Compare commits

..

5 Commits

Author SHA1 Message Date
Andrew Kane
915a630d00 Revert "Free tuple"
This reverts commit 466ec1ac46.
2023-10-22 16:09:07 -07:00
Andrew Kane
20c24e2d18 Added comment [skip ci] 2023-10-22 11:18:58 -07:00
Andrew Kane
466ec1ac46 Free tuple 2023-10-22 11:15:14 -07:00
Andrew Kane
ff2bdfb3b4 Improved check [skip ci] 2023-10-22 11:15:00 -07:00
Andrew Kane
0a3669a680 Added support for index-only scans to HNSW 2023-09-02 18:54:46 -07:00
16 changed files with 161 additions and 234 deletions

View File

@@ -1,7 +1,3 @@
## 0.5.1 (unreleased)
- Fixed locking for index scans for HNSW
## 0.5.0 (2023-08-28) ## 0.5.0 (2023-08-28)
- Added HNSW index type - Added HNSW index type

View File

@@ -8,7 +8,7 @@ HEADERS = src/vector.h
TESTS = $(wildcard test/sql/*.sql) TESTS = $(wildcard test/sql/*.sql)
REGRESS = $(patsubst test/sql/%.sql,%,$(TESTS)) REGRESS = $(patsubst test/sql/%.sql,%,$(TESTS))
REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION) REGRESS_OPTS = --inputdir=test --load-extension=vector
OPTFLAGS = -march=native OPTFLAGS = -march=native

View File

@@ -5,7 +5,7 @@ OBJS = src\hnsw.obj src\hnswbuild.obj src\hnswinsert.obj src\hnswscan.obj src\hn
HEADERS = src\vector.h HEADERS = src\vector.h
REGRESS = btree cast copy functions input ivfflat_cosine ivfflat_ip ivfflat_l2 ivfflat_options ivfflat_unlogged REGRESS = btree cast copy functions input ivfflat_cosine ivfflat_ip ivfflat_l2 ivfflat_options ivfflat_unlogged
REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION) REGRESS_OPTS = --inputdir=test --load-extension=vector
# For /arch flags # For /arch flags
# https://learn.microsoft.com/en-us/cpp/build/reference/arch-minimum-cpu-architecture # https://learn.microsoft.com/en-us/cpp/build/reference/arch-minimum-cpu-architecture

View File

@@ -595,18 +595,12 @@ pgvector is available on [these providers](https://github.com/pgvector/pgvector/
## Upgrading ## Upgrading
Install the latest version. Then in each database you want to upgrade, run: Install the latest version and run:
```sql ```sql
ALTER EXTENSION vector UPDATE; ALTER EXTENSION vector UPDATE;
``` ```
You can check the version in the current database with:
```sql
SELECT extversion FROM pg_extension WHERE extname = 'vector';
```
## Upgrade Notes ## Upgrade Notes
### 0.4.0 ### 0.4.0

View File

@@ -91,7 +91,7 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
MemSet(&costs, 0, sizeof(costs)); MemSet(&costs, 0, sizeof(costs));
index = index_open(path->indexinfo->indexoid, NoLock); index = index_open(path->indexinfo->indexoid, NoLock);
HnswGetMetaPageInfo(index, &m, NULL); m = HnswGetM(index);
index_close(index, NoLock); index_close(index, NoLock);
/* Approximate entry level */ /* Approximate entry level */
@@ -155,6 +155,15 @@ hnswvalidate(Oid opclassoid)
return true; return true;
} }
/*
* Checks if index-only scan is supported
*/
static bool
hnswcanreturn(Relation indexRelation, int attno)
{
return attno == 1 && !OidIsValid(index_getprocid(indexRelation, 1, HNSW_NORM_PROC));
}
/* /*
* Define index handler * Define index handler
* *
@@ -196,7 +205,7 @@ hnswhandler(PG_FUNCTION_ARGS)
amroutine->aminsert = hnswinsert; amroutine->aminsert = hnswinsert;
amroutine->ambulkdelete = hnswbulkdelete; amroutine->ambulkdelete = hnswbulkdelete;
amroutine->amvacuumcleanup = hnswvacuumcleanup; amroutine->amvacuumcleanup = hnswvacuumcleanup;
amroutine->amcanreturn = NULL; amroutine->amcanreturn = hnswcanreturn;
amroutine->amcostestimate = hnswcostestimate; amroutine->amcostestimate = hnswcostestimate;
amroutine->amoptions = hnswoptions; amroutine->amoptions = hnswoptions;
amroutine->amproperty = NULL; /* TODO AMPROP_DISTANCE_ORDERABLE */ amroutine->amproperty = NULL; /* TODO AMPROP_DISTANCE_ORDERABLE */

