock); |
1da177e4 LT |
696 | } |
697 | ||
1da177e4 LT |
698 | /* |
699 | * Check for the cursor referring to the last block at the given level. | |
700 | */ | |
701 | int /* 1=is last block, 0=not last block */ | |
702 | xfs_btree_islastblock( | |
703 | xfs_btree_cur_t *cur, /* btree cursor */ | |
704 | int level) /* level to check */ | |
705 | { | |
7cc95a82 | 706 | struct xfs_btree_block *block; /* generic btree block pointer */ |
1da177e4 LT |
707 | xfs_buf_t *bp; /* buffer containing block */ |
708 | ||
709 | block = xfs_btree_get_block(cur, level, &bp); | |
710 | xfs_btree_check_block(cur, block, level, bp); | |
e99ab90d | 711 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) |
d5cf09ba | 712 | return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK); |
1da177e4 | 713 | else |
69ef921b | 714 | return block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK); |
1da177e4 LT |
715 | } |
716 | ||
cdcf4333 CH |
717 | /* |
718 | * Change the cursor to point to the first record at the given level. | |
719 | * Other levels are unaffected. | |
720 | */ | |
3cc7524c | 721 | STATIC int /* success=1, failure=0 */ |
cdcf4333 CH |
722 | xfs_btree_firstrec( |
723 | xfs_btree_cur_t *cur, /* btree cursor */ | |
724 | int level) /* level to change */ | |
725 | { | |
7cc95a82 | 726 | struct xfs_btree_block *block; /* generic btree block pointer */ |
cdcf4333 CH |
727 | xfs_buf_t *bp; /* buffer containing block */ |
728 | ||
729 | /* | |
730 | * Get the block pointer for this level. | |
731 | */ | |
732 | block = xfs_btree_get_block(cur, level, &bp); | |
733 | xfs_btree_check_block(cur, block, level, bp); | |
734 | /* | |
735 | * It's empty, there is no such record. | |
736 | */ | |
f2277f06 | 737 | if (!block->bb_numrecs) |
cdcf4333 CH |
738 | return 0; |
739 | /* | |
740 | * Set the ptr value to 1, that's the first record/key. | |
741 | */ | |
742 | cur->bc_ptrs[level] = 1; | |
743 | return 1; | |
744 | } | |
745 | ||
1da177e4 LT |
746 | /* |
747 | * Change the cursor to point to the last record in the current block | |
748 | * at the given level. Other levels are unaffected. | |
749 | */ | |
3cc7524c | 750 | STATIC int /* success=1, failure=0 */ |
1da177e4 LT |
751 | xfs_btree_lastrec( |
752 | xfs_btree_cur_t *cur, /* btree cursor */ | |
753 | int level) /* level to change */ | |
754 | { | |
7cc95a82 | 755 | struct xfs_btree_block *block; /* generic btree block pointer */ |
1da177e4 LT |
756 | xfs_buf_t *bp; /* buffer containing block */ |
757 | ||
758 | /* | |
759 | * Get the block pointer for this level. | |
760 | */ | |
761 | block = xfs_btree_get_block(cur, level, &bp); | |
762 | xfs_btree_check_block(cur, block, level, bp); | |
763 | /* | |
764 | * It's empty, there is no such record. | |
765 | */ | |
f2277f06 | 766 | if (!block->bb_numrecs) |
1da177e4 LT |
767 | return 0; |
768 | /* | |
769 | * Set the ptr value to numrecs, that's the last record/key. | |
770 | */ | |
f2277f06 | 771 | cur->bc_ptrs[level] = be16_to_cpu(block->bb_numrecs); |
1da177e4 LT |
772 | return 1; |
773 | } | |
774 | ||
775 | /* | |
776 | * Compute first and last byte offsets for the fields given. | |
777 | * Interprets the offsets table, which contains struct field offsets. | |
778 | */ | |
779 | void | |
780 | xfs_btree_offsets( | |
781 | __int64_t fields, /* bitmask of fields */ | |
782 | const short *offsets, /* table of field offsets */ | |
783 | int nbits, /* number of bits to inspect */ | |
784 | int *first, /* output: first byte offset */ | |
785 | int *last) /* output: last byte offset */ | |
786 | { | |
787 | int i; /* current bit number */ | |
788 | __int64_t imask; /* mask for current bit number */ | |
789 | ||
790 | ASSERT(fields != 0); | |
791 | /* | |
792 | * Find the lowest bit, so the first byte offset. | |
793 | */ | |
794 | for (i = 0, imask = 1LL; ; i++, imask <<= 1) { | |
795 | if (imask & fields) { | |
796 | *first = offsets[i]; | |
797 | break; | |
798 | } | |
799 | } | |
800 | /* | |
801 | * Find the highest bit, so the last byte offset. | |
802 | */ | |
803 | for (i = nbits - 1, imask = 1LL << i; ; i--, imask >>= 1) { | |
804 | if (imask & fields) { | |
805 | *last = offsets[i + 1] - 1; | |
806 | break; | |
807 | } | |
808 | } | |
809 | } | |
810 | ||
811 | /* | |
812 | * Get a buffer for the block, return it read in. | |
813 | * Long-form addressing. | |
814 | */ | |
3d3e6f64 | 815 | int |
1da177e4 | 816 | xfs_btree_read_bufl( |
3d3e6f64 DC |
817 | struct xfs_mount *mp, /* file system mount point */ |
818 | struct xfs_trans *tp, /* transaction pointer */ | |
819 | xfs_fsblock_t fsbno, /* file system block number */ | |
820 | uint lock, /* lock flags for read_buf */ | |
821 | struct xfs_buf **bpp, /* buffer for fsbno */ | |
822 | int refval, /* ref count value for buffer */ | |
1813dd64 | 823 | const struct xfs_buf_ops *ops) |
1da177e4 | 824 | { |
3d3e6f64 | 825 | struct xfs_buf *bp; /* return value */ |
1da177e4 | 826 | xfs_daddr_t d; /* real disk block address */ |
3d3e6f64 | 827 | int error; |
1da177e4 | 828 | |
d5a91bae DW |
829 | if (!XFS_FSB_SANITY_CHECK(mp, fsbno)) |
830 | return -EFSCORRUPTED; | |
1da177e4 | 831 | d = XFS_FSB_TO_DADDR(mp, fsbno); |
c3f8fc73 | 832 | error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, d, |
1813dd64 | 833 | mp->m_bsize, lock, &bp, ops); |
c3f8fc73 | 834 | if (error) |
1da177e4 | 835 | return error; |
821eb21d | 836 | if (bp) |
38f23232 | 837 | xfs_buf_set_ref(bp, refval); |
1da177e4 LT |
838 | *bpp = bp; |
839 | return 0; | |
840 | } | |
841 | ||
1da177e4 LT |
842 | /* |
843 | * Read-ahead the block, don't wait for it, don't return a buffer. | |
844 | * Long-form addressing. | |
845 | */ | |
846 | /* ARGSUSED */ | |
847 | void | |
848 | xfs_btree_reada_bufl( | |
3d3e6f64 DC |
849 | struct xfs_mount *mp, /* file system mount point */ |
850 | xfs_fsblock_t fsbno, /* file system block number */ | |
851 | xfs_extlen_t count, /* count of filesystem blocks */ | |
1813dd64 | 852 | const struct xfs_buf_ops *ops) |
1da177e4 LT |
853 | { |
854 | xfs_daddr_t d; | |
855 | ||
856 | ASSERT(fsbno != NULLFSBLOCK); | |
857 | d = XFS_FSB_TO_DADDR(mp, fsbno); | |
1813dd64 | 858 | xfs_buf_readahead(mp->m_ddev_targp, d, mp->m_bsize * count, ops); |
1da177e4 LT |
859 | } |
860 | ||
861 | /* | |
862 | * Read-ahead the block, don't wait for it, don't return a buffer. | |
863 | * Short-form addressing. | |
864 | */ | |
865 | /* ARGSUSED */ | |
866 | void | |
867 | xfs_btree_reada_bufs( | |
3d3e6f64 DC |
868 | struct xfs_mount *mp, /* file system mount point */ |
869 | xfs_agnumber_t agno, /* allocation group number */ | |
870 | xfs_agblock_t agbno, /* allocation group block number */ | |
871 | xfs_extlen_t count, /* count of filesystem blocks */ | |
1813dd64 | 872 | const struct xfs_buf_ops *ops) |
1da177e4 LT |
873 | { |
874 | xfs_daddr_t d; | |
875 | ||
876 | ASSERT(agno != NULLAGNUMBER); | |
877 | ASSERT(agbno != NULLAGBLOCK); | |
878 | d = XFS_AGB_TO_DADDR(mp, agno, agbno); | |
1813dd64 | 879 | xfs_buf_readahead(mp->m_ddev_targp, d, mp->m_bsize * count, ops); |
1da177e4 LT |
880 | } |
881 | ||
b524bfee CH |
882 | STATIC int |
883 | xfs_btree_readahead_lblock( | |
884 | struct xfs_btree_cur *cur, | |
885 | int lr, | |
886 | struct xfs_btree_block *block) | |
887 | { | |
888 | int rval = 0; | |
d5cf09ba CH |
889 | xfs_fsblock_t left = be64_to_cpu(block->bb_u.l.bb_leftsib); |
890 | xfs_fsblock_t right = be64_to_cpu(block->bb_u.l.bb_rightsib); | |
b524bfee | 891 | |
d5cf09ba | 892 | if ((lr & XFS_BTCUR_LEFTRA) && left != NULLFSBLOCK) { |
3d3e6f64 | 893 | xfs_btree_reada_bufl(cur->bc_mp, left, 1, |
1813dd64 | 894 | cur->bc_ops->buf_ops); |
b524bfee CH |
895 | rval++; |
896 | } | |
897 | ||
d5cf09ba | 898 | if ((lr & XFS_BTCUR_RIGHTRA) && right != NULLFSBLOCK) { |
3d3e6f64 | 899 | xfs_btree_reada_bufl(cur->bc_mp, right, 1, |
1813dd64 | 900 | cur->bc_ops->buf_ops); |
b524bfee CH |
901 | rval++; |
902 | } | |
903 | ||
904 | return rval; | |
905 | } | |
906 | ||
907 | STATIC int | |
908 | xfs_btree_readahead_sblock( | |
909 | struct xfs_btree_cur *cur, | |
910 | int lr, | |
911 | struct xfs_btree_block *block) | |
912 | { | |
913 | int rval = 0; | |
914 | xfs_agblock_t left = be32_to_cpu(block->bb_u.s.bb_leftsib); | |
915 | xfs_agblock_t right = be32_to_cpu(block->bb_u.s.bb_rightsib); | |
916 | ||
917 | ||
918 | if ((lr & XFS_BTCUR_LEFTRA) && left != NULLAGBLOCK) { | |
919 | xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno, | |
1813dd64 | 920 | left, 1, cur->bc_ops->buf_ops); |
b524bfee CH |
921 | rval++; |
922 | } | |
923 | ||
924 | if ((lr & XFS_BTCUR_RIGHTRA) && right != NULLAGBLOCK) { | |
925 | xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno, | |
1813dd64 | 926 | right, 1, cur->bc_ops->buf_ops); |
b524bfee CH |
927 | rval++; |
928 | } | |
929 | ||
930 | return rval; | |
931 | } | |
932 | ||
1da177e4 LT |
933 | /* |
934 | * Read-ahead btree blocks, at the given level. | |
935 | * Bits in lr are set from XFS_BTCUR_{LEFT,RIGHT}RA. | |
936 | */ | |
3cc7524c | 937 | STATIC int |
b524bfee CH |
938 | xfs_btree_readahead( |
939 | struct xfs_btree_cur *cur, /* btree cursor */ | |
1da177e4 LT |
940 | int lev, /* level in btree */ |
941 | int lr) /* left/right bits */ | |
942 | { | |
b524bfee CH |
943 | struct xfs_btree_block *block; |
944 | ||
945 | /* | |
946 | * No readahead needed if we are at the root level and the | |
947 | * btree root is stored in the inode. | |
948 | */ | |
949 | if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) && | |
950 | (lev == cur->bc_nlevels - 1)) | |
951 | return 0; | |
952 | ||
953 | if ((cur->bc_ra[lev] | lr) == cur->bc_ra[lev]) | |
954 | return 0; | |
1da177e4 | 955 | |
1da177e4 | 956 | cur->bc_ra[lev] |= lr; |
b524bfee CH |
957 | block = XFS_BUF_TO_BLOCK(cur->bc_bufs[lev]); |
958 | ||
959 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) | |
960 | return xfs_btree_readahead_lblock(cur, lr, block); | |
961 | return xfs_btree_readahead_sblock(cur, lr, block); | |
1da177e4 LT |
962 | } |
963 | ||
21b5c978 DC |
964 | STATIC xfs_daddr_t |
965 | xfs_btree_ptr_to_daddr( | |
966 | struct xfs_btree_cur *cur, | |
967 | union xfs_btree_ptr *ptr) | |
968 | { | |
969 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) { | |
d5cf09ba | 970 | ASSERT(ptr->l != cpu_to_be64(NULLFSBLOCK)); |
21b5c978 DC |
971 | |
972 | return XFS_FSB_TO_DADDR(cur->bc_mp, be64_to_cpu(ptr->l)); | |
973 | } else { | |
974 | ASSERT(cur->bc_private.a.agno != NULLAGNUMBER); | |
975 | ASSERT(ptr->s != cpu_to_be32(NULLAGBLOCK)); | |
976 | ||
977 | return XFS_AGB_TO_DADDR(cur->bc_mp, cur->bc_private.a.agno, | |
978 | be32_to_cpu(ptr->s)); | |
979 | } | |
980 | } | |
981 | ||
982 | /* | |
983 | * Readahead @count btree blocks at the given @ptr location. | |
984 | * | |
985 | * We don't need to care about long or short form btrees here as we have a | |
986 | * method of converting the ptr directly to a daddr available to us. | |
987 | */ | |
988 | STATIC void | |
989 | xfs_btree_readahead_ptr( | |
990 | struct xfs_btree_cur *cur, | |
991 | union xfs_btree_ptr *ptr, | |
992 | xfs_extlen_t count) | |
993 | { | |
994 | xfs_buf_readahead(cur->bc_mp->m_ddev_targp, | |
995 | xfs_btree_ptr_to_daddr(cur, ptr), | |
996 | cur->bc_mp->m_bsize * count, cur->bc_ops->buf_ops); | |
997 | } | |
998 | ||
1da177e4 LT |
999 | /* |
1000 | * Set the buffer for level "lev" in the cursor to bp, releasing | |
1001 | * any previous buffer. | |
1002 | */ | |
c0e59e1a | 1003 | STATIC void |
1da177e4 LT |
1004 | xfs_btree_setbuf( |
1005 | xfs_btree_cur_t *cur, /* btree cursor */ | |
1006 | int lev, /* level in btree */ | |
1007 | xfs_buf_t *bp) /* new buffer to set */ | |
1008 | { | |
7cc95a82 | 1009 | struct xfs_btree_block *b; /* btree block */ |
1da177e4 | 1010 | |
c0e59e1a CH |
1011 | if (cur->bc_bufs[lev]) |
1012 | xfs_trans_brelse(cur->bc_tp, cur->bc_bufs[lev]); | |
1da177e4 LT |
1013 | cur->bc_bufs[lev] = bp; |
1014 | cur->bc_ra[lev] = 0; | |
c0e59e1a | 1015 | |
1da177e4 | 1016 | b = XFS_BUF_TO_BLOCK(bp); |
e99ab90d | 1017 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) { |
d5cf09ba | 1018 | if (b->bb_u.l.bb_leftsib == cpu_to_be64(NULLFSBLOCK)) |
1da177e4 | 1019 | cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA; |
d5cf09ba | 1020 | if (b->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK)) |
1da177e4 LT |
1021 | cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA; |
1022 | } else { | |
69ef921b | 1023 | if (b->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK)) |
1da177e4 | 1024 | cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA; |
69ef921b | 1025 | if (b->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK)) |
1da177e4 LT |
1026 | cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA; |
1027 | } | |
1028 | } | |
637aa50f CH |
1029 | |
1030 | STATIC int | |
1031 | xfs_btree_ptr_is_null( | |
1032 | struct xfs_btree_cur *cur, | |
1033 | union xfs_btree_ptr *ptr) | |
1034 | { | |
1035 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) | |
d5cf09ba | 1036 | return ptr->l == cpu_to_be64(NULLFSBLOCK); |
637aa50f | 1037 | else |
69ef921b | 1038 | return ptr->s == cpu_to_be32(NULLAGBLOCK); |
637aa50f CH |
1039 | } |
1040 | ||
4b22a571 CH |
1041 | STATIC void |
1042 | xfs_btree_set_ptr_null( | |
1043 | struct xfs_btree_cur *cur, | |
1044 | union xfs_btree_ptr *ptr) | |
1045 | { | |
1046 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) | |
d5cf09ba | 1047 | ptr->l = cpu_to_be64(NULLFSBLOCK); |
4b22a571 CH |
1048 | else |
1049 | ptr->s = cpu_to_be32(NULLAGBLOCK); | |
1050 | } | |
1051 | ||
637aa50f CH |
1052 | /* |
1053 | * Get/set/init sibling pointers | |
1054 | */ | |
1055 | STATIC void | |
1056 | xfs_btree_get_sibling( | |
1057 | struct xfs_btree_cur *cur, | |
1058 | struct xfs_btree_block *block, | |
1059 | union xfs_btree_ptr *ptr, | |
1060 | int lr) | |
1061 | { | |
1062 | ASSERT(lr == XFS_BB_LEFTSIB || lr == XFS_BB_RIGHTSIB); | |
1063 | ||
1064 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) { | |
1065 | if (lr == XFS_BB_RIGHTSIB) | |
1066 | ptr->l = block->bb_u.l.bb_rightsib; | |
1067 | else | |
1068 | ptr->l = block->bb_u.l.bb_leftsib; | |
1069 | } else { | |
1070 | if (lr == XFS_BB_RIGHTSIB) | |
1071 | ptr->s = block->bb_u.s.bb_rightsib; | |
1072 | else | |
1073 | ptr->s = block->bb_u.s.bb_leftsib; | |
1074 | } | |
1075 | } | |
1076 | ||
f5eb8e7c CH |
1077 | STATIC void |
1078 | xfs_btree_set_sibling( | |
1079 | struct xfs_btree_cur *cur, | |
1080 | struct xfs_btree_block *block, | |
1081 | union xfs_btree_ptr *ptr, | |
1082 | int lr) | |
1083 | { | |
1084 | ASSERT(lr == XFS_BB_LEFTSIB || lr == XFS_BB_RIGHTSIB); | |
1085 | ||
1086 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) { | |
1087 | if (lr == XFS_BB_RIGHTSIB) | |
1088 | block->bb_u.l.bb_rightsib = ptr->l; | |
1089 | else | |
1090 | block->bb_u.l.bb_leftsib = ptr->l; | |
1091 | } else { | |
1092 | if (lr == XFS_BB_RIGHTSIB) | |
1093 | block->bb_u.s.bb_rightsib = ptr->s; | |
1094 | else | |
1095 | block->bb_u.s.bb_leftsib = ptr->s; | |
1096 | } | |
1097 | } | |
1098 | ||
ee1a47ab CH |
1099 | void |
1100 | xfs_btree_init_block_int( | |
1101 | struct xfs_mount *mp, | |
1102 | struct xfs_btree_block *buf, | |
1103 | xfs_daddr_t blkno, | |
b6f41e44 | 1104 | xfs_btnum_t btnum, |
ee1a47ab CH |
1105 | __u16 level, |
1106 | __u16 numrecs, | |
1107 | __u64 owner, | |
1108 | unsigned int flags) | |
1109 | { | |
f88ae46b | 1110 | int crc = xfs_sb_version_hascrc(&mp->m_sb); |
b6f41e44 | 1111 | __u32 magic = xfs_btree_magic(crc, btnum); |
f88ae46b | 1112 | |
ee1a47ab CH |
1113 | buf->bb_magic = cpu_to_be32(magic); |
1114 | buf->bb_level = cpu_to_be16(level); | |
1115 | buf->bb_numrecs = cpu_to_be16(numrecs); | |
1116 | ||
1117 | if (flags & XFS_BTREE_LONG_PTRS) { | |
d5cf09ba CH |
1118 | buf->bb_u.l.bb_leftsib = cpu_to_be64(NULLFSBLOCK); |
1119 | buf->bb_u.l.bb_rightsib = cpu_to_be64(NULLFSBLOCK); | |
f88ae46b | 1120 | if (crc) { |
ee1a47ab CH |
1121 | buf->bb_u.l.bb_blkno = cpu_to_be64(blkno); |
1122 | buf->bb_u.l.bb_owner = cpu_to_be64(owner); | |
ce748eaa | 1123 | uuid_copy(&buf->bb_u.l.bb_uuid, &mp->m_sb.sb_meta_uuid); |
ee1a47ab | 1124 | buf->bb_u.l.bb_pad = 0; |
b58fa554 | 1125 | buf->bb_u.l.bb_lsn = 0; |
ee1a47ab CH |
1126 | } |
1127 | } else { | |
1128 | /* owner is a 32 bit value on short blocks */ | |
1129 | __u32 __owner = (__u32)owner; | |
1130 | ||
1131 | buf->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK); | |
1132 | buf->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK); | |
f88ae46b | 1133 | if (crc) { |
ee1a47ab CH |
1134 | buf->bb_u.s.bb_blkno = cpu_to_be64(blkno); |
1135 | buf->bb_u.s.bb_owner = cpu_to_be32(__owner); | |
ce748eaa | 1136 | uuid_copy(&buf->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid); |
b58fa554 | 1137 | buf->bb_u.s.bb_lsn = 0; |
ee1a47ab CH |
1138 | } |
1139 | } | |
1140 | } | |
1141 | ||
b64f3a39 | 1142 | void |
f5eb8e7c | 1143 | xfs_btree_init_block( |
b64f3a39 DC |
1144 | struct xfs_mount *mp, |
1145 | struct xfs_buf *bp, | |
b6f41e44 | 1146 | xfs_btnum_t btnum, |
b64f3a39 DC |
1147 | __u16 level, |
1148 | __u16 numrecs, | |
ee1a47ab | 1149 | __u64 owner, |
b64f3a39 | 1150 | unsigned int flags) |
f5eb8e7c | 1151 | { |
ee1a47ab | 1152 | xfs_btree_init_block_int(mp, XFS_BUF_TO_BLOCK(bp), bp->b_bn, |
b6f41e44 | 1153 | btnum, level, numrecs, owner, flags); |
f5eb8e7c CH |
1154 | } |
1155 | ||
b64f3a39 DC |
1156 | STATIC void |
1157 | xfs_btree_init_block_cur( | |
1158 | struct xfs_btree_cur *cur, | |
ee1a47ab | 1159 | struct xfs_buf *bp, |
b64f3a39 | 1160 | int level, |
ee1a47ab | 1161 | int numrecs) |
b64f3a39 | 1162 | { |
af7d20fd | 1163 | __u64 owner; |
ee1a47ab CH |
1164 | |
1165 | /* | |
1166 | * we can pull the owner from the cursor right now as the different | |
1167 | * owners align directly with the pointer size of the btree. This may | |
1168 | * change in future, but is safe for current users of the generic btree | |
1169 | * code. | |
1170 | */ | |
1171 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) | |
1172 | owner = cur->bc_private.b.ip->i_ino; | |
1173 | else | |
1174 | owner = cur->bc_private.a.agno; | |
1175 | ||
1176 | xfs_btree_init_block_int(cur->bc_mp, XFS_BUF_TO_BLOCK(bp), bp->b_bn, | |
b6f41e44 | 1177 | cur->bc_btnum, level, numrecs, |
ee1a47ab | 1178 | owner, cur->bc_flags); |
b64f3a39 DC |
1179 | } |
1180 | ||
278d0ca1 CH |
1181 | /* |
1182 | * Return true if ptr is the last record in the btree and | |
ee1a47ab | 1183 | * we need to track updates to this record. The decision |
278d0ca1 CH |
1184 | * will be further refined in the update_lastrec method. |
1185 | */ | |
1186 | STATIC int | |
1187 | xfs_btree_is_lastrec( | |
1188 | struct xfs_btree_cur *cur, | |
1189 | struct xfs_btree_block *block, | |
1190 | int level) | |
1191 | { | |
1192 | union xfs_btree_ptr ptr; | |
1193 | ||
1194 | if (level > 0) | |
1195 | return 0; | |
1196 | if (!(cur->bc_flags & XFS_BTREE_LASTREC_UPDATE)) | |
1197 | return 0; | |
1198 | ||
1199 | xfs_btree_get_sibling(cur, block, &ptr, XFS_BB_RIGHTSIB); | |
1200 | if (!xfs_btree_ptr_is_null(cur, &ptr)) | |
1201 | return 0; | |
1202 | return 1; | |
1203 | } | |
1204 | ||
f5eb8e7c CH |
1205 | STATIC void |
1206 | xfs_btree_buf_to_ptr( | |
1207 | struct xfs_btree_cur *cur, | |
1208 | struct xfs_buf *bp, | |
1209 | union xfs_btree_ptr *ptr) | |
1210 | { | |
1211 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) | |
1212 | ptr->l = cpu_to_be64(XFS_DADDR_TO_FSB(cur->bc_mp, | |
1213 | XFS_BUF_ADDR(bp))); | |
1214 | else { | |
9d87c319 | 1215 | ptr->s = cpu_to_be32(xfs_daddr_to_agbno(cur->bc_mp, |
f5eb8e7c CH |
1216 | XFS_BUF_ADDR(bp))); |
1217 | } | |
1218 | } | |
1219 | ||
637aa50f CH |
1220 | STATIC void |
1221 | xfs_btree_set_refs( | |
