Skip to content

Commit fd267c1

Browse files
tglsfdctglsfdc
committed
Skip ambulkdelete scan if there's nothing to delete and the index is not
partial. None of the existing AMs do anything useful except counting tuples when there's nothing to delete, and we can get a tuple count from the heap as long as it's not a partial index. (hash actually can skip anyway because it maintains a tuple count in the index metapage.) GIST is not currently able to exploit this optimization because, due to failure to index NULLs, GIST is always effectively partial. Possibly we should fix that sometime. Simon Riggs w/ some review by Tom Lane.
1 parent a02f6ce commit fd267c1

File tree

8 files changed

+108
-31
lines changed

8 files changed

+108
-31
lines changed

‎doc/src/sgml/indexam.sgml

Lines changed: 7 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
<!--
2-
$PostgreSQL: pgsql/doc/src/sgml/indexam.sgml,v 2.7 2005/11/04 23:14:00 petere Exp $
2+
$PostgreSQL: pgsql/doc/src/sgml/indexam.sgml,v 2.8 2006/02/11 23:31:32 tgl Exp $
33
-->
44

55
<chapter id="indexam">
@@ -200,6 +200,12 @@ ambulkdelete (Relation indexRelation,
200200
struct containing statistics about the effects of the deletion operation.
201201
</para>
202202

203+
<para>
204+
If <literal>callback_state</> is NULL then no tuples are to be deleted.
205+
The index AM may choose to optimize this case (eg by not scanning the
206+
index) but it is still expected to deliver accurate statistics.
207+
</para>
208+
203209
<para>
204210
<programlisting>
205211
IndexBulkDeleteResult *

‎src/backend/access/gist/gistvacuum.c

Lines changed: 50 additions & 23 deletions
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@
88
* Portions Copyright (c) 1994, Regents of the University of California
99
*
1010
* IDENTIFICATION
11-
* $PostgreSQL: pgsql/src/backend/access/gist/gistvacuum.c,v 1.13 2006/02/11 17:14:08 momjian Exp $
11+
* $PostgreSQL: pgsql/src/backend/access/gist/gistvacuum.c,v 1.14 2006/02/11 23:31:33 tgl Exp $
1212
*
1313
*-------------------------------------------------------------------------
1414
*/
@@ -25,16 +25,19 @@
2525
#include "storage/freespace.h"
2626
#include "storage/smgr.h"
2727

28-
/* filled by gistbulkdelete, cleared by gistvacuumpcleanup */
29-
static bool needFullVacuum = false;
3028

29+
typedef struct GistBulkDeleteResult
30+
{
31+
IndexBulkDeleteResult std; /* common state */
32+
bool needFullVacuum;
33+
} GistBulkDeleteResult;
3134

3235
typedef struct
3336
{
3437
GISTSTATE giststate;
3538
Relation index;
3639
MemoryContext opCtx;
37-
IndexBulkDeleteResult *result;
40+
GistBulkDeleteResult *result;
3841
} GistVacuum;
3942

4043
typedef struct
@@ -44,6 +47,7 @@ typedef struct
4447
bool emptypage;
4548
} ArrayTuple;
4649

50+
4751
static ArrayTuple
4852
gistVacuumUpdate(GistVacuum *gv, BlockNumber blkno, bool needunion)
4953
{
@@ -125,7 +129,7 @@ gistVacuumUpdate(GistVacuum *gv, BlockNumber blkno, bool needunion)
125129
if (chldtuple.ituplen > 1)
126130
{
127131
/*
128-
* child was splitted, so we need mark completion
132+
* child was split, so we need mark completion
129133
* insert(split)
130134
*/
131135
int j;
@@ -262,7 +266,7 @@ gistVacuumUpdate(GistVacuum *gv, BlockNumber blkno, bool needunion)
262266
needwrite = true;
263267
res.emptypage = true;
264268
GistPageSetDeleted(page);
265-
gv->result->pages_deleted++;
269+
gv->result->std.pages_deleted++;
266270
}
267271
}
268272
else
@@ -329,9 +333,9 @@ gistVacuumUpdate(GistVacuum *gv, BlockNumber blkno, bool needunion)
329333
}
330334

331335
/*
332-
* For usial vacuum just update FSM, for full vacuum
336+
* For usual vacuum just update FSM, for full vacuum
333337
* reforms parent tuples if some of childs was deleted or changed,
334-
* update invalid tuples (they can exsist from last crash recovery only),
338+
* update invalid tuples (they can exist from last crash recovery only),
335339
* tries to get smaller index
336340
*/
337341

