Fixed CI [skip ci]

This reverts commit d032726976.

.github/workflows/build.yml

@@ -39,6 +39,10 @@ jobs:
           sudo apt-get update
           sudo apt-get install libipc-run-perl
       - run: make prove_installcheck
+      - if: ${{ matrix.os == 'ubuntu-22.04' }}
+        run: make clean && make
+        env:
+          PG_CFLAGS: -Werror -fanalyzer
   mac:
     runs-on: macos-latest
     if: ${{ !startsWith(github.ref_name, 'windows') }}

CHANGELOG.md

@@ -1,3 +1,7 @@
+## 0.5.1 (unreleased)
+
+- Improved performance of index scans for IVFFlat after updates and deletes
+
 ## 0.5.0 (2023-08-28)
 
 - Added HNSW index type

Makefile

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

Makefile.win

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

README.md

@@ -595,12 +595,18 @@ pgvector is available on [these providers](https://github.com/pgvector/pgvector/
 
 ## Upgrading
 
-Install the latest version and run:
+Install the latest version. Then in each database you want to upgrade, run:
 
 ```sql
 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
 
 ### 0.4.0

src/hnsw.c

@@ -91,7 +91,7 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
 	MemSet(&costs, 0, sizeof(costs));
 
 	index = index_open(path->indexinfo->indexoid, NoLock);
-	m = HnswGetM(index);
+	HnswGetMetaPageInfo(index, &m, NULL);
 	index_close(index, NoLock);
 
 	/* Approximate entry level */
@@ -196,7 +196,7 @@ hnswhandler(PG_FUNCTION_ARGS)
 	amroutine->aminsert = hnswinsert;
 	amroutine->ambulkdelete = hnswbulkdelete;
 	amroutine->amvacuumcleanup = hnswvacuumcleanup;
-	amroutine->amcanreturn = NULL;	/* tuple not included in heapsort */
+	amroutine->amcanreturn = NULL;
 	amroutine->amcostestimate = hnswcostestimate;
 	amroutine->amoptions = hnswoptions;
 	amroutine->amproperty = NULL;	/* TODO AMPROP_DISTANCE_ORDERABLE */

src/hnsw.h

@@ -218,10 +218,6 @@ typedef HnswNeighborTupleData * HnswNeighborTuple;
 typedef struct HnswScanOpaqueData
 {
 	bool		first;
-	Buffer		buf;
-	ItemPointerData heaptid;
-	OffsetNumber offno;
-	int			removedCount;
 	List	   *w;
 	MemoryContext tmpCtx;
 
@@ -262,15 +258,16 @@ typedef struct HnswVacuumState
 /* Methods */
 int			HnswGetM(Relation index);
 int			HnswGetEfConstruction(Relation index);
-FmgrInfo   *HnswOptionalProcInfo(Relation rel, uint16 procnum);
+FmgrInfo   *HnswOptionalProcInfo(Relation index, uint16 procnum);
 bool		HnswNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
 void		HnswCommitBuffer(Buffer buf, GenericXLogState *state);
 Buffer		HnswNewBuffer(Relation index, ForkNumber forkNum);
 void		HnswInitPage(Buffer buf, Page page);
 void		HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state);
 void		HnswInit(void);
-List	   *HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, bool inserting, HnswElement skipElement);
+List	   *HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, int m, bool inserting, HnswElement skipElement);
 HnswElement HnswGetEntryPoint(Relation index);
+void		HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint);
 HnswElement HnswInitElement(ItemPointer tid, int m, double ml, int maxLevel);
 void		HnswFreeElement(HnswElement element);
 HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno);
@@ -287,7 +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		HnswSetElementTuple(HnswElementTuple etup, HnswElement element);
 void		HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation);
-void		HnswLoadNeighbors(HnswElement element, Relation index);
+void		HnswLoadNeighbors(HnswElement element, Relation index, int m);
 
 /* Index access methods */
 IndexBuildResult *hnswbuild(Relation heap, Relation index, IndexInfo *indexInfo);

src/hnswbuild.c

@@ -81,7 +81,6 @@ HnswBuildAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **
 	HnswPageGetOpaque(*page)->nextblkno = BufferGetBlockNumber(newbuf);
 
 	/* Commit */
-	MarkBufferDirty(*buf);
 	GenericXLogFinish(*state);
 	UnlockReleaseBuffer(*buf);
 
@@ -179,7 +178,6 @@ CreateElementPages(HnswBuildState * buildstate)
 	insertPage = BufferGetBlockNumber(buf);
 
 	/* Commit */
-	MarkBufferDirty(buf);
 	GenericXLogFinish(state);
 	UnlockReleaseBuffer(buf);
 
@@ -227,7 +225,6 @@ CreateNeighborPages(HnswBuildState * buildstate)
 			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
 
 		/* Commit */
-		MarkBufferDirty(buf);