1222 | struct xfs_btree_cur *cur, | |
1223 | struct xfs_buf *bp) | |
1224 | { | |
1225 | switch (cur->bc_btnum) { | |
1226 | case XFS_BTNUM_BNO: | |
1227 | case XFS_BTNUM_CNT: | |
38f23232 | 1228 | xfs_buf_set_ref(bp, XFS_ALLOC_BTREE_REF); |
637aa50f CH |
1229 | break; |
1230 | case XFS_BTNUM_INO: | |
aafc3c24 | 1231 | case XFS_BTNUM_FINO: |
38f23232 | 1232 | xfs_buf_set_ref(bp, XFS_INO_BTREE_REF); |
637aa50f CH |
1233 | break; |
1234 | case XFS_BTNUM_BMAP: | |
38f23232 | 1235 | xfs_buf_set_ref(bp, XFS_BMAP_BTREE_REF); |
637aa50f | 1236 | break; |
035e00ac DW |
1237 | case XFS_BTNUM_RMAP: |
1238 | xfs_buf_set_ref(bp, XFS_RMAP_BTREE_REF); | |
1239 | break; | |
1946b91c DW |
1240 | case XFS_BTNUM_REFC: |
1241 | xfs_buf_set_ref(bp, XFS_REFC_BTREE_REF); | |
1242 | break; | |
637aa50f CH |
1243 | default: |
1244 | ASSERT(0); | |
1245 | } | |
1246 | } | |
1247 | ||
f5eb8e7c CH |
1248 | STATIC int |
1249 | xfs_btree_get_buf_block( | |
1250 | struct xfs_btree_cur *cur, | |
1251 | union xfs_btree_ptr *ptr, | |
1252 | int flags, | |
1253 | struct xfs_btree_block **block, | |
1254 | struct xfs_buf **bpp) | |
1255 | { | |
1256 | struct xfs_mount *mp = cur->bc_mp; | |
1257 | xfs_daddr_t d; | |
1258 | ||
1259 | /* need to sort out how callers deal with failures first */ | |
0cadda1c | 1260 | ASSERT(!(flags & XBF_TRYLOCK)); |
f5eb8e7c CH |
1261 | |
1262 | d = xfs_btree_ptr_to_daddr(cur, ptr); | |
1263 | *bpp = xfs_trans_get_buf(cur->bc_tp, mp->m_ddev_targp, d, | |
1264 | mp->m_bsize, flags); | |
1265 | ||
2a30f36d | 1266 | if (!*bpp) |
2451337d | 1267 | return -ENOMEM; |
f5eb8e7c | 1268 | |
1813dd64 | 1269 | (*bpp)->b_ops = cur->bc_ops->buf_ops; |
f5eb8e7c CH |
1270 | *block = XFS_BUF_TO_BLOCK(*bpp); |
1271 | return 0; | |
1272 | } | |
1273 | ||
637aa50f CH |
1274 | /* |
1275 | * Read in the buffer at the given ptr and return the buffer and | |
1276 | * the block pointer within the buffer. | |
1277 | */ | |
1278 | STATIC int | |
1279 | xfs_btree_read_buf_block( | |
1280 | struct xfs_btree_cur *cur, | |
1281 | union xfs_btree_ptr *ptr, | |
637aa50f CH |
1282 | int flags, |
1283 | struct xfs_btree_block **block, | |
1284 | struct xfs_buf **bpp) | |
1285 | { | |
1286 | struct xfs_mount *mp = cur->bc_mp; | |
1287 | xfs_daddr_t d; | |
1288 | int error; | |
1289 | ||
1290 | /* need to sort out how callers deal with failures first */ | |
0cadda1c | 1291 | ASSERT(!(flags & XBF_TRYLOCK)); |
637aa50f CH |
1292 | |
1293 | d = xfs_btree_ptr_to_daddr(cur, ptr); | |
1294 | error = xfs_trans_read_buf(mp, cur->bc_tp, mp->m_ddev_targp, d, | |
3d3e6f64 | 1295 | mp->m_bsize, flags, bpp, |
1813dd64 | 1296 | cur->bc_ops->buf_ops); |
637aa50f CH |
1297 | if (error) |
1298 | return error; | |
1299 | ||
637aa50f CH |
1300 | xfs_btree_set_refs(cur, *bpp); |
1301 | *block = XFS_BUF_TO_BLOCK(*bpp); | |
3d3e6f64 | 1302 | return 0; |
637aa50f CH |
1303 | } |
1304 | ||
38bb7423 CH |
1305 | /* |
1306 | * Copy keys from one btree block to another. | |
1307 | */ | |
1308 | STATIC void | |
1309 | xfs_btree_copy_keys( | |
1310 | struct xfs_btree_cur *cur, | |
1311 | union xfs_btree_key *dst_key, | |
1312 | union xfs_btree_key *src_key, | |
1313 | int numkeys) | |
1314 | { | |
1315 | ASSERT(numkeys >= 0); | |
1316 | memcpy(dst_key, src_key, numkeys * cur->bc_ops->key_len); | |
1317 | } | |
1318 | ||
278d0ca1 CH |
1319 | /* |
1320 | * Copy records from one btree block to another. | |
1321 | */ | |
1322 | STATIC void | |
1323 | xfs_btree_copy_recs( | |
1324 | struct xfs_btree_cur *cur, | |
1325 | union xfs_btree_rec *dst_rec, | |
1326 | union xfs_btree_rec *src_rec, | |
1327 | int numrecs) | |
1328 | { | |
1329 | ASSERT(numrecs >= 0); | |
1330 | memcpy(dst_rec, src_rec, numrecs * cur->bc_ops->rec_len); | |
1331 | } | |
1332 | ||
9eaead51 CH |
1333 | /* |
1334 | * Copy block pointers from one btree block to another. | |
1335 | */ | |
1336 | STATIC void | |
1337 | xfs_btree_copy_ptrs( | |
1338 | struct xfs_btree_cur *cur, | |
1339 | union xfs_btree_ptr *dst_ptr, | |
1340 | union xfs_btree_ptr *src_ptr, | |
1341 | int numptrs) | |
1342 | { | |
1343 | ASSERT(numptrs >= 0); | |
1344 | memcpy(dst_ptr, src_ptr, numptrs * xfs_btree_ptr_len(cur)); | |
1345 | } | |
1346 | ||
1347 | /* | |
1348 | * Shift keys one index left/right inside a single btree block. | |
1349 | */ | |
1350 | STATIC void | |
1351 | xfs_btree_shift_keys( | |
1352 | struct xfs_btree_cur *cur, | |
1353 | union xfs_btree_key *key, | |
1354 | int dir, | |
1355 | int numkeys) | |
1356 | { | |
1357 | char *dst_key; | |
1358 | ||
1359 | ASSERT(numkeys >= 0); | |
1360 | ASSERT(dir == 1 || dir == -1); | |
1361 | ||
1362 | dst_key = (char *)key + (dir * cur->bc_ops->key_len); | |
1363 | memmove(dst_key, key, numkeys * cur->bc_ops->key_len); | |
1364 | } | |
1365 | ||
1366 | /* | |
1367 | * Shift records one index left/right inside a single btree block. | |
1368 | */ | |
1369 | STATIC void | |
1370 | xfs_btree_shift_recs( | |
1371 | struct xfs_btree_cur *cur, | |
1372 | union xfs_btree_rec *rec, | |
1373 | int dir, | |
1374 | int numrecs) | |
1375 | { | |
1376 | char *dst_rec; | |
1377 | ||
1378 | ASSERT(numrecs >= 0); | |
1379 | ASSERT(dir == 1 || dir == -1); | |
1380 | ||
1381 | dst_rec = (char *)rec + (dir * cur->bc_ops->rec_len); | |
1382 | memmove(dst_rec, rec, numrecs * cur->bc_ops->rec_len); | |
1383 | } | |
1384 | ||
1385 | /* | |
1386 | * Shift block pointers one index left/right inside a single btree block. | |
1387 | */ | |
1388 | STATIC void | |
1389 | xfs_btree_shift_ptrs( | |
1390 | struct xfs_btree_cur *cur, | |
1391 | union xfs_btree_ptr *ptr, | |
1392 | int dir, | |
1393 | int numptrs) | |
1394 | { | |
1395 | char *dst_ptr; | |
1396 | ||
1397 | ASSERT(numptrs >= 0); | |
1398 | ASSERT(dir == 1 || dir == -1); | |
1399 | ||
1400 | dst_ptr = (char *)ptr + (dir * xfs_btree_ptr_len(cur)); | |
1401 | memmove(dst_ptr, ptr, numptrs * xfs_btree_ptr_len(cur)); | |
1402 | } | |
1403 | ||
38bb7423 CH |
1404 | /* |
1405 | * Log key values from the btree block. | |
1406 | */ | |
1407 | STATIC void | |
1408 | xfs_btree_log_keys( | |
1409 | struct xfs_btree_cur *cur, | |
1410 | struct xfs_buf *bp, | |
1411 | int first, | |
1412 | int last) | |
1413 | { | |
1414 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
1415 | XFS_BTREE_TRACE_ARGBII(cur, bp, first, last); | |
1416 | ||
1417 | if (bp) { | |
61fe135c | 1418 | xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLFT_BTREE_BUF); |
38bb7423 CH |
1419 | xfs_trans_log_buf(cur->bc_tp, bp, |
1420 | xfs_btree_key_offset(cur, first), | |
1421 | xfs_btree_key_offset(cur, last + 1) - 1); | |
1422 | } else { | |
1423 | xfs_trans_log_inode(cur->bc_tp, cur->bc_private.b.ip, | |
1424 | xfs_ilog_fbroot(cur->bc_private.b.whichfork)); | |
1425 | } | |
1426 | ||
1427 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1428 | } | |
1429 | ||
278d0ca1 CH |
1430 | /* |
1431 | * Log record values from the btree block. | |
1432 | */ | |
fd6bcc5b | 1433 | void |
278d0ca1 CH |
1434 | xfs_btree_log_recs( |
1435 | struct xfs_btree_cur *cur, | |
1436 | struct xfs_buf *bp, | |
1437 | int first, | |
1438 | int last) | |
1439 | { | |
1440 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
1441 | XFS_BTREE_TRACE_ARGBII(cur, bp, first, last); | |
1442 | ||
61fe135c | 1443 | xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLFT_BTREE_BUF); |
278d0ca1 CH |
1444 | xfs_trans_log_buf(cur->bc_tp, bp, |
1445 | xfs_btree_rec_offset(cur, first), | |
1446 | xfs_btree_rec_offset(cur, last + 1) - 1); | |
1447 | ||
1448 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1449 | } | |
1450 | ||
9eaead51 CH |
1451 | /* |
1452 | * Log block pointer fields from a btree block (nonleaf). | |
1453 | */ | |
1454 | STATIC void | |
1455 | xfs_btree_log_ptrs( | |
1456 | struct xfs_btree_cur *cur, /* btree cursor */ | |
1457 | struct xfs_buf *bp, /* buffer containing btree block */ | |
1458 | int first, /* index of first pointer to log */ | |
1459 | int last) /* index of last pointer to log */ | |
1460 | { | |
1461 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
1462 | XFS_BTREE_TRACE_ARGBII(cur, bp, first, last); | |
1463 | ||
1464 | if (bp) { | |
1465 | struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); | |
1466 | int level = xfs_btree_get_level(block); | |
1467 | ||
61fe135c | 1468 | xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLFT_BTREE_BUF); |
9eaead51 CH |
1469 | xfs_trans_log_buf(cur->bc_tp, bp, |
1470 | xfs_btree_ptr_offset(cur, first, level), | |
1471 | xfs_btree_ptr_offset(cur, last + 1, level) - 1); | |
1472 | } else { | |
1473 | xfs_trans_log_inode(cur->bc_tp, cur->bc_private.b.ip, | |
1474 | xfs_ilog_fbroot(cur->bc_private.b.whichfork)); | |
1475 | } | |
1476 | ||
1477 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1478 | } | |
1479 | ||
1480 | /* | |
1481 | * Log fields from a btree block header. | |
1482 | */ | |
fd6bcc5b | 1483 | void |
9eaead51 CH |
1484 | xfs_btree_log_block( |
1485 | struct xfs_btree_cur *cur, /* btree cursor */ | |
1486 | struct xfs_buf *bp, /* buffer containing btree block */ | |
1487 | int fields) /* mask of fields: XFS_BB_... */ | |
1488 | { | |
1489 | int first; /* first byte offset logged */ | |
1490 | int last; /* last byte offset logged */ | |
1491 | static const short soffsets[] = { /* table of offsets (short) */ | |
7cc95a82 CH |
1492 | offsetof(struct xfs_btree_block, bb_magic), |
1493 | offsetof(struct xfs_btree_block, bb_level), | |
1494 | offsetof(struct xfs_btree_block, bb_numrecs), | |
1495 | offsetof(struct xfs_btree_block, bb_u.s.bb_leftsib), | |
1496 | offsetof(struct xfs_btree_block, bb_u.s.bb_rightsib), | |
ee1a47ab CH |
1497 | offsetof(struct xfs_btree_block, bb_u.s.bb_blkno), |
1498 | offsetof(struct xfs_btree_block, bb_u.s.bb_lsn), | |
1499 | offsetof(struct xfs_btree_block, bb_u.s.bb_uuid), | |
1500 | offsetof(struct xfs_btree_block, bb_u.s.bb_owner), | |
1501 | offsetof(struct xfs_btree_block, bb_u.s.bb_crc), | |
1502 | XFS_BTREE_SBLOCK_CRC_LEN | |
9eaead51 CH |
1503 | }; |
1504 | static const short loffsets[] = { /* table of offsets (long) */ | |
7cc95a82 CH |
1505 | offsetof(struct xfs_btree_block, bb_magic), |
1506 | offsetof(struct xfs_btree_block, bb_level), | |
1507 | offsetof(struct xfs_btree_block, bb_numrecs), | |
1508 | offsetof(struct xfs_btree_block, bb_u.l.bb_leftsib), | |
1509 | offsetof(struct xfs_btree_block, bb_u.l.bb_rightsib), | |
ee1a47ab CH |
1510 | offsetof(struct xfs_btree_block, bb_u.l.bb_blkno), |
1511 | offsetof(struct xfs_btree_block, bb_u.l.bb_lsn), | |
1512 | offsetof(struct xfs_btree_block, bb_u.l.bb_uuid), | |
1513 | offsetof(struct xfs_btree_block, bb_u.l.bb_owner), | |
1514 | offsetof(struct xfs_btree_block, bb_u.l.bb_crc), | |
1515 | offsetof(struct xfs_btree_block, bb_u.l.bb_pad), | |
1516 | XFS_BTREE_LBLOCK_CRC_LEN | |
9eaead51 CH |
1517 | }; |
1518 | ||
1519 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
1520 | XFS_BTREE_TRACE_ARGBI(cur, bp, fields); | |
1521 | ||
1522 | if (bp) { | |
ee1a47ab CH |
1523 | int nbits; |
1524 | ||
1525 | if (cur->bc_flags & XFS_BTREE_CRC_BLOCKS) { | |
1526 | /* | |
1527 | * We don't log the CRC when updating a btree | |
1528 | * block but instead recreate it during log | |
1529 | * recovery. As the log buffers have checksums | |
1530 | * of their own this is safe and avoids logging a crc | |
1531 | * update in a lot of places. | |
1532 | */ | |
1533 | if (fields == XFS_BB_ALL_BITS) | |
1534 | fields = XFS_BB_ALL_BITS_CRC; | |
1535 | nbits = XFS_BB_NUM_BITS_CRC; | |
1536 | } else { | |
1537 | nbits = XFS_BB_NUM_BITS; | |
1538 | } | |
9eaead51 CH |
1539 | xfs_btree_offsets(fields, |
1540 | (cur->bc_flags & XFS_BTREE_LONG_PTRS) ? | |
1541 | loffsets : soffsets, | |
ee1a47ab | 1542 | nbits, &first, &last); |
61fe135c | 1543 | xfs_trans_buf_set_type(cur->bc_tp, bp, XFS_BLFT_BTREE_BUF); |
9eaead51 CH |
1544 | xfs_trans_log_buf(cur->bc_tp, bp, first, last); |
1545 | } else { | |
1546 | xfs_trans_log_inode(cur->bc_tp, cur->bc_private.b.ip, | |
1547 | xfs_ilog_fbroot(cur->bc_private.b.whichfork)); | |
1548 | } | |
1549 | ||
1550 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1551 | } | |
1552 | ||
637aa50f CH |
1553 | /* |
1554 | * Increment cursor by one record at the level. | |
1555 | * For nonzero levels the leaf-ward information is untouched. | |
1556 | */ | |
1557 | int /* error */ | |
1558 | xfs_btree_increment( | |
1559 | struct xfs_btree_cur *cur, | |
1560 | int level, | |
1561 | int *stat) /* success/failure */ | |
1562 | { | |
1563 | struct xfs_btree_block *block; | |
1564 | union xfs_btree_ptr ptr; | |
1565 | struct xfs_buf *bp; | |
1566 | int error; /* error return value */ | |
1567 | int lev; | |
1568 | ||
1569 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
1570 | XFS_BTREE_TRACE_ARGI(cur, level); | |
1571 | ||
1572 | ASSERT(level < cur->bc_nlevels); | |
1573 | ||
1574 | /* Read-ahead to the right at this level. */ | |
1575 | xfs_btree_readahead(cur, level, XFS_BTCUR_RIGHTRA); | |
1576 | ||
1577 | /* Get a pointer to the btree block. */ | |
1578 | block = xfs_btree_get_block(cur, level, &bp); | |
1579 | ||
1580 | #ifdef DEBUG | |
1581 | error = xfs_btree_check_block(cur, block, level, bp); | |
1582 | if (error) | |
1583 | goto error0; | |
1584 | #endif | |
1585 | ||
1586 | /* We're done if we remain in the block after the increment. */ | |
1587 | if (++cur->bc_ptrs[level] <= xfs_btree_get_numrecs(block)) | |
1588 | goto out1; | |
1589 | ||
1590 | /* Fail if we just went off the right edge of the tree. */ | |
1591 | xfs_btree_get_sibling(cur, block, &ptr, XFS_BB_RIGHTSIB); | |
1592 | if (xfs_btree_ptr_is_null(cur, &ptr)) | |
1593 | goto out0; | |
1594 | ||
1595 | XFS_BTREE_STATS_INC(cur, increment); | |
1596 | ||
1597 | /* | |
1598 | * March up the tree incrementing pointers. | |
1599 | * Stop when we don't go off the right edge of a block. | |
1600 | */ | |
1601 | for (lev = level + 1; lev < cur->bc_nlevels; lev++) { | |
1602 | block = xfs_btree_get_block(cur, lev, &bp); | |
1603 | ||
1604 | #ifdef DEBUG | |
1605 | error = xfs_btree_check_block(cur, block, lev, bp); | |
1606 | if (error) | |
1607 | goto error0; | |
1608 | #endif | |
1609 | ||
1610 | if (++cur->bc_ptrs[lev] <= xfs_btree_get_numrecs(block)) | |
1611 | break; | |
1612 | ||
1613 | /* Read-ahead the right block for the next loop. */ | |
1614 | xfs_btree_readahead(cur, lev, XFS_BTCUR_RIGHTRA); | |
1615 | } | |
1616 | ||
1617 | /* | |
1618 | * If we went off the root then we are either seriously | |
1619 | * confused or have the tree root in an inode. | |
1620 | */ | |
1621 | if (lev == cur->bc_nlevels) { | |
1622 | if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) | |
1623 | goto out0; | |
1624 | ASSERT(0); | |
2451337d | 1625 | error = -EFSCORRUPTED; |
637aa50f CH |
1626 | goto error0; |
1627 | } | |
1628 | ASSERT(lev < cur->bc_nlevels); | |
1629 | ||
1630 | /* | |
1631 | * Now walk back down the tree, fixing up the cursor's buffer | |
1632 | * pointers and key numbers. | |
1633 | */ | |
1634 | for (block = xfs_btree_get_block(cur, lev, &bp); lev > level; ) { | |
1635 | union xfs_btree_ptr *ptrp; | |
1636 | ||
1637 | ptrp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[lev], block); | |
0d7409b1 ES |
1638 | --lev; |
1639 | error = xfs_btree_read_buf_block(cur, ptrp, 0, &block, &bp); | |
637aa50f CH |
1640 | if (error) |
1641 | goto error0; | |
1642 | ||
1643 | xfs_btree_setbuf(cur, lev, bp); | |
1644 | cur->bc_ptrs[lev] = 1; | |
1645 | } | |
1646 | out1: | |
1647 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1648 | *stat = 1; | |
1649 | return 0; | |
1650 | ||
1651 | out0: | |
1652 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1653 | *stat = 0; | |
1654 | return 0; | |
1655 | ||
1656 | error0: | |
1657 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
1658 | return error; | |
1659 | } | |
8df4da4a CH |
1660 | |
1661 | /* | |
1662 | * Decrement cursor by one record at the level. | |
1663 | * For nonzero levels the leaf-ward information is untouched. | |
1664 | */ | |
1665 | int /* error */ | |
1666 | xfs_btree_decrement( | |
1667 | struct xfs_btree_cur *cur, | |
1668 | int level, | |
1669 | int *stat) /* success/failure */ | |
1670 | { | |
1671 | struct xfs_btree_block *block; | |
1672 | xfs_buf_t *bp; | |
1673 | int error; /* error return value */ | |
1674 | int lev; | |
1675 | union xfs_btree_ptr ptr; | |
1676 | ||
1677 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
1678 | XFS_BTREE_TRACE_ARGI(cur, level); | |
1679 | ||
1680 | ASSERT(level < cur->bc_nlevels); | |
1681 | ||
1682 | /* Read-ahead to the left at this level. */ | |
1683 | xfs_btree_readahead(cur, level, XFS_BTCUR_LEFTRA); | |
1684 | ||
1685 | /* We're done if we remain in the block after the decrement. */ | |
1686 | if (--cur->bc_ptrs[level] > 0) | |
1687 | goto out1; | |
1688 | ||
1689 | /* Get a pointer to the btree block. */ | |
1690 | block = xfs_btree_get_block(cur, level, &bp); | |
1691 | ||
1692 | #ifdef DEBUG | |
1693 | error = xfs_btree_check_block(cur, block, level, bp); | |
1694 | if (error) | |
1695 | goto error0; | |
1696 | #endif | |
1697 | ||
1698 | /* Fail if we just went off the left edge of the tree. */ | |
1699 | xfs_btree_get_sibling(cur, block, &ptr, XFS_BB_LEFTSIB); | |
1700 | if (xfs_btree_ptr_is_null(cur, &ptr)) | |
1701 | goto out0; | |
1702 | ||
1703 | XFS_BTREE_STATS_INC(cur, decrement); | |
1704 | ||
1705 | /* | |
1706 | * March up the tree decrementing pointers. | |
1707 | * Stop when we don't go off the left edge of a block. | |
1708 | */ | |
1709 | for (lev = level + 1; lev < cur->bc_nlevels; lev++) { | |
1710 | if (--cur->bc_ptrs[lev] > 0) | |
1711 | break; | |
1712 | /* Read-ahead the left block for the next loop. */ | |
1713 | xfs_btree_readahead(cur, lev, XFS_BTCUR_LEFTRA); | |
1714 | } | |
1715 | ||
1716 | /* | |
1717 | * If we went off the root then we are seriously confused. | |
1718 | * or the root of the tree is in an inode. | |
1719 | */ | |
1720 | if (lev == cur->bc_nlevels) { | |
1721 | if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) | |
1722 | goto out0; | |
1723 | ASSERT(0); | |
2451337d | 1724 | error = -EFSCORRUPTED; |
8df4da4a CH |
1725 | goto error0; |
1726 | } | |
1727 | ASSERT(lev < cur->bc_nlevels); | |
1728 | ||
1729 | /* | |
1730 | * Now walk back down the tree, fixing up the cursor's buffer | |
1731 | * pointers and key numbers. | |
1732 | */ | |
1733 | for (block = xfs_btree_get_block(cur, lev, &bp); lev > level; ) { | |
1734 | union xfs_btree_ptr *ptrp; | |
1735 | ||
1736 | ptrp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[lev], block); | |
0d7409b1 ES |
1737 | --lev; |
1738 | error = xfs_btree_read_buf_block(cur, ptrp, 0, &block, &bp); | |
8df4da4a CH |
1739 | if (error) |
1740 | goto error0; | |
1741 | xfs_btree_setbuf(cur, lev, bp); | |
1742 | cur->bc_ptrs[lev] = xfs_btree_get_numrecs(block); | |
1743 | } | |
1744 | out1: | |
1745 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1746 | *stat = 1; | |
1747 | return 0; | |
1748 | ||
1749 | out0: | |
1750 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1751 | *stat = 0; | |
1752 | return 0; | |
1753 | ||
1754 | error0: | |
1755 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
1756 | return error; | |
1757 | } | |
1758 | ||
fe033cc8 CH |
1759 | STATIC int |
1760 | xfs_btree_lookup_get_block( | |
1761 | struct xfs_btree_cur *cur, /* btree cursor */ | |
1762 | int level, /* level in the btree */ | |
1763 | union xfs_btree_ptr *pp, /* ptr to btree block */ | |
1764 | struct xfs_btree_block **blkp) /* return btree block */ | |
1765 | { | |
1766 | struct xfs_buf *bp; /* buffer pointer for btree block */ | |
1767 | int error = 0; | |
1768 | ||
1769 | /* special case the root block if in an inode */ | |
