Line data Source code
1 : /*-------------------------------------------------------------------------
2 : *
3 : * visibilitymap.c
4 : * bitmap for tracking visibility of heap tuples
5 : *
6 : * Portions Copyright (c) 1996-2024, PostgreSQL Global Development Group
7 : * Portions Copyright (c) 1994, Regents of the University of California
8 : *
9 : *
10 : * IDENTIFICATION
11 : * src/backend/access/heap/visibilitymap.c
12 : *
13 : * INTERFACE ROUTINES
14 : * visibilitymap_clear - clear bits for one page in the visibility map
15 : * visibilitymap_pin - pin a map page for setting a bit
16 : * visibilitymap_pin_ok - check whether correct map page is already pinned
17 : * visibilitymap_set - set a bit in a previously pinned page
18 : * visibilitymap_get_status - get status of bits
19 : * visibilitymap_count - count number of bits set in visibility map
20 : * visibilitymap_prepare_truncate -
21 : * prepare for truncation of the visibility map
22 : *
23 : * NOTES
24 : *
25 : * The visibility map is a bitmap with two bits (all-visible and all-frozen)
26 : * per heap page. A set all-visible bit means that all tuples on the page are
27 : * known visible to all transactions, and therefore the page doesn't need to
28 : * be vacuumed. A set all-frozen bit means that all tuples on the page are
29 : * completely frozen, and therefore the page doesn't need to be vacuumed even
30 : * if whole table scanning vacuum is required (e.g. anti-wraparound vacuum).
31 : * The all-frozen bit must be set only when the page is already all-visible.
32 : *
33 : * The map is conservative in the sense that we make sure that whenever a bit
34 : * is set, we know the condition is true, but if a bit is not set, it might or
35 : * might not be true.
36 : *
37 : * Clearing visibility map bits is not separately WAL-logged. The callers
38 : * must make sure that whenever a bit is cleared, the bit is cleared on WAL
39 : * replay of the updating operation as well.
40 : *
41 : * When we *set* a visibility map during VACUUM, we must write WAL. This may
42 : * seem counterintuitive, since the bit is basically a hint: if it is clear,
43 : * it may still be the case that every tuple on the page is visible to all
44 : * transactions; we just don't know that for certain. The difficulty is that
45 : * there are two bits which are typically set together: the PD_ALL_VISIBLE bit
46 : * on the page itself, and the visibility map bit. If a crash occurs after the
47 : * visibility map page makes it to disk and before the updated heap page makes
48 : * it to disk, redo must set the bit on the heap page. Otherwise, the next
49 : * insert, update, or delete on the heap page will fail to realize that the
50 : * visibility map bit must be cleared, possibly causing index-only scans to
51 : * return wrong answers.
52 : *
53 : * VACUUM will normally skip pages for which the visibility map bit is set;
54 : * such pages can't contain any dead tuples and therefore don't need vacuuming.
55 : *
56 : * LOCKING
57 : *
58 : * In heapam.c, whenever a page is modified so that not all tuples on the
59 : * page are visible to everyone anymore, the corresponding bit in the
60 : * visibility map is cleared. In order to be crash-safe, we need to do this
61 : * while still holding a lock on the heap page and in the same critical
62 : * section that logs the page modification. However, we don't want to hold
63 : * the buffer lock over any I/O that may be required to read in the visibility
64 : * map page. To avoid this, we examine the heap page before locking it;
65 : * if the page-level PD_ALL_VISIBLE bit is set, we pin the visibility map
66 : * bit. Then, we lock the buffer. But this creates a race condition: there
67 : * is a possibility that in the time it takes to lock the buffer, the
68 : * PD_ALL_VISIBLE bit gets set. If that happens, we have to unlock the
69 : * buffer, pin the visibility map page, and relock the buffer. This shouldn't
70 : * happen often, because only VACUUM currently sets visibility map bits,
71 : * and the race will only occur if VACUUM processes a given page at almost
72 : * exactly the same time that someone tries to further modify it.
73 : *
74 : * To set a bit, you need to hold a lock on the heap page. That prevents
75 : * the race condition where VACUUM sees that all tuples on the page are
76 : * visible to everyone, but another backend modifies the page before VACUUM
77 : * sets the bit in the visibility map.