View File

@@ -266,9 +266,8 @@ Buffer HnswNewBuffer(Relation index, ForkNumber forkNum);
void HnswInitPage(Buffer buf, Page page); void HnswInitPage(Buffer buf, Page page);
void HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state); void HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state);
void HnswInit(void); void HnswInit(void);
List *HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, int m, bool inserting, HnswElement skipElement); List *HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec, HnswElement skipElement);
HnswElement HnswGetEntryPoint(Relation index); HnswElement HnswGetEntryPoint(Relation index);
void HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint);
HnswElement HnswInitElement(ItemPointer tid, int m, double ml, int maxLevel); HnswElement HnswInitElement(ItemPointer tid, int m, double ml, int maxLevel);
void HnswFreeElement(HnswElement element); void HnswFreeElement(HnswElement element);
HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno); HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno);
@@ -285,7 +284,7 @@ void HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool
void HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec); void HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec);
void HnswSetElementTuple(HnswElementTuple etup, HnswElement element); void HnswSetElementTuple(HnswElementTuple etup, HnswElement element);
void HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation); void HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation);
void HnswLoadNeighbors(HnswElement element, Relation index, int m); void HnswLoadNeighbors(HnswElement element, Relation index);
/* Index access methods */ /* Index access methods */
IndexBuildResult *hnswbuild(Relation heap, Relation index, IndexInfo *indexInfo); IndexBuildResult *hnswbuild(Relation heap, Relation index, IndexInfo *indexInfo);

View File

@@ -329,7 +329,7 @@ HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswE
/* Get latest neighbors since they may have changed */ /* Get latest neighbors since they may have changed */
/* Do not lock yet since selecting neighbors can take time */ /* Do not lock yet since selecting neighbors can take time */
HnswLoadNeighbors(hc->element, index, m); HnswLoadNeighbors(hc->element, index);
/* /*
* Could improve performance for vacuuming by checking neighbors * Could improve performance for vacuuming by checking neighbors
@@ -492,8 +492,9 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
FmgrInfo *normprocinfo; FmgrInfo *normprocinfo;
HnswElement entryPoint; HnswElement entryPoint;
HnswElement element; HnswElement element;
int m; int m = HnswGetM(index);
int efConstruction = HnswGetEfConstruction(index); int efConstruction = HnswGetEfConstruction(index);
double ml = HnswGetMl(m);
FmgrInfo *procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC); FmgrInfo *procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
Oid collation = index->rd_indcollation[0]; Oid collation = index->rd_indcollation[0];
HnswElement dup; HnswElement dup;
@@ -510,6 +511,10 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
return false; return false;
} }
/* Create an element */
element = HnswInitElement(heap_tid, m, ml, HnswGetMaxLevel(m));
element->vec = DatumGetVector(value);
/* /*
* Get a shared lock. This allows vacuum to ensure no in-flight inserts * Get a shared lock. This allows vacuum to ensure no in-flight inserts
* before repairing graph. Use a page lock so it does not interfere with * before repairing graph. Use a page lock so it does not interfere with
@@ -517,12 +522,8 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
*/ */
LockPage(index, HNSW_UPDATE_LOCK, lockmode); LockPage(index, HNSW_UPDATE_LOCK, lockmode);
/* Get m and entry point */ /* Get entry point */
HnswGetMetaPageInfo(index, &m, &entryPoint); entryPoint = HnswGetEntryPoint(index);
/* Create an element */
element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m));
element->vec = DatumGetVector(value);
/* Prevent concurrent inserts when likely updating entry point */ /* Prevent concurrent inserts when likely updating entry point */
if (entryPoint == NULL || element->level > entryPoint->level) if (entryPoint == NULL || element->level > entryPoint->level)