1770 | if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) && | |
1771 | (level == cur->bc_nlevels - 1)) { | |
1772 | *blkp = xfs_btree_get_iroot(cur); | |
1773 | return 0; | |
1774 | } | |
1775 | ||
1776 | /* | |
1777 | * If the old buffer at this level for the disk address we are | |
1778 | * looking for re-use it. | |
1779 | * | |
1780 | * Otherwise throw it away and get a new one. | |
1781 | */ | |
1782 | bp = cur->bc_bufs[level]; | |
1783 | if (bp && XFS_BUF_ADDR(bp) == xfs_btree_ptr_to_daddr(cur, pp)) { | |
1784 | *blkp = XFS_BUF_TO_BLOCK(bp); | |
1785 | return 0; | |
1786 | } | |
1787 | ||
0d7409b1 | 1788 | error = xfs_btree_read_buf_block(cur, pp, 0, blkp, &bp); |
fe033cc8 CH |
1789 | if (error) |
1790 | return error; | |
1791 | ||
bb3be7e7 DW |
1792 | /* Check the inode owner since the verifiers don't. */ |
1793 | if (xfs_sb_version_hascrc(&cur->bc_mp->m_sb) && | |
1794 | (cur->bc_flags & XFS_BTREE_LONG_PTRS) && | |
1795 | be64_to_cpu((*blkp)->bb_u.l.bb_owner) != | |
1796 | cur->bc_private.b.ip->i_ino) | |
1797 | goto out_bad; | |
1798 | ||
1799 | /* Did we get the level we were looking for? */ | |
1800 | if (be16_to_cpu((*blkp)->bb_level) != level) | |
1801 | goto out_bad; | |
1802 | ||
1803 | /* Check that internal nodes have at least one record. */ | |
1804 | if (level != 0 && be16_to_cpu((*blkp)->bb_numrecs) == 0) | |
1805 | goto out_bad; | |
1806 | ||
fe033cc8 CH |
1807 | xfs_btree_setbuf(cur, level, bp); |
1808 | return 0; | |
bb3be7e7 DW |
1809 | |
1810 | out_bad: | |
1811 | *blkp = NULL; | |
1812 | xfs_trans_brelse(cur->bc_tp, bp); | |
1813 | return -EFSCORRUPTED; | |
fe033cc8 CH |
1814 | } |
1815 | ||
1816 | /* | |
1817 | * Get current search key. For level 0 we don't actually have a key | |
1818 | * structure so we make one up from the record. For all other levels | |
1819 | * we just return the right key. | |
1820 | */ | |
1821 | STATIC union xfs_btree_key * | |
1822 | xfs_lookup_get_search_key( | |
1823 | struct xfs_btree_cur *cur, | |
1824 | int level, | |
1825 | int keyno, | |
1826 | struct xfs_btree_block *block, | |
1827 | union xfs_btree_key *kp) | |
1828 | { | |
1829 | if (level == 0) { | |
1830 | cur->bc_ops->init_key_from_rec(kp, | |
1831 | xfs_btree_rec_addr(cur, keyno, block)); | |
1832 | return kp; | |
1833 | } | |
1834 | ||
1835 | return xfs_btree_key_addr(cur, keyno, block); | |
1836 | } | |
1837 | ||
1838 | /* | |
1839 | * Lookup the record. The cursor is made to point to it, based on dir. | |
49d3da14 | 1840 | * stat is set to 0 if can't find any such record, 1 for success. |
fe033cc8 CH |
1841 | */ |
1842 | int /* error */ | |
1843 | xfs_btree_lookup( | |
1844 | struct xfs_btree_cur *cur, /* btree cursor */ | |
1845 | xfs_lookup_t dir, /* <=, ==, or >= */ | |
1846 | int *stat) /* success/failure */ | |
1847 | { | |
1848 | struct xfs_btree_block *block; /* current btree block */ | |
1849 | __int64_t diff; /* difference for the current key */ | |
1850 | int error; /* error return value */ | |
1851 | int keyno; /* current key number */ | |
1852 | int level; /* level in the btree */ | |
1853 | union xfs_btree_ptr *pp; /* ptr to btree block */ | |
1854 | union xfs_btree_ptr ptr; /* ptr to btree block */ | |
1855 | ||
1856 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
1857 | XFS_BTREE_TRACE_ARGI(cur, dir); | |
1858 | ||
1859 | XFS_BTREE_STATS_INC(cur, lookup); | |
1860 | ||
ed150e1a DW |
1861 | /* No such thing as a zero-level tree. */ |
1862 | if (cur->bc_nlevels == 0) | |
1863 | return -EFSCORRUPTED; | |
1864 | ||
fe033cc8 CH |
1865 | block = NULL; |
1866 | keyno = 0; | |
1867 | ||
1868 | /* initialise start pointer from cursor */ | |
1869 | cur->bc_ops->init_ptr_from_cur(cur, &ptr); | |
1870 | pp = &ptr; | |
1871 | ||
1872 | /* | |
1873 | * Iterate over each level in the btree, starting at the root. | |
1874 | * For each level above the leaves, find the key we need, based | |
1875 | * on the lookup record, then follow the corresponding block | |
1876 | * pointer down to the next level. | |
1877 | */ | |
1878 | for (level = cur->bc_nlevels - 1, diff = 1; level >= 0; level--) { | |
1879 | /* Get the block we need to do the lookup on. */ | |
1880 | error = xfs_btree_lookup_get_block(cur, level, pp, &block); | |
1881 | if (error) | |
1882 | goto error0; | |
1883 | ||
1884 | if (diff == 0) { | |
1885 | /* | |
1886 | * If we already had a key match at a higher level, we | |
1887 | * know we need to use the first entry in this block. | |
1888 | */ | |
1889 | keyno = 1; | |
1890 | } else { | |
1891 | /* Otherwise search this block. Do a binary search. */ | |
1892 | ||
1893 | int high; /* high entry number */ | |
1894 | int low; /* low entry number */ | |
1895 | ||
1896 | /* Set low and high entry numbers, 1-based. */ | |
1897 | low = 1; | |
1898 | high = xfs_btree_get_numrecs(block); | |
1899 | if (!high) { | |
1900 | /* Block is empty, must be an empty leaf. */ | |
1901 | ASSERT(level == 0 && cur->bc_nlevels == 1); | |
1902 | ||
1903 | cur->bc_ptrs[0] = dir != XFS_LOOKUP_LE; | |
1904 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1905 | *stat = 0; | |
1906 | return 0; | |
1907 | } | |
1908 | ||
1909 | /* Binary search the block. */ | |
1910 | while (low <= high) { | |
1911 | union xfs_btree_key key; | |
1912 | union xfs_btree_key *kp; | |
1913 | ||
1914 | XFS_BTREE_STATS_INC(cur, compare); | |
1915 | ||
1916 | /* keyno is average of low and high. */ | |
1917 | keyno = (low + high) >> 1; | |
1918 | ||
1919 | /* Get current search key */ | |
1920 | kp = xfs_lookup_get_search_key(cur, level, | |
1921 | keyno, block, &key); | |
1922 | ||
1923 | /* | |
1924 | * Compute difference to get next direction: | |
1925 | * - less than, move right | |
1926 | * - greater than, move left | |
1927 | * - equal, we're done | |
1928 | */ | |
1929 | diff = cur->bc_ops->key_diff(cur, kp); | |
1930 | if (diff < 0) | |
1931 | low = keyno + 1; | |
1932 | else if (diff > 0) | |
1933 | high = keyno - 1; | |
1934 | else | |
1935 | break; | |
1936 | } | |
1937 | } | |
1938 | ||
1939 | /* | |
1940 | * If there are more levels, set up for the next level | |
1941 | * by getting the block number and filling in the cursor. | |
1942 | */ | |
1943 | if (level > 0) { | |
1944 | /* | |
1945 | * If we moved left, need the previous key number, | |
1946 | * unless there isn't one. | |
1947 | */ | |
1948 | if (diff > 0 && --keyno < 1) | |
1949 | keyno = 1; | |
1950 | pp = xfs_btree_ptr_addr(cur, keyno, block); | |
1951 | ||
1952 | #ifdef DEBUG | |
1953 | error = xfs_btree_check_ptr(cur, pp, 0, level); | |
1954 | if (error) | |
1955 | goto error0; | |
1956 | #endif | |
1957 | cur->bc_ptrs[level] = keyno; | |
1958 | } | |
1959 | } | |
1960 | ||
1961 | /* Done with the search. See if we need to adjust the results. */ | |
1962 | if (dir != XFS_LOOKUP_LE && diff < 0) { | |
1963 | keyno++; | |
1964 | /* | |
1965 | * If ge search and we went off the end of the block, but it's | |
1966 | * not the last block, we're in the wrong block. | |
1967 | */ | |
1968 | xfs_btree_get_sibling(cur, block, &ptr, XFS_BB_RIGHTSIB); | |
1969 | if (dir == XFS_LOOKUP_GE && | |
1970 | keyno > xfs_btree_get_numrecs(block) && | |
1971 | !xfs_btree_ptr_is_null(cur, &ptr)) { | |
1972 | int i; | |
1973 | ||
1974 | cur->bc_ptrs[0] = keyno; | |
1975 | error = xfs_btree_increment(cur, 0, &i); | |
1976 | if (error) | |
1977 | goto error0; | |
5fb5aeee | 1978 | XFS_WANT_CORRUPTED_RETURN(cur->bc_mp, i == 1); |
fe033cc8 CH |
1979 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); |
1980 | *stat = 1; | |
1981 | return 0; | |
1982 | } | |
1983 | } else if (dir == XFS_LOOKUP_LE && diff > 0) | |
1984 | keyno--; | |
1985 | cur->bc_ptrs[0] = keyno; | |
1986 | ||
1987 | /* Return if we succeeded or not. */ | |
1988 | if (keyno == 0 || keyno > xfs_btree_get_numrecs(block)) | |
1989 | *stat = 0; | |
1990 | else if (dir != XFS_LOOKUP_EQ || diff == 0) | |
1991 | *stat = 1; | |
1992 | else | |
1993 | *stat = 0; | |
1994 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
1995 | return 0; | |
1996 | ||
1997 | error0: | |
1998 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
1999 | return error; | |
2000 | } | |
38bb7423 | 2001 | |
2c813ad6 DW |
2002 | /* Find the high key storage area from a regular key. */ |
2003 | STATIC union xfs_btree_key * | |
2004 | xfs_btree_high_key_from_key( | |
2005 | struct xfs_btree_cur *cur, | |
2006 | union xfs_btree_key *key) | |
2007 | { | |
2008 | ASSERT(cur->bc_flags & XFS_BTREE_OVERLAPPING); | |
2009 | return (union xfs_btree_key *)((char *)key + | |
2010 | (cur->bc_ops->key_len / 2)); | |
2011 | } | |
2012 | ||
973b8319 DW |
2013 | /* Determine the low (and high if overlapped) keys of a leaf block */ |
2014 | STATIC void | |
2015 | xfs_btree_get_leaf_keys( | |
2c813ad6 DW |
2016 | struct xfs_btree_cur *cur, |
2017 | struct xfs_btree_block *block, | |
2018 | union xfs_btree_key *key) | |
2019 | { | |
2c813ad6 DW |
2020 | union xfs_btree_key max_hkey; |
2021 | union xfs_btree_key hkey; | |
973b8319 | 2022 | union xfs_btree_rec *rec; |
2c813ad6 | 2023 | union xfs_btree_key *high; |
973b8319 | 2024 | int n; |
2c813ad6 | 2025 | |
2c813ad6 DW |
2026 | rec = xfs_btree_rec_addr(cur, 1, block); |
2027 | cur->bc_ops->init_key_from_rec(key, rec); | |
2028 | ||
973b8319 DW |
2029 | if (cur->bc_flags & XFS_BTREE_OVERLAPPING) { |
2030 | ||
2031 | cur->bc_ops->init_high_key_from_rec(&max_hkey, rec); | |
2032 | for (n = 2; n <= xfs_btree_get_numrecs(block); n++) { | |
2033 | rec = xfs_btree_rec_addr(cur, n, block); | |
2034 | cur->bc_ops->init_high_key_from_rec(&hkey, rec); | |
2035 | if (cur->bc_ops->diff_two_keys(cur, &hkey, &max_hkey) | |
2036 | > 0) | |
2037 | max_hkey = hkey; | |
2038 | } | |
2c813ad6 | 2039 | |
973b8319 DW |
2040 | high = xfs_btree_high_key_from_key(cur, key); |
2041 | memcpy(high, &max_hkey, cur->bc_ops->key_len / 2); | |
2042 | } | |
2c813ad6 DW |
2043 | } |
2044 | ||
973b8319 DW |
2045 | /* Determine the low (and high if overlapped) keys of a node block */ |
2046 | STATIC void | |
2047 | xfs_btree_get_node_keys( | |
2c813ad6 DW |
2048 | struct xfs_btree_cur *cur, |
2049 | struct xfs_btree_block *block, | |
2050 | union xfs_btree_key *key) | |
2051 | { | |
2c813ad6 DW |
2052 | union xfs_btree_key *hkey; |
2053 | union xfs_btree_key *max_hkey; | |
2054 | union xfs_btree_key *high; | |
973b8319 | 2055 | int n; |
2c813ad6 | 2056 | |
973b8319 DW |
2057 | if (cur->bc_flags & XFS_BTREE_OVERLAPPING) { |
2058 | memcpy(key, xfs_btree_key_addr(cur, 1, block), | |
2059 | cur->bc_ops->key_len / 2); | |
2060 | ||
2061 | max_hkey = xfs_btree_high_key_addr(cur, 1, block); | |
2062 | for (n = 2; n <= xfs_btree_get_numrecs(block); n++) { | |
2063 | hkey = xfs_btree_high_key_addr(cur, n, block); | |
2064 | if (cur->bc_ops->diff_two_keys(cur, hkey, max_hkey) > 0) | |
2065 | max_hkey = hkey; | |
2066 | } | |
2c813ad6 | 2067 | |
973b8319 DW |
2068 | high = xfs_btree_high_key_from_key(cur, key); |
2069 | memcpy(high, max_hkey, cur->bc_ops->key_len / 2); | |
2070 | } else { | |
2071 | memcpy(key, xfs_btree_key_addr(cur, 1, block), | |
2072 | cur->bc_ops->key_len); | |
2c813ad6 | 2073 | } |
2c813ad6 DW |
2074 | } |
2075 | ||
70b22659 DW |
2076 | /* Derive the keys for any btree block. */ |
2077 | STATIC void | |
2078 | xfs_btree_get_keys( | |
2079 | struct xfs_btree_cur *cur, | |
2080 | struct xfs_btree_block *block, | |
2081 | union xfs_btree_key *key) | |
2082 | { | |
2083 | if (be16_to_cpu(block->bb_level) == 0) | |
973b8319 | 2084 | xfs_btree_get_leaf_keys(cur, block, key); |
70b22659 | 2085 | else |
973b8319 | 2086 | xfs_btree_get_node_keys(cur, block, key); |
70b22659 DW |
2087 | } |
2088 | ||
38bb7423 | 2089 | /* |
70b22659 DW |
2090 | * Decide if we need to update the parent keys of a btree block. For |
2091 | * a standard btree this is only necessary if we're updating the first | |
2c813ad6 DW |
2092 | * record/key. For an overlapping btree, we must always update the |
2093 | * keys because the highest key can be in any of the records or keys | |
2094 | * in the block. | |
38bb7423 | 2095 | */ |
70b22659 DW |
2096 | static inline bool |
2097 | xfs_btree_needs_key_update( | |
2098 | struct xfs_btree_cur *cur, | |
2099 | int ptr) | |
2100 | { | |
2c813ad6 DW |
2101 | return (cur->bc_flags & XFS_BTREE_OVERLAPPING) || ptr == 1; |
2102 | } | |
2103 | ||
2104 | /* | |
2105 | * Update the low and high parent keys of the given level, progressing | |
2106 | * towards the root. If force_all is false, stop if the keys for a given | |
2107 | * level do not need updating. | |
2108 | */ | |
2109 | STATIC int | |
2110 | __xfs_btree_updkeys( | |
2111 | struct xfs_btree_cur *cur, | |
2112 | int level, | |
2113 | struct xfs_btree_block *block, | |
2114 | struct xfs_buf *bp0, | |
2115 | bool force_all) | |
2116 | { | |
a1d46cff | 2117 | union xfs_btree_key key; /* keys from current level */ |
2c813ad6 DW |
2118 | union xfs_btree_key *lkey; /* keys from the next level up */ |
2119 | union xfs_btree_key *hkey; | |
2120 | union xfs_btree_key *nlkey; /* keys from the next level up */ | |
2121 | union xfs_btree_key *nhkey; | |
2122 | struct xfs_buf *bp; | |
2123 | int ptr; | |
2124 | ||
2125 | ASSERT(cur->bc_flags & XFS_BTREE_OVERLAPPING); | |
2126 | ||
2127 | /* Exit if there aren't any parent levels to update. */ | |
2128 | if (level + 1 >= cur->bc_nlevels) | |
2129 | return 0; | |
2130 | ||
2131 | trace_xfs_btree_updkeys(cur, level, bp0); | |
2132 | ||
a1d46cff | 2133 | lkey = &key; |
2c813ad6 DW |
2134 | hkey = xfs_btree_high_key_from_key(cur, lkey); |
2135 | xfs_btree_get_keys(cur, block, lkey); | |
2136 | for (level++; level < cur->bc_nlevels; level++) { | |
2137 | #ifdef DEBUG | |
2138 | int error; | |
2139 | #endif | |
2140 | block = xfs_btree_get_block(cur, level, &bp); | |
2141 | trace_xfs_btree_updkeys(cur, level, bp); | |
2142 | #ifdef DEBUG | |
2143 | error = xfs_btree_check_block(cur, block, level, bp); | |
2144 | if (error) { | |
2145 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
2146 | return error; | |
2147 | } | |
2148 | #endif | |
2149 | ptr = cur->bc_ptrs[level]; | |
2150 | nlkey = xfs_btree_key_addr(cur, ptr, block); | |
2151 | nhkey = xfs_btree_high_key_addr(cur, ptr, block); | |
2152 | if (!force_all && | |
2153 | !(cur->bc_ops->diff_two_keys(cur, nlkey, lkey) != 0 || | |
2154 | cur->bc_ops->diff_two_keys(cur, nhkey, hkey) != 0)) | |
2155 | break; | |
2156 | xfs_btree_copy_keys(cur, nlkey, lkey, 1); | |
2157 | xfs_btree_log_keys(cur, bp, ptr, ptr); | |
2158 | if (level + 1 >= cur->bc_nlevels) | |
2159 | break; | |
973b8319 | 2160 | xfs_btree_get_node_keys(cur, block, lkey); |
2c813ad6 DW |
2161 | } |
2162 | ||
2163 | return 0; | |
2164 | } | |
2165 | ||
2c813ad6 DW |
2166 | /* Update all the keys from some level in cursor back to the root. */ |
2167 | STATIC int | |
2168 | xfs_btree_updkeys_force( | |
2169 | struct xfs_btree_cur *cur, | |
2170 | int level) | |
2171 | { | |
2172 | struct xfs_buf *bp; | |
2173 | struct xfs_btree_block *block; | |
2174 | ||
2175 | block = xfs_btree_get_block(cur, level, &bp); | |
2176 | return __xfs_btree_updkeys(cur, level, block, bp, true); | |
70b22659 DW |
2177 | } |
2178 | ||
2179 | /* | |
2180 | * Update the parent keys of the given level, progressing towards the root. | |
2181 | */ | |
973b8319 | 2182 | STATIC int |
70b22659 | 2183 | xfs_btree_update_keys( |
38bb7423 | 2184 | struct xfs_btree_cur *cur, |
38bb7423 CH |
2185 | int level) |
2186 | { | |
2187 | struct xfs_btree_block *block; | |
2188 | struct xfs_buf *bp; | |
2189 | union xfs_btree_key *kp; | |
70b22659 | 2190 | union xfs_btree_key key; |
38bb7423 CH |
2191 | int ptr; |
2192 | ||
973b8319 DW |
2193 | ASSERT(level >= 0); |
2194 | ||
2195 | block = xfs_btree_get_block(cur, level, &bp); | |
2196 | if (cur->bc_flags & XFS_BTREE_OVERLAPPING) | |
2197 | return __xfs_btree_updkeys(cur, level, block, bp, false); | |
2c813ad6 | 2198 | |
38bb7423 CH |
2199 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); |
2200 | XFS_BTREE_TRACE_ARGIK(cur, level, keyp); | |
2201 | ||
38bb7423 CH |
2202 | /* |
2203 | * Go up the tree from this level toward the root. | |
2204 | * At each level, update the key value to the value input. | |
2205 | * Stop when we reach a level where the cursor isn't pointing | |
2206 | * at the first entry in the block. | |
2207 | */ | |
70b22659 DW |
2208 | xfs_btree_get_keys(cur, block, &key); |
2209 | for (level++, ptr = 1; ptr == 1 && level < cur->bc_nlevels; level++) { | |
38bb7423 CH |
2210 | #ifdef DEBUG |
2211 | int error; | |
2212 | #endif | |
2213 | block = xfs_btree_get_block(cur, level, &bp); | |
2214 | #ifdef DEBUG | |
2215 | error = xfs_btree_check_block(cur, block, level, bp); | |
2216 | if (error) { | |
2217 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
2218 | return error; | |
2219 | } | |
2220 | #endif | |
2221 | ptr = cur->bc_ptrs[level]; | |
2222 | kp = xfs_btree_key_addr(cur, ptr, block); | |
70b22659 | 2223 | xfs_btree_copy_keys(cur, kp, &key, 1); |
38bb7423 CH |
2224 | xfs_btree_log_keys(cur, bp, ptr, ptr); |
2225 | } | |
2226 | ||
2227 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
2228 | return 0; | |
2229 | } | |
278d0ca1 CH |
2230 | |
2231 | /* | |
2232 | * Update the record referred to by cur to the value in the | |
2233 | * given record. This either works (return 0) or gets an | |
2234 | * EFSCORRUPTED error. | |
2235 | */ | |
2236 | int | |
2237 | xfs_btree_update( | |
2238 | struct xfs_btree_cur *cur, | |
2239 | union xfs_btree_rec *rec) | |
2240 | { | |
2241 | struct xfs_btree_block *block; | |
2242 | struct xfs_buf *bp; | |
2243 | int error; | |
2244 | int ptr; | |
2245 | union xfs_btree_rec *rp; | |
2246 | ||
2247 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
2248 | XFS_BTREE_TRACE_ARGR(cur, rec); | |
2249 | ||
2250 | /* Pick up the current block. */ | |
2251 | block = xfs_btree_get_block(cur, 0, &bp); | |
2252 | ||
2253 | #ifdef DEBUG | |
2254 | error = xfs_btree_check_block(cur, block, 0, bp); | |
2255 | if (error) | |
2256 | goto error0; | |
2257 | #endif | |
2258 | /* Get the address of the rec to be updated. */ | |
2259 | ptr = cur->bc_ptrs[0]; | |
2260 | rp = xfs_btree_rec_addr(cur, ptr, block); | |
2261 | ||
2262 | /* Fill in the new contents and log them. */ | |
2263 | xfs_btree_copy_recs(cur, rp, rec, 1); | |
2264 | xfs_btree_log_recs(cur, bp, ptr, ptr); | |
2265 | ||
2266 | /* | |
2267 | * If we are tracking the last record in the tree and | |
2268 | * we are at the far right edge of the tree, update it. | |
2269 | */ | |
2270 | if (xfs_btree_is_lastrec(cur, block, 0)) { | |
2271 | cur->bc_ops->update_lastrec(cur, block, rec, | |
2272 | ptr, LASTREC_UPDATE); | |
2273 | } | |
2274 | ||
2c813ad6 | 2275 | /* Pass new key value up to our parent. */ |
70b22659 | 2276 | if (xfs_btree_needs_key_update(cur, ptr)) { |
973b8319 | 2277 | error = xfs_btree_update_keys(cur, 0); |
278d0ca1 CH |
2278 | if (error) |
2279 | goto error0; | |
2280 | } | |
2281 | ||
2282 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
2283 | return 0; | |
2284 | ||
2285 | error0: | |
2286 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
2287 | return error; | |
2288 | } | |
2289 | ||
687b890a CH |
2290 | /* |
2291 | * Move 1 record left from cur/level if possible. | |
2292 | * Update cur to reflect the new path. | |
2293 | */ | |
3cc7524c | 2294 | STATIC int /* error */ |
687b890a CH |
2295 | xfs_btree_lshift( |
2296 | struct xfs_btree_cur *cur, | |
2297 | int level, | |
2298 | int *stat) /* success/failure */ | |
2299 | { | |
687b890a CH |
2300 | struct xfs_buf *lbp; /* left buffer pointer */ |
2301 | struct xfs_btree_block *left; /* left btree block */ | |
2302 | int lrecs; /* left record count */ | |
2303 | struct xfs_buf *rbp; /* right buffer pointer */ | |