78 : *
79 : * When a bit is set, the LSN of the visibility map page is updated to make
80 : * sure that the visibility map update doesn't get written to disk before the
81 : * WAL record of the changes that made it possible to set the bit is flushed.
82 : * But when a bit is cleared, we don't have to do that because it's always
83 : * safe to clear a bit in the map from correctness point of view.
84 : *
85 : *-------------------------------------------------------------------------
86 : */
87 : #include "postgres.h"
88 :
89 : #include "access/heapam_xlog.h"
90 : #include "access/visibilitymap.h"
91 : #include "access/xloginsert.h"
92 : #include "access/xlogutils.h"
93 : #include "miscadmin.h"
94 : #include "port/pg_bitutils.h"
95 : #include "storage/bufmgr.h"
96 : #include "storage/smgr.h"
97 : #include "utils/inval.h"
98 : #include "utils/rel.h"
99 :
100 :
101 : /*#define TRACE_VISIBILITYMAP */
102 :
103 : /*
104 : * Size of the bitmap on each visibility map page, in bytes. There's no
105 : * extra headers, so the whole page minus the standard page header is
106 : * used for the bitmap.
107 : */
108 : #define MAPSIZE (BLCKSZ - MAXALIGN(SizeOfPageHeaderData))
109 :
110 : /* Number of heap blocks we can represent in one byte */
111 : #define HEAPBLOCKS_PER_BYTE (BITS_PER_BYTE / BITS_PER_HEAPBLOCK)
112 :
113 : /* Number of heap blocks we can represent in one visibility map page. */
114 : #define HEAPBLOCKS_PER_PAGE (MAPSIZE * HEAPBLOCKS_PER_BYTE)
115 :
116 : /* Mapping from heap block number to the right bit in the visibility map */
117 : #define HEAPBLK_TO_MAPBLOCK(x) ((x) / HEAPBLOCKS_PER_PAGE)
118 : #define HEAPBLK_TO_MAPBYTE(x) (((x) % HEAPBLOCKS_PER_PAGE) / HEAPBLOCKS_PER_BYTE)
119 : #define HEAPBLK_TO_OFFSET(x) (((x) % HEAPBLOCKS_PER_BYTE) * BITS_PER_HEAPBLOCK)
120 :
121 : /* Masks for counting subsets of bits in the visibility map. */
122 : #define VISIBLE_MASK8 (0x55) /* The lower bit of each bit pair */
123 : #define FROZEN_MASK8 (0xaa) /* The upper bit of each bit pair */
124 :
125 : /* prototypes for internal routines */
126 : static Buffer vm_readbuf(Relation rel, BlockNumber blkno, bool extend);
127 : static Buffer vm_extend(Relation rel, BlockNumber vm_nblocks);
128 :
129 :
130 : /*
131 : * visibilitymap_clear - clear specified bits for one page in visibility map
132 : *
133 : * You must pass a buffer containing the correct map page to this function.
134 : * Call visibilitymap_pin first to pin the right one. This function doesn't do
135 : * any I/O. Returns true if any bits have been cleared and false otherwise.
136 : */
137 : bool
138 26740 : visibilitymap_clear(Relation rel, BlockNumber heapBlk, Buffer vmbuf, uint8 flags)
139 : {
140 26740 : BlockNumber mapBlock = HEAPBLK_TO_MAPBLOCK(heapBlk);
141 26740 : int mapByte = HEAPBLK_TO_MAPBYTE(heapBlk);
142 26740 : int mapOffset = HEAPBLK_TO_OFFSET(heapBlk);
143 26740 : uint8 mask = flags << mapOffset;
144 : char *map;
145 26740 : bool cleared = false;
146 :
147 : /* Must never clear all_visible bit while leaving all_frozen bit set */
148 : Assert(flags & VISIBILITYMAP_VALID_BITS);
149 : Assert(flags != VISIBILITYMAP_ALL_VISIBLE);
150 :
151 : #ifdef TRACE_VISIBILITYMAP
152 : elog(DEBUG1, "vm_clear %s %d", RelationGetRelationName(rel), heapBlk);
153 : #endif
154 :
155 26740 : if (!BufferIsValid(vmbuf) || BufferGetBlockNumber(vmbuf) != mapBlock)
156 0 : elog(ERROR, "wrong buffer passed to visibilitymap_clear");
157 :
158 26740 : LockBuffer(vmbuf, BUFFER_LOCK_EXCLUSIVE);
159 26740 : map = PageGetContents(BufferGetPage(vmbuf));
160 :
161 26740 : if (map[mapByte] & mask)
162 : {
163 22702 : map[mapByte] &= ~mask;
164 :
165 22702 : MarkBufferDirty(vmbuf);
166 22702 : cleared = true;
167 : }
168 :
169 26740 : LockBuffer(vmbuf, BUFFER_LOCK_UNLOCK);
170 :
171 26740 : return cleared;
172 : }
173 :
174 : /*
175 : * visibilitymap_pin - pin a map page for setting a bit
176 : *
177 : * Setting a bit in the visibility map is a two-phase operation. First, call
178 : * visibilitymap_pin, to pin the visibility map page containing the bit for
179 : * the heap page. Because that can require I/O to read the map page, you
180 : * shouldn't hold a lock on the heap page while doing that. Then, call
181 : * visibilitymap_set to actually set the bit.