View File

@@ -17,26 +17,23 @@ GetScanItems(IndexScanDesc scan, Datum q)
Relation index = scan->indexRelation; Relation index = scan->indexRelation;
FmgrInfo *procinfo = so->procinfo; FmgrInfo *procinfo = so->procinfo;
Oid collation = so->collation; Oid collation = so->collation;
bool loadVec = scan->xs_want_itup;
List *ep; List *ep;
List *w; List *w;
int m; HnswElement entryPoint = HnswGetEntryPoint(index);
HnswElement entryPoint;
/* Get m and entry point */
HnswGetMetaPageInfo(index, &m, &entryPoint);
if (entryPoint == NULL) if (entryPoint == NULL)
return NIL; return NIL;
ep = list_make1(HnswEntryCandidate(entryPoint, q, index, procinfo, collation, false)); ep = list_make1(HnswEntryCandidate(entryPoint, q, index, procinfo, collation, loadVec));
for (int lc = entryPoint->level; lc >= 1; lc--) for (int lc = entryPoint->level; lc >= 1; lc--)
{ {
w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, m, false, NULL); w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, loadVec, NULL);
ep = w; ep = w;
} }
return HnswSearchLayer(q, ep, hnsw_ef_search, 0, index, procinfo, collation, m, false, NULL); return HnswSearchLayer(q, ep, hnsw_ef_search, 0, index, procinfo, collation, loadVec, NULL);
} }
/* /*
@@ -62,33 +59,6 @@ GetDimensions(Relation index)
return dimensions; return dimensions;
} }
/*
* Get scan value
*/
static Datum
GetScanValue(IndexScanDesc scan)
{
HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
Datum value;
if (scan->orderByData->sk_flags & SK_ISNULL)
value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
else
{
value = scan->orderByData->sk_argument;
/* Value should not be compressed or toasted */
Assert(!VARATT_IS_COMPRESSED(DatumGetPointer(value)));
Assert(!VARATT_IS_EXTENDED(DatumGetPointer(value)));
/* Fine if normalization fails */
if (so->normprocinfo != NULL)
HnswNormValue(so->normprocinfo, so->collation, &value, NULL);
}
return value;
}
/* /*
* Prepare for an index scan * Prepare for an index scan
*/ */
@@ -114,11 +84,8 @@ hnswbeginscan(Relation index, int nkeys, int norderbys)
scan->opaque = so; scan->opaque = so;
/* /* OK to always set since cheap */
* Get a shared lock. This allows vacuum to ensure no in-flight scans scan->xs_itupdesc = RelationGetDescr(index);
* before marking tuples as deleted.
*/
LockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
return scan; return scan;
} }
@@ -167,18 +134,39 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
if (scan->orderByData == NULL) if (scan->orderByData == NULL)
elog(ERROR, "cannot scan hnsw index without order"); elog(ERROR, "cannot scan hnsw index without order");
/* Get scan value */ if (scan->orderByData->sk_flags & SK_ISNULL)
value = GetScanValue(scan); value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
else
{
value = scan->orderByData->sk_argument;
/* Value should not be compressed or toasted */
Assert(!VARATT_IS_COMPRESSED(DatumGetPointer(value)));
Assert(!VARATT_IS_EXTENDED(DatumGetPointer(value)));
/* Fine if normalization fails */
if (so->normprocinfo != NULL)
HnswNormValue(so->normprocinfo, so->collation, &value, NULL);
}
/*
* Get a shared lock. This allows vacuum to ensure no in-flight scans
* before marking tuples as deleted.
*/
LockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
so->w = GetScanItems(scan, value); so->w = GetScanItems(scan, value);
/* Release shared lock */
UnlockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
so->first = false; so->first = false;
} }
while (list_length(so->w) > 0) while (list_length(so->w) > 0)
{ {
HnswCandidate *hc = llast(so->w); HnswCandidate *hc = llast(so->w);
ItemPointer heaptid; ItemPointer tid;
BlockNumber indexblkno; BlockNumber indexblkno;
/* Move to next element if no valid heap TIDs */ /* Move to next element if no valid heap TIDs */
@@ -188,17 +176,26 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
continue; continue;
} }
heaptid = llast(hc->element->heaptids); tid = llast(hc->element->heaptids);
indexblkno = hc->element->blkno; indexblkno = hc->element->blkno;
hc->element->heaptids = list_delete_last(hc->element->heaptids); hc->element->heaptids = list_delete_last(hc->element->heaptids);
if (scan->xs_want_itup)
{
Datum value = PointerGetDatum(hc->element->vec);
bool isnull = false;
scan->xs_itup = index_form_tuple(scan->xs_itupdesc, &value, &isnull);
scan->xs_itup->t_tid = *tid;
}
MemoryContextSwitchTo(oldCtx); MemoryContextSwitchTo(oldCtx);
#if PG_VERSION_NUM >= 120000 #if PG_VERSION_NUM >= 120000
scan->xs_heaptid = *heaptid; scan->xs_heaptid = *tid;
#else #else
scan->xs_ctup.t_self = *heaptid; scan->xs_ctup.t_self = *tid;
#endif #endif
/* Unpin buffer */ /* Unpin buffer */
@@ -233,9 +230,6 @@ hnswendscan(IndexScanDesc scan)
if (BufferIsValid(so->buf)) if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf); ReleaseBuffer(so->buf);
/* Release shared lock */
UnlockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
MemoryContextDelete(so->tmpCtx); MemoryContextDelete(so->tmpCtx);
pfree(so); pfree(so);