2304 | struct xfs_btree_block *right; /* right btree block */ | |
2c813ad6 | 2305 | struct xfs_btree_cur *tcur; /* temporary btree cursor */ |
687b890a CH |
2306 | int rrecs; /* right record count */ |
2307 | union xfs_btree_ptr lptr; /* left btree pointer */ | |
2308 | union xfs_btree_key *rkp = NULL; /* right btree key */ | |
2309 | union xfs_btree_ptr *rpp = NULL; /* right address pointer */ | |
2310 | union xfs_btree_rec *rrp = NULL; /* right record pointer */ | |
2311 | int error; /* error return value */ | |
2c813ad6 | 2312 | int i; |
687b890a CH |
2313 | |
2314 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
2315 | XFS_BTREE_TRACE_ARGI(cur, level); | |
2316 | ||
2317 | if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) && | |
2318 | level == cur->bc_nlevels - 1) | |
2319 | goto out0; | |
2320 | ||
2321 | /* Set up variables for this block as "right". */ | |
2322 | right = xfs_btree_get_block(cur, level, &rbp); | |
2323 | ||
2324 | #ifdef DEBUG | |
2325 | error = xfs_btree_check_block(cur, right, level, rbp); | |
2326 | if (error) | |
2327 | goto error0; | |
2328 | #endif | |
2329 | ||
2330 | /* If we've got no left sibling then we can't shift an entry left. */ | |
2331 | xfs_btree_get_sibling(cur, right, &lptr, XFS_BB_LEFTSIB); | |
2332 | if (xfs_btree_ptr_is_null(cur, &lptr)) | |
2333 | goto out0; | |
2334 | ||
2335 | /* | |
2336 | * If the cursor entry is the one that would be moved, don't | |
2337 | * do it... it's too complicated. | |
2338 | */ | |
2339 | if (cur->bc_ptrs[level] <= 1) | |
2340 | goto out0; | |
2341 | ||
2342 | /* Set up the left neighbor as "left". */ | |
0d7409b1 | 2343 | error = xfs_btree_read_buf_block(cur, &lptr, 0, &left, &lbp); |
687b890a CH |
2344 | if (error) |
2345 | goto error0; | |
2346 | ||
2347 | /* If it's full, it can't take another entry. */ | |
2348 | lrecs = xfs_btree_get_numrecs(left); | |
2349 | if (lrecs == cur->bc_ops->get_maxrecs(cur, level)) | |
2350 | goto out0; | |
2351 | ||
2352 | rrecs = xfs_btree_get_numrecs(right); | |
2353 | ||
2354 | /* | |
2355 | * We add one entry to the left side and remove one for the right side. | |
9da096fd | 2356 | * Account for it here, the changes will be updated on disk and logged |
687b890a CH |
2357 | * later. |
2358 | */ | |
2359 | lrecs++; | |
2360 | rrecs--; | |
2361 | ||
2362 | XFS_BTREE_STATS_INC(cur, lshift); | |
2363 | XFS_BTREE_STATS_ADD(cur, moves, 1); | |
2364 | ||
2365 | /* | |
2366 | * If non-leaf, copy a key and a ptr to the left block. | |
2367 | * Log the changes to the left block. | |
2368 | */ | |
2369 | if (level > 0) { | |
2370 | /* It's a non-leaf. Move keys and pointers. */ | |
2371 | union xfs_btree_key *lkp; /* left btree key */ | |
2372 | union xfs_btree_ptr *lpp; /* left address pointer */ | |
2373 | ||
2374 | lkp = xfs_btree_key_addr(cur, lrecs, left); | |
2375 | rkp = xfs_btree_key_addr(cur, 1, right); | |
2376 | ||
2377 | lpp = xfs_btree_ptr_addr(cur, lrecs, left); | |
2378 | rpp = xfs_btree_ptr_addr(cur, 1, right); | |
2379 | #ifdef DEBUG | |
2380 | error = xfs_btree_check_ptr(cur, rpp, 0, level); | |
2381 | if (error) | |
2382 | goto error0; | |
2383 | #endif | |
2384 | xfs_btree_copy_keys(cur, lkp, rkp, 1); | |
2385 | xfs_btree_copy_ptrs(cur, lpp, rpp, 1); | |
2386 | ||
2387 | xfs_btree_log_keys(cur, lbp, lrecs, lrecs); | |
2388 | xfs_btree_log_ptrs(cur, lbp, lrecs, lrecs); | |
2389 | ||
4a26e66e CH |
2390 | ASSERT(cur->bc_ops->keys_inorder(cur, |
2391 | xfs_btree_key_addr(cur, lrecs - 1, left), lkp)); | |
687b890a CH |
2392 | } else { |
2393 | /* It's a leaf. Move records. */ | |
2394 | union xfs_btree_rec *lrp; /* left record pointer */ | |
2395 | ||
2396 | lrp = xfs_btree_rec_addr(cur, lrecs, left); | |
2397 | rrp = xfs_btree_rec_addr(cur, 1, right); | |
2398 | ||
2399 | xfs_btree_copy_recs(cur, lrp, rrp, 1); | |
2400 | xfs_btree_log_recs(cur, lbp, lrecs, lrecs); | |
2401 | ||
4a26e66e CH |
2402 | ASSERT(cur->bc_ops->recs_inorder(cur, |
2403 | xfs_btree_rec_addr(cur, lrecs - 1, left), lrp)); | |
687b890a CH |
2404 | } |
2405 | ||
2406 | xfs_btree_set_numrecs(left, lrecs); | |
2407 | xfs_btree_log_block(cur, lbp, XFS_BB_NUMRECS); | |
2408 | ||
2409 | xfs_btree_set_numrecs(right, rrecs); | |
2410 | xfs_btree_log_block(cur, rbp, XFS_BB_NUMRECS); | |
2411 | ||
2412 | /* | |
2413 | * Slide the contents of right down one entry. | |
2414 | */ | |
2415 | XFS_BTREE_STATS_ADD(cur, moves, rrecs - 1); | |
2416 | if (level > 0) { | |
2417 | /* It's a nonleaf. operate on keys and ptrs */ | |
2418 | #ifdef DEBUG | |
2419 | int i; /* loop index */ | |
2420 | ||
2421 | for (i = 0; i < rrecs; i++) { | |
2422 | error = xfs_btree_check_ptr(cur, rpp, i + 1, level); | |
2423 | if (error) | |
2424 | goto error0; | |
2425 | } | |
2426 | #endif | |
2427 | xfs_btree_shift_keys(cur, | |
2428 | xfs_btree_key_addr(cur, 2, right), | |
2429 | -1, rrecs); | |
2430 | xfs_btree_shift_ptrs(cur, | |
2431 | xfs_btree_ptr_addr(cur, 2, right), | |
2432 | -1, rrecs); | |
2433 | ||
2434 | xfs_btree_log_keys(cur, rbp, 1, rrecs); | |
2435 | xfs_btree_log_ptrs(cur, rbp, 1, rrecs); | |
2436 | } else { | |
2437 | /* It's a leaf. operate on records */ | |
2438 | xfs_btree_shift_recs(cur, | |
2439 | xfs_btree_rec_addr(cur, 2, right), | |
2440 | -1, rrecs); | |
2441 | xfs_btree_log_recs(cur, rbp, 1, rrecs); | |
687b890a CH |
2442 | } |
2443 | ||
2c813ad6 DW |
2444 | /* |
2445 | * Using a temporary cursor, update the parent key values of the | |
2446 | * block on the left. | |
2447 | */ | |
c1d22ae8 DW |
2448 | if (cur->bc_flags & XFS_BTREE_OVERLAPPING) { |
2449 | error = xfs_btree_dup_cursor(cur, &tcur); | |
2450 | if (error) | |
2451 | goto error0; | |
2452 | i = xfs_btree_firstrec(tcur, level); | |
2453 | XFS_WANT_CORRUPTED_GOTO(tcur->bc_mp, i == 1, error0); | |
2c813ad6 | 2454 | |
c1d22ae8 DW |
2455 | error = xfs_btree_decrement(tcur, level, &i); |
2456 | if (error) | |
2457 | goto error1; | |
2c813ad6 | 2458 | |
c1d22ae8 | 2459 | /* Update the parent high keys of the left block, if needed. */ |
973b8319 | 2460 | error = xfs_btree_update_keys(tcur, level); |
2c813ad6 DW |
2461 | if (error) |
2462 | goto error1; | |
c1d22ae8 DW |
2463 | |
2464 | xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); | |
2c813ad6 DW |
2465 | } |
2466 | ||
c1d22ae8 DW |
2467 | /* Update the parent keys of the right block. */ |
2468 | error = xfs_btree_update_keys(cur, level); | |
2469 | if (error) | |
2470 | goto error0; | |
687b890a CH |
2471 | |
2472 | /* Slide the cursor value left one. */ | |
2473 | cur->bc_ptrs[level]--; | |
2474 | ||
2475 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
2476 | *stat = 1; | |
2477 | return 0; | |
2478 | ||
2479 | out0: | |
2480 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
2481 | *stat = 0; | |
2482 | return 0; | |
2483 | ||
2484 | error0: | |
2485 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
2486 | return error; | |
2c813ad6 DW |
2487 | |
2488 | error1: | |
2489 | XFS_BTREE_TRACE_CURSOR(tcur, XBT_ERROR); | |
2490 | xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR); | |
2491 | return error; | |
687b890a CH |
2492 | } |
2493 | ||
9eaead51 CH |
2494 | /* |
2495 | * Move 1 record right from cur/level if possible. | |
2496 | * Update cur to reflect the new path. | |
2497 | */ | |
3cc7524c | 2498 | STATIC int /* error */ |
9eaead51 CH |
2499 | xfs_btree_rshift( |
2500 | struct xfs_btree_cur *cur, | |
2501 | int level, | |
2502 | int *stat) /* success/failure */ | |
2503 | { | |
9eaead51 CH |
2504 | struct xfs_buf *lbp; /* left buffer pointer */ |
2505 | struct xfs_btree_block *left; /* left btree block */ | |
2506 | struct xfs_buf *rbp; /* right buffer pointer */ | |
2507 | struct xfs_btree_block *right; /* right btree block */ | |
2508 | struct xfs_btree_cur *tcur; /* temporary btree cursor */ | |
2509 | union xfs_btree_ptr rptr; /* right block pointer */ | |
2510 | union xfs_btree_key *rkp; /* right btree key */ | |
2511 | int rrecs; /* right record count */ | |
2512 | int lrecs; /* left record count */ | |
2513 | int error; /* error return value */ | |
2514 | int i; /* loop counter */ | |
2515 | ||
2516 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
2517 | XFS_BTREE_TRACE_ARGI(cur, level); | |
2518 | ||
2519 | if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) && | |
2520 | (level == cur->bc_nlevels - 1)) | |
2521 | goto out0; | |
2522 | ||
2523 | /* Set up variables for this block as "left". */ | |
2524 | left = xfs_btree_get_block(cur, level, &lbp); | |
2525 | ||
2526 | #ifdef DEBUG | |
2527 | error = xfs_btree_check_block(cur, left, level, lbp); | |
2528 | if (error) | |
2529 | goto error0; | |
2530 | #endif | |
2531 | ||
2532 | /* If we've got no right sibling then we can't shift an entry right. */ | |
2533 | xfs_btree_get_sibling(cur, left, &rptr, XFS_BB_RIGHTSIB); | |
2534 | if (xfs_btree_ptr_is_null(cur, &rptr)) | |
2535 | goto out0; | |
2536 | ||
2537 | /* | |
2538 | * If the cursor entry is the one that would be moved, don't | |
2539 | * do it... it's too complicated. | |
2540 | */ | |
2541 | lrecs = xfs_btree_get_numrecs(left); | |
2542 | if (cur->bc_ptrs[level] >= lrecs) | |
2543 | goto out0; | |
2544 | ||
2545 | /* Set up the right neighbor as "right". */ | |
0d7409b1 | 2546 | error = xfs_btree_read_buf_block(cur, &rptr, 0, &right, &rbp); |
9eaead51 CH |
2547 | if (error) |
2548 | goto error0; | |
2549 | ||
2550 | /* If it's full, it can't take another entry. */ | |
2551 | rrecs = xfs_btree_get_numrecs(right); | |
2552 | if (rrecs == cur->bc_ops->get_maxrecs(cur, level)) | |
2553 | goto out0; | |
2554 | ||
2555 | XFS_BTREE_STATS_INC(cur, rshift); | |
2556 | XFS_BTREE_STATS_ADD(cur, moves, rrecs); | |
2557 | ||
2558 | /* | |
2559 | * Make a hole at the start of the right neighbor block, then | |
2560 | * copy the last left block entry to the hole. | |
2561 | */ | |
2562 | if (level > 0) { | |
2563 | /* It's a nonleaf. make a hole in the keys and ptrs */ | |
2564 | union xfs_btree_key *lkp; | |
2565 | union xfs_btree_ptr *lpp; | |
2566 | union xfs_btree_ptr *rpp; | |
2567 | ||
2568 | lkp = xfs_btree_key_addr(cur, lrecs, left); | |
2569 | lpp = xfs_btree_ptr_addr(cur, lrecs, left); | |
2570 | rkp = xfs_btree_key_addr(cur, 1, right); | |
2571 | rpp = xfs_btree_ptr_addr(cur, 1, right); | |
2572 | ||
2573 | #ifdef DEBUG | |
2574 | for (i = rrecs - 1; i >= 0; i--) { | |
2575 | error = xfs_btree_check_ptr(cur, rpp, i, level); | |
2576 | if (error) | |
2577 | goto error0; | |
2578 | } | |
2579 | #endif | |
2580 | ||
2581 | xfs_btree_shift_keys(cur, rkp, 1, rrecs); | |
2582 | xfs_btree_shift_ptrs(cur, rpp, 1, rrecs); | |
2583 | ||
2584 | #ifdef DEBUG | |
2585 | error = xfs_btree_check_ptr(cur, lpp, 0, level); | |
2586 | if (error) | |
2587 | goto error0; | |
2588 | #endif | |
2589 | ||
2590 | /* Now put the new data in, and log it. */ | |
2591 | xfs_btree_copy_keys(cur, rkp, lkp, 1); | |
2592 | xfs_btree_copy_ptrs(cur, rpp, lpp, 1); | |
2593 | ||
2594 | xfs_btree_log_keys(cur, rbp, 1, rrecs + 1); | |
2595 | xfs_btree_log_ptrs(cur, rbp, 1, rrecs + 1); | |
2596 | ||
4a26e66e CH |
2597 | ASSERT(cur->bc_ops->keys_inorder(cur, rkp, |
2598 | xfs_btree_key_addr(cur, 2, right))); | |
9eaead51 CH |
2599 | } else { |
2600 | /* It's a leaf. make a hole in the records */ | |
2601 | union xfs_btree_rec *lrp; | |
2602 | union xfs_btree_rec *rrp; | |
2603 | ||
2604 | lrp = xfs_btree_rec_addr(cur, lrecs, left); | |
2605 | rrp = xfs_btree_rec_addr(cur, 1, right); | |
2606 | ||
2607 | xfs_btree_shift_recs(cur, rrp, 1, rrecs); | |
2608 | ||
2609 | /* Now put the new data in, and log it. */ | |
2610 | xfs_btree_copy_recs(cur, rrp, lrp, 1); | |
2611 | xfs_btree_log_recs(cur, rbp, 1, rrecs + 1); | |
9eaead51 CH |
2612 | } |
2613 | ||
2614 | /* | |
2615 | * Decrement and log left's numrecs, bump and log right's numrecs. | |
2616 | */ | |
2617 | xfs_btree_set_numrecs(left, --lrecs); | |
2618 | xfs_btree_log_block(cur, lbp, XFS_BB_NUMRECS); | |
2619 | ||
2620 | xfs_btree_set_numrecs(right, ++rrecs); | |
2621 | xfs_btree_log_block(cur, rbp, XFS_BB_NUMRECS); | |
2622 | ||
2623 | /* | |
2624 | * Using a temporary cursor, update the parent key values of the | |
2625 | * block on the right. | |
2626 | */ | |
2627 | error = xfs_btree_dup_cursor(cur, &tcur); | |
2628 | if (error) | |
2629 | goto error0; | |
2630 | i = xfs_btree_lastrec(tcur, level); | |
c1d22ae8 | 2631 | XFS_WANT_CORRUPTED_GOTO(tcur->bc_mp, i == 1, error0); |
9eaead51 CH |
2632 | |
2633 | error = xfs_btree_increment(tcur, level, &i); | |
2634 | if (error) | |
2635 | goto error1; | |
2636 | ||
2c813ad6 DW |
2637 | /* Update the parent high keys of the left block, if needed. */ |
2638 | if (cur->bc_flags & XFS_BTREE_OVERLAPPING) { | |
973b8319 | 2639 | error = xfs_btree_update_keys(cur, level); |
2c813ad6 DW |
2640 | if (error) |
2641 | goto error1; | |
2642 | } | |
2643 | ||
70b22659 | 2644 | /* Update the parent keys of the right block. */ |
973b8319 | 2645 | error = xfs_btree_update_keys(tcur, level); |
9eaead51 CH |
2646 | if (error) |
2647 | goto error1; | |
2648 | ||
2649 | xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); | |
2650 | ||
2651 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
2652 | *stat = 1; | |
2653 | return 0; | |
2654 | ||
2655 | out0: | |
2656 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
2657 | *stat = 0; | |
2658 | return 0; | |
2659 | ||
2660 | error0: | |
2661 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
2662 | return error; | |
2663 | ||
2664 | error1: | |
2665 | XFS_BTREE_TRACE_CURSOR(tcur, XBT_ERROR); | |
2666 | xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR); | |
2667 | return error; | |
2668 | } | |
f5eb8e7c CH |
2669 | |
2670 | /* | |
2671 | * Split cur/level block in half. | |
2672 | * Return new block number and the key to its first | |
2673 | * record (to be inserted into parent). | |
2674 | */ | |
3cc7524c | 2675 | STATIC int /* error */ |
cf11da9c | 2676 | __xfs_btree_split( |
f5eb8e7c CH |
2677 | struct xfs_btree_cur *cur, |
2678 | int level, | |
2679 | union xfs_btree_ptr *ptrp, | |
2680 | union xfs_btree_key *key, | |
2681 | struct xfs_btree_cur **curp, | |
2682 | int *stat) /* success/failure */ | |
2683 | { | |
2684 | union xfs_btree_ptr lptr; /* left sibling block ptr */ | |
2685 | struct xfs_buf *lbp; /* left buffer pointer */ | |
2686 | struct xfs_btree_block *left; /* left btree block */ | |
2687 | union xfs_btree_ptr rptr; /* right sibling block ptr */ | |
2688 | struct xfs_buf *rbp; /* right buffer pointer */ | |
2689 | struct xfs_btree_block *right; /* right btree block */ | |
2690 | union xfs_btree_ptr rrptr; /* right-right sibling ptr */ | |
2691 | struct xfs_buf *rrbp; /* right-right buffer pointer */ | |
2692 | struct xfs_btree_block *rrblock; /* right-right btree block */ | |
2693 | int lrecs; | |
2694 | int rrecs; | |
2695 | int src_index; | |
2696 | int error; /* error return value */ | |
2697 | #ifdef DEBUG | |
2698 | int i; | |
2699 | #endif | |
2700 | ||
2701 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
2702 | XFS_BTREE_TRACE_ARGIPK(cur, level, *ptrp, key); | |
2703 | ||
2704 | XFS_BTREE_STATS_INC(cur, split); | |
2705 | ||
2706 | /* Set up left block (current one). */ | |
2707 | left = xfs_btree_get_block(cur, level, &lbp); | |
2708 | ||
2709 | #ifdef DEBUG | |
2710 | error = xfs_btree_check_block(cur, left, level, lbp); | |
2711 | if (error) | |
2712 | goto error0; | |
2713 | #endif | |
2714 | ||
2715 | xfs_btree_buf_to_ptr(cur, lbp, &lptr); | |
2716 | ||
2717 | /* Allocate the new block. If we can't do it, we're toast. Give up. */ | |
6f8950cd | 2718 | error = cur->bc_ops->alloc_block(cur, &lptr, &rptr, stat); |
f5eb8e7c CH |
2719 | if (error) |
2720 | goto error0; | |
2721 | if (*stat == 0) | |
2722 | goto out0; | |
2723 | XFS_BTREE_STATS_INC(cur, alloc); | |
2724 | ||
2725 | /* Set up the new block as "right". */ | |
2726 | error = xfs_btree_get_buf_block(cur, &rptr, 0, &right, &rbp); | |
2727 | if (error) | |
2728 | goto error0; | |
2729 | ||
2730 | /* Fill in the btree header for the new right block. */ | |
ee1a47ab | 2731 | xfs_btree_init_block_cur(cur, rbp, xfs_btree_get_level(left), 0); |
f5eb8e7c CH |
2732 | |
2733 | /* | |
2734 | * Split the entries between the old and the new block evenly. | |
2735 | * Make sure that if there's an odd number of entries now, that | |
2736 | * each new block will have the same number of entries. | |
2737 | */ | |
2738 | lrecs = xfs_btree_get_numrecs(left); | |
2739 | rrecs = lrecs / 2; | |
2740 | if ((lrecs & 1) && cur->bc_ptrs[level] <= rrecs + 1) | |
2741 | rrecs++; | |
2742 | src_index = (lrecs - rrecs + 1); | |
2743 | ||
2744 | XFS_BTREE_STATS_ADD(cur, moves, rrecs); | |
2745 | ||
70b22659 DW |
2746 | /* Adjust numrecs for the later get_*_keys() calls. */ |
2747 | lrecs -= rrecs; | |
2748 | xfs_btree_set_numrecs(left, lrecs); | |
2749 | xfs_btree_set_numrecs(right, xfs_btree_get_numrecs(right) + rrecs); | |
2750 | ||
f5eb8e7c CH |
2751 | /* |
2752 | * Copy btree block entries from the left block over to the | |
2753 | * new block, the right. Update the right block and log the | |
2754 | * changes. | |
2755 | */ | |
2756 | if (level > 0) { | |
2757 | /* It's a non-leaf. Move keys and pointers. */ | |
2758 | union xfs_btree_key *lkp; /* left btree key */ | |
2759 | union xfs_btree_ptr *lpp; /* left address pointer */ | |
2760 | union xfs_btree_key *rkp; /* right btree key */ | |
2761 | union xfs_btree_ptr *rpp; /* right address pointer */ | |
2762 | ||
2763 | lkp = xfs_btree_key_addr(cur, src_index, left); | |
2764 | lpp = xfs_btree_ptr_addr(cur, src_index, left); | |
2765 | rkp = xfs_btree_key_addr(cur, 1, right); | |
2766 | rpp = xfs_btree_ptr_addr(cur, 1, right); | |
2767 | ||
2768 | #ifdef DEBUG | |
2769 | for (i = src_index; i < rrecs; i++) { | |
2770 | error = xfs_btree_check_ptr(cur, lpp, i, level); | |
2771 | if (error) | |
2772 | goto error0; | |
2773 | } | |
2774 | #endif | |
2775 | ||
70b22659 | 2776 | /* Copy the keys & pointers to the new block. */ |
f5eb8e7c CH |
2777 | xfs_btree_copy_keys(cur, rkp, lkp, rrecs); |
2778 | xfs_btree_copy_ptrs(cur, rpp, lpp, rrecs); | |
2779 | ||
2780 | xfs_btree_log_keys(cur, rbp, 1, rrecs); | |
2781 | xfs_btree_log_ptrs(cur, rbp, 1, rrecs); | |
2782 | ||
70b22659 | 2783 | /* Stash the keys of the new block for later insertion. */ |
973b8319 | 2784 | xfs_btree_get_node_keys(cur, right, key); |
f5eb8e7c CH |
2785 | } else { |
2786 | /* It's a leaf. Move records. */ | |
2787 | union xfs_btree_rec *lrp; /* left record pointer */ | |
2788 | union xfs_btree_rec *rrp; /* right record pointer */ | |
2789 | ||
2790 | lrp = xfs_btree_rec_addr(cur, src_index, left); | |
2791 | rrp = xfs_btree_rec_addr(cur, 1, right); | |
2792 | ||
70b22659 | 2793 | /* Copy records to the new block. */ |
f5eb8e7c CH |
2794 | xfs_btree_copy_recs(cur, rrp, lrp, rrecs); |
2795 | xfs_btree_log_recs(cur, rbp, 1, rrecs); | |
2796 | ||
70b22659 | 2797 | /* Stash the keys of the new block for later insertion. */ |
973b8319 | 2798 | xfs_btree_get_leaf_keys(cur, right, key); |
f5eb8e7c CH |
2799 | } |
2800 | ||
f5eb8e7c CH |
2801 | /* |
2802 | * Find the left block number by looking in the buffer. | |
70b22659 | 2803 | * Adjust sibling pointers. |
f5eb8e7c CH |
2804 | */ |
2805 | xfs_btree_get_sibling(cur, left, &rrptr, XFS_BB_RIGHTSIB); | |
2806 | xfs_btree_set_sibling(cur, right, &rrptr, XFS_BB_RIGHTSIB); | |
2807 | xfs_btree_set_sibling(cur, right, &lptr, XFS_BB_LEFTSIB); | |
2808 | xfs_btree_set_sibling(cur, left, &rptr, XFS_BB_RIGHTSIB); | |
2809 | ||
f5eb8e7c CH |
2810 | xfs_btree_log_block(cur, rbp, XFS_BB_ALL_BITS); |
2811 | xfs_btree_log_block(cur, lbp, XFS_BB_NUMRECS | XFS_BB_RIGHTSIB); | |
2812 | ||
2813 | /* | |