182 : *
183 : * On entry, *vmbuf should be InvalidBuffer or a valid buffer returned by
184 : * an earlier call to visibilitymap_pin or visibilitymap_get_status on the same
185 : * relation. On return, *vmbuf is a valid buffer with the map page containing
186 : * the bit for heapBlk.
187 : *
188 : * If the page doesn't exist in the map file yet, it is extended.
189 : */
190 : void
191 155768 : visibilitymap_pin(Relation rel, BlockNumber heapBlk, Buffer *vmbuf)
192 : {
193 155768 : BlockNumber mapBlock = HEAPBLK_TO_MAPBLOCK(heapBlk);
194 :
195 : /* Reuse the old pinned buffer if possible */
196 155768 : if (BufferIsValid(*vmbuf))
197 : {
198 120658 : if (BufferGetBlockNumber(*vmbuf) == mapBlock)
199 120658 : return;
200 :
201 0 : ReleaseBuffer(*vmbuf);
202 : }
203 35110 : *vmbuf = vm_readbuf(rel, mapBlock, true);
204 : }
205 :
206 : /*
207 : * visibilitymap_pin_ok - do we already have the correct page pinned?
208 : *
209 : * On entry, vmbuf should be InvalidBuffer or a valid buffer returned by
210 : * an earlier call to visibilitymap_pin or visibilitymap_get_status on the same
211 : * relation. The return value indicates whether the buffer covers the
212 : * given heapBlk.
213 : */
214 : bool
215 22220 : visibilitymap_pin_ok(BlockNumber heapBlk, Buffer vmbuf)
216 : {
217 22220 : BlockNumber mapBlock = HEAPBLK_TO_MAPBLOCK(heapBlk);
218 :
219 22220 : return BufferIsValid(vmbuf) && BufferGetBlockNumber(vmbuf) == mapBlock;
220 : }
221 :
222 : /*
223 : * visibilitymap_set - set bit(s) on a previously pinned page
224 : *
225 : * recptr is the LSN of the XLOG record we're replaying, if we're in recovery,
226 : * or InvalidXLogRecPtr in normal running. The VM page LSN is advanced to the
227 : * one provided; in normal running, we generate a new XLOG record and set the
228 : * page LSN to that value (though the heap page's LSN may *not* be updated;
229 : * see below). cutoff_xid is the largest xmin on the page being marked
230 : * all-visible; it is needed for Hot Standby, and can be InvalidTransactionId
231 : * if the page contains no tuples. It can also be set to InvalidTransactionId
232 : * when a page that is already all-visible is being marked all-frozen.
233 : *
234 : * Caller is expected to set the heap page's PD_ALL_VISIBLE bit before calling
235 : * this function. Except in recovery, caller should also pass the heap
236 : * buffer. When checksums are enabled and we're not in recovery, we must add
237 : * the heap buffer to the WAL chain to protect it from being torn.
238 : *
239 : * You must pass a buffer containing the correct map page to this function.
240 : * Call visibilitymap_pin first to pin the right one. This function doesn't do
241 : * any I/O.