View File

@@ -210,43 +210,25 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
} }
/* /*
* Get the metapage info * Get the entry point
*/ */
void HnswElement
HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint) HnswGetEntryPoint(Relation index)
{ {
Buffer buf; Buffer buf;
Page page; Page page;
HnswMetaPage metap; HnswMetaPage metap;
HnswElement entryPoint = NULL;
buf = ReadBuffer(index, HNSW_METAPAGE_BLKNO); buf = ReadBuffer(index, HNSW_METAPAGE_BLKNO);
LockBuffer(buf, BUFFER_LOCK_SHARE); LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf); page = BufferGetPage(buf);
metap = HnswPageGetMeta(page); metap = HnswPageGetMeta(page);
if (m != NULL) if (BlockNumberIsValid(metap->entryBlkno))
*m = metap->m; entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
if (entryPoint != NULL)
{
if (BlockNumberIsValid(metap->entryBlkno))
*entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
else
*entryPoint = NULL;
}
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
}
/*
* Get the entry point
*/
HnswElement
HnswGetEntryPoint(Relation index)
{
HnswElement entryPoint;
HnswGetMetaPageInfo(index, NULL, &entryPoint);
return entryPoint; return entryPoint;
} }
@@ -355,9 +337,10 @@ HnswSetNeighborTuple(HnswNeighborTuple ntup, HnswElement e, int m)
* Load neighbors from page * Load neighbors from page
*/ */
static void static void
LoadNeighborsFromPage(HnswElement element, Relation index, Page page, int m) LoadNeighborsFromPage(HnswElement element, Relation index, Page page)
{ {
HnswNeighborTuple ntup = (HnswNeighborTuple) PageGetItem(page, PageGetItemId(page, element->neighborOffno)); HnswNeighborTuple ntup = (HnswNeighborTuple) PageGetItem(page, PageGetItemId(page, element->neighborOffno));
int m = HnswGetM(index);
int neighborCount = (element->level + 2) * m; int neighborCount = (element->level + 2) * m;
Assert(HnswIsNeighborTuple(ntup)); Assert(HnswIsNeighborTuple(ntup));
@@ -398,7 +381,7 @@ LoadNeighborsFromPage(HnswElement element, Relation index, Page page, int m)
* Load neighbors * Load neighbors
*/ */
void void
HnswLoadNeighbors(HnswElement element, Relation index, int m) HnswLoadNeighbors(HnswElement element, Relation index)
{ {
Buffer buf; Buffer buf;
Page page; Page page;
@@ -407,7 +390,7 @@ HnswLoadNeighbors(HnswElement element, Relation index, int m)
LockBuffer(buf, BUFFER_LOCK_SHARE); LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf); page = BufferGetPage(buf);
LoadNeighborsFromPage(element, index, page, m); LoadNeighborsFromPage(element, index, page);
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
} }
@@ -560,7 +543,7 @@ AddToVisited(HTAB *v, HnswCandidate * hc, Relation index, bool *found)
* Algorithm 2 from paper * Algorithm 2 from paper
*/ */
List * List *
HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, int m, bool inserting, HnswElement skipElement) HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec, HnswElement skipElement)
{ {
ListCell *lc2; ListCell *lc2;
@@ -615,7 +598,7 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
break; break;
if (c->element->neighbors == NULL) if (c->element->neighbors == NULL)
HnswLoadNeighbors(c->element, index, m); HnswLoadNeighbors(c->element, index);
/* Get the neighborhood at layer lc */ /* Get the neighborhood at layer lc */
neighborhood = &c->element->neighbors[lc]; neighborhood = &c->element->neighbors[lc];
@@ -636,7 +619,7 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
if (index == NULL) if (index == NULL)
eDistance = GetCandidateDistance(e, q, procinfo, collation); eDistance = GetCandidateDistance(e, q, procinfo, collation);
else else
HnswLoadElement(e->element, &eDistance, &q, index, procinfo, collation, inserting); HnswLoadElement(e->element, &eDistance, &q, index, procinfo, collation, loadVec);
Assert(!e->element->deleted); Assert(!e->element->deleted);
@@ -973,7 +956,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
/* 1st phase: greedy search to insert level */ /* 1st phase: greedy search to insert level */
for (int lc = entryLevel; lc >= level + 1; lc--) for (int lc = entryLevel; lc >= level + 1; lc--)
{ {
w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, m, true, skipElement); w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, true, skipElement);
ep = w; ep = w;
} }
@@ -991,7 +974,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
List *neighbors; List *neighbors;
List *lw; List *lw;
w = HnswSearchLayer(q, ep, efConstruction, lc, index, procinfo, collation, m, true, skipElement); w = HnswSearchLayer(q, ep, efConstruction, lc, index, procinfo, collation, true, skipElement);
/* Elements being deleted or skipped can help with search */ /* Elements being deleted or skipped can help with search */
/* but should be removed before selecting neighbors */ /* but should be removed before selecting neighbors */