2814 | * If there's a block to the new block's right, make that block | |
2815 | * point back to right instead of to left. | |
2816 | */ | |
2817 | if (!xfs_btree_ptr_is_null(cur, &rrptr)) { | |
0d7409b1 | 2818 | error = xfs_btree_read_buf_block(cur, &rrptr, |
f5eb8e7c CH |
2819 | 0, &rrblock, &rrbp); |
2820 | if (error) | |
2821 | goto error0; | |
2822 | xfs_btree_set_sibling(cur, rrblock, &rptr, XFS_BB_LEFTSIB); | |
2823 | xfs_btree_log_block(cur, rrbp, XFS_BB_LEFTSIB); | |
2824 | } | |
2c813ad6 DW |
2825 | |
2826 | /* Update the parent high keys of the left block, if needed. */ | |
2827 | if (cur->bc_flags & XFS_BTREE_OVERLAPPING) { | |
973b8319 | 2828 | error = xfs_btree_update_keys(cur, level); |
2c813ad6 DW |
2829 | if (error) |
2830 | goto error0; | |
2831 | } | |
2832 | ||
f5eb8e7c CH |
2833 | /* |
2834 | * If the cursor is really in the right block, move it there. | |
2835 | * If it's just pointing past the last entry in left, then we'll | |
2836 | * insert there, so don't change anything in that case. | |
2837 | */ | |
2838 | if (cur->bc_ptrs[level] > lrecs + 1) { | |
2839 | xfs_btree_setbuf(cur, level, rbp); | |
2840 | cur->bc_ptrs[level] -= lrecs; | |
2841 | } | |
2842 | /* | |
2843 | * If there are more levels, we'll need another cursor which refers | |
2844 | * the right block, no matter where this cursor was. | |
2845 | */ | |
2846 | if (level + 1 < cur->bc_nlevels) { | |
2847 | error = xfs_btree_dup_cursor(cur, curp); | |
2848 | if (error) | |
2849 | goto error0; | |
2850 | (*curp)->bc_ptrs[level + 1]++; | |
2851 | } | |
2852 | *ptrp = rptr; | |
2853 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
2854 | *stat = 1; | |
2855 | return 0; | |
2856 | out0: | |
2857 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
2858 | *stat = 0; | |
2859 | return 0; | |
2860 | ||
2861 | error0: | |
2862 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
2863 | return error; | |
2864 | } | |
344207ce | 2865 | |
cf11da9c DC |
2866 | struct xfs_btree_split_args { |
2867 | struct xfs_btree_cur *cur; | |
2868 | int level; | |
2869 | union xfs_btree_ptr *ptrp; | |
2870 | union xfs_btree_key *key; | |
2871 | struct xfs_btree_cur **curp; | |
2872 | int *stat; /* success/failure */ | |
2873 | int result; | |
2874 | bool kswapd; /* allocation in kswapd context */ | |
2875 | struct completion *done; | |
2876 | struct work_struct work; | |
2877 | }; | |
2878 | ||
2879 | /* | |
2880 | * Stack switching interfaces for allocation | |
2881 | */ | |
2882 | static void | |
2883 | xfs_btree_split_worker( | |
2884 | struct work_struct *work) | |
2885 | { | |
2886 | struct xfs_btree_split_args *args = container_of(work, | |
2887 | struct xfs_btree_split_args, work); | |
2888 | unsigned long pflags; | |
9070733b | 2889 | unsigned long new_pflags = PF_MEMALLOC_NOFS; |
cf11da9c DC |
2890 | |
2891 | /* | |
2892 | * we are in a transaction context here, but may also be doing work | |
2893 | * in kswapd context, and hence we may need to inherit that state | |
2894 | * temporarily to ensure that we don't block waiting for memory reclaim | |
2895 | * in any way. | |
2896 | */ | |
2897 | if (args->kswapd) | |
2898 | new_pflags |= PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD; | |
2899 | ||
2900 | current_set_flags_nested(&pflags, new_pflags); | |
2901 | ||
2902 | args->result = __xfs_btree_split(args->cur, args->level, args->ptrp, | |
2903 | args->key, args->curp, args->stat); | |
2904 | complete(args->done); | |
2905 | ||
2906 | current_restore_flags_nested(&pflags, new_pflags); | |
2907 | } | |
2908 | ||
2909 | /* | |
2910 | * BMBT split requests often come in with little stack to work on. Push | |
2911 | * them off to a worker thread so there is lots of stack to use. For the other | |
2912 | * btree types, just call directly to avoid the context switch overhead here. | |
2913 | */ | |
2914 | STATIC int /* error */ | |
2915 | xfs_btree_split( | |
2916 | struct xfs_btree_cur *cur, | |
2917 | int level, | |
2918 | union xfs_btree_ptr *ptrp, | |
2919 | union xfs_btree_key *key, | |
2920 | struct xfs_btree_cur **curp, | |
2921 | int *stat) /* success/failure */ | |
2922 | { | |
2923 | struct xfs_btree_split_args args; | |
2924 | DECLARE_COMPLETION_ONSTACK(done); | |
2925 | ||
2926 | if (cur->bc_btnum != XFS_BTNUM_BMAP) | |
2927 | return __xfs_btree_split(cur, level, ptrp, key, curp, stat); | |
2928 | ||
2929 | args.cur = cur; | |
2930 | args.level = level; | |
2931 | args.ptrp = ptrp; | |
2932 | args.key = key; | |
2933 | args.curp = curp; | |
2934 | args.stat = stat; | |
2935 | args.done = &done; | |
2936 | args.kswapd = current_is_kswapd(); | |
2937 | INIT_WORK_ONSTACK(&args.work, xfs_btree_split_worker); | |
2938 | queue_work(xfs_alloc_wq, &args.work); | |
2939 | wait_for_completion(&done); | |
2940 | destroy_work_on_stack(&args.work); | |
2941 | return args.result; | |
2942 | } | |
2943 | ||
2944 | ||
ea77b0a6 CH |
2945 | /* |
2946 | * Copy the old inode root contents into a real block and make the | |
2947 | * broot point to it. | |
2948 | */ | |
2949 | int /* error */ | |
2950 | xfs_btree_new_iroot( | |
2951 | struct xfs_btree_cur *cur, /* btree cursor */ | |
2952 | int *logflags, /* logging flags for inode */ | |
2953 | int *stat) /* return status - 0 fail */ | |
2954 | { | |
2955 | struct xfs_buf *cbp; /* buffer for cblock */ | |
2956 | struct xfs_btree_block *block; /* btree block */ | |
2957 | struct xfs_btree_block *cblock; /* child btree block */ | |
2958 | union xfs_btree_key *ckp; /* child key pointer */ | |
2959 | union xfs_btree_ptr *cpp; /* child ptr pointer */ | |
2960 | union xfs_btree_key *kp; /* pointer to btree key */ | |
2961 | union xfs_btree_ptr *pp; /* pointer to block addr */ | |
2962 | union xfs_btree_ptr nptr; /* new block addr */ | |
2963 | int level; /* btree level */ | |
2964 | int error; /* error return code */ | |
2965 | #ifdef DEBUG | |
2966 | int i; /* loop counter */ | |
2967 | #endif | |
2968 | ||
2969 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
2970 | XFS_BTREE_STATS_INC(cur, newroot); | |
2971 | ||
2972 | ASSERT(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE); | |
2973 | ||
2974 | level = cur->bc_nlevels - 1; | |
2975 | ||
2976 | block = xfs_btree_get_iroot(cur); | |
2977 | pp = xfs_btree_ptr_addr(cur, 1, block); | |
2978 | ||
2979 | /* Allocate the new block. If we can't do it, we're toast. Give up. */ | |
6f8950cd | 2980 | error = cur->bc_ops->alloc_block(cur, pp, &nptr, stat); |
ea77b0a6 CH |
2981 | if (error) |
2982 | goto error0; | |
2983 | if (*stat == 0) { | |
2984 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
2985 | return 0; | |
2986 | } | |
2987 | XFS_BTREE_STATS_INC(cur, alloc); | |
2988 | ||
2989 | /* Copy the root into a real block. */ | |
2990 | error = xfs_btree_get_buf_block(cur, &nptr, 0, &cblock, &cbp); | |
2991 | if (error) | |
2992 | goto error0; | |
2993 | ||
088c9f67 DC |
2994 | /* |
2995 | * we can't just memcpy() the root in for CRC enabled btree blocks. | |
2996 | * In that case have to also ensure the blkno remains correct | |
2997 | */ | |
ea77b0a6 | 2998 | memcpy(cblock, block, xfs_btree_block_len(cur)); |
088c9f67 DC |
2999 | if (cur->bc_flags & XFS_BTREE_CRC_BLOCKS) { |
3000 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) | |
3001 | cblock->bb_u.l.bb_blkno = cpu_to_be64(cbp->b_bn); | |
3002 | else | |
3003 | cblock->bb_u.s.bb_blkno = cpu_to_be64(cbp->b_bn); | |
3004 | } | |
ea77b0a6 CH |
3005 | |
3006 | be16_add_cpu(&block->bb_level, 1); | |
3007 | xfs_btree_set_numrecs(block, 1); | |
3008 | cur->bc_nlevels++; | |
3009 | cur->bc_ptrs[level + 1] = 1; | |
3010 | ||
3011 | kp = xfs_btree_key_addr(cur, 1, block); | |
3012 | ckp = xfs_btree_key_addr(cur, 1, cblock); | |
3013 | xfs_btree_copy_keys(cur, ckp, kp, xfs_btree_get_numrecs(cblock)); | |
3014 | ||
3015 | cpp = xfs_btree_ptr_addr(cur, 1, cblock); | |
3016 | #ifdef DEBUG | |
3017 | for (i = 0; i < be16_to_cpu(cblock->bb_numrecs); i++) { | |
3018 | error = xfs_btree_check_ptr(cur, pp, i, level); | |
3019 | if (error) | |
3020 | goto error0; | |
3021 | } | |
3022 | #endif | |
3023 | xfs_btree_copy_ptrs(cur, cpp, pp, xfs_btree_get_numrecs(cblock)); | |
3024 | ||
3025 | #ifdef DEBUG | |
3026 | error = xfs_btree_check_ptr(cur, &nptr, 0, level); | |
3027 | if (error) | |
3028 | goto error0; | |
3029 | #endif | |
3030 | xfs_btree_copy_ptrs(cur, pp, &nptr, 1); | |
3031 | ||
3032 | xfs_iroot_realloc(cur->bc_private.b.ip, | |
3033 | 1 - xfs_btree_get_numrecs(cblock), | |
3034 | cur->bc_private.b.whichfork); | |
3035 | ||
3036 | xfs_btree_setbuf(cur, level, cbp); | |
3037 | ||
3038 | /* | |
3039 | * Do all this logging at the end so that | |
3040 | * the root is at the right level. | |
3041 | */ | |
3042 | xfs_btree_log_block(cur, cbp, XFS_BB_ALL_BITS); | |
3043 | xfs_btree_log_keys(cur, cbp, 1, be16_to_cpu(cblock->bb_numrecs)); | |
3044 | xfs_btree_log_ptrs(cur, cbp, 1, be16_to_cpu(cblock->bb_numrecs)); | |
3045 | ||
3046 | *logflags |= | |
9d87c319 | 3047 | XFS_ILOG_CORE | xfs_ilog_fbroot(cur->bc_private.b.whichfork); |
ea77b0a6 CH |
3048 | *stat = 1; |
3049 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3050 | return 0; | |
3051 | error0: | |
3052 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
3053 | return error; | |
3054 | } | |
3055 | ||
344207ce CH |
3056 | /* |
3057 | * Allocate a new root block, fill it in. | |
3058 | */ | |
3cc7524c | 3059 | STATIC int /* error */ |
344207ce CH |
3060 | xfs_btree_new_root( |
3061 | struct xfs_btree_cur *cur, /* btree cursor */ | |
3062 | int *stat) /* success/failure */ | |
3063 | { | |
3064 | struct xfs_btree_block *block; /* one half of the old root block */ | |
3065 | struct xfs_buf *bp; /* buffer containing block */ | |
3066 | int error; /* error return value */ | |
3067 | struct xfs_buf *lbp; /* left buffer pointer */ | |
3068 | struct xfs_btree_block *left; /* left btree block */ | |
3069 | struct xfs_buf *nbp; /* new (root) buffer */ | |
3070 | struct xfs_btree_block *new; /* new (root) btree block */ | |
3071 | int nptr; /* new value for key index, 1 or 2 */ | |
3072 | struct xfs_buf *rbp; /* right buffer pointer */ | |
3073 | struct xfs_btree_block *right; /* right btree block */ | |
3074 | union xfs_btree_ptr rptr; | |
3075 | union xfs_btree_ptr lptr; | |
3076 | ||
3077 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
3078 | XFS_BTREE_STATS_INC(cur, newroot); | |
3079 | ||
3080 | /* initialise our start point from the cursor */ | |
3081 | cur->bc_ops->init_ptr_from_cur(cur, &rptr); | |
3082 | ||
3083 | /* Allocate the new block. If we can't do it, we're toast. Give up. */ | |
6f8950cd | 3084 | error = cur->bc_ops->alloc_block(cur, &rptr, &lptr, stat); |
344207ce CH |
3085 | if (error) |
3086 | goto error0; | |
3087 | if (*stat == 0) | |
3088 | goto out0; | |
3089 | XFS_BTREE_STATS_INC(cur, alloc); | |
3090 | ||
3091 | /* Set up the new block. */ | |
3092 | error = xfs_btree_get_buf_block(cur, &lptr, 0, &new, &nbp); | |
3093 | if (error) | |
3094 | goto error0; | |
3095 | ||
3096 | /* Set the root in the holding structure increasing the level by 1. */ | |
3097 | cur->bc_ops->set_root(cur, &lptr, 1); | |
3098 | ||
3099 | /* | |
3100 | * At the previous root level there are now two blocks: the old root, | |
3101 | * and the new block generated when it was split. We don't know which | |
3102 | * one the cursor is pointing at, so we set up variables "left" and | |
3103 | * "right" for each case. | |
3104 | */ | |
3105 | block = xfs_btree_get_block(cur, cur->bc_nlevels - 1, &bp); | |
3106 | ||
3107 | #ifdef DEBUG | |
3108 | error = xfs_btree_check_block(cur, block, cur->bc_nlevels - 1, bp); | |
3109 | if (error) | |
3110 | goto error0; | |
3111 | #endif | |
3112 | ||
3113 | xfs_btree_get_sibling(cur, block, &rptr, XFS_BB_RIGHTSIB); | |
3114 | if (!xfs_btree_ptr_is_null(cur, &rptr)) { | |
3115 | /* Our block is left, pick up the right block. */ | |
3116 | lbp = bp; | |
3117 | xfs_btree_buf_to_ptr(cur, lbp, &lptr); | |
3118 | left = block; | |
0d7409b1 | 3119 | error = xfs_btree_read_buf_block(cur, &rptr, 0, &right, &rbp); |
344207ce CH |
3120 | if (error) |
3121 | goto error0; | |
3122 | bp = rbp; | |
3123 | nptr = 1; | |
3124 | } else { | |
3125 | /* Our block is right, pick up the left block. */ | |
3126 | rbp = bp; | |
3127 | xfs_btree_buf_to_ptr(cur, rbp, &rptr); | |
3128 | right = block; | |
3129 | xfs_btree_get_sibling(cur, right, &lptr, XFS_BB_LEFTSIB); | |
0d7409b1 | 3130 | error = xfs_btree_read_buf_block(cur, &lptr, 0, &left, &lbp); |
344207ce CH |
3131 | if (error) |
3132 | goto error0; | |
3133 | bp = lbp; | |
3134 | nptr = 2; | |
3135 | } | |
70b22659 | 3136 | |
344207ce | 3137 | /* Fill in the new block's btree header and log it. */ |
ee1a47ab | 3138 | xfs_btree_init_block_cur(cur, nbp, cur->bc_nlevels, 2); |
344207ce CH |
3139 | xfs_btree_log_block(cur, nbp, XFS_BB_ALL_BITS); |
3140 | ASSERT(!xfs_btree_ptr_is_null(cur, &lptr) && | |
3141 | !xfs_btree_ptr_is_null(cur, &rptr)); | |
3142 | ||
3143 | /* Fill in the key data in the new root. */ | |
3144 | if (xfs_btree_get_level(left) > 0) { | |
70b22659 DW |
3145 | /* |
3146 | * Get the keys for the left block's keys and put them directly | |
3147 | * in the parent block. Do the same for the right block. | |
3148 | */ | |
973b8319 | 3149 | xfs_btree_get_node_keys(cur, left, |
70b22659 | 3150 | xfs_btree_key_addr(cur, 1, new)); |
973b8319 | 3151 | xfs_btree_get_node_keys(cur, right, |
70b22659 | 3152 | xfs_btree_key_addr(cur, 2, new)); |
344207ce | 3153 | } else { |
70b22659 DW |
3154 | /* |
3155 | * Get the keys for the left block's records and put them | |
3156 | * directly in the parent block. Do the same for the right | |
3157 | * block. | |
3158 | */ | |
973b8319 | 3159 | xfs_btree_get_leaf_keys(cur, left, |
70b22659 | 3160 | xfs_btree_key_addr(cur, 1, new)); |
973b8319 | 3161 | xfs_btree_get_leaf_keys(cur, right, |
70b22659 | 3162 | xfs_btree_key_addr(cur, 2, new)); |
344207ce CH |
3163 | } |
3164 | xfs_btree_log_keys(cur, nbp, 1, 2); | |
3165 | ||
3166 | /* Fill in the pointer data in the new root. */ | |
3167 | xfs_btree_copy_ptrs(cur, | |
3168 | xfs_btree_ptr_addr(cur, 1, new), &lptr, 1); | |
3169 | xfs_btree_copy_ptrs(cur, | |
3170 | xfs_btree_ptr_addr(cur, 2, new), &rptr, 1); | |
3171 | xfs_btree_log_ptrs(cur, nbp, 1, 2); | |
3172 | ||
3173 | /* Fix up the cursor. */ | |
3174 | xfs_btree_setbuf(cur, cur->bc_nlevels, nbp); | |
3175 | cur->bc_ptrs[cur->bc_nlevels] = nptr; | |
3176 | cur->bc_nlevels++; | |
3177 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3178 | *stat = 1; | |
3179 | return 0; | |
3180 | error0: | |
3181 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
3182 | return error; | |
3183 | out0: | |
3184 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3185 | *stat = 0; | |
3186 | return 0; | |
3187 | } | |
4b22a571 CH |
3188 | |
3189 | STATIC int | |
3190 | xfs_btree_make_block_unfull( | |
3191 | struct xfs_btree_cur *cur, /* btree cursor */ | |
3192 | int level, /* btree level */ | |
3193 | int numrecs,/* # of recs in block */ | |
3194 | int *oindex,/* old tree index */ | |
3195 | int *index, /* new tree index */ | |
3196 | union xfs_btree_ptr *nptr, /* new btree ptr */ | |
3197 | struct xfs_btree_cur **ncur, /* new btree cursor */ | |
70b22659 | 3198 | union xfs_btree_key *key, /* key of new block */ |
4b22a571 CH |
3199 | int *stat) |
3200 | { | |
4b22a571 CH |
3201 | int error = 0; |
3202 | ||
3203 | if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) && | |
3204 | level == cur->bc_nlevels - 1) { | |
3205 | struct xfs_inode *ip = cur->bc_private.b.ip; | |
3206 | ||
3207 | if (numrecs < cur->bc_ops->get_dmaxrecs(cur, level)) { | |
3208 | /* A root block that can be made bigger. */ | |
4b22a571 | 3209 | xfs_iroot_realloc(ip, 1, cur->bc_private.b.whichfork); |
0d309791 | 3210 | *stat = 1; |
4b22a571 CH |
3211 | } else { |
3212 | /* A root block that needs replacing */ | |
3213 | int logflags = 0; | |
3214 | ||
3215 | error = xfs_btree_new_iroot(cur, &logflags, stat); | |
3216 | if (error || *stat == 0) | |
3217 | return error; | |
3218 | ||
3219 | xfs_trans_log_inode(cur->bc_tp, ip, logflags); | |
3220 | } | |
3221 | ||
3222 | return 0; | |
3223 | } | |
3224 | ||
3225 | /* First, try shifting an entry to the right neighbor. */ | |
3226 | error = xfs_btree_rshift(cur, level, stat); | |
3227 | if (error || *stat) | |
3228 | return error; | |
3229 | ||
3230 | /* Next, try shifting an entry to the left neighbor. */ | |
3231 | error = xfs_btree_lshift(cur, level, stat); | |
3232 | if (error) | |
3233 | return error; | |
3234 | ||
3235 | if (*stat) { | |
3236 | *oindex = *index = cur->bc_ptrs[level]; | |
3237 | return 0; | |
3238 | } | |
3239 | ||
3240 | /* | |
3241 | * Next, try splitting the current block in half. | |
3242 | * | |
3243 | * If this works we have to re-set our variables because we | |
3244 | * could be in a different block now. | |
3245 | */ | |
e5821e57 | 3246 | error = xfs_btree_split(cur, level, nptr, key, ncur, stat); |
4b22a571 CH |
3247 | if (error || *stat == 0) |
3248 | return error; | |
3249 | ||
3250 | ||
3251 | *index = cur->bc_ptrs[level]; | |
4b22a571 CH |
3252 | return 0; |
3253 | } | |
3254 | ||
3255 | /* | |
3256 | * Insert one record/level. Return information to the caller | |
3257 | * allowing the next level up to proceed if necessary. | |
3258 | */ | |
3259 | STATIC int | |
3260 | xfs_btree_insrec( | |
3261 | struct xfs_btree_cur *cur, /* btree cursor */ | |
3262 | int level, /* level to insert record at */ | |
3263 | union xfs_btree_ptr *ptrp, /* i/o: block number inserted */ | |
e5821e57 DW |
3264 | union xfs_btree_rec *rec, /* record to insert */ |
3265 | union xfs_btree_key *key, /* i/o: block key for ptrp */ | |
4b22a571 CH |
3266 | struct xfs_btree_cur **curp, /* output: new cursor replacing cur */ |
3267 | int *stat) /* success/failure */ | |
3268 | { | |
3269 | struct xfs_btree_block *block; /* btree block */ | |
3270 | struct xfs_buf *bp; /* buffer for block */ | |
4b22a571 CH |
3271 | union xfs_btree_ptr nptr; /* new block ptr */ |
3272 | struct xfs_btree_cur *ncur; /* new btree cursor */ | |
a1d46cff | 3273 | union xfs_btree_key nkey; /* new block key */ |
2c813ad6 | 3274 | union xfs_btree_key *lkey; |
4b22a571 CH |
3275 | int optr; /* old key/record index */ |
3276 | int ptr; /* key/record index */ | |
3277 | int numrecs;/* number of records */ | |
3278 | int error; /* error return value */ | |
3279 | #ifdef DEBUG | |
3280 | int i; | |
3281 | #endif | |
2c813ad6 | 3282 | xfs_daddr_t old_bn; |
4b22a571 CH |
3283 | |
3284 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
e5821e57 | 3285 | XFS_BTREE_TRACE_ARGIPR(cur, level, *ptrp, &rec); |
4b22a571 CH |
3286 | |
3287 | ncur = NULL; | |
a1d46cff | 3288 | lkey = &nkey; |
4b22a571 CH |
3289 | |
3290 | /* | |
3291 | * If we have an external root pointer, and we've made it to the | |
3292 | * root level, allocate a new root block and we're done. | |
3293 | */ | |
3294 | if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) && | |
3295 | (level >= cur->bc_nlevels)) { | |
3296 | error = xfs_btree_new_root(cur, stat); | |
3297 | xfs_btree_set_ptr_null(cur, ptrp); | |
3298 | ||
3299 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3300 | return error; | |
3301 | } | |
3302 | ||
3303 | /* If we're off the left edge, return failure. */ | |
3304 | ptr = cur->bc_ptrs[level]; | |
3305 | if (ptr == 0) { | |
3306 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3307 | *stat = 0; | |
3308 | return 0; | |
3309 | } | |
3310 | ||
4b22a571 CH |
3311 | optr = ptr; |
3312 | ||
3313 | XFS_BTREE_STATS_INC(cur, insrec); | |
3314 | ||
3315 | /* Get pointers to the btree buffer and block. */ | |
3316 | block = xfs_btree_get_block(cur, level, &bp); | |