242 : */
243 : void
244 79476 : visibilitymap_set(Relation rel, BlockNumber heapBlk, Buffer heapBuf,
245 : XLogRecPtr recptr, Buffer vmBuf, TransactionId cutoff_xid,
246 : uint8 flags)
247 : {
248 79476 : BlockNumber mapBlock = HEAPBLK_TO_MAPBLOCK(heapBlk);
249 79476 : uint32 mapByte = HEAPBLK_TO_MAPBYTE(heapBlk);
250 79476 : uint8 mapOffset = HEAPBLK_TO_OFFSET(heapBlk);
251 : Page page;
252 : uint8 *map;
253 :
254 : #ifdef TRACE_VISIBILITYMAP
255 : elog(DEBUG1, "vm_set %s %d", RelationGetRelationName(rel), heapBlk);
256 : #endif
257 :
258 : Assert(InRecovery || XLogRecPtrIsInvalid(recptr));
259 : Assert(InRecovery || PageIsAllVisible((Page) BufferGetPage(heapBuf)));
260 : Assert((flags & VISIBILITYMAP_VALID_BITS) == flags);
261 :
262 : /* Must never set all_frozen bit without also setting all_visible bit */
263 : Assert(flags != VISIBILITYMAP_ALL_FROZEN);
264 :
265 : /* Check that we have the right heap page pinned, if present */
266 79476 : if (BufferIsValid(heapBuf) && BufferGetBlockNumber(heapBuf) != heapBlk)
267 0 : elog(ERROR, "wrong heap buffer passed to visibilitymap_set");
268 :
269 : /* Check that we have the right VM page pinned */
270 79476 : if (!BufferIsValid(vmBuf) || BufferGetBlockNumber(vmBuf) != mapBlock)
271 0 : elog(ERROR, "wrong VM buffer passed to visibilitymap_set");
272 :
273 79476 : page = BufferGetPage(vmBuf);
274 79476 : map = (uint8 *) PageGetContents(page);
275 79476 : LockBuffer(vmBuf, BUFFER_LOCK_EXCLUSIVE);
276 :
277 79476 : if (flags != (map[mapByte] >> mapOffset & VISIBILITYMAP_VALID_BITS))
278 : {
279 79476 : START_CRIT_SECTION();
280 :
281 79476 : map[mapByte] |= (flags << mapOffset);
282 79476 : MarkBufferDirty(vmBuf);
283 :
284 79476 : if (RelationNeedsWAL(rel))
285 : {
286 74664 : if (XLogRecPtrIsInvalid(recptr))
287 : {
288 : Assert(!InRecovery);
289 67778 : recptr = log_heap_visible(rel, heapBuf, vmBuf, cutoff_xid, flags);
290 :
291 : /*
292 : * If data checksums are enabled (or wal_log_hints=on), we
293 : * need to protect the heap page from being torn.
294 : *
295 : * If not, then we must *not* update the heap page's LSN. In
296 : * this case, the FPI for the heap page was omitted from the
297 : * WAL record inserted above, so it would be incorrect to
298 : * update the heap page's LSN.
299 : */
300 67778 : if (XLogHintBitIsNeeded())
301 : {
302 11818 : Page heapPage = BufferGetPage(heapBuf);
303 :
304 11818 : PageSetLSN(heapPage, recptr);
305 : }
306 : }
307 74664 : PageSetLSN(page, recptr);
308 : }
309 :
310 79476 : END_CRIT_SECTION();
311 : }
312 :
313 79476 : LockBuffer(vmBuf, BUFFER_LOCK_UNLOCK);
314 79476 : }
315 :
316 : /*
317 : * visibilitymap_get_status - get status of bits
318 : *
319 : * Are all tuples on heapBlk visible to all or are marked frozen, according
320 : * to the visibility map?
321 : *
322 : * On entry, *vmbuf should be InvalidBuffer or a valid buffer returned by an
323 : * earlier call to visibilitymap_pin or visibilitymap_get_status on the same
324 : * relation. On return, *vmbuf is a valid buffer with the map page containing
325 : * the bit for heapBlk, or InvalidBuffer. The caller is responsible for
326 : * releasing *vmbuf after it's done testing and setting bits.
327 : *
328 : * NOTE: This function is typically called without a lock on the heap page,
329 : * so somebody else could change the bit just after we look at it. In fact,
330 : * since we don't lock the visibility map page either, it's even possible that
331 : * someone else could have changed the bit just before we look at it, but yet
332 : * we might see the old value. It is the caller's responsibility to deal with
333 : * all concurrency issues!