@@ -340,7 +344,7 @@ gistvacuumcleanup(PG_FUNCTION_ARGS)
340344
{
341345
Relation rel = (Relation) PG_GETARG_POINTER(0);
342346
IndexVacuumCleanupInfo *info = (IndexVacuumCleanupInfo *) PG_GETARG_POINTER(1);
343-
IndexBulkDeleteResult *stats = (IndexBulkDeleteResult *) PG_GETARG_POINTER(2);
347+
GistBulkDeleteResult *stats = (GistBulkDeleteResult *) PG_GETARG_POINTER(2);
344348
BlockNumber npages,
345349
blkno;
346350
BlockNumber nFreePages,
@@ -377,13 +381,11 @@ gistvacuumcleanup(PG_FUNCTION_ARGS)
377381
freeGISTstate(&(gv.giststate));
378382
MemoryContextDelete(gv.opCtx);
379383
}
380-
else if (needFullVacuum)
384+
else if (stats->needFullVacuum)
381385
ereport(NOTICE,
382386
(errmsg("index \"%s\" needs VACUUM FULL or REINDEX to finish crash recovery",
383387
RelationGetRelationName(rel))));
384388

385-
needFullVacuum = false;
386-
387389
if (info->vacuum_full)
388390
needLock = false; /* relation locked with AccessExclusiveLock */
389391
else
@@ -438,23 +440,30 @@ gistvacuumcleanup(PG_FUNCTION_ARGS)
438440

439441
if (lastBlock > lastFilledBlock)
440442
RelationTruncate(rel, lastFilledBlock + 1);
441-
stats->pages_removed = lastBlock - lastFilledBlock;
443+
stats->std.pages_removed = lastBlock - lastFilledBlock;
442444
}
443445

444446
RecordIndexFreeSpace(&rel->rd_node, nFreePages, freePages);
445447
pfree(freePages);
446448

447449
/* return statistics */
448-
stats->pages_free = nFreePages;
450+
stats->std.pages_free = nFreePages;
449451
if (needLock)
450452
LockRelationForExtension(rel, ExclusiveLock);
451-
stats->num_pages = RelationGetNumberOfBlocks(rel);
453+
stats->std.num_pages = RelationGetNumberOfBlocks(rel);
452454
if (needLock)
453455
UnlockRelationForExtension(rel, ExclusiveLock);
454456

455457
if (info->vacuum_full)
456458
UnlockRelation(rel, AccessExclusiveLock);
457459

460+
/* if gistbulkdelete skipped the scan, use heap's tuple count */
461+
if (stats->std.num_index_tuples < 0)
462+
{
463+
Assert(info->num_heap_tuples >= 0);
464+
stats->std.num_index_tuples = info->num_heap_tuples;
465+
}
466+
458467
PG_RETURN_POINTER(stats);
459468
}
460469

@@ -500,15 +509,33 @@ gistbulkdelete(PG_FUNCTION_ARGS)
500509
Relation rel = (Relation) PG_GETARG_POINTER(0);
501510
IndexBulkDeleteCallback callback = (IndexBulkDeleteCallback) PG_GETARG_POINTER(1);
502511
void *callback_state = (void *) PG_GETARG_POINTER(2);
503-
IndexBulkDeleteResult *result = (IndexBulkDeleteResult *) palloc0(sizeof(IndexBulkDeleteResult));
512+
GistBulkDeleteResult *result;
504513
GistBDItem *stack,
505514
*ptr;
506515
bool needLock;
507516

508-
stack = (GistBDItem *) palloc0(sizeof(GistBDItem));
517+
result = (GistBulkDeleteResult *) palloc0(sizeof(GistBulkDeleteResult));
509518

510-
stack->blkno = GIST_ROOT_BLKNO;
511-
needFullVacuum = false;
519+
/*
520+
* We can skip the scan entirely if there's nothing to delete (indicated
521+
* by callback_state == NULL) and the index isn't partial. For a partial
522+
* index we must scan in order to derive a trustworthy tuple count.
523+
*
524+
* XXX as of PG 8.2 this is dead code because GIST indexes are always
525+
* effectively partial ... but keep it anyway in case our null-handling
526+
* gets fixed.
527+
*/
528+
if (callback_state || vac_is_partial_index(rel))
529+
{
530+
stack = (GistBDItem *) palloc0(sizeof(GistBDItem));
531+
stack->blkno = GIST_ROOT_BLKNO;
532+
}
533+
else
534+
{
535+
/* skip scan and set flag for gistvacuumcleanup */
536+
stack = NULL;
537+
result->std.num_index_tuples = -1;
538+
}
512539