2c813ad6 | 3317 | old_bn = bp ? bp->b_bn : XFS_BUF_DADDR_NULL; |
4b22a571 CH |
3318 | numrecs = xfs_btree_get_numrecs(block); |
3319 | ||
3320 | #ifdef DEBUG | |
3321 | error = xfs_btree_check_block(cur, block, level, bp); | |
3322 | if (error) | |
3323 | goto error0; | |
3324 | ||
3325 | /* Check that the new entry is being inserted in the right place. */ | |
3326 | if (ptr <= numrecs) { | |
3327 | if (level == 0) { | |
e5821e57 | 3328 | ASSERT(cur->bc_ops->recs_inorder(cur, rec, |
4a26e66e | 3329 | xfs_btree_rec_addr(cur, ptr, block))); |
4b22a571 | 3330 | } else { |
e5821e57 | 3331 | ASSERT(cur->bc_ops->keys_inorder(cur, key, |
4a26e66e | 3332 | xfs_btree_key_addr(cur, ptr, block))); |
4b22a571 CH |
3333 | } |
3334 | } | |
3335 | #endif | |
3336 | ||
3337 | /* | |
3338 | * If the block is full, we can't insert the new entry until we | |
3339 | * make the block un-full. | |
3340 | */ | |
3341 | xfs_btree_set_ptr_null(cur, &nptr); | |
3342 | if (numrecs == cur->bc_ops->get_maxrecs(cur, level)) { | |
3343 | error = xfs_btree_make_block_unfull(cur, level, numrecs, | |
2c813ad6 | 3344 | &optr, &ptr, &nptr, &ncur, lkey, stat); |
4b22a571 CH |
3345 | if (error || *stat == 0) |
3346 | goto error0; | |
3347 | } | |
3348 | ||
3349 | /* | |
3350 | * The current block may have changed if the block was | |
3351 | * previously full and we have just made space in it. | |
3352 | */ | |
3353 | block = xfs_btree_get_block(cur, level, &bp); | |
3354 | numrecs = xfs_btree_get_numrecs(block); | |
3355 | ||
3356 | #ifdef DEBUG | |
3357 | error = xfs_btree_check_block(cur, block, level, bp); | |
3358 | if (error) | |
3359 | return error; | |
3360 | #endif | |
3361 | ||
3362 | /* | |
3363 | * At this point we know there's room for our new entry in the block | |
3364 | * we're pointing at. | |
3365 | */ | |
3366 | XFS_BTREE_STATS_ADD(cur, moves, numrecs - ptr + 1); | |
3367 | ||
3368 | if (level > 0) { | |
3369 | /* It's a nonleaf. make a hole in the keys and ptrs */ | |
3370 | union xfs_btree_key *kp; | |
3371 | union xfs_btree_ptr *pp; | |
3372 | ||
3373 | kp = xfs_btree_key_addr(cur, ptr, block); | |
3374 | pp = xfs_btree_ptr_addr(cur, ptr, block); | |
3375 | ||
3376 | #ifdef DEBUG | |
3377 | for (i = numrecs - ptr; i >= 0; i--) { | |
3378 | error = xfs_btree_check_ptr(cur, pp, i, level); | |
3379 | if (error) | |
3380 | return error; | |
3381 | } | |
3382 | #endif | |
3383 | ||
3384 | xfs_btree_shift_keys(cur, kp, 1, numrecs - ptr + 1); | |
3385 | xfs_btree_shift_ptrs(cur, pp, 1, numrecs - ptr + 1); | |
3386 | ||
3387 | #ifdef DEBUG | |
3388 | error = xfs_btree_check_ptr(cur, ptrp, 0, level); | |
3389 | if (error) | |
3390 | goto error0; | |
3391 | #endif | |
3392 | ||
3393 | /* Now put the new data in, bump numrecs and log it. */ | |
e5821e57 | 3394 | xfs_btree_copy_keys(cur, kp, key, 1); |
4b22a571 CH |
3395 | xfs_btree_copy_ptrs(cur, pp, ptrp, 1); |
3396 | numrecs++; | |
3397 | xfs_btree_set_numrecs(block, numrecs); | |
3398 | xfs_btree_log_ptrs(cur, bp, ptr, numrecs); | |
3399 | xfs_btree_log_keys(cur, bp, ptr, numrecs); | |
3400 | #ifdef DEBUG | |
3401 | if (ptr < numrecs) { | |
4a26e66e CH |
3402 | ASSERT(cur->bc_ops->keys_inorder(cur, kp, |
3403 | xfs_btree_key_addr(cur, ptr + 1, block))); | |
4b22a571 CH |
3404 | } |
3405 | #endif | |
3406 | } else { | |
3407 | /* It's a leaf. make a hole in the records */ | |
3408 | union xfs_btree_rec *rp; | |
3409 | ||
3410 | rp = xfs_btree_rec_addr(cur, ptr, block); | |
3411 | ||
3412 | xfs_btree_shift_recs(cur, rp, 1, numrecs - ptr + 1); | |
3413 | ||
3414 | /* Now put the new data in, bump numrecs and log it. */ | |
e5821e57 | 3415 | xfs_btree_copy_recs(cur, rp, rec, 1); |
4b22a571 CH |
3416 | xfs_btree_set_numrecs(block, ++numrecs); |
3417 | xfs_btree_log_recs(cur, bp, ptr, numrecs); | |
3418 | #ifdef DEBUG | |
3419 | if (ptr < numrecs) { | |
4a26e66e CH |
3420 | ASSERT(cur->bc_ops->recs_inorder(cur, rp, |
3421 | xfs_btree_rec_addr(cur, ptr + 1, block))); | |
4b22a571 CH |
3422 | } |
3423 | #endif | |
3424 | } | |
3425 | ||
3426 | /* Log the new number of records in the btree header. */ | |
3427 | xfs_btree_log_block(cur, bp, XFS_BB_NUMRECS); | |
3428 | ||
2c813ad6 DW |
3429 | /* |
3430 | * If we just inserted into a new tree block, we have to | |
3431 | * recalculate nkey here because nkey is out of date. | |
3432 | * | |
3433 | * Otherwise we're just updating an existing block (having shoved | |
3434 | * some records into the new tree block), so use the regular key | |
3435 | * update mechanism. | |
3436 | */ | |
3437 | if (bp && bp->b_bn != old_bn) { | |
3438 | xfs_btree_get_keys(cur, block, lkey); | |
3439 | } else if (xfs_btree_needs_key_update(cur, optr)) { | |
973b8319 | 3440 | error = xfs_btree_update_keys(cur, level); |
4b22a571 CH |
3441 | if (error) |
3442 | goto error0; | |
3443 | } | |
3444 | ||
3445 | /* | |
3446 | * If we are tracking the last record in the tree and | |
3447 | * we are at the far right edge of the tree, update it. | |
3448 | */ | |
3449 | if (xfs_btree_is_lastrec(cur, block, level)) { | |
e5821e57 | 3450 | cur->bc_ops->update_lastrec(cur, block, rec, |
4b22a571 CH |
3451 | ptr, LASTREC_INSREC); |
3452 | } | |
3453 | ||
3454 | /* | |
3455 | * Return the new block number, if any. | |
3456 | * If there is one, give back a record value and a cursor too. | |
3457 | */ | |
3458 | *ptrp = nptr; | |
3459 | if (!xfs_btree_ptr_is_null(cur, &nptr)) { | |
2c813ad6 | 3460 | xfs_btree_copy_keys(cur, key, lkey, 1); |
4b22a571 CH |
3461 | *curp = ncur; |
3462 | } | |
3463 | ||
3464 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3465 | *stat = 1; | |
3466 | return 0; | |
3467 | ||
3468 | error0: | |
3469 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
3470 | return error; | |
3471 | } | |
3472 | ||
3473 | /* | |
3474 | * Insert the record at the point referenced by cur. | |
3475 | * | |
3476 | * A multi-level split of the tree on insert will invalidate the original | |
3477 | * cursor. All callers of this function should assume that the cursor is | |
3478 | * no longer valid and revalidate it. | |
3479 | */ | |
3480 | int | |
3481 | xfs_btree_insert( | |
3482 | struct xfs_btree_cur *cur, | |
3483 | int *stat) | |
3484 | { | |
3485 | int error; /* error return value */ | |
3486 | int i; /* result value, 0 for failure */ | |
3487 | int level; /* current level number in btree */ | |
3488 | union xfs_btree_ptr nptr; /* new block number (split result) */ | |
3489 | struct xfs_btree_cur *ncur; /* new cursor (split result) */ | |
3490 | struct xfs_btree_cur *pcur; /* previous level's cursor */ | |
a1d46cff | 3491 | union xfs_btree_key bkey; /* key of block to insert */ |
2c813ad6 | 3492 | union xfs_btree_key *key; |
4b22a571 CH |
3493 | union xfs_btree_rec rec; /* record to insert */ |
3494 | ||
3495 | level = 0; | |
3496 | ncur = NULL; | |
3497 | pcur = cur; | |
a1d46cff | 3498 | key = &bkey; |
4b22a571 CH |
3499 | |
3500 | xfs_btree_set_ptr_null(cur, &nptr); | |
e5821e57 DW |
3501 | |
3502 | /* Make a key out of the record data to be inserted, and save it. */ | |
4b22a571 | 3503 | cur->bc_ops->init_rec_from_cur(cur, &rec); |
2c813ad6 | 3504 | cur->bc_ops->init_key_from_rec(key, &rec); |
4b22a571 CH |
3505 | |
3506 | /* | |
3507 | * Loop going up the tree, starting at the leaf level. | |
3508 | * Stop when we don't get a split block, that must mean that | |
3509 | * the insert is finished with this level. | |
3510 | */ | |
3511 | do { | |
3512 | /* | |
3513 | * Insert nrec/nptr into this level of the tree. | |
3514 | * Note if we fail, nptr will be null. | |
3515 | */ | |
2c813ad6 | 3516 | error = xfs_btree_insrec(pcur, level, &nptr, &rec, key, |
e5821e57 | 3517 | &ncur, &i); |
4b22a571 CH |
3518 | if (error) { |
3519 | if (pcur != cur) | |
3520 | xfs_btree_del_cursor(pcur, XFS_BTREE_ERROR); | |
3521 | goto error0; | |
3522 | } | |
3523 | ||
c29aad41 | 3524 | XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0); |
4b22a571 CH |
3525 | level++; |
3526 | ||
3527 | /* | |
3528 | * See if the cursor we just used is trash. | |
3529 | * Can't trash the caller's cursor, but otherwise we should | |
3530 | * if ncur is a new cursor or we're about to be done. | |
3531 | */ | |
3532 | if (pcur != cur && | |
3533 | (ncur || xfs_btree_ptr_is_null(cur, &nptr))) { | |
3534 | /* Save the state from the cursor before we trash it */ | |
3535 | if (cur->bc_ops->update_cursor) | |
3536 | cur->bc_ops->update_cursor(pcur, cur); | |
3537 | cur->bc_nlevels = pcur->bc_nlevels; | |
3538 | xfs_btree_del_cursor(pcur, XFS_BTREE_NOERROR); | |
3539 | } | |
3540 | /* If we got a new cursor, switch to it. */ | |
3541 | if (ncur) { | |
3542 | pcur = ncur; | |
3543 | ncur = NULL; | |
3544 | } | |
3545 | } while (!xfs_btree_ptr_is_null(cur, &nptr)); | |
3546 | ||
3547 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3548 | *stat = i; | |
3549 | return 0; | |
3550 | error0: | |
3551 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
3552 | return error; | |
3553 | } | |
d4b3a4b7 CH |
3554 | |
3555 | /* | |
3556 | * Try to merge a non-leaf block back into the inode root. | |
3557 | * | |
3558 | * Note: the killroot names comes from the fact that we're effectively | |
3559 | * killing the old root block. But because we can't just delete the | |
3560 | * inode we have to copy the single block it was pointing to into the | |
3561 | * inode. | |
3562 | */ | |
d96f8f89 | 3563 | STATIC int |
d4b3a4b7 CH |
3564 | xfs_btree_kill_iroot( |
3565 | struct xfs_btree_cur *cur) | |
3566 | { | |
3567 | int whichfork = cur->bc_private.b.whichfork; | |
3568 | struct xfs_inode *ip = cur->bc_private.b.ip; | |
3569 | struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); | |
3570 | struct xfs_btree_block *block; | |
3571 | struct xfs_btree_block *cblock; | |
3572 | union xfs_btree_key *kp; | |
3573 | union xfs_btree_key *ckp; | |
3574 | union xfs_btree_ptr *pp; | |
3575 | union xfs_btree_ptr *cpp; | |
3576 | struct xfs_buf *cbp; | |
3577 | int level; | |
3578 | int index; | |
3579 | int numrecs; | |
196328ec | 3580 | int error; |
d4b3a4b7 CH |
3581 | #ifdef DEBUG |
3582 | union xfs_btree_ptr ptr; | |
3583 | int i; | |
3584 | #endif | |
3585 | ||
3586 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
3587 | ||
3588 | ASSERT(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE); | |
3589 | ASSERT(cur->bc_nlevels > 1); | |
3590 | ||
3591 | /* | |
3592 | * Don't deal with the root block needs to be a leaf case. | |
3593 | * We're just going to turn the thing back into extents anyway. | |
3594 | */ | |
3595 | level = cur->bc_nlevels - 1; | |
3596 | if (level == 1) | |
3597 | goto out0; | |
3598 | ||
3599 | /* | |
3600 | * Give up if the root has multiple children. | |
3601 | */ | |
3602 | block = xfs_btree_get_iroot(cur); | |
3603 | if (xfs_btree_get_numrecs(block) != 1) | |
3604 | goto out0; | |
3605 | ||
3606 | cblock = xfs_btree_get_block(cur, level - 1, &cbp); | |
3607 | numrecs = xfs_btree_get_numrecs(cblock); | |
3608 | ||
3609 | /* | |
3610 | * Only do this if the next level will fit. | |
3611 | * Then the data must be copied up to the inode, | |
3612 | * instead of freeing the root you free the next level. | |
3613 | */ | |
3614 | if (numrecs > cur->bc_ops->get_dmaxrecs(cur, level)) | |
3615 | goto out0; | |
3616 | ||
3617 | XFS_BTREE_STATS_INC(cur, killroot); | |
3618 | ||
3619 | #ifdef DEBUG | |
3620 | xfs_btree_get_sibling(cur, block, &ptr, XFS_BB_LEFTSIB); | |
3621 | ASSERT(xfs_btree_ptr_is_null(cur, &ptr)); | |
3622 | xfs_btree_get_sibling(cur, block, &ptr, XFS_BB_RIGHTSIB); | |
3623 | ASSERT(xfs_btree_ptr_is_null(cur, &ptr)); | |
3624 | #endif | |
3625 | ||
3626 | index = numrecs - cur->bc_ops->get_maxrecs(cur, level); | |
3627 | if (index) { | |
3628 | xfs_iroot_realloc(cur->bc_private.b.ip, index, | |
3629 | cur->bc_private.b.whichfork); | |
7cc95a82 | 3630 | block = ifp->if_broot; |
d4b3a4b7 CH |
3631 | } |
3632 | ||
3633 | be16_add_cpu(&block->bb_numrecs, index); | |
3634 | ASSERT(block->bb_numrecs == cblock->bb_numrecs); | |
3635 | ||
3636 | kp = xfs_btree_key_addr(cur, 1, block); | |
3637 | ckp = xfs_btree_key_addr(cur, 1, cblock); | |
3638 | xfs_btree_copy_keys(cur, kp, ckp, numrecs); | |
3639 | ||
3640 | pp = xfs_btree_ptr_addr(cur, 1, block); | |
3641 | cpp = xfs_btree_ptr_addr(cur, 1, cblock); | |
3642 | #ifdef DEBUG | |
3643 | for (i = 0; i < numrecs; i++) { | |
d4b3a4b7 CH |
3644 | error = xfs_btree_check_ptr(cur, cpp, i, level - 1); |
3645 | if (error) { | |
3646 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
3647 | return error; | |
3648 | } | |
3649 | } | |
3650 | #endif | |
3651 | xfs_btree_copy_ptrs(cur, pp, cpp, numrecs); | |
3652 | ||
c46ee8ad | 3653 | error = xfs_btree_free_block(cur, cbp); |
196328ec CH |
3654 | if (error) { |
3655 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
3656 | return error; | |
3657 | } | |
d4b3a4b7 CH |
3658 | |
3659 | cur->bc_bufs[level - 1] = NULL; | |
3660 | be16_add_cpu(&block->bb_level, -1); | |
3661 | xfs_trans_log_inode(cur->bc_tp, ip, | |
9d87c319 | 3662 | XFS_ILOG_CORE | xfs_ilog_fbroot(cur->bc_private.b.whichfork)); |
d4b3a4b7 CH |
3663 | cur->bc_nlevels--; |
3664 | out0: | |
3665 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3666 | return 0; | |
3667 | } | |
91cca5df | 3668 | |
c0e59e1a CH |
3669 | /* |
3670 | * Kill the current root node, and replace it with it's only child node. | |
3671 | */ | |
3672 | STATIC int | |
3673 | xfs_btree_kill_root( | |
3674 | struct xfs_btree_cur *cur, | |
3675 | struct xfs_buf *bp, | |
3676 | int level, | |
3677 | union xfs_btree_ptr *newroot) | |
3678 | { | |
3679 | int error; | |
3680 | ||
3681 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
3682 | XFS_BTREE_STATS_INC(cur, killroot); | |
3683 | ||
3684 | /* | |
3685 | * Update the root pointer, decreasing the level by 1 and then | |
3686 | * free the old root. | |
3687 | */ | |
3688 | cur->bc_ops->set_root(cur, newroot, -1); | |
3689 | ||
c46ee8ad | 3690 | error = xfs_btree_free_block(cur, bp); |
c0e59e1a CH |
3691 | if (error) { |
3692 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
3693 | return error; | |
3694 | } | |
3695 | ||
c0e59e1a CH |
3696 | cur->bc_bufs[level] = NULL; |
3697 | cur->bc_ra[level] = 0; | |
3698 | cur->bc_nlevels--; | |
3699 | ||
3700 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3701 | return 0; | |
3702 | } | |
3703 | ||
91cca5df CH |
3704 | STATIC int |
3705 | xfs_btree_dec_cursor( | |
3706 | struct xfs_btree_cur *cur, | |
3707 | int level, | |
3708 | int *stat) | |
3709 | { | |
3710 | int error; | |
3711 | int i; | |
3712 | ||
3713 | if (level > 0) { | |
3714 | error = xfs_btree_decrement(cur, level, &i); | |
3715 | if (error) | |
3716 | return error; | |
3717 | } | |
3718 | ||
3719 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3720 | *stat = 1; | |
3721 | return 0; | |
3722 | } | |
3723 | ||
3724 | /* | |
3725 | * Single level of the btree record deletion routine. | |
3726 | * Delete record pointed to by cur/level. | |
3727 | * Remove the record from its block then rebalance the tree. | |
3728 | * Return 0 for error, 1 for done, 2 to go on to the next level. | |
3729 | */ | |
3730 | STATIC int /* error */ | |
3731 | xfs_btree_delrec( | |
3732 | struct xfs_btree_cur *cur, /* btree cursor */ | |
3733 | int level, /* level removing record from */ | |
3734 | int *stat) /* fail/done/go-on */ | |
3735 | { | |
3736 | struct xfs_btree_block *block; /* btree block */ | |
3737 | union xfs_btree_ptr cptr; /* current block ptr */ | |
3738 | struct xfs_buf *bp; /* buffer for block */ | |
3739 | int error; /* error return value */ | |
3740 | int i; /* loop counter */ | |
91cca5df CH |
3741 | union xfs_btree_ptr lptr; /* left sibling block ptr */ |
3742 | struct xfs_buf *lbp; /* left buffer pointer */ | |
3743 | struct xfs_btree_block *left; /* left btree block */ | |
3744 | int lrecs = 0; /* left record count */ | |
3745 | int ptr; /* key/record index */ | |
3746 | union xfs_btree_ptr rptr; /* right sibling block ptr */ | |
3747 | struct xfs_buf *rbp; /* right buffer pointer */ | |
3748 | struct xfs_btree_block *right; /* right btree block */ | |
3749 | struct xfs_btree_block *rrblock; /* right-right btree block */ | |
3750 | struct xfs_buf *rrbp; /* right-right buffer pointer */ | |
3751 | int rrecs = 0; /* right record count */ | |
3752 | struct xfs_btree_cur *tcur; /* temporary btree cursor */ | |
3753 | int numrecs; /* temporary numrec count */ | |
3754 | ||
3755 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
3756 | XFS_BTREE_TRACE_ARGI(cur, level); | |
3757 | ||
3758 | tcur = NULL; | |
3759 | ||
3760 | /* Get the index of the entry being deleted, check for nothing there. */ | |
3761 | ptr = cur->bc_ptrs[level]; | |
3762 | if (ptr == 0) { | |
3763 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3764 | *stat = 0; | |
3765 | return 0; | |
3766 | } | |
3767 | ||
3768 | /* Get the buffer & block containing the record or key/ptr. */ | |
3769 | block = xfs_btree_get_block(cur, level, &bp); | |
3770 | numrecs = xfs_btree_get_numrecs(block); | |
3771 | ||
3772 | #ifdef DEBUG | |
3773 | error = xfs_btree_check_block(cur, block, level, bp); | |
3774 | if (error) | |
3775 | goto error0; | |
3776 | #endif | |
3777 | ||
3778 | /* Fail if we're off the end of the block. */ | |
3779 | if (ptr > numrecs) { | |
3780 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
3781 | *stat = 0; | |
3782 | return 0; | |
3783 | } | |
3784 | ||
3785 | XFS_BTREE_STATS_INC(cur, delrec); | |
3786 | XFS_BTREE_STATS_ADD(cur, moves, numrecs - ptr); | |
3787 | ||
3788 | /* Excise the entries being deleted. */ | |
3789 | if (level > 0) { | |
3790 | /* It's a nonleaf. operate on keys and ptrs */ | |
3791 | union xfs_btree_key *lkp; | |
3792 | union xfs_btree_ptr *lpp; | |
3793 | ||
3794 | lkp = xfs_btree_key_addr(cur, ptr + 1, block); | |
3795 | lpp = xfs_btree_ptr_addr(cur, ptr + 1, block); | |
3796 | ||
3797 | #ifdef DEBUG | |
3798 | for (i = 0; i < numrecs - ptr; i++) { | |
3799 | error = xfs_btree_check_ptr(cur, lpp, i, level); | |
3800 | if (error) | |
3801 | goto error0; | |
3802 | } | |
3803 | #endif | |
3804 | ||
3805 | if (ptr < numrecs) { | |
3806 | xfs_btree_shift_keys(cur, lkp, -1, numrecs - ptr); | |
3807 | xfs_btree_shift_ptrs(cur, lpp, -1, numrecs - ptr); | |
3808 | xfs_btree_log_keys(cur, bp, ptr, numrecs - 1); | |
3809 | xfs_btree_log_ptrs(cur, bp, ptr, numrecs - 1); | |
3810 | } | |
91cca5df CH |
3811 | } else { |
3812 | /* It's a leaf. operate on records */ | |
3813 | if (ptr < numrecs) { | |
3814 | xfs_btree_shift_recs(cur, | |
3815 | xfs_btree_rec_addr(cur, ptr + 1, block), | |
3816 | -1, numrecs - ptr); | |
3817 | xfs_btree_log_recs(cur, bp, ptr, numrecs - 1); | |
3818 | } | |
91cca5df CH |
3819 | } |
3820 | ||
3821 | /* | |
3822 | * Decrement and log the number of entries in the block. | |
3823 | */ | |
3824 | xfs_btree_set_numrecs(block, --numrecs); | |
3825 | xfs_btree_log_block(cur, bp, XFS_BB_NUMRECS); | |
3826 | ||
3827 | /* | |
3828 | * If we are tracking the last record in the tree and | |
3829 | * we are at the far right edge of the tree, update it. | |
3830 | */ | |
3831 | if (xfs_btree_is_lastrec(cur, block, level)) { | |
3832 | cur->bc_ops->update_lastrec(cur, block, NULL, | |
3833 | ptr, LASTREC_DELREC); | |
3834 | } | |
3835 | ||
3836 | /* | |
3837 | * We're at the root level. First, shrink the root block in-memory. | |
3838 | * Try to get rid of the next level down. If we can't then there's | |
3839 | * nothing left to do. | |
3840 | */ | |
3841 | if (level == cur->bc_nlevels - 1) { | |
3842 | if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) { | |
3843 | xfs_iroot_realloc(cur->bc_private.b.ip, -1, | |