334 : */
335 : uint8
336 5585976 : visibilitymap_get_status(Relation rel, BlockNumber heapBlk, Buffer *vmbuf)
337 : {
338 5585976 : BlockNumber mapBlock = HEAPBLK_TO_MAPBLOCK(heapBlk);
339 5585976 : uint32 mapByte = HEAPBLK_TO_MAPBYTE(heapBlk);
340 5585976 : uint8 mapOffset = HEAPBLK_TO_OFFSET(heapBlk);
341 : char *map;
342 : uint8 result;
343 :
344 : #ifdef TRACE_VISIBILITYMAP
345 : elog(DEBUG1, "vm_get_status %s %d", RelationGetRelationName(rel), heapBlk);
346 : #endif
347 :
348 : /* Reuse the old pinned buffer if possible */
349 5585976 : if (BufferIsValid(*vmbuf))
350 : {
351 3927054 : if (BufferGetBlockNumber(*vmbuf) != mapBlock)
352 : {
353 0 : ReleaseBuffer(*vmbuf);
354 0 : *vmbuf = InvalidBuffer;
355 : }
356 : }
357 :
358 5585976 : if (!BufferIsValid(*vmbuf))
359 : {
360 1658922 : *vmbuf = vm_readbuf(rel, mapBlock, false);
361 1658922 : if (!BufferIsValid(*vmbuf))
362 1551532 : return false;
363 : }
364 :
365 4034444 : map = PageGetContents(BufferGetPage(*vmbuf));
366 :
367 : /*
368 : * A single byte read is atomic. There could be memory-ordering effects
369 : * here, but for performance reasons we make it the caller's job to worry
370 : * about that.
371 : */
372 4034444 : result = ((map[mapByte] >> mapOffset) & VISIBILITYMAP_VALID_BITS);
373 4034444 : return result;
374 : }
375 :
376 : /*
377 : * visibilitymap_count - count number of bits set in visibility map
378 : *
379 : * Note: we ignore the possibility of race conditions when the table is being
380 : * extended concurrently with the call. New pages added to the table aren't
381 : * going to be marked all-visible or all-frozen, so they won't affect the result.
382 : */
383 : void
384 42696 : visibilitymap_count(Relation rel, BlockNumber *all_visible, BlockNumber *all_frozen)
385 : {
386 : BlockNumber mapBlock;
387 42696 : BlockNumber nvisible = 0;
388 42696 : BlockNumber nfrozen = 0;
389 :
390 : /* all_visible must be specified */
391 : Assert(all_visible);
392 :
393 42696 : for (mapBlock = 0;; mapBlock++)
394 11864 : {
395 : Buffer mapBuffer;
396 : uint64 *map;
397 :
398 : /*
399 : * Read till we fall off the end of the map. We assume that any extra
400 : * bytes in the last page are zeroed, so we don't bother excluding
401 : * them from the count.
402 : */
403 54560 : mapBuffer = vm_readbuf(rel, mapBlock, false);
404 54560 : if (!BufferIsValid(mapBuffer))
405 42696 : break;
406 :
407 : /*
408 : * We choose not to lock the page, since the result is going to be
409 : * immediately stale anyway if anyone is concurrently setting or
410 : * clearing bits, and we only really need an approximate value.
411 : */
412 11864 : map = (uint64 *) PageGetContents(BufferGetPage(mapBuffer));
413 :
414 11864 : nvisible += pg_popcount_masked((const char *) map, MAPSIZE, VISIBLE_MASK8);
415 11864 : if (all_frozen)
416 0 : nfrozen += pg_popcount_masked((const char *) map, MAPSIZE, FROZEN_MASK8);
417 :
418 11864 : ReleaseBuffer(mapBuffer);
419 : }
420 :
421 42696 : *all_visible = nvisible;
422 42696 : if (all_frozen)
423 0 : *all_frozen = nfrozen;
424 42696 : }
425 :
426 : /*
427 : * visibilitymap_prepare_truncate -
428 : * prepare for truncation of the visibility map
429 : *
430 : * nheapblocks is the new size of the heap.
431 : *
432 : * Return the number of blocks of new visibility map.
433 : * If it's InvalidBlockNumber, there is nothing to truncate;
434 : * otherwise the caller is responsible for calling smgrtruncate()
435 : * to truncate the visibility map pages.