 		GenericXLogFinish(state);
 		UnlockReleaseBuffer(buf);
 	}

src/hnswinsert.c

@@ -202,8 +202,6 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
 			HnswInsertAppendPage(index, &newbuf, &newpage, state, page);
 
 			/* Commit */
-			MarkBufferDirty(newbuf);
-			MarkBufferDirty(buf);
 			GenericXLogFinish(state);
 
 			/* Unlock previous buffer */
@@ -270,9 +268,6 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
 	}
 
 	/* Commit */
-	MarkBufferDirty(buf);
-	if (nbuf != buf)
-		MarkBufferDirty(nbuf);
 	GenericXLogFinish(state);
 	UnlockReleaseBuffer(buf);
 	if (nbuf != buf)
@@ -329,7 +324,7 @@ HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswE
 
 			/* Get latest neighbors since they may have changed */
 			/* Do not lock yet since selecting neighbors can take time */
-			HnswLoadNeighbors(hc->element, index);
+			HnswLoadNeighbors(hc->element, index, m);
 
 			/*
 			 * Could improve performance for vacuuming by checking neighbors
@@ -391,7 +386,6 @@ HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswE
 					elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
 
 				/* Commit */
-				MarkBufferDirty(buf);
 				GenericXLogFinish(state);
 			}
 			else
@@ -445,7 +439,6 @@ HnswAddDuplicate(Relation index, HnswElement element, HnswElement dup)
 		elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
 
 	/* Commit */
-	MarkBufferDirty(buf);
 	GenericXLogFinish(state);
 	UnlockReleaseBuffer(buf);
 
@@ -492,9 +485,8 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
 	FmgrInfo   *normprocinfo;
 	HnswElement entryPoint;
 	HnswElement element;
-	int			m = HnswGetM(index);
+	int			m;
 	int			efConstruction = HnswGetEfConstruction(index);
-	double		ml = HnswGetMl(m);
 	FmgrInfo   *procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
 	Oid			collation = index->rd_indcollation[0];
 	HnswElement dup;
@@ -511,10 +503,6 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
 			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
 	 * before repairing graph. Use a page lock so it does not interfere with
@@ -522,8 +510,12 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
 	 */
 	LockPage(index, HNSW_UPDATE_LOCK, lockmode);
 
-	/* Get entry point */
+	/* Get m and entry point */
-	entryPoint = HnswGetEntryPoint(index);
+	HnswGetMetaPageInfo(index, &m, &entryPoint);
+
+	/* Create an element */
+	element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m));
+	element->vec = DatumGetVector(value);
 
 	/* Prevent concurrent inserts when likely updating entry point */
 	if (entryPoint == NULL || element->level > entryPoint->level)

src/hnswscan.c

@@ -19,7 +19,11 @@ GetScanItems(IndexScanDesc scan, Datum q)
 	Oid			collation = so->collation;
 	List	   *ep;
 	List	   *w;
-	HnswElement entryPoint = HnswGetEntryPoint(index);
+	int			m;
+	HnswElement entryPoint;
+
+	/* Get m and entry point */
+	HnswGetMetaPageInfo(index, &m, &entryPoint);
 
 	if (entryPoint == NULL)
 		return NIL;
@@ -28,11 +32,11 @@ GetScanItems(IndexScanDesc scan, Datum q)
 
 	for (int lc = entryPoint->level; lc >= 1; lc--)
 	{
-		w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, false, NULL);
+		w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, m, false, NULL);
 		ep = w;
 	}
 
-	return HnswSearchLayer(q, ep, hnsw_ef_search, 0, index, procinfo, collation, false, NULL);
+	return HnswSearchLayer(q, ep, hnsw_ef_search, 0, index, procinfo, collation, m, false, NULL);
 }
 
 /*
@@ -59,72 +63,30 @@ GetDimensions(Relation index)
 }
 
 /*
- * Remove deleted heap TID
+ * Get scan value
  */
-static void
+static Datum
-RemoveHeapTid(IndexScanDesc scan)
+GetScanValue(IndexScanDesc scan)
 {
 	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
-	Relation	index = scan->indexRelation;
+	Datum		value;
-	Buffer		buf = so->buf;
-	Page		page;
-	GenericXLogState *state;
-	ItemId		itemid;
-	HnswElementTuple etup;
-	Size		etupSize;
-	int			idx = -1;
 
-	/* Safety check */
+	if (scan->orderByData->sk_flags & SK_ISNULL)
-	if (!BufferIsValid(buf) || !OffsetNumberIsValid(so->offno) || !ItemPointerIsValid(&so->heaptid))
+		value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
-		return;
-
-	/* Use WAL rather than hint */
-	LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
-	state = GenericXLogStart(index);
-	page = GenericXLogRegisterBuffer(state, buf, 0);