3844 | cur->bc_private.b.whichfork); | |
3845 | ||
3846 | error = xfs_btree_kill_iroot(cur); | |
3847 | if (error) | |
3848 | goto error0; | |
3849 | ||
3850 | error = xfs_btree_dec_cursor(cur, level, stat); | |
3851 | if (error) | |
3852 | goto error0; | |
3853 | *stat = 1; | |
3854 | return 0; | |
3855 | } | |
3856 | ||
3857 | /* | |
3858 | * If this is the root level, and there's only one entry left, | |
3859 | * and it's NOT the leaf level, then we can get rid of this | |
3860 | * level. | |
3861 | */ | |
3862 | if (numrecs == 1 && level > 0) { | |
3863 | union xfs_btree_ptr *pp; | |
3864 | /* | |
3865 | * pp is still set to the first pointer in the block. | |
3866 | * Make it the new root of the btree. | |
3867 | */ | |
3868 | pp = xfs_btree_ptr_addr(cur, 1, block); | |
c0e59e1a | 3869 | error = xfs_btree_kill_root(cur, bp, level, pp); |
91cca5df CH |
3870 | if (error) |
3871 | goto error0; | |
3872 | } else if (level > 0) { | |
3873 | error = xfs_btree_dec_cursor(cur, level, stat); | |
3874 | if (error) | |
3875 | goto error0; | |
3876 | } | |
3877 | *stat = 1; | |
3878 | return 0; | |
3879 | } | |
3880 | ||
3881 | /* | |
3882 | * If we deleted the leftmost entry in the block, update the | |
3883 | * key values above us in the tree. | |
3884 | */ | |
70b22659 | 3885 | if (xfs_btree_needs_key_update(cur, ptr)) { |
973b8319 | 3886 | error = xfs_btree_update_keys(cur, level); |
91cca5df CH |
3887 | if (error) |
3888 | goto error0; | |
3889 | } | |
3890 | ||
3891 | /* | |
3892 | * If the number of records remaining in the block is at least | |
3893 | * the minimum, we're done. | |
3894 | */ | |
3895 | if (numrecs >= cur->bc_ops->get_minrecs(cur, level)) { | |
3896 | error = xfs_btree_dec_cursor(cur, level, stat); | |
3897 | if (error) | |
3898 | goto error0; | |
3899 | return 0; | |
3900 | } | |
3901 | ||
3902 | /* | |
3903 | * Otherwise, we have to move some records around to keep the | |
3904 | * tree balanced. Look at the left and right sibling blocks to | |
3905 | * see if we can re-balance by moving only one record. | |
3906 | */ | |
3907 | xfs_btree_get_sibling(cur, block, &rptr, XFS_BB_RIGHTSIB); | |
3908 | xfs_btree_get_sibling(cur, block, &lptr, XFS_BB_LEFTSIB); | |
3909 | ||
3910 | if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) { | |
3911 | /* | |
3912 | * One child of root, need to get a chance to copy its contents | |
3913 | * into the root and delete it. Can't go up to next level, | |
3914 | * there's nothing to delete there. | |
3915 | */ | |
3916 | if (xfs_btree_ptr_is_null(cur, &rptr) && | |
3917 | xfs_btree_ptr_is_null(cur, &lptr) && | |
3918 | level == cur->bc_nlevels - 2) { | |
3919 | error = xfs_btree_kill_iroot(cur); | |
3920 | if (!error) | |
3921 | error = xfs_btree_dec_cursor(cur, level, stat); | |
3922 | if (error) | |
3923 | goto error0; | |
3924 | return 0; | |
3925 | } | |
3926 | } | |
3927 | ||
3928 | ASSERT(!xfs_btree_ptr_is_null(cur, &rptr) || | |
3929 | !xfs_btree_ptr_is_null(cur, &lptr)); | |
3930 | ||
3931 | /* | |
3932 | * Duplicate the cursor so our btree manipulations here won't | |
3933 | * disrupt the next level up. | |
3934 | */ | |
3935 | error = xfs_btree_dup_cursor(cur, &tcur); | |
3936 | if (error) | |
3937 | goto error0; | |
3938 | ||
3939 | /* | |
3940 | * If there's a right sibling, see if it's ok to shift an entry | |
3941 | * out of it. | |
3942 | */ | |
3943 | if (!xfs_btree_ptr_is_null(cur, &rptr)) { | |
3944 | /* | |
3945 | * Move the temp cursor to the last entry in the next block. | |
3946 | * Actually any entry but the first would suffice. | |
3947 | */ | |
3948 | i = xfs_btree_lastrec(tcur, level); | |
c29aad41 | 3949 | XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0); |
91cca5df CH |
3950 | |
3951 | error = xfs_btree_increment(tcur, level, &i); | |
3952 | if (error) | |
3953 | goto error0; | |
c29aad41 | 3954 | XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0); |
91cca5df CH |
3955 | |
3956 | i = xfs_btree_lastrec(tcur, level); | |
c29aad41 | 3957 | XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0); |
91cca5df CH |
3958 | |
3959 | /* Grab a pointer to the block. */ | |
3960 | right = xfs_btree_get_block(tcur, level, &rbp); | |
3961 | #ifdef DEBUG | |
3962 | error = xfs_btree_check_block(tcur, right, level, rbp); | |
3963 | if (error) | |
3964 | goto error0; | |
3965 | #endif | |
3966 | /* Grab the current block number, for future use. */ | |
3967 | xfs_btree_get_sibling(tcur, right, &cptr, XFS_BB_LEFTSIB); | |
3968 | ||
3969 | /* | |
3970 | * If right block is full enough so that removing one entry | |
3971 | * won't make it too empty, and left-shifting an entry out | |
3972 | * of right to us works, we're done. | |
3973 | */ | |
3974 | if (xfs_btree_get_numrecs(right) - 1 >= | |
3975 | cur->bc_ops->get_minrecs(tcur, level)) { | |
3976 | error = xfs_btree_lshift(tcur, level, &i); | |
3977 | if (error) | |
3978 | goto error0; | |
3979 | if (i) { | |
3980 | ASSERT(xfs_btree_get_numrecs(block) >= | |
3981 | cur->bc_ops->get_minrecs(tcur, level)); | |
3982 | ||
3983 | xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); | |
3984 | tcur = NULL; | |
3985 | ||
3986 | error = xfs_btree_dec_cursor(cur, level, stat); | |
3987 | if (error) | |
3988 | goto error0; | |
3989 | return 0; | |
3990 | } | |
3991 | } | |
3992 | ||
3993 | /* | |
3994 | * Otherwise, grab the number of records in right for | |
3995 | * future reference, and fix up the temp cursor to point | |
3996 | * to our block again (last record). | |
3997 | */ | |
3998 | rrecs = xfs_btree_get_numrecs(right); | |
3999 | if (!xfs_btree_ptr_is_null(cur, &lptr)) { | |
4000 | i = xfs_btree_firstrec(tcur, level); | |
c29aad41 | 4001 | XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0); |
91cca5df CH |
4002 | |
4003 | error = xfs_btree_decrement(tcur, level, &i); | |
4004 | if (error) | |
4005 | goto error0; | |
c29aad41 | 4006 | XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0); |
91cca5df CH |
4007 | } |
4008 | } | |
4009 | ||
4010 | /* | |
4011 | * If there's a left sibling, see if it's ok to shift an entry | |
4012 | * out of it. | |
4013 | */ | |
4014 | if (!xfs_btree_ptr_is_null(cur, &lptr)) { | |
4015 | /* | |
4016 | * Move the temp cursor to the first entry in the | |
4017 | * previous block. | |
4018 | */ | |
4019 | i = xfs_btree_firstrec(tcur, level); | |
c29aad41 | 4020 | XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0); |
91cca5df CH |
4021 | |
4022 | error = xfs_btree_decrement(tcur, level, &i); | |
4023 | if (error) | |
4024 | goto error0; | |
4025 | i = xfs_btree_firstrec(tcur, level); | |
c29aad41 | 4026 | XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, i == 1, error0); |
91cca5df CH |
4027 | |
4028 | /* Grab a pointer to the block. */ | |
4029 | left = xfs_btree_get_block(tcur, level, &lbp); | |
4030 | #ifdef DEBUG | |
4031 | error = xfs_btree_check_block(cur, left, level, lbp); | |
4032 | if (error) | |
4033 | goto error0; | |
4034 | #endif | |
4035 | /* Grab the current block number, for future use. */ | |
4036 | xfs_btree_get_sibling(tcur, left, &cptr, XFS_BB_RIGHTSIB); | |
4037 | ||
4038 | /* | |
4039 | * If left block is full enough so that removing one entry | |
4040 | * won't make it too empty, and right-shifting an entry out | |
4041 | * of left to us works, we're done. | |
4042 | */ | |
4043 | if (xfs_btree_get_numrecs(left) - 1 >= | |
4044 | cur->bc_ops->get_minrecs(tcur, level)) { | |
4045 | error = xfs_btree_rshift(tcur, level, &i); | |
4046 | if (error) | |
4047 | goto error0; | |
4048 | if (i) { | |
4049 | ASSERT(xfs_btree_get_numrecs(block) >= | |
4050 | cur->bc_ops->get_minrecs(tcur, level)); | |
4051 | xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); | |
4052 | tcur = NULL; | |
4053 | if (level == 0) | |
4054 | cur->bc_ptrs[0]++; | |
4055 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
4056 | *stat = 1; | |
4057 | return 0; | |
4058 | } | |
4059 | } | |
4060 | ||
4061 | /* | |
4062 | * Otherwise, grab the number of records in right for | |
4063 | * future reference. | |
4064 | */ | |
4065 | lrecs = xfs_btree_get_numrecs(left); | |
4066 | } | |
4067 | ||
4068 | /* Delete the temp cursor, we're done with it. */ | |
4069 | xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); | |
4070 | tcur = NULL; | |
4071 | ||
4072 | /* If here, we need to do a join to keep the tree balanced. */ | |
4073 | ASSERT(!xfs_btree_ptr_is_null(cur, &cptr)); | |
4074 | ||
4075 | if (!xfs_btree_ptr_is_null(cur, &lptr) && | |
4076 | lrecs + xfs_btree_get_numrecs(block) <= | |
4077 | cur->bc_ops->get_maxrecs(cur, level)) { | |
4078 | /* | |
4079 | * Set "right" to be the starting block, | |
4080 | * "left" to be the left neighbor. | |
4081 | */ | |
4082 | rptr = cptr; | |
4083 | right = block; | |
4084 | rbp = bp; | |
0d7409b1 | 4085 | error = xfs_btree_read_buf_block(cur, &lptr, 0, &left, &lbp); |
91cca5df CH |
4086 | if (error) |
4087 | goto error0; | |
4088 | ||
4089 | /* | |
4090 | * If that won't work, see if we can join with the right neighbor block. | |
4091 | */ | |
4092 | } else if (!xfs_btree_ptr_is_null(cur, &rptr) && | |
4093 | rrecs + xfs_btree_get_numrecs(block) <= | |
4094 | cur->bc_ops->get_maxrecs(cur, level)) { | |
4095 | /* | |
4096 | * Set "left" to be the starting block, | |
4097 | * "right" to be the right neighbor. | |
4098 | */ | |
4099 | lptr = cptr; | |
4100 | left = block; | |
4101 | lbp = bp; | |
0d7409b1 | 4102 | error = xfs_btree_read_buf_block(cur, &rptr, 0, &right, &rbp); |
91cca5df CH |
4103 | if (error) |
4104 | goto error0; | |
4105 | ||
4106 | /* | |
4107 | * Otherwise, we can't fix the imbalance. | |
4108 | * Just return. This is probably a logic error, but it's not fatal. | |
4109 | */ | |
4110 | } else { | |
4111 | error = xfs_btree_dec_cursor(cur, level, stat); | |
4112 | if (error) | |
4113 | goto error0; | |
4114 | return 0; | |
4115 | } | |
4116 | ||
4117 | rrecs = xfs_btree_get_numrecs(right); | |
4118 | lrecs = xfs_btree_get_numrecs(left); | |
4119 | ||
4120 | /* | |
4121 | * We're now going to join "left" and "right" by moving all the stuff | |
4122 | * in "right" to "left" and deleting "right". | |
4123 | */ | |
4124 | XFS_BTREE_STATS_ADD(cur, moves, rrecs); | |
4125 | if (level > 0) { | |
4126 | /* It's a non-leaf. Move keys and pointers. */ | |
4127 | union xfs_btree_key *lkp; /* left btree key */ | |
4128 | union xfs_btree_ptr *lpp; /* left address pointer */ | |
4129 | union xfs_btree_key *rkp; /* right btree key */ | |
4130 | union xfs_btree_ptr *rpp; /* right address pointer */ | |
4131 | ||
4132 | lkp = xfs_btree_key_addr(cur, lrecs + 1, left); | |
4133 | lpp = xfs_btree_ptr_addr(cur, lrecs + 1, left); | |
4134 | rkp = xfs_btree_key_addr(cur, 1, right); | |
4135 | rpp = xfs_btree_ptr_addr(cur, 1, right); | |
4136 | #ifdef DEBUG | |
4137 | for (i = 1; i < rrecs; i++) { | |
4138 | error = xfs_btree_check_ptr(cur, rpp, i, level); | |
4139 | if (error) | |
4140 | goto error0; | |
4141 | } | |
4142 | #endif | |
4143 | xfs_btree_copy_keys(cur, lkp, rkp, rrecs); | |
4144 | xfs_btree_copy_ptrs(cur, lpp, rpp, rrecs); | |
4145 | ||
4146 | xfs_btree_log_keys(cur, lbp, lrecs + 1, lrecs + rrecs); | |
4147 | xfs_btree_log_ptrs(cur, lbp, lrecs + 1, lrecs + rrecs); | |
4148 | } else { | |
4149 | /* It's a leaf. Move records. */ | |
4150 | union xfs_btree_rec *lrp; /* left record pointer */ | |
4151 | union xfs_btree_rec *rrp; /* right record pointer */ | |
4152 | ||
4153 | lrp = xfs_btree_rec_addr(cur, lrecs + 1, left); | |
4154 | rrp = xfs_btree_rec_addr(cur, 1, right); | |
4155 | ||
4156 | xfs_btree_copy_recs(cur, lrp, rrp, rrecs); | |
4157 | xfs_btree_log_recs(cur, lbp, lrecs + 1, lrecs + rrecs); | |
4158 | } | |
4159 | ||
4160 | XFS_BTREE_STATS_INC(cur, join); | |
4161 | ||
4162 | /* | |
9da096fd | 4163 | * Fix up the number of records and right block pointer in the |
91cca5df CH |
4164 | * surviving block, and log it. |
4165 | */ | |
4166 | xfs_btree_set_numrecs(left, lrecs + rrecs); | |
4167 | xfs_btree_get_sibling(cur, right, &cptr, XFS_BB_RIGHTSIB), | |
4168 | xfs_btree_set_sibling(cur, left, &cptr, XFS_BB_RIGHTSIB); | |
4169 | xfs_btree_log_block(cur, lbp, XFS_BB_NUMRECS | XFS_BB_RIGHTSIB); | |
4170 | ||
4171 | /* If there is a right sibling, point it to the remaining block. */ | |
4172 | xfs_btree_get_sibling(cur, left, &cptr, XFS_BB_RIGHTSIB); | |
4173 | if (!xfs_btree_ptr_is_null(cur, &cptr)) { | |
0d7409b1 | 4174 | error = xfs_btree_read_buf_block(cur, &cptr, 0, &rrblock, &rrbp); |
91cca5df CH |
4175 | if (error) |
4176 | goto error0; | |
4177 | xfs_btree_set_sibling(cur, rrblock, &lptr, XFS_BB_LEFTSIB); | |
4178 | xfs_btree_log_block(cur, rrbp, XFS_BB_LEFTSIB); | |
4179 | } | |
4180 | ||
4181 | /* Free the deleted block. */ | |
c46ee8ad | 4182 | error = xfs_btree_free_block(cur, rbp); |
91cca5df CH |
4183 | if (error) |
4184 | goto error0; | |
91cca5df CH |
4185 | |
4186 | /* | |
4187 | * If we joined with the left neighbor, set the buffer in the | |
4188 | * cursor to the left block, and fix up the index. | |
4189 | */ | |
4190 | if (bp != lbp) { | |
4191 | cur->bc_bufs[level] = lbp; | |
4192 | cur->bc_ptrs[level] += lrecs; | |
4193 | cur->bc_ra[level] = 0; | |
4194 | } | |
4195 | /* | |
4196 | * If we joined with the right neighbor and there's a level above | |
4197 | * us, increment the cursor at that level. | |
4198 | */ | |
4199 | else if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) || | |
4200 | (level + 1 < cur->bc_nlevels)) { | |
4201 | error = xfs_btree_increment(cur, level + 1, &i); | |
4202 | if (error) | |
4203 | goto error0; | |
4204 | } | |
4205 | ||
4206 | /* | |
4207 | * Readjust the ptr at this level if it's not a leaf, since it's | |
4208 | * still pointing at the deletion point, which makes the cursor | |
4209 | * inconsistent. If this makes the ptr 0, the caller fixes it up. | |
4210 | * We can't use decrement because it would change the next level up. | |
4211 | */ | |
4212 | if (level > 0) | |
4213 | cur->bc_ptrs[level]--; | |
4214 | ||
2c813ad6 DW |
4215 | /* |
4216 | * We combined blocks, so we have to update the parent keys if the | |
4217 | * btree supports overlapped intervals. However, bc_ptrs[level + 1] | |
4218 | * points to the old block so that the caller knows which record to | |
4219 | * delete. Therefore, the caller must be savvy enough to call updkeys | |
4220 | * for us if we return stat == 2. The other exit points from this | |
4221 | * function don't require deletions further up the tree, so they can | |
4222 | * call updkeys directly. | |
4223 | */ | |
4224 | ||
91cca5df CH |
4225 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); |
4226 | /* Return value means the next level up has something to do. */ | |
4227 | *stat = 2; | |
4228 | return 0; | |
4229 | ||
4230 | error0: | |
4231 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
4232 | if (tcur) | |
4233 | xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR); | |
4234 | return error; | |
4235 | } | |
4236 | ||
4237 | /* | |
4238 | * Delete the record pointed to by cur. | |
4239 | * The cursor refers to the place where the record was (could be inserted) | |
4240 | * when the operation returns. | |
4241 | */ | |
4242 | int /* error */ | |
4243 | xfs_btree_delete( | |
4244 | struct xfs_btree_cur *cur, | |
4245 | int *stat) /* success/failure */ | |
4246 | { | |
4247 | int error; /* error return value */ | |
4248 | int level; | |
4249 | int i; | |
2c813ad6 | 4250 | bool joined = false; |
91cca5df CH |
4251 | |
4252 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ENTRY); | |
4253 | ||
4254 | /* | |
4255 | * Go up the tree, starting at leaf level. | |
4256 | * | |
4257 | * If 2 is returned then a join was done; go to the next level. | |
4258 | * Otherwise we are done. | |
4259 | */ | |
4260 | for (level = 0, i = 2; i == 2; level++) { | |
4261 | error = xfs_btree_delrec(cur, level, &i); | |
4262 | if (error) | |
4263 | goto error0; | |
2c813ad6 DW |
4264 | if (i == 2) |
4265 | joined = true; | |
4266 | } | |
4267 | ||
4268 | /* | |
4269 | * If we combined blocks as part of deleting the record, delrec won't | |
4270 | * have updated the parent high keys so we have to do that here. | |
4271 | */ | |
4272 | if (joined && (cur->bc_flags & XFS_BTREE_OVERLAPPING)) { | |
4273 | error = xfs_btree_updkeys_force(cur, 0); | |
4274 | if (error) | |
4275 | goto error0; | |
91cca5df CH |
4276 | } |
4277 | ||
4278 | if (i == 0) { | |
4279 | for (level = 1; level < cur->bc_nlevels; level++) { | |
4280 | if (cur->bc_ptrs[level] == 0) { | |
4281 | error = xfs_btree_decrement(cur, level, &i); | |
4282 | if (error) | |
4283 | goto error0; | |
4284 | break; | |
4285 | } | |
4286 | } | |
4287 | } | |
4288 | ||
4289 | XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT); | |
4290 | *stat = i; | |
4291 | return 0; | |
4292 | error0: | |
4293 | XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR); | |
4294 | return error; | |
4295 | } | |
8cc938fe CH |
4296 | |
4297 | /* | |
4298 | * Get the data from the pointed-to record. | |
4299 | */ | |
4300 | int /* error */ | |
4301 | xfs_btree_get_rec( | |
4302 | struct xfs_btree_cur *cur, /* btree cursor */ | |
4303 | union xfs_btree_rec **recp, /* output: btree record */ | |
4304 | int *stat) /* output: success/failure */ | |
4305 | { | |
4306 | struct xfs_btree_block *block; /* btree block */ | |
4307 | struct xfs_buf *bp; /* buffer pointer */ | |
4308 | int ptr; /* record number */ | |
4309 | #ifdef DEBUG | |
4310 | int error; /* error return value */ | |
4311 | #endif | |
4312 | ||
4313 | ptr = cur->bc_ptrs[0]; | |
4314 | block = xfs_btree_get_block(cur, 0, &bp); | |
4315 | ||
4316 | #ifdef DEBUG | |
4317 | error = xfs_btree_check_block(cur, block, 0, bp); | |
4318 | if (error) | |
4319 | return error; | |
4320 | #endif | |
4321 | ||
4322 | /* | |
4323 | * Off the right end or left end, return failure. | |
4324 | */ | |
4325 | if (ptr > xfs_btree_get_numrecs(block) || ptr <= 0) { | |
4326 | *stat = 0; | |
4327 | return 0; | |
4328 | } | |
4329 | ||
4330 | /* | |
4331 | * Point to the record and extract its data. | |
4332 | */ | |
4333 | *recp = xfs_btree_rec_addr(cur, ptr, block); | |
4334 | *stat = 1; | |
4335 | return 0; | |
4336 | } | |
21b5c978 | 4337 | |
28a89567 DW |
4338 | /* Visit a block in a btree. */ |
4339 | STATIC int | |
4340 | xfs_btree_visit_block( | |
4341 | struct xfs_btree_cur *cur, | |
4342 | int level, | |
4343 | xfs_btree_visit_blocks_fn fn, | |
4344 | void *data) | |
4345 | { | |
4346 | struct xfs_btree_block *block; | |
4347 | struct xfs_buf *bp; | |
4348 | union xfs_btree_ptr rptr; | |
4349 | int error; | |
4350 | ||
4351 | /* do right sibling readahead */ | |
4352 | xfs_btree_readahead(cur, level, XFS_BTCUR_RIGHTRA); | |
4353 | block = xfs_btree_get_block(cur, level, &bp); | |
4354 | ||
4355 | /* process the block */ | |
4356 | error = fn(cur, level, data); | |
4357 | if (error) | |
4358 | return error; | |
4359 | ||
4360 | /* now read rh sibling block for next iteration */ | |
4361 | xfs_btree_get_sibling(cur, block, &rptr, XFS_BB_RIGHTSIB); | |
4362 | if (xfs_btree_ptr_is_null(cur, &rptr)) | |
4363 | return -ENOENT; | |
4364 | ||
4365 | return xfs_btree_lookup_get_block(cur, level, &rptr, &block); | |
4366 | } | |
4367 | ||
4368 | ||
4369 | /* Visit every block in a btree. */ | |
4370 | int | |
4371 | xfs_btree_visit_blocks( | |
4372 | struct xfs_btree_cur *cur, | |
4373 | xfs_btree_visit_blocks_fn fn, | |
4374 | void *data) | |
4375 | { | |
4376 | union xfs_btree_ptr lptr; | |
4377 | int level; | |
4378 | struct xfs_btree_block *block = NULL; | |
4379 | int error = 0; | |
4380 | ||
4381 | cur->bc_ops->init_ptr_from_cur(cur, &lptr); | |
4382 | ||
4383 | /* for each level */ | |