View File

@@ -330,10 +330,7 @@ RepairGraph(HnswVacuumState * vacuumstate)
BufferAccessStrategy bas = vacuumstate->bas; BufferAccessStrategy bas = vacuumstate->bas;
BlockNumber blkno = HNSW_HEAD_BLKNO; BlockNumber blkno = HNSW_HEAD_BLKNO;
/* /* Wait for inserts to complete */
* Wait for inserts to complete. Inserts before this point may have
* neighbors about to be deleted. Inserts after this point will not.
*/
LockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock); LockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
UnlockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock); UnlockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
@@ -446,11 +443,7 @@ MarkDeleted(HnswVacuumState * vacuumstate)
Relation index = vacuumstate->index; Relation index = vacuumstate->index;
BufferAccessStrategy bas = vacuumstate->bas; BufferAccessStrategy bas = vacuumstate->bas;
/* /* Wait for selects to complete */
* Wait for index scans to complete. Scans before this point may contain
* tuples about to be deleted. Scans after this point will not, since the
* graph has been repaired.
*/
LockPage(index, HNSW_SCAN_LOCK, ExclusiveLock); LockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
UnlockPage(index, HNSW_SCAN_LOCK, ExclusiveLock); UnlockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
@@ -589,6 +582,7 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
vacuumstate->stats = stats; vacuumstate->stats = stats;
vacuumstate->callback = callback; vacuumstate->callback = callback;
vacuumstate->callback_state = callback_state; vacuumstate->callback_state = callback_state;
vacuumstate->m = HnswGetM(index);
vacuumstate->efConstruction = HnswGetEfConstruction(index); vacuumstate->efConstruction = HnswGetEfConstruction(index);
vacuumstate->bas = GetAccessStrategy(BAS_BULKREAD); vacuumstate->bas = GetAccessStrategy(BAS_BULKREAD);
vacuumstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC); vacuumstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
@@ -598,9 +592,6 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
"Hnsw vacuum temporary context", "Hnsw vacuum temporary context",
ALLOCSET_DEFAULT_SIZES); ALLOCSET_DEFAULT_SIZES);
/* Get m from metapage */
HnswGetMetaPageInfo(index, &vacuumstate->m, NULL);
/* Create hash table */ /* Create hash table */
hash_ctl.keysize = sizeof(ItemPointerData); hash_ctl.keysize = sizeof(ItemPointerData);
hash_ctl.entrysize = sizeof(ItemPointerData); hash_ctl.entrysize = sizeof(ItemPointerData);

View File

@@ -71,7 +71,7 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
int lists; int lists;
double ratio; double ratio;
double spc_seq_page_cost; double spc_seq_page_cost;
Relation index; Relation indexRel;
#if PG_VERSION_NUM < 120000 #if PG_VERSION_NUM < 120000
List *qinfos; List *qinfos;
#endif #endif
@@ -89,9 +89,9 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
MemSet(&costs, 0, sizeof(costs)); MemSet(&costs, 0, sizeof(costs));
index = index_open(path->indexinfo->indexoid, NoLock); indexRel = index_open(path->indexinfo->indexoid, NoLock);
IvfflatGetMetaPageInfo(index, &lists, NULL); lists = IvfflatGetLists(indexRel);
index_close(index, NoLock); index_close(indexRel, NoLock);
/* Get the ratio of lists that we need to visit */ /* Get the ratio of lists that we need to visit */
ratio = ((double) ivfflat_probes) / lists; ratio = ((double) ivfflat_probes) / lists;

View File

@@ -244,8 +244,8 @@ typedef struct IvfflatScanList
typedef struct IvfflatScanOpaqueData typedef struct IvfflatScanOpaqueData
{ {
int probes; int probes;
int dimensions;
bool first; bool first;
Buffer buf;
/* Sorting */ /* Sorting */
Tuplesortstate *sortstate; Tuplesortstate *sortstate;
@@ -278,7 +278,6 @@ void IvfflatKmeans(Relation index, VectorArray samples, VectorArray centers);
FmgrInfo *IvfflatOptionalProcInfo(Relation rel, uint16 procnum); FmgrInfo *IvfflatOptionalProcInfo(Relation rel, uint16 procnum);
bool IvfflatNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result); bool IvfflatNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
int IvfflatGetLists(Relation index); int IvfflatGetLists(Relation index);
void IvfflatGetMetaPageInfo(Relation index, int *lists, int *dimensions);
void IvfflatUpdateList(Relation index, ListInfo listInfo, BlockNumber insertPage, BlockNumber originalInsertPage, BlockNumber startPage, ForkNumber forkNum); void IvfflatUpdateList(Relation index, ListInfo listInfo, BlockNumber insertPage, BlockNumber originalInsertPage, BlockNumber startPage, ForkNumber forkNum);
void IvfflatCommitBuffer(Buffer buf, GenericXLogState *state); void IvfflatCommitBuffer(Buffer buf, GenericXLogState *state);
void IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state, ForkNumber forkNum); void IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state, ForkNumber forkNum);