-	itemid = PageGetItemId(page, so->offno);
-	etup = (HnswElementTuple) PageGetItem(page, itemid);
-	etupSize = ItemIdGetLength(itemid);
-
-	Assert(HnswIsElementTuple(etup));
-
-	/* Find index */
-	for (int i = 0; i < HNSW_HEAPTIDS; i++)
-	{
-		if (!ItemPointerIsValid(&etup->heaptids[i]))
-			break;
-
-		if (ItemPointerEquals(&etup->heaptids[i], &so->heaptid))
-		{
-			idx = i;
-			break;
-		}
-	}
-
-	if (idx == -1)
-		GenericXLogAbort(state);
 	else
 	{
-		/* Move pointers forward */
+		value = scan->orderByData->sk_argument;
-		for (int i = idx; i < HNSW_HEAPTIDS; i++)
-		{
-			if (i + 1 == HNSW_HEAPTIDS || !ItemPointerIsValid(&etup->heaptids[i + 1]))
-				ItemPointerSetInvalid(&etup->heaptids[i]);
-			else
-				ItemPointerCopy(&etup->heaptids[i + 1], &etup->heaptids[i]);
-		}
 
-		/* Overwrite tuple */
+		/* Value should not be compressed or toasted */
-		if (!PageIndexTupleOverwrite(page, so->offno, (Item) etup, etupSize))
+		Assert(!VARATT_IS_COMPRESSED(DatumGetPointer(value)));
-			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+		Assert(!VARATT_IS_EXTENDED(DatumGetPointer(value)));
 
-		/* Commit */
+		/* Fine if normalization fails */
-		MarkBufferDirty(buf);
+		if (so->normprocinfo != NULL)
-		GenericXLogFinish(state);
+			HnswNormValue(so->normprocinfo, so->collation, &value, NULL);
 	}
 
-	/* Unlock buffer */
+	return value;
-	LockBuffer(buf, BUFFER_LOCK_UNLOCK);
 }
 
 /*
@@ -139,10 +101,6 @@ hnswbeginscan(Relation index, int nkeys, int norderbys)
 	scan = RelationGetIndexScan(index, nkeys, norderbys);
 
 	so = (HnswScanOpaque) palloc(sizeof(HnswScanOpaqueData));
-	so->buf = InvalidBuffer;
-	ItemPointerSetInvalid(&so->heaptid);
-	so->offno = InvalidOffsetNumber;
-	so->removedCount = 0;
 	so->first = true;
 	so->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
 									   "Hnsw scan temporary context",
@@ -167,7 +125,6 @@ hnswrescan(IndexScanDesc scan, ScanKey keys, int nkeys, ScanKey orderbys, int no
 	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
 
 	so->first = true;
-	ItemPointerSetInvalid(&so->heaptid);
 	MemoryContextReset(so->tmpCtx);
 
 	if (keys && scan->numberOfKeys > 0)
@@ -203,20 +160,8 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 		if (scan->orderByData == NULL)
 			elog(ERROR, "cannot scan hnsw index without order");
 
-		if (scan->orderByData->sk_flags & SK_ISNULL)
+		/* Get scan value */
-			value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
+		value = GetScanValue(scan);
-		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
@@ -231,62 +176,39 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 
 		so->first = false;
 	}
-	else
-	{
-		/*
-		 * Remove dead tuples. kill_prior_tuple will only be true if not in
-		 * recovery. Limit the number removed per scan for performance.
-		 */
-		if (scan->kill_prior_tuple && so->removedCount < 3)
-		{
-			RemoveHeapTid(scan);
-			so->removedCount++;
-		}
-	}
 
 	while (list_length(so->w) > 0)
 	{
 		HnswCandidate *hc = llast(so->w);
-		ItemPointer tid;
+		ItemPointer heaptid;
-		BlockNumber indexblkno;
-		OffsetNumber indexoffno;
 
-		/* Move to next element if no valid heap tids */
+		/* Move to next element if no valid heap TIDs */
 		if (list_length(hc->element->heaptids) == 0)
 		{
 			so->w = list_delete_last(so->w);
 			continue;
 		}
 
-		tid = llast(hc->element->heaptids);
+		heaptid = llast(hc->element->heaptids);
-		indexblkno = hc->element->blkno;
-		indexoffno = hc->element->offno;
 
 		hc->element->heaptids = list_delete_last(hc->element->heaptids);
 
 		MemoryContextSwitchTo(oldCtx);
 
 #if PG_VERSION_NUM >= 120000
-		scan->xs_heaptid = *tid;
+		scan->xs_heaptid = *heaptid;
 #else
-		scan->xs_ctup.t_self = *tid;
+		scan->xs_ctup.t_self = *heaptid;
 #endif
 
-		/* Keep track of info needed to remove dead tuples */
-		so->heaptid = *tid;
-		so->offno = indexoffno;
-
-		/* Unpin buffer */
-		if (BufferIsValid(so->buf))
-			ReleaseBuffer(so->buf);
-
 		/*
-		 * An index scan must maintain a pin on the index page holding the
+		 * Typically, an index scan must maintain a pin on the index page
-		 * item last returned by amgettuple
+		 * holding the item last returned by amgettuple. However, this is not
+		 * needed with the current vacuum strategy, which ensures scans do not
+		 * visit tuples in danger of being marked as deleted.