4384 | for (level = cur->bc_nlevels - 1; level >= 0; level--) { | |
4385 | /* grab the left hand block */ | |
4386 | error = xfs_btree_lookup_get_block(cur, level, &lptr, &block); | |
4387 | if (error) | |
4388 | return error; | |
4389 | ||
4390 | /* readahead the left most block for the next level down */ | |
4391 | if (level > 0) { | |
4392 | union xfs_btree_ptr *ptr; | |
4393 | ||
4394 | ptr = xfs_btree_ptr_addr(cur, 1, block); | |
4395 | xfs_btree_readahead_ptr(cur, ptr, 1); | |
4396 | ||
4397 | /* save for the next iteration of the loop */ | |
a4d768e7 | 4398 | xfs_btree_copy_ptrs(cur, &lptr, ptr, 1); |
28a89567 DW |
4399 | } |
4400 | ||
4401 | /* for each buffer in the level */ | |
4402 | do { | |
4403 | error = xfs_btree_visit_block(cur, level, fn, data); | |
4404 | } while (!error); | |
4405 | ||
4406 | if (error != -ENOENT) | |
4407 | return error; | |
4408 | } | |
4409 | ||
4410 | return 0; | |
4411 | } | |
4412 | ||
21b5c978 DC |
4413 | /* |
4414 | * Change the owner of a btree. | |
4415 | * | |
4416 | * The mechanism we use here is ordered buffer logging. Because we don't know | |
4417 | * how many buffers were are going to need to modify, we don't really want to | |
4418 | * have to make transaction reservations for the worst case of every buffer in a | |
4419 | * full size btree as that may be more space that we can fit in the log.... | |
4420 | * | |
4421 | * We do the btree walk in the most optimal manner possible - we have sibling | |
4422 | * pointers so we can just walk all the blocks on each level from left to right | |
4423 | * in a single pass, and then move to the next level and do the same. We can | |
4424 | * also do readahead on the sibling pointers to get IO moving more quickly, | |
4425 | * though for slow disks this is unlikely to make much difference to performance | |
4426 | * as the amount of CPU work we have to do before moving to the next block is | |
4427 | * relatively small. | |
4428 | * | |
4429 | * For each btree block that we load, modify the owner appropriately, set the | |
4430 | * buffer as an ordered buffer and log it appropriately. We need to ensure that | |
4431 | * we mark the region we change dirty so that if the buffer is relogged in | |
4432 | * a subsequent transaction the changes we make here as an ordered buffer are | |
638f4416 DC |
4433 | * correctly relogged in that transaction. If we are in recovery context, then |
4434 | * just queue the modified buffer as delayed write buffer so the transaction | |
4435 | * recovery completion writes the changes to disk. | |
21b5c978 | 4436 | */ |
28a89567 DW |
4437 | struct xfs_btree_block_change_owner_info { |
4438 | __uint64_t new_owner; | |
4439 | struct list_head *buffer_list; | |
4440 | }; | |
4441 | ||
21b5c978 DC |
4442 | static int |
4443 | xfs_btree_block_change_owner( | |
4444 | struct xfs_btree_cur *cur, | |
4445 | int level, | |
28a89567 | 4446 | void *data) |
21b5c978 | 4447 | { |
28a89567 | 4448 | struct xfs_btree_block_change_owner_info *bbcoi = data; |
21b5c978 DC |
4449 | struct xfs_btree_block *block; |
4450 | struct xfs_buf *bp; | |
21b5c978 DC |
4451 | |
4452 | /* modify the owner */ | |
4453 | block = xfs_btree_get_block(cur, level, &bp); | |
4454 | if (cur->bc_flags & XFS_BTREE_LONG_PTRS) | |
28a89567 | 4455 | block->bb_u.l.bb_owner = cpu_to_be64(bbcoi->new_owner); |
21b5c978 | 4456 | else |
28a89567 | 4457 | block->bb_u.s.bb_owner = cpu_to_be32(bbcoi->new_owner); |
21b5c978 DC |
4458 | |
4459 | /* | |
638f4416 DC |
4460 | * If the block is a root block hosted in an inode, we might not have a |
4461 | * buffer pointer here and we shouldn't attempt to log the change as the | |
4462 | * information is already held in the inode and discarded when the root | |
4463 | * block is formatted into the on-disk inode fork. We still change it, | |
4464 | * though, so everything is consistent in memory. | |
21b5c978 DC |
4465 | */ |
4466 | if (bp) { | |
638f4416 DC |
4467 | if (cur->bc_tp) { |
4468 | xfs_trans_ordered_buf(cur->bc_tp, bp); | |
4469 | xfs_btree_log_block(cur, bp, XFS_BB_OWNER); | |
4470 | } else { | |
28a89567 | 4471 | xfs_buf_delwri_queue(bp, bbcoi->buffer_list); |
638f4416 | 4472 | } |
21b5c978 DC |
4473 | } else { |
4474 | ASSERT(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE); | |
4475 | ASSERT(level == cur->bc_nlevels - 1); | |
4476 | } | |
4477 | ||
28a89567 | 4478 | return 0; |
21b5c978 DC |
4479 | } |
4480 | ||
4481 | int | |
4482 | xfs_btree_change_owner( | |
4483 | struct xfs_btree_cur *cur, | |
638f4416 DC |
4484 | __uint64_t new_owner, |
4485 | struct list_head *buffer_list) | |
21b5c978 | 4486 | { |
28a89567 | 4487 | struct xfs_btree_block_change_owner_info bbcoi; |
21b5c978 | 4488 | |
28a89567 DW |
4489 | bbcoi.new_owner = new_owner; |
4490 | bbcoi.buffer_list = buffer_list; | |
21b5c978 | 4491 | |
28a89567 DW |
4492 | return xfs_btree_visit_blocks(cur, xfs_btree_block_change_owner, |
4493 | &bbcoi); | |
21b5c978 | 4494 | } |
c5ab131b DW |
4495 | |
4496 | /** | |
4497 | * xfs_btree_sblock_v5hdr_verify() -- verify the v5 fields of a short-format | |
4498 | * btree block | |
4499 | * | |
4500 | * @bp: buffer containing the btree block | |
4501 | * @max_recs: pointer to the m_*_mxr max records field in the xfs mount | |
4502 | * @pag_max_level: pointer to the per-ag max level field | |
4503 | */ | |
4504 | bool | |
4505 | xfs_btree_sblock_v5hdr_verify( | |
4506 | struct xfs_buf *bp) | |
4507 | { | |
4508 | struct xfs_mount *mp = bp->b_target->bt_mount; | |
4509 | struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); | |
4510 | struct xfs_perag *pag = bp->b_pag; | |
4511 | ||
4512 | if (!xfs_sb_version_hascrc(&mp->m_sb)) | |
4513 | return false; | |
4514 | if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid)) | |
4515 | return false; | |
4516 | if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn)) | |
4517 | return false; | |
4518 | if (pag && be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno) | |
4519 | return false; | |
4520 | return true; | |
4521 | } | |
4522 | ||
4523 | /** | |
4524 | * xfs_btree_sblock_verify() -- verify a short-format btree block | |
4525 | * | |
4526 | * @bp: buffer containing the btree block | |
4527 | * @max_recs: maximum records allowed in this btree node | |
4528 | */ | |
4529 | bool | |
4530 | xfs_btree_sblock_verify( | |
4531 | struct xfs_buf *bp, | |
4532 | unsigned int max_recs) | |
4533 | { | |
4534 | struct xfs_mount *mp = bp->b_target->bt_mount; | |
4535 | struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); | |
4536 | ||
4537 | /* numrecs verification */ | |
4538 | if (be16_to_cpu(block->bb_numrecs) > max_recs) | |
4539 | return false; | |
4540 | ||
4541 | /* sibling pointer verification */ | |
4542 | if (!block->bb_u.s.bb_leftsib || | |
4543 | (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks && | |
4544 | block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK))) | |
4545 | return false; | |
4546 | if (!block->bb_u.s.bb_rightsib || | |
4547 | (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks && | |
4548 | block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK))) | |
4549 | return false; | |
4550 | ||
4551 | return true; | |
4552 | } | |
19b54ee6 DW |
4553 | |
4554 | /* | |
4555 | * Calculate the number of btree levels needed to store a given number of | |
4556 | * records in a short-format btree. | |
4557 | */ | |
4558 | uint | |
4559 | xfs_btree_compute_maxlevels( | |
4560 | struct xfs_mount *mp, | |
4561 | uint *limits, | |
4562 | unsigned long len) | |
4563 | { | |
4564 | uint level; | |
4565 | unsigned long maxblocks; | |
4566 | ||
4567 | maxblocks = (len + limits[0] - 1) / limits[0]; | |
4568 | for (level = 1; maxblocks > 1; level++) | |
4569 | maxblocks = (maxblocks + limits[1] - 1) / limits[1]; | |
4570 | return level; | |
4571 | } | |
105f7d83 DW |
4572 | |
4573 | /* | |
4574 | * Query a regular btree for all records overlapping a given interval. | |
4575 | * Start with a LE lookup of the key of low_rec and return all records | |
4576 | * until we find a record with a key greater than the key of high_rec. | |
4577 | */ | |
4578 | STATIC int | |
4579 | xfs_btree_simple_query_range( | |
4580 | struct xfs_btree_cur *cur, | |
4581 | union xfs_btree_key *low_key, | |
4582 | union xfs_btree_key *high_key, | |
4583 | xfs_btree_query_range_fn fn, | |
4584 | void *priv) | |
4585 | { | |
4586 | union xfs_btree_rec *recp; | |
4587 | union xfs_btree_key rec_key; | |
4588 | __int64_t diff; | |
4589 | int stat; | |
4590 | bool firstrec = true; | |
4591 | int error; | |
4592 | ||
4593 | ASSERT(cur->bc_ops->init_high_key_from_rec); | |
4594 | ASSERT(cur->bc_ops->diff_two_keys); | |
4595 | ||
4596 | /* | |
4597 | * Find the leftmost record. The btree cursor must be set | |
4598 | * to the low record used to generate low_key. | |
4599 | */ | |
4600 | stat = 0; | |
4601 | error = xfs_btree_lookup(cur, XFS_LOOKUP_LE, &stat); | |
4602 | if (error) | |
4603 | goto out; | |
4604 | ||
5b5c2dbd DW |
4605 | /* Nothing? See if there's anything to the right. */ |
4606 | if (!stat) { | |
4607 | error = xfs_btree_increment(cur, 0, &stat); | |
4608 | if (error) | |
4609 | goto out; | |
4610 | } | |
4611 | ||
105f7d83 DW |
4612 | while (stat) { |
4613 | /* Find the record. */ | |
4614 | error = xfs_btree_get_rec(cur, &recp, &stat); | |
4615 | if (error || !stat) | |
4616 | break; | |
105f7d83 DW |
4617 | |
4618 | /* Skip if high_key(rec) < low_key. */ | |
4619 | if (firstrec) { | |
72227899 | 4620 | cur->bc_ops->init_high_key_from_rec(&rec_key, recp); |
105f7d83 DW |
4621 | firstrec = false; |
4622 | diff = cur->bc_ops->diff_two_keys(cur, low_key, | |
4623 | &rec_key); | |
4624 | if (diff > 0) | |
4625 | goto advloop; | |
4626 | } | |
4627 | ||
4628 | /* Stop if high_key < low_key(rec). */ | |
72227899 | 4629 | cur->bc_ops->init_key_from_rec(&rec_key, recp); |
105f7d83 DW |
4630 | diff = cur->bc_ops->diff_two_keys(cur, &rec_key, high_key); |
4631 | if (diff > 0) | |
4632 | break; | |
4633 | ||
4634 | /* Callback */ | |
4635 | error = fn(cur, recp, priv); | |
4636 | if (error < 0 || error == XFS_BTREE_QUERY_RANGE_ABORT) | |
4637 | break; | |
4638 | ||
4639 | advloop: | |
4640 | /* Move on to the next record. */ | |
4641 | error = xfs_btree_increment(cur, 0, &stat); | |
4642 | if (error) | |
4643 | break; | |
4644 | } | |
4645 | ||
4646 | out: | |
4647 | return error; | |
4648 | } | |
4649 | ||
4650 | /* | |
4651 | * Query an overlapped interval btree for all records overlapping a given | |
4652 | * interval. This function roughly follows the algorithm given in | |
4653 | * "Interval Trees" of _Introduction to Algorithms_, which is section | |
4654 | * 14.3 in the 2nd and 3rd editions. | |
4655 | * | |
4656 | * First, generate keys for the low and high records passed in. | |
4657 | * | |
4658 | * For any leaf node, generate the high and low keys for the record. | |
4659 | * If the record keys overlap with the query low/high keys, pass the | |
4660 | * record to the function iterator. | |
4661 | * | |
4662 | * For any internal node, compare the low and high keys of each | |
4663 | * pointer against the query low/high keys. If there's an overlap, | |
4664 | * follow the pointer. | |
4665 | * | |
4666 | * As an optimization, we stop scanning a block when we find a low key | |
4667 | * that is greater than the query's high key. | |
4668 | */ | |
4669 | STATIC int | |
4670 | xfs_btree_overlapped_query_range( | |
4671 | struct xfs_btree_cur *cur, | |
4672 | union xfs_btree_key *low_key, | |
4673 | union xfs_btree_key *high_key, | |
4674 | xfs_btree_query_range_fn fn, | |
4675 | void *priv) | |
4676 | { | |
4677 | union xfs_btree_ptr ptr; | |
4678 | union xfs_btree_ptr *pp; | |
4679 | union xfs_btree_key rec_key; | |
4680 | union xfs_btree_key rec_hkey; | |
4681 | union xfs_btree_key *lkp; | |
4682 | union xfs_btree_key *hkp; | |
4683 | union xfs_btree_rec *recp; | |
4684 | struct xfs_btree_block *block; | |
4685 | __int64_t ldiff; | |
4686 | __int64_t hdiff; | |
4687 | int level; | |
4688 | struct xfs_buf *bp; | |
4689 | int i; | |
4690 | int error; | |
4691 | ||
4692 | /* Load the root of the btree. */ | |
4693 | level = cur->bc_nlevels - 1; | |
4694 | cur->bc_ops->init_ptr_from_cur(cur, &ptr); | |
4695 | error = xfs_btree_lookup_get_block(cur, level, &ptr, &block); | |
4696 | if (error) | |
4697 | return error; | |
4698 | xfs_btree_get_block(cur, level, &bp); | |
4699 | trace_xfs_btree_overlapped_query_range(cur, level, bp); | |
4700 | #ifdef DEBUG | |
4701 | error = xfs_btree_check_block(cur, block, level, bp); | |
4702 | if (error) | |
4703 | goto out; | |
4704 | #endif | |
4705 | cur->bc_ptrs[level] = 1; | |
4706 | ||
4707 | while (level < cur->bc_nlevels) { | |
4708 | block = xfs_btree_get_block(cur, level, &bp); | |
4709 | ||
4710 | /* End of node, pop back towards the root. */ | |
4711 | if (cur->bc_ptrs[level] > be16_to_cpu(block->bb_numrecs)) { | |
4712 | pop_up: | |
4713 | if (level < cur->bc_nlevels - 1) | |
4714 | cur->bc_ptrs[level + 1]++; | |
4715 | level++; | |
4716 | continue; | |
4717 | } | |
4718 | ||
4719 | if (level == 0) { | |
4720 | /* Handle a leaf node. */ | |
4721 | recp = xfs_btree_rec_addr(cur, cur->bc_ptrs[0], block); | |
4722 | ||
4723 | cur->bc_ops->init_high_key_from_rec(&rec_hkey, recp); | |
4724 | ldiff = cur->bc_ops->diff_two_keys(cur, &rec_hkey, | |
4725 | low_key); | |
4726 | ||
4727 | cur->bc_ops->init_key_from_rec(&rec_key, recp); | |
4728 | hdiff = cur->bc_ops->diff_two_keys(cur, high_key, | |
4729 | &rec_key); | |
4730 | ||
4731 | /* | |
4732 | * If (record's high key >= query's low key) and | |
4733 | * (query's high key >= record's low key), then | |
4734 | * this record overlaps the query range; callback. | |
4735 | */ | |
4736 | if (ldiff >= 0 && hdiff >= 0) { | |
4737 | error = fn(cur, recp, priv); | |
4738 | if (error < 0 || | |
4739 | error == XFS_BTREE_QUERY_RANGE_ABORT) | |
4740 | break; | |
4741 | } else if (hdiff < 0) { | |
4742 | /* Record is larger than high key; pop. */ | |
4743 | goto pop_up; | |
4744 | } | |
4745 | cur->bc_ptrs[level]++; | |
4746 | continue; | |
4747 | } | |
4748 | ||
4749 | /* Handle an internal node. */ | |
4750 | lkp = xfs_btree_key_addr(cur, cur->bc_ptrs[level], block); | |
4751 | hkp = xfs_btree_high_key_addr(cur, cur->bc_ptrs[level], block); | |
4752 | pp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[level], block); | |
4753 | ||
4754 | ldiff = cur->bc_ops->diff_two_keys(cur, hkp, low_key); | |
4755 | hdiff = cur->bc_ops->diff_two_keys(cur, high_key, lkp); | |
4756 | ||
4757 | /* | |
4758 | * If (pointer's high key >= query's low key) and | |
4759 | * (query's high key >= pointer's low key), then | |
4760 | * this record overlaps the query range; follow pointer. | |
4761 | */ | |
4762 | if (ldiff >= 0 && hdiff >= 0) { | |
4763 | level--; | |
4764 | error = xfs_btree_lookup_get_block(cur, level, pp, | |
4765 | &block); | |
4766 | if (error) | |
4767 | goto out; | |
4768 | xfs_btree_get_block(cur, level, &bp); | |
4769 | trace_xfs_btree_overlapped_query_range(cur, level, bp); | |
4770 | #ifdef DEBUG | |
4771 | error = xfs_btree_check_block(cur, block, level, bp); | |
4772 | if (error) | |
4773 | goto out; | |
4774 | #endif | |
4775 | cur->bc_ptrs[level] = 1; | |
4776 | continue; | |
4777 | } else if (hdiff < 0) { | |
4778 | /* The low key is larger than the upper range; pop. */ | |
4779 | goto pop_up; | |
4780 | } | |
4781 | cur->bc_ptrs[level]++; | |
4782 | } | |
4783 | ||
4784 | out: | |
4785 | /* | |
4786 | * If we don't end this function with the cursor pointing at a record | |
4787 | * block, a subsequent non-error cursor deletion will not release | |
4788 | * node-level buffers, causing a buffer leak. This is quite possible | |
4789 | * with a zero-results range query, so release the buffers if we | |
4790 | * failed to return any results. | |
4791 | */ | |
4792 | if (cur->bc_bufs[0] == NULL) { | |
4793 | for (i = 0; i < cur->bc_nlevels; i++) { | |
4794 | if (cur->bc_bufs[i]) { | |
4795 | xfs_trans_brelse(cur->bc_tp, cur->bc_bufs[i]); | |
4796 | cur->bc_bufs[i] = NULL; | |
4797 | cur->bc_ptrs[i] = 0; | |
4798 | cur->bc_ra[i] = 0; | |
4799 | } | |
4800 | } | |
4801 | } | |
4802 | ||
4803 | return error; | |
4804 | } | |
4805 | ||
4806 | /* | |
4807 | * Query a btree for all records overlapping a given interval of keys. The | |
4808 | * supplied function will be called with each record found; return one of the | |
4809 | * XFS_BTREE_QUERY_RANGE_{CONTINUE,ABORT} values or the usual negative error | |
4810 | * code. This function returns XFS_BTREE_QUERY_RANGE_ABORT, zero, or a | |
4811 | * negative error code. | |
4812 | */ | |
4813 | int | |
4814 | xfs_btree_query_range( | |
4815 | struct xfs_btree_cur *cur, | |
4816 | union xfs_btree_irec *low_rec, | |
4817 | union xfs_btree_irec *high_rec, | |
4818 | xfs_btree_query_range_fn fn, | |
4819 | void *priv) | |
4820 | { | |
4821 | union xfs_btree_rec rec; | |
4822 | union xfs_btree_key low_key; | |
4823 | union xfs_btree_key high_key; | |
4824 | ||
4825 | /* Find the keys of both ends of the interval. */ | |
4826 | cur->bc_rec = *high_rec; | |
4827 | cur->bc_ops->init_rec_from_cur(cur, &rec); | |
4828 | cur->bc_ops->init_key_from_rec(&high_key, &rec); | |
4829 | ||
4830 | cur->bc_rec = *low_rec; | |
4831 | cur->bc_ops->init_rec_from_cur(cur, &rec); | |
4832 | cur->bc_ops->init_key_from_rec(&low_key, &rec); | |
4833 | ||
4834 | /* Enforce low key < high key. */ | |
4835 | if (cur->bc_ops->diff_two_keys(cur, &low_key, &high_key) > 0) | |
4836 | return -EINVAL; | |
4837 | ||
4838 | if (!(cur->bc_flags & XFS_BTREE_OVERLAPPING)) | |
4839 | return xfs_btree_simple_query_range(cur, &low_key, | |
4840 | &high_key, fn, priv); | |
4841 | return xfs_btree_overlapped_query_range(cur, &low_key, &high_key, | |
4842 | fn, priv); | |
4843 | } | |
4ed3f687 | 4844 | |
e9a2599a DW |
4845 | /* Query a btree for all records. */ |
4846 | int | |
4847 | xfs_btree_query_all( | |
4848 | struct xfs_btree_cur *cur, | |
4849 | xfs_btree_query_range_fn fn, | |
4850 | void *priv) | |
4851 | { | |
4852 | union xfs_btree_irec low_rec; | |
4853 | union xfs_btree_irec high_rec; | |
4854 | ||
4855 | memset(&low_rec, 0, sizeof(low_rec)); | |
4856 | memset(&high_rec, 0xFF, sizeof(high_rec)); | |
4857 | return xfs_btree_query_range(cur, &low_rec, &high_rec, fn, priv); | |
4858 | } | |
4859 | ||
4ed3f687 DW |
4860 | /* |
4861 | * Calculate the number of blocks needed to store a given number of records | |
4862 | * in a short-format (per-AG metadata) btree. | |
4863 | */ | |
4864 | xfs_extlen_t | |
4865 | xfs_btree_calc_size( | |
4866 | struct xfs_mount *mp, | |
4867 | uint *limits, | |
4868 | unsigned long long len) | |
4869 | { | |
4870 | int level; | |
4871 | int maxrecs; | |
4872 | xfs_extlen_t rval; | |
4873 | ||
4874 | maxrecs = limits[0]; | |
4875 | for (level = 0, rval = 0; len > 1; level++) { | |
4876 | len += maxrecs - 1; | |
4877 | do_div(len, maxrecs); | |
4878 | maxrecs = limits[1]; | |
4879 | rval += len; | |
4880 | } | |
4881 | return rval; | |
4882 | } | |
c611cc03 | 4883 | |
f1b8243c | 4884 | static int |
c611cc03 DW |
4885 | xfs_btree_count_blocks_helper( |
4886 | struct xfs_btree_cur *cur, | |
4887 | int level, | |
4888 | void *data) | |
4889 | { | |
4890 | xfs_extlen_t *blocks = data; | |
4891 | (*blocks)++; | |
4892 | ||
4893 | return 0; | |
4894 | } | |
4895 | ||
4896 | /* Count the blocks in a btree and return the result in *blocks. */ | |
4897 | int | |
4898 | xfs_btree_count_blocks( | |
4899 | struct xfs_btree_cur *cur, | |
4900 | xfs_extlen_t *blocks) | |
4901 | { | |
4902 | *blocks = 0; | |
4903 | return xfs_btree_visit_blocks(cur, xfs_btree_count_blocks_helper, | |
4904 | blocks); | |
4905 | } |