436 : */
437 : BlockNumber
438 292 : visibilitymap_prepare_truncate(Relation rel, BlockNumber nheapblocks)
439 : {
440 : BlockNumber newnblocks;
441 :
442 : /* last remaining block, byte, and bit */
443 292 : BlockNumber truncBlock = HEAPBLK_TO_MAPBLOCK(nheapblocks);
444 292 : uint32 truncByte = HEAPBLK_TO_MAPBYTE(nheapblocks);
445 292 : uint8 truncOffset = HEAPBLK_TO_OFFSET(nheapblocks);
446 :
447 : #ifdef TRACE_VISIBILITYMAP
448 : elog(DEBUG1, "vm_truncate %s %d", RelationGetRelationName(rel), nheapblocks);
449 : #endif
450 :
451 : /*
452 : * If no visibility map has been created yet for this relation, there's
453 : * nothing to truncate.
454 : */
455 292 : if (!smgrexists(RelationGetSmgr(rel), VISIBILITYMAP_FORKNUM))
456 0 : return InvalidBlockNumber;
457 :
458 : /*
459 : * Unless the new size is exactly at a visibility map page boundary, the
460 : * tail bits in the last remaining map page, representing truncated heap
461 : * blocks, need to be cleared. This is not only tidy, but also necessary
462 : * because we don't get a chance to clear the bits if the heap is extended
463 : * again.
464 : */
465 292 : if (truncByte != 0 || truncOffset != 0)
466 180 : {
467 : Buffer mapBuffer;
468 : Page page;
469 : char *map;
470 :
471 180 : newnblocks = truncBlock + 1;
472 :
473 180 : mapBuffer = vm_readbuf(rel, truncBlock, false);
474 180 : if (!BufferIsValid(mapBuffer))
475 : {
476 : /* nothing to do, the file was already smaller */
477 0 : return InvalidBlockNumber;
478 : }
479 :
480 180 : page = BufferGetPage(mapBuffer);
481 180 : map = PageGetContents(page);
482 :
483 180 : LockBuffer(mapBuffer, BUFFER_LOCK_EXCLUSIVE);
484 :
485 : /* NO EREPORT(ERROR) from here till changes are logged */
486 180 : START_CRIT_SECTION();
487 :
488 : /* Clear out the unwanted bytes. */
489 180 : MemSet(&map[truncByte + 1], 0, MAPSIZE - (truncByte + 1));
490 :
491 : /*----
492 : * Mask out the unwanted bits of the last remaining byte.
493 : *
494 : * ((1 << 0) - 1) = 00000000
495 : * ((1 << 1) - 1) = 00000001
496 : * ...
497 : * ((1 << 6) - 1) = 00111111
498 : * ((1 << 7) - 1) = 01111111
499 : *----
500 : */
501 180 : map[truncByte] &= (1 << truncOffset) - 1;
502 :
503 : /*
504 : * Truncation of a relation is WAL-logged at a higher-level, and we
505 : * will be called at WAL replay. But if checksums are enabled, we need
506 : * to still write a WAL record to protect against a torn page, if the
507 : * page is flushed to disk before the truncation WAL record. We cannot
508 : * use MarkBufferDirtyHint here, because that will not dirty the page
509 : * during recovery.
510 : */
511 180 : MarkBufferDirty(mapBuffer);
512 180 : if (!InRecovery && RelationNeedsWAL(rel) && XLogHintBitIsNeeded())
513 32 : log_newpage_buffer(mapBuffer, false);
514 :
515 180 : END_CRIT_SECTION();
516 :
517 180 : UnlockReleaseBuffer(mapBuffer);
518 : }
519 : else
520 112 : newnblocks = truncBlock;
521 :
522 292 : if (smgrnblocks(RelationGetSmgr(rel), VISIBILITYMAP_FORKNUM) <= newnblocks)
523 : {
524 : /* nothing to do, the file was already smaller than requested size */
525 180 : return InvalidBlockNumber;
526 : }
527 :
528 112 : return newnblocks;
529 : }
530 :
531 : /*
532 : * Read a visibility map page.
533 : *
534 : * If the page doesn't exist, InvalidBuffer is returned, or if 'extend' is
535 : * true, the visibility map file is extended.