View File

@@ -31,36 +31,36 @@ CompareLists(const pairingheap_node *a, const pairingheap_node *b, void *arg)
static void static void
GetScanLists(IndexScanDesc scan, Datum value) GetScanLists(IndexScanDesc scan, Datum value)
{ {
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque; Buffer cbuf;
Page cpage;
IvfflatList list;
OffsetNumber offno;
OffsetNumber maxoffno;
BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO; BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
int listCount = 0; int listCount = 0;
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
double distance;
IvfflatScanList *scanlist;
double maxDistance = DBL_MAX; double maxDistance = DBL_MAX;
/* Search all list pages */ /* Search all list pages */
while (BlockNumberIsValid(nextblkno)) while (BlockNumberIsValid(nextblkno))
{ {
Buffer cbuf;
Page cpage;
OffsetNumber maxoffno;
cbuf = ReadBuffer(scan->indexRelation, nextblkno); cbuf = ReadBuffer(scan->indexRelation, nextblkno);
LockBuffer(cbuf, BUFFER_LOCK_SHARE); LockBuffer(cbuf, BUFFER_LOCK_SHARE);
cpage = BufferGetPage(cbuf); cpage = BufferGetPage(cbuf);
maxoffno = PageGetMaxOffsetNumber(cpage); maxoffno = PageGetMaxOffsetNumber(cpage);
for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno)) for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
{ {
IvfflatList list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno)); list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
double distance;
/* Use procinfo from the index instead of scan key for performance */ /* Use procinfo from the index instead of scan key for performance */
distance = DatumGetFloat8(FunctionCall2Coll(so->procinfo, so->collation, PointerGetDatum(&list->center), value)); distance = DatumGetFloat8(FunctionCall2Coll(so->procinfo, so->collation, PointerGetDatum(&list->center), value));
if (listCount < so->probes) if (listCount < so->probes)
{ {
IvfflatScanList *scanlist;
scanlist = &so->lists[listCount]; scanlist = &so->lists[listCount];
scanlist->startPage = list->startPage; scanlist->startPage = list->startPage;
scanlist->distance = distance; scanlist->distance = distance;
@@ -75,8 +75,6 @@ GetScanLists(IndexScanDesc scan, Datum value)
} }
else if (distance < maxDistance) else if (distance < maxDistance)
{ {
IvfflatScanList *scanlist;
/* Remove */ /* Remove */
scanlist = (IvfflatScanList *) pairingheap_remove_first(so->listQueue); scanlist = (IvfflatScanList *) pairingheap_remove_first(so->listQueue);
@@ -103,6 +101,14 @@ static void
GetScanItems(IndexScanDesc scan, Datum value) GetScanItems(IndexScanDesc scan, Datum value)
{ {
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque; IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
Buffer buf;
Page page;
IndexTuple itup;
BlockNumber searchPage;
OffsetNumber offno;
OffsetNumber maxoffno;
Datum datum;
bool isnull;
TupleDesc tupdesc = RelationGetDescr(scan->indexRelation); TupleDesc tupdesc = RelationGetDescr(scan->indexRelation);
double tuples = 0; double tuples = 0;
@@ -122,28 +128,19 @@ GetScanItems(IndexScanDesc scan, Datum value)
/* Search closest probes lists */ /* Search closest probes lists */
while (!pairingheap_is_empty(so->listQueue)) while (!pairingheap_is_empty(so->listQueue))
{ {
BlockNumber searchPage = ((IvfflatScanList *) pairingheap_remove_first(so->listQueue))->startPage; searchPage = ((IvfflatScanList *) pairingheap_remove_first(so->listQueue))->startPage;
/* Search all entry pages for list */ /* Search all entry pages for list */
while (BlockNumberIsValid(searchPage)) while (BlockNumberIsValid(searchPage))
{ {
Buffer buf;
Page page;
OffsetNumber maxoffno;
buf = ReadBufferExtended(scan->indexRelation, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas); buf = ReadBufferExtended(scan->indexRelation, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas);
LockBuffer(buf, BUFFER_LOCK_SHARE); LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf); page = BufferGetPage(buf);
maxoffno = PageGetMaxOffsetNumber(page); maxoffno = PageGetMaxOffsetNumber(page);
for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno)) for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
{ {
IndexTuple itup; itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offno));
Datum datum;
bool isnull;
ItemId itemid = PageGetItemId(page, offno);
itup = (IndexTuple) PageGetItem(page, itemid);
datum = index_getattr(itup, 1, tupdesc, &isnull); datum = index_getattr(itup, 1, tupdesc, &isnull);
/* /*
@@ -157,6 +154,8 @@ GetScanItems(IndexScanDesc scan, Datum value)
slot->tts_isnull[0] = false; slot->tts_isnull[0] = false;
slot->tts_values[1] = PointerGetDatum(&itup->t_tid); slot->tts_values[1] = PointerGetDatum(&itup->t_tid);
slot->tts_isnull[1] = false; slot->tts_isnull[1] = false;
slot->tts_values[2] = Int32GetDatum((int) searchPage);
slot->tts_isnull[2] = false;
ExecStoreVirtualTuple(slot); ExecStoreVirtualTuple(slot);
tuplesort_puttupleslot(so->sortstate, slot); tuplesort_puttupleslot(so->sortstate, slot);
@@ -181,6 +180,29 @@ GetScanItems(IndexScanDesc scan, Datum value)
tuplesort_performsort(so->sortstate); tuplesort_performsort(so->sortstate);
} }
/*
* Get dimensions from metapage
*/
static int