 		 *
 		 * https://www.postgresql.org/docs/current/index-locking.html
 		 */
-		so->buf = ReadBuffer(scan->indexRelation, indexblkno);
 
 		scan->xs_recheckorderby = false;
 		return true;
@@ -304,10 +226,6 @@ hnswendscan(IndexScanDesc scan)
 {
 	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
 
-	/* Release pin */
-	if (BufferIsValid(so->buf))
-		ReleaseBuffer(so->buf);
-
 	MemoryContextDelete(so->tmpCtx);
 
 	pfree(so);

src/hnswutils.c

@@ -38,12 +38,12 @@ HnswGetEfConstruction(Relation index)
  * Get proc
  */
 FmgrInfo *
-HnswOptionalProcInfo(Relation rel, uint16 procnum)
+HnswOptionalProcInfo(Relation index, uint16 procnum)
 {
-	if (!OidIsValid(index_getprocid(rel, 1, procnum)))
+	if (!OidIsValid(index_getprocid(index, 1, procnum)))
 		return NULL;
 
-	return index_getprocinfo(rel, 1, procnum);
+	return index_getprocinfo(index, 1, procnum);
 }
 
 /*
@@ -117,7 +117,6 @@ HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState *
 void
 HnswCommitBuffer(Buffer buf, GenericXLogState *state)
 {
-	MarkBufferDirty(buf);
 	GenericXLogFinish(state);
 	UnlockReleaseBuffer(buf);
 }
@@ -143,6 +142,17 @@ HnswInitNeighbors(HnswElement element, int m)
 	}
 }
 
+/*
+ * Free neighbors
+ */
+static void
+HnswFreeNeighbors(HnswElement element)
+{
+	for (int lc = 0; lc <= element->level; lc++)
+		pfree(element->neighbors[lc].items);
+	pfree(element->neighbors);
+}
+
 /*
  * Allocate an element
  */
@@ -174,10 +184,8 @@ HnswInitElement(ItemPointer heaptid, int m, double ml, int maxLevel)
 void
 HnswFreeElement(HnswElement element)
 {
+	HnswFreeNeighbors(element);
 	list_free_deep(element->heaptids);
-	for (int lc = 0; lc <= element->level; lc++)
-		pfree(element->neighbors[lc].items);
-	pfree(element->neighbors);
 	pfree(element->vec);
 	pfree(element);
 }
@@ -210,25 +218,43 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
 }
 
 /*
- * Get the entry point
+ * Get the metapage info
  */
-HnswElement
+void
-HnswGetEntryPoint(Relation index)
+HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint)
 {
 	Buffer		buf;
 	Page		page;
 	HnswMetaPage metap;
-	HnswElement entryPoint = NULL;
 
 	buf = ReadBuffer(index, HNSW_METAPAGE_BLKNO);
 	LockBuffer(buf, BUFFER_LOCK_SHARE);
 	page = BufferGetPage(buf);
 	metap = HnswPageGetMeta(page);
 
-	if (BlockNumberIsValid(metap->entryBlkno))
+	if (m != NULL)
-		entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
+		*m = metap->m;
+
+	if (entryPoint != NULL)
+	{
+		if (BlockNumberIsValid(metap->entryBlkno))
+			*entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
+		else
+			*entryPoint = NULL;
+	}
 
 	UnlockReleaseBuffer(buf);
+}
+
+/*
+ * Get the entry point
+ */
+HnswElement
+HnswGetEntryPoint(Relation index)
+{
+	HnswElement entryPoint;
+
+	HnswGetMetaPageInfo(index, NULL, &entryPoint);
 
 	return entryPoint;
 }
@@ -337,10 +363,9 @@ HnswSetNeighborTuple(HnswNeighborTuple ntup, HnswElement e, int m)
  * Load neighbors from page
  */
 static void
-LoadNeighborsFromPage(HnswElement element, Relation index, Page page)
+LoadNeighborsFromPage(HnswElement element, Relation index, Page page, int m)
 {
 	HnswNeighborTuple ntup = (HnswNeighborTuple) PageGetItem(page, PageGetItemId(page, element->neighborOffno));
-	int			m = HnswGetM(index);
 	int			neighborCount = (element->level + 2) * m;
 
 	Assert(HnswIsNeighborTuple(ntup));
@@ -381,7 +406,7 @@ LoadNeighborsFromPage(HnswElement element, Relation index, Page page)
  * Load neighbors
  */
 void
-HnswLoadNeighbors(HnswElement element, Relation index)
+HnswLoadNeighbors(HnswElement element, Relation index, int m)
 {
 	Buffer		buf;
 	Page		page;
@@ -390,7 +415,7 @@ HnswLoadNeighbors(HnswElement element, Relation index)
 	LockBuffer(buf, BUFFER_LOCK_SHARE);
 	page = BufferGetPage(buf);
 
-	LoadNeighborsFromPage(element, index, page);
+	LoadNeighborsFromPage(element, index, page, m);
 
 	UnlockReleaseBuffer(buf);
 }
@@ -543,7 +568,7 @@ AddToVisited(HTAB *v, HnswCandidate * hc, Relation index, bool *found)
  * Algorithm 2 from paper
  */
 List *
-HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, bool inserting, HnswElement skipElement)
+HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, int m, bool inserting, HnswElement skipElement)
 {
 	ListCell   *lc2;
 
@@ -551,8 +576,6 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
 	pairingheap *C = pairingheap_allocate(CompareNearestCandidates, NULL);
 	pairingheap *W = pairingheap_allocate(CompareFurthestCandidates, NULL);
 	int			wlen = 0;
-	uint64		dead = 0;
-	uint64		maxAdditional = skipElement == NULL ? ef : PG_UINT64_MAX;
 	HASHCTL		hash_ctl;
 	HTAB	   *v;
 
@@ -581,14 +604,13 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
 		pairingheap_add(C, &(CreatePairingHeapNode(hc)->ph_node));
 		pairingheap_add(W, &(CreatePairingHeapNode(hc)->ph_node));
 
-		/* Do not count certain number of dead elements towards ef */
+		/*
-		if (list_length(hc->element->heaptids) == 0)
+		 * Do not count elements being deleted towards ef when vacuuming. It
-		{
+		 * would be ideal to do this for inserts as well, but this could
-			if ((++dead) <= maxAdditional)
+		 * affect insert performance.
-				continue;
+		 */
-		}
+		if (skipElement == NULL || list_length(hc->element->heaptids) != 0)
-
+			wlen++;
-		wlen++;
 	}
 
 	while (!pairingheap_is_empty(C))
@@ -601,7 +623,7 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
 			break;
 
 		if (c->element->neighbors == NULL)
-			HnswLoadNeighbors(c->element, index);
+			HnswLoadNeighbors(c->element, index, m);
 
 		/* Get the neighborhood at layer lc */
 		neighborhood = &c->element->neighbors[lc];
@@ -641,18 +663,19 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
 					pairingheap_add(C, &(CreatePairingHeapNode(ec)->ph_node));
 					pairingheap_add(W, &(CreatePairingHeapNode(ec)->ph_node));
 
-					/* Do not count certain number of dead elements towards ef */
+					/*
-					if (list_length(e->element->heaptids) == 0)
+					 * Do not count elements being deleted towards ef when
+					 * vacuuming. It would be ideal to do this for inserts as
+					 * well, but this could affect insert performance.
+					 */
+					if (skipElement == NULL || list_length(e->element->heaptids) != 0)
 					{
-						if ((++dead) <= maxAdditional)
+						wlen++;
-							continue;
+
+						/* No need to decrement wlen */
+						if (wlen > ef)
+							pairingheap_remove_first(W);
 					}
-
-					wlen++;
-
-					/* No need to decrement wlen */
-					if (wlen > ef)
-						pairingheap_remove_first(W);
 				}
 			}
 		}
@@ -958,7 +981,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
 	/* 1st phase: greedy search to insert level */
 	for (int lc = entryLevel; lc >= level + 1; lc--)
 	{
-		w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, true, skipElement);
+		w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, m, true, skipElement);
 		ep = w;
 	}
 
@@ -976,7 +999,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
 		List	   *neighbors;
 		List	   *lw;
 
-		w = HnswSearchLayer(q, ep, efConstruction, lc, index, procinfo, collation, true, skipElement);
+		w = HnswSearchLayer(q, ep, efConstruction, lc, index, procinfo, collation, m, true, skipElement);
 
 		/* Elements being deleted or skipped can help with search */
 		/* but should be removed before selecting neighbors */

src/hnswvacuum.c

@@ -128,10 +128,7 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
 		blkno = HnswPageGetOpaque(page)->nextblkno;
 
 		if (updated)
-		{
-			MarkBufferDirty(buf);
 			GenericXLogFinish(state);
-		}
 		else
 			GenericXLogAbort(state);
 
@@ -229,7 +226,6 @@ RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswEleme
 		elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
 
 	/* Commit */
-	MarkBufferDirty(buf);
 	GenericXLogFinish(state);
 	UnlockReleaseBuffer(buf);
 
@@ -330,7 +326,10 @@ RepairGraph(HnswVacuumState * vacuumstate)
 	BufferAccessStrategy bas = vacuumstate->bas;
 	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);
 	UnlockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
 
@@ -443,7 +442,11 @@ MarkDeleted(HnswVacuumState * vacuumstate)
 	Relation	index = vacuumstate->index;