513540
while (stack)
514541
{
@@ -561,11 +588,11 @@ gistbulkdelete(PG_FUNCTION_ARGS)
561588
i--;
562589
maxoff--;
563590
ntodelete++;
564-
result->tuples_removed += 1;
591+
result->std.tuples_removed += 1;
565592
Assert(maxoff == PageGetMaxOffsetNumber(page));
566593
}
567594
else
568-
result->num_index_tuples += 1;
595+
result->std.num_index_tuples += 1;
569596
}
570597

571598
if (ntodelete)
@@ -615,7 +642,7 @@ gistbulkdelete(PG_FUNCTION_ARGS)
615642
stack->next = ptr;
616643

617644
if (GistTupleIsInvalid(idxtuple))
618-
needFullVacuum = true;
645+
result->needFullVacuum = true;
619646
}
620647
}
621648

@@ -634,7 +661,7 @@ gistbulkdelete(PG_FUNCTION_ARGS)
634661

635662
if (needLock)
636663
LockRelationForExtension(rel, ExclusiveLock);
637-
result->num_pages = RelationGetNumberOfBlocks(rel);
664+
result->std.num_pages = RelationGetNumberOfBlocks(rel);
638665
if (needLock)
639666
UnlockRelationForExtension(rel, ExclusiveLock);
640667

‎src/backend/access/hash/hash.c

Lines changed: 13 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@
88
*
99
*
1010
* IDENTIFICATION
11-
* $PostgreSQL: pgsql/src/backend/access/hash/hash.c,v 1.85 2006/02/11 17:14:08 momjian Exp $
11+
* $PostgreSQL: pgsql/src/backend/access/hash/hash.c,v 1.86 2006/02/11 23:31:33 tgl Exp $
1212
*
1313
* NOTES
1414
* This file contains only the public interface routines.
@@ -517,6 +517,18 @@ hashbulkdelete(PG_FUNCTION_ARGS)
517517
cur_maxbucket = orig_maxbucket;
518518

519519
loop_top:
520+
521+
/*
522+
* If we don't have anything to delete, skip the scan, and report the
523+
* number of tuples shown in the metapage. (Unlike btree and gist,
524+
* we can trust this number even for a partial index.)
525+
*/
526+
if (!callback_state)
527+
{
528+
cur_bucket = cur_maxbucket + 1;
529+
num_index_tuples = local_metapage.hashm_ntuples;
530+
}
531+
520532
while (cur_bucket <= cur_maxbucket)
521533
{
522534
BlockNumber bucket_blkno;

‎src/backend/access/index/indexam.c

Lines changed: 5 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@
88
*
99
*
1010
* IDENTIFICATION
11-
* $PostgreSQL: pgsql/src/backend/access/index/indexam.c,v 1.89 2006/02/11 17:14:08 momjian Exp $
11+
* $PostgreSQL: pgsql/src/backend/access/index/indexam.c,v 1.90 2006/02/11 23:31:33 tgl Exp $
1212
*
1313
* INTERFACE ROUTINES
1414
* index_open - open an index relation by relation OID
@@ -685,6 +685,10 @@ index_getmulti(IndexScanDesc scan,
685685
* callback routine tells whether a given main-heap tuple is
686686
* to be deleted
687687
*
688+
* if callback_state is NULL then there are no tuples to be deleted;
689+
* index AM can choose to avoid work in this case, but must still
690+
* follow the protocol of returning statistical info.
691+
*
688692
* return value is an optional palloc'd struct of statistics
689693
* ----------------
690694
*/

‎src/backend/access/nbtree/nbtree.c

Lines changed: 23 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -12,7 +12,7 @@
1212
* Portions Copyright (c) 1994, Regents of the University of California
1313
*
1414
* IDENTIFICATION
15-
* $PostgreSQL: pgsql/src/backend/access/nbtree/nbtree.c,v 1.138 2006/02/11 17:14:08 momjian Exp $
15+
* $PostgreSQL: pgsql/src/backend/access/nbtree/nbtree.c,v 1.139 2006/02/11 23:31:33 tgl Exp $
1616
*
1717
*-------------------------------------------------------------------------
1818
*/
@@ -564,8 +564,22 @@ btbulkdelete(PG_FUNCTION_ARGS)
564564
* further to its right, which the indexscan will have no pin on.) We can
565565
* skip obtaining exclusive lock on empty pages though, since no indexscan
566566
* could be stopped on those.
567+
*
568+
* We can skip the scan entirely if there's nothing to delete (indicated
569+
* by callback_state == NULL) and the index isn't partial. For a partial
570+
* index we must scan in order to derive a trustworthy tuple count.
567571
*/
568-
buf = _bt_get_endpoint(rel, 0, false);
572+
if (callback_state || vac_is_partial_index(rel))
573+
{
574+
buf = _bt_get_endpoint(rel, 0, false);
575+
}
576+
else
577+
{
578+
/* skip scan and set flag for btvacuumcleanup */
579+
buf = InvalidBuffer;
580+
num_index_tuples = -1;
581+
}
582+
569583
if (BufferIsValid(buf)) /* check for empty index */
570584
{
571585
for (;;)
@@ -836,6 +850,13 @@ btvacuumcleanup(PG_FUNCTION_ARGS)
836850
stats->pages_deleted = pages_deleted;
837851
stats->pages_free = nFreePages;
838852