536 : */
537 : static Buffer
538 1748772 : vm_readbuf(Relation rel, BlockNumber blkno, bool extend)
539 : {
540 : Buffer buf;
541 : SMgrRelation reln;
542 :
543 : /*
544 : * Caution: re-using this smgr pointer could fail if the relcache entry
545 : * gets closed. It's safe as long as we only do smgr-level operations
546 : * between here and the last use of the pointer.
547 : */
548 1748772 : reln = RelationGetSmgr(rel);
549 :
550 : /*
551 : * If we haven't cached the size of the visibility map fork yet, check it
552 : * first.
553 : */
554 1748772 : if (reln->smgr_cached_nblocks[VISIBILITYMAP_FORKNUM] == InvalidBlockNumber)
555 : {
556 59878 : if (smgrexists(reln, VISIBILITYMAP_FORKNUM))
557 25932 : smgrnblocks(reln, VISIBILITYMAP_FORKNUM);
558 : else
559 33946 : reln->smgr_cached_nblocks[VISIBILITYMAP_FORKNUM] = 0;
560 : }
561 :
562 : /*
563 : * For reading we use ZERO_ON_ERROR mode, and initialize the page if
564 : * necessary. It's always safe to clear bits, so it's better to clear
565 : * corrupt pages than error out.
566 : *
567 : * We use the same path below to initialize pages when extending the
568 : * relation, as a concurrent extension can end up with vm_extend()
569 : * returning an already-initialized page.
570 : */
571 1748772 : if (blkno >= reln->smgr_cached_nblocks[VISIBILITYMAP_FORKNUM])
572 : {
573 1598874 : if (extend)
574 4646 : buf = vm_extend(rel, blkno + 1);
575 : else
576 1594228 : return InvalidBuffer;
577 : }
578 : else
579 149898 : buf = ReadBufferExtended(rel, VISIBILITYMAP_FORKNUM, blkno,
580 : RBM_ZERO_ON_ERROR, NULL);
581 :
582 : /*
583 : * Initializing the page when needed is trickier than it looks, because of
584 : * the possibility of multiple backends doing this concurrently, and our
585 : * desire to not uselessly take the buffer lock in the normal path where
586 : * the page is OK. We must take the lock to initialize the page, so
587 : * recheck page newness after we have the lock, in case someone else
588 : * already did it. Also, because we initially check PageIsNew with no
589 : * lock, it's possible to fall through and return the buffer while someone
590 : * else is still initializing the page (i.e., we might see pd_upper as set
591 : * but other page header fields are still zeroes). This is harmless for
592 : * callers that will take a buffer lock themselves, but some callers
593 : * inspect the page without any lock at all. The latter is OK only so
594 : * long as it doesn't depend on the page header having correct contents.
595 : * Current usage is safe because PageGetContents() does not require that.
596 : */
597 154544 : if (PageIsNew(BufferGetPage(buf)))
598 : {
599 4736 : LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
600 4736 : if (PageIsNew(BufferGetPage(buf)))
601 4736 : PageInit(BufferGetPage(buf), BLCKSZ, 0);
602 4736 : LockBuffer(buf, BUFFER_LOCK_UNLOCK);
603 : }
604 154544 : return buf;
605 : }
606 :
607 : /*
608 : * Ensure that the visibility map fork is at least vm_nblocks long, extending
609 : * it if necessary with zeroed pages.
610 : */
611 : static Buffer
612 4646 : vm_extend(Relation rel, BlockNumber vm_nblocks)
613 : {
614 : Buffer buf;
615 :
616 4646 : buf = ExtendBufferedRelTo(BMR_REL(rel), VISIBILITYMAP_FORKNUM, NULL,
617 : EB_CREATE_FORK_IF_NEEDED |
618 : EB_CLEAR_SIZE_CACHE,
619 : vm_nblocks,
620 : RBM_ZERO_ON_ERROR);
621 :
622 : /*
623 : * Send a shared-inval message to force other backends to close any smgr
624 : * references they may have for this rel, which we are about to change.
625 : * This is a useful optimization because it means that backends don't have
626 : * to keep checking for creation or extension of the file, which happens
627 : * infrequently.
628 : */
629 4646 : CacheInvalidateSmgr(RelationGetSmgr(rel)->smgr_rlocator);
630 :
631 4646 : return buf;
632 : }
|