 	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);
 	UnlockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
 
@@ -540,9 +543,6 @@ MarkDeleted(HnswVacuumState * vacuumstate)
 				elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
 
 			/* Commit */
-			MarkBufferDirty(buf);
-			if (nbuf != buf)
-				MarkBufferDirty(nbuf);
 			GenericXLogFinish(state);
 			if (nbuf != buf)
 				UnlockReleaseBuffer(nbuf);
@@ -582,7 +582,6 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
 	vacuumstate->stats = stats;
 	vacuumstate->callback = callback;
 	vacuumstate->callback_state = callback_state;
-	vacuumstate->m = HnswGetM(index);
 	vacuumstate->efConstruction = HnswGetEfConstruction(index);
 	vacuumstate->bas = GetAccessStrategy(BAS_BULKREAD);
 	vacuumstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
@@ -592,6 +591,9 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
 												"Hnsw vacuum temporary context",
 												ALLOCSET_DEFAULT_SIZES);
 
+	/* Get m from metapage */
+	HnswGetMetaPageInfo(index, &vacuumstate->m, NULL);
+
 	/* Create hash table */
 	hash_ctl.keysize = sizeof(ItemPointerData);
 	hash_ctl.entrysize = sizeof(ItemPointerData);

src/ivfbuild.c

@@ -506,29 +506,30 @@ CreateListPages(Relation index, VectorArray centers, int dimensions,
 	Buffer		buf;
 	Page		page;
 	GenericXLogState *state;
-	OffsetNumber offno;
+	Size		listSize;
-	Size		itemsz;
 	IvfflatList list;
 
-	itemsz = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
+	listSize = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
-	list = palloc(itemsz);
+	list = palloc(listSize);
 
 	buf = IvfflatNewBuffer(index, forkNum);
 	IvfflatInitRegisterPage(index, &buf, &page, &state);
 
 	for (int i = 0; i < lists; i++)
 	{
+		OffsetNumber offno;
+
 		/* Load list */
 		list->startPage = InvalidBlockNumber;
 		list->insertPage = InvalidBlockNumber;
 		memcpy(&list->center, VectorArrayGet(centers, i), VECTOR_SIZE(dimensions));
 
 		/* Ensure free space */
-		if (PageGetFreeSpace(page) < itemsz)
+		if (PageGetFreeSpace(page) < listSize)
 			IvfflatAppendPage(index, &buf, &page, &state, forkNum);
 
 		/* Add the item */
-		offno = PageAddItem(page, (Item) list, itemsz, InvalidOffsetNumber, false, false);
+		offno = PageAddItem(page, (Item) list, listSize, InvalidOffsetNumber, false, false);
 		if (offno == InvalidOffsetNumber)
 			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
 

src/ivfflat.c

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

src/ivfflat.h

@@ -244,8 +244,10 @@ typedef struct IvfflatScanList
 typedef struct IvfflatScanOpaqueData
 {
 	int			probes;
+	int			dimensions;
 	bool		first;
 	Buffer		buf;
+	ItemPointerData heaptid;
 
 	/* Sorting */
 	Tuplesortstate *sortstate;
@@ -275,9 +277,10 @@ VectorArray VectorArrayInit(int maxlen, int dimensions);
 void		VectorArrayFree(VectorArray arr);
 void		PrintVectorArray(char *msg, VectorArray arr);
 void		IvfflatKmeans(Relation index, VectorArray samples, VectorArray centers);
-FmgrInfo   *IvfflatOptionalProcInfo(Relation rel, uint16 procnum);
+FmgrInfo   *IvfflatOptionalProcInfo(Relation index, uint16 procnum);
 bool		IvfflatNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
 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		IvfflatCommitBuffer(Buffer buf, GenericXLogState *state);
 void		IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state, ForkNumber forkNum);

src/ivfinsert.c

@@ -11,36 +11,37 @@
  * Find the list that minimizes the distance function
  */
 static void
-FindInsertPage(Relation rel, Datum *values, BlockNumber *insertPage, ListInfo * listInfo)
+FindInsertPage(Relation index, Datum *values, BlockNumber *insertPage, ListInfo * listInfo)
 {
-	Buffer		cbuf;
-	Page		cpage;
-	IvfflatList list;
-	double		distance;
 	double		minDistance = DBL_MAX;
 	BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
 	FmgrInfo   *procinfo;
 	Oid			collation;
-	OffsetNumber offno;
-	OffsetNumber maxoffno;
 
 	/* Avoid compiler warning */
 	listInfo->blkno = nextblkno;
 	listInfo->offno = FirstOffsetNumber;
 
-	procinfo = index_getprocinfo(rel, 1, IVFFLAT_DISTANCE_PROC);
+	procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC);
-	collation = rel->rd_indcollation[0];
+	collation = index->rd_indcollation[0];
 
 	/* Search all list pages */
 	while (BlockNumberIsValid(nextblkno))
 	{
-		cbuf = ReadBuffer(rel, nextblkno);
+		Buffer		cbuf;
+		Page		cpage;
+		OffsetNumber maxoffno;
+
+		cbuf = ReadBuffer(index, nextblkno);
 		LockBuffer(cbuf, BUFFER_LOCK_SHARE);
 		cpage = BufferGetPage(cbuf);
 		maxoffno = PageGetMaxOffsetNumber(cpage);
 
-		for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+		for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
 		{
+			IvfflatList list;
+			double		distance;
+
 			list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
 			distance = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, values[0], PointerGetDatum(&list->center)));
 
@@ -63,7 +64,7 @@ FindInsertPage(Relation rel, Datum *values, BlockNumber *insertPage, ListInfo *
  * Insert a tuple into the index
  */
 static void
-InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
+InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
 {
 	IndexTuple	itup;
 	Datum		value;
@@ -80,20 +81,20 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
 	value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
 
 	/* Normalize if needed */
-	normprocinfo = IvfflatOptionalProcInfo(rel, IVFFLAT_NORM_PROC);
+	normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC);
 	if (normprocinfo != NULL)
 	{
-		if (!IvfflatNormValue(normprocinfo, rel->rd_indcollation[0], &value, NULL))
+		if (!IvfflatNormValue(normprocinfo, index->rd_indcollation[0], &value, NULL))
 			return;
 	}
 
 	/* Find the insert page - sets the page and list info */
-	FindInsertPage(rel, values, &insertPage, &listInfo);
+	FindInsertPage(index, values, &insertPage, &listInfo);
 	Assert(BlockNumberIsValid(insertPage));
 	originalInsertPage = insertPage;
 
 	/* Form tuple */
-	itup = index_form_tuple(RelationGetDescr(rel), &value, isnull);
+	itup = index_form_tuple(RelationGetDescr(index), &value, isnull);
 	itup->t_tid = *heap_tid;
 
 	/* Get tuple size */
@@ -103,10 +104,10 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
 	/* Find a page to insert the item */
 	for (;;)
 	{
-		buf = ReadBuffer(rel, insertPage);
+		buf = ReadBuffer(index, insertPage);
 		LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
 
-		state = GenericXLogStart(rel);
+		state = GenericXLogStart(index);
 		page = GenericXLogRegisterBuffer(state, buf, 0);
 
 		if (PageGetFreeSpace(page) >= itemsz)
@@ -126,9 +127,9 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
 			Page		newpage;
 
 			/* Add a new page */
-			LockRelationForExtension(rel, ExclusiveLock);
+			LockRelationForExtension(index, ExclusiveLock);
-			newbuf = IvfflatNewBuffer(rel, MAIN_FORKNUM);
+			newbuf = IvfflatNewBuffer(index, MAIN_FORKNUM);
-			UnlockRelationForExtension(rel, ExclusiveLock);
+			UnlockRelationForExtension(index, ExclusiveLock);
 
 			/* Init new page */
 			newpage = GenericXLogRegisterBuffer(state, newbuf, GENERIC_XLOG_FULL_IMAGE);
@@ -141,15 +142,13 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
 			IvfflatPageGetOpaque(page)->nextblkno = insertPage;
 
 			/* Commit */
-			MarkBufferDirty(newbuf);
-			MarkBufferDirty(buf);
 			GenericXLogFinish(state);
 
 			/* Unlock previous buffer */
 			UnlockReleaseBuffer(buf);
 
 			/* Prepare new buffer */
-			state = GenericXLogStart(rel);
+			state = GenericXLogStart(index);
 			buf = newbuf;
 			page = GenericXLogRegisterBuffer(state, buf, 0);
 			break;
@@ -158,13 +157,13 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
 
 	/* Add to next offset */
 	if (PageAddItem(page, (Item) itup, itemsz, InvalidOffsetNumber, false, false) == InvalidOffsetNumber)
-		elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(rel));
+		elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
 
 	IvfflatCommitBuffer(buf, state);
 
 	/* Update the insert page */
 	if (insertPage != originalInsertPage)
-		IvfflatUpdateList(rel, listInfo, insertPage, originalInsertPage, InvalidBlockNumber, MAIN_FORKNUM);