853+
/* if btbulkdelete skipped the scan, use heap's tuple count */
854+
if (stats->num_index_tuples < 0)
855+
{
856+
Assert(info->num_heap_tuples >= 0);
857+
stats->num_index_tuples = info->num_heap_tuples;
858+
}
859+
839860
PG_RETURN_POINTER(stats);
840861
}
841862

‎src/backend/commands/vacuum.c

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -13,7 +13,7 @@
1313
*
1414
*
1515
* IDENTIFICATION
16-
* $PostgreSQL: pgsql/src/backend/commands/vacuum.c,v 1.323 2006/02/11 17:14:09 momjian Exp $
16+
* $PostgreSQL: pgsql/src/backend/commands/vacuum.c,v 1.324 2006/02/11 23:31:33 tgl Exp $
1717
*
1818
*-------------------------------------------------------------------------
1919
*/
@@ -2955,6 +2955,7 @@ scan_index(Relation indrel, double num_tuples)
29552955
/* Do post-VACUUM cleanup, even though we deleted nothing */
29562956
vcinfo.vacuum_full = true;
29572957
vcinfo.message_level = elevel;
2958+
vcinfo.num_heap_tuples = num_tuples;
29582959

29592960
stats = index_vacuum_cleanup(indrel, &vcinfo, stats);
29602961

@@ -3022,6 +3023,7 @@ vacuum_index(VacPageList vacpagelist, Relation indrel,
30223023
/* Do post-VACUUM cleanup */
30233024
vcinfo.vacuum_full = true;
30243025
vcinfo.message_level = elevel;
3026+
vcinfo.num_heap_tuples = num_tuples + keep_tuples;
30253027

30263028
stats = index_vacuum_cleanup(indrel, &vcinfo, stats);
30273029

‎src/backend/commands/vacuumlazy.c

Lines changed: 5 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -31,7 +31,7 @@
3131
*
3232
*
3333
* IDENTIFICATION
34-
* $PostgreSQL: pgsql/src/backend/commands/vacuumlazy.c,v 1.65 2006/02/11 17:14:09 momjian Exp $
34+
* $PostgreSQL: pgsql/src/backend/commands/vacuumlazy.c,v 1.66 2006/02/11 23:31:34 tgl Exp $
3535
*
3636
*-------------------------------------------------------------------------
3737
*/
@@ -625,6 +625,7 @@ lazy_scan_index(Relation indrel, LVRelStats *vacrelstats)
625625
/* Do post-VACUUM cleanup, even though we deleted nothing */
626626
vcinfo.vacuum_full = false;
627627
vcinfo.message_level = elevel;
628+
vcinfo.num_heap_tuples = vacrelstats->rel_tuples;
628629

629630
stats = index_vacuum_cleanup(indrel, &vcinfo, stats);
630631

@@ -697,6 +698,9 @@ lazy_vacuum_index(Relation indrel,
697698
/* Do post-VACUUM cleanup */
698699
vcinfo.vacuum_full = false;
699700
vcinfo.message_level = elevel;
701+
/* We don't yet know rel_tuples, so pass -1 */
702+
/* index_bulk_delete can't have skipped scan anyway ... */
703+
vcinfo.num_heap_tuples = -1;
700704

701705
stats = index_vacuum_cleanup(indrel, &vcinfo, stats);
702706

‎src/include/access/genam.h

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,7 @@
77
* Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
88
* Portions Copyright (c) 1994, Regents of the University of California
99
*
10-
* $PostgreSQL: pgsql/src/include/access/genam.h,v 1.56 2006/02/11 17:14:09 momjian Exp $
10+
* $PostgreSQL: pgsql/src/include/access/genam.h,v 1.57 2006/02/11 23:31:34 tgl Exp $
1111
*
1212
*-------------------------------------------------------------------------
1313
*/
@@ -51,6 +51,7 @@ typedef struct IndexVacuumCleanupInfo
5151
{
5252
bool vacuum_full; /* VACUUM FULL (we have exclusive lock) */
5353
int message_level; /* ereport level for progress messages */
54+
double num_heap_tuples; /* tuples remaining in heap */
5455
} IndexVacuumCleanupInfo;
5556

5657
/* Struct for heap-or-index scans of system tables */

0 commit comments

Comments
 (0)