GetDimensions(Relation index)
{
Buffer buf;
Page page;
IvfflatMetaPage metap;
int dimensions;
buf = ReadBuffer(index, IVFFLAT_METAPAGE_BLKNO);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
metap = IvfflatPageGetMeta(page);
dimensions = metap->dimensions;
UnlockReleaseBuffer(buf);
return dimensions;
}
/* /*
* Prepare for an index scan * Prepare for an index scan
*/ */
@@ -190,7 +212,6 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
IndexScanDesc scan; IndexScanDesc scan;
IvfflatScanOpaque so; IvfflatScanOpaque so;
int lists; int lists;
int dimensions;
AttrNumber attNums[] = {1}; AttrNumber attNums[] = {1};
Oid sortOperators[] = {Float8LessOperator}; Oid sortOperators[] = {Float8LessOperator};
Oid sortCollations[] = {InvalidOid}; Oid sortCollations[] = {InvalidOid};
@@ -198,17 +219,15 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
int probes = ivfflat_probes; int probes = ivfflat_probes;
scan = RelationGetIndexScan(index, nkeys, norderbys); scan = RelationGetIndexScan(index, nkeys, norderbys);
lists = IvfflatGetLists(scan->indexRelation);
/* Get lists and dimensions from metapage */
IvfflatGetMetaPageInfo(index, &lists, &dimensions);
if (probes > lists) if (probes > lists)
probes = lists; probes = lists;
so = (IvfflatScanOpaque) palloc(offsetof(IvfflatScanOpaqueData, lists) + probes * sizeof(IvfflatScanList)); so = (IvfflatScanOpaque) palloc(offsetof(IvfflatScanOpaqueData, lists) + probes * sizeof(IvfflatScanList));
so->buf = InvalidBuffer;
so->first = true; so->first = true;
so->probes = probes; so->probes = probes;
so->dimensions = dimensions;
/* Set support functions */ /* Set support functions */
so->procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC); so->procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC);
@@ -217,12 +236,13 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
/* Create tuple description for sorting */ /* Create tuple description for sorting */
#if PG_VERSION_NUM >= 120000 #if PG_VERSION_NUM >= 120000
so->tupdesc = CreateTemplateTupleDesc(2); so->tupdesc = CreateTemplateTupleDesc(3);
#else #else
so->tupdesc = CreateTemplateTupleDesc(2, false); so->tupdesc = CreateTemplateTupleDesc(3, false);
#endif #endif
TupleDescInitEntry(so->tupdesc, (AttrNumber) 1, "distance", FLOAT8OID, -1, 0); TupleDescInitEntry(so->tupdesc, (AttrNumber) 1, "distance", FLOAT8OID, -1, 0);
TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "heaptid", TIDOID, -1, 0); TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "tid", TIDOID, -1, 0);
TupleDescInitEntry(so->tupdesc, (AttrNumber) 3, "indexblkno", INT4OID, -1, 0);
/* Prep sort */ /* Prep sort */
so->sortstate = tuplesort_begin_heap(so->tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, work_mem, NULL, false); so->sortstate = tuplesort_begin_heap(so->tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, work_mem, NULL, false);
@@ -288,12 +308,8 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
if (scan->orderByData == NULL) if (scan->orderByData == NULL)
elog(ERROR, "cannot scan ivfflat index without order"); elog(ERROR, "cannot scan ivfflat index without order");
/* Safety check */
if (!IsMVCCSnapshot(scan->xs_snapshot))
elog(ERROR, "non-MVCC snapshots are not supported with ivfflat");
if (scan->orderByData->sk_flags & SK_ISNULL) if (scan->orderByData->sk_flags & SK_ISNULL)
value = PointerGetDatum(InitVector(so->dimensions)); value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
else else
{ {
value = scan->orderByData->sk_argument; value = scan->orderByData->sk_argument;
@@ -318,19 +334,26 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
if (tuplesort_gettupleslot(so->sortstate, true, false, so->slot, NULL)) if (tuplesort_gettupleslot(so->sortstate, true, false, so->slot, NULL))
{ {
ItemPointer heaptid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull)); ItemPointer tid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
BlockNumber indexblkno = DatumGetInt32(slot_getattr(so->slot, 3, &so->isnull));
#if PG_VERSION_NUM >= 120000 #if PG_VERSION_NUM >= 120000
scan->xs_heaptid = *heaptid; scan->xs_heaptid = *tid;
#else #else
scan->xs_ctup.t_self = *heaptid; scan->xs_ctup.t_self = *tid;
#endif #endif
/* Unpin buffer */
if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf);
/* /*
* Since no pinning during sorting, requires MVCC-compliant snapshot. * An index scan must maintain a pin on the index page holding the
* Also, cannot kill prior tuples, since heap TID may be recycled. * item last returned by amgettuple
*
* https://www.postgresql.org/docs/current/index-locking.html * https://www.postgresql.org/docs/current/index-locking.html
*/ */
so->buf = ReadBuffer(scan->indexRelation, indexblkno);
scan->xs_recheckorderby = false; scan->xs_recheckorderby = false;
return true; return true;
@@ -347,6 +370,10 @@ ivfflatendscan(IndexScanDesc scan)
{ {
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque; IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
/* Release pin */
if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf);
pairingheap_free(so->listQueue); pairingheap_free(so->listQueue);
tuplesort_end(so->sortstate); tuplesort_end(so->sortstate);