+		IvfflatUpdateList(index, listInfo, insertPage, originalInsertPage, InvalidBlockNumber, MAIN_FORKNUM);
 }
 
 /*

src/ivfkmeans.c

@@ -17,10 +17,6 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
 	FmgrInfo   *procinfo;
 	Oid			collation;
 	int64		j;
-	double		distance;
-	double		sum;
-	double		choice;
-	Vector	   *vec;
 	float	   *weight = palloc(samples->length * sizeof(float));
 	int			numCenters = centers->maxlen;
 	int			numSamples = samples->length;
@@ -33,17 +29,21 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
 	centers->length++;
 
 	for (j = 0; j < numSamples; j++)
-		weight[j] = DBL_MAX;
+		weight[j] = FLT_MAX;
 
 	for (int i = 0; i < numCenters; i++)
 	{
+		double		sum;
+		double		choice;
+
 		CHECK_FOR_INTERRUPTS();
 
 		sum = 0.0;
 
 		for (j = 0; j < numSamples; j++)
 		{
-			vec = VectorArrayGet(samples, j);
+			Vector	   *vec = VectorArrayGet(samples, j);
+			double		distance;
 
 			/* Only need to compute distance for new center */
 			/* TODO Use triangle inequality to reduce distance calculations */
@@ -112,7 +112,6 @@ CompareVectors(const void *a, const void *b)
 static void
 QuickCenters(Relation index, VectorArray samples, VectorArray centers)
 {
-	Vector	   *vec;
 	int			dimensions = centers->dim;
 	Oid			collation = index->rd_indcollation[0];
 	FmgrInfo   *normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_KMEANS_NORM_PROC);
@@ -123,7 +122,7 @@ QuickCenters(Relation index, VectorArray samples, VectorArray centers)
 		qsort(samples->items, samples->length, VECTOR_SIZE(samples->dim), CompareVectors);
 		for (int i = 0; i < samples->length; i++)
 		{
-			vec = VectorArrayGet(samples, i);
+			Vector	   *vec = VectorArrayGet(samples, i);
 
 			if (i == 0 || CompareVectors(vec, VectorArrayGet(samples, i - 1)) != 0)
 			{
@@ -136,7 +135,7 @@ QuickCenters(Relation index, VectorArray samples, VectorArray centers)
 	/* Fill remaining with random data */
 	while (centers->length < centers->maxlen)
 	{
-		vec = VectorArrayGet(centers, centers->length);
+		Vector	   *vec = VectorArrayGet(centers, centers->length);
 
 		SET_VARSIZE(vec, VECTOR_SIZE(dimensions));
 		vec->dim = dimensions;
@@ -168,7 +167,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 	Oid			collation;
 	Vector	   *vec;
 	Vector	   *newCenter;
-	int			iteration;
 	int64		j;
 	int64		k;
 	int			dimensions = centers->dim;
@@ -182,14 +180,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 	float	   *s;
 	float	   *halfcdist;
 	float	   *newcdist;
-	int			changes;
-	double		minDistance;
-	int			closestCenter;
-	double		distance;
-	bool		rj;
-	bool		rjreset;
-	double		dxcx;
-	double		dxc;
 
 	/* Calculate allocation sizes */
 	Size		samplesSize = VECTOR_ARRAY_SIZE(samples->maxlen, samples->dim);
@@ -247,14 +237,14 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 	/* Assign each x to its closest initial center c(x) = argmin d(x,c) */
 	for (j = 0; j < numSamples; j++)
 	{
-		minDistance = DBL_MAX;
+		float		minDistance = FLT_MAX;
-		closestCenter = 0;
+		int			closestCenter = 0;
 
 		/* Find closest center */
 		for (k = 0; k < numCenters; k++)
 		{
 			/* TODO Use Lemma 1 in k-means++ initialization */
-			distance = lowerBound[j * numCenters + k];
+			float		distance = lowerBound[j * numCenters + k];
 
 			if (distance < minDistance)
 			{
@@ -268,13 +258,14 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 	}
 
 	/* Give 500 iterations to converge */
-	for (iteration = 0; iteration < 500; iteration++)
+	for (int iteration = 0; iteration < 500; iteration++)
 	{
+		int			changes = 0;
+		bool		rjreset;
+
 		/* Can take a while, so ensure we can interrupt */
 		CHECK_FOR_INTERRUPTS();
 
-		changes = 0;
-
 		/* Step 1: For all centers, compute distance */
 		for (j = 0; j < numCenters; j++)
 		{
@@ -282,7 +273,8 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 			for (k = j + 1; k < numCenters; k++)
 			{
-				distance = 0.5 * DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
+				float		distance = 0.5 * DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