View File

@@ -172,29 +172,6 @@ IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **st
*buf = newbuf; *buf = newbuf;
} }
/*
* Get the metapage info
*/
void
IvfflatGetMetaPageInfo(Relation index, int *lists, int *dimensions)
{
Buffer buf;
Page page;
IvfflatMetaPage metap;
buf = ReadBuffer(index, IVFFLAT_METAPAGE_BLKNO);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
metap = IvfflatPageGetMeta(page);
*lists = metap->lists;
if (dimensions != NULL)
*dimensions = metap->dimensions;
UnlockReleaseBuffer(buf);
}
/* /*
* Update the start or insert page of a list * Update the start or insert page of a list
*/ */

View File

@@ -1,43 +0,0 @@
use strict;
use warnings;
use PostgresNode;
use TestLib;
use Test::More;
my $dim = 3;
my $array_sql = join(",", ('random()') x $dim);
# Initialize node
my $node = get_new_node('node');
$node->init;
$node->start;
# Create table and index
$node->safe_psql("postgres", "CREATE EXTENSION vector;");
$node->safe_psql("postgres", "CREATE TABLE tst (i serial, v vector($dim));");
$node->safe_psql("postgres",
"INSERT INTO tst (v) SELECT ARRAY[$array_sql] FROM generate_series(1, 10000) i;"
);
$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops);");
# Delete data
$node->safe_psql("postgres", "DELETE FROM tst WHERE i % 100 != 0;");
my $exp = $node->safe_psql("postgres", qq(
SET enable_indexscan = off;
SELECT i FROM tst ORDER BY v <-> '[0,0,0]';
));
# Run twice to make sure correct tuples marked as dead
for (1 .. 2)
{
my $res = $node->safe_psql("postgres", qq(
SET enable_seqscan = off;
SET ivfflat.probes = 100;
SELECT i FROM tst ORDER BY v <-> '[0,0,0]';
));
is($res, $exp);
}
done_testing();

View File

@@ -1,4 +1,4 @@
comment = 'vector data type and ivfflat and hnsw access methods' comment = 'vector data type and ivfflat access method'
default_version = '0.5.0' default_version = '0.5.0'
module_pathname = '$libdir/vector' module_pathname = '$libdir/vector'
relocatable = true relocatable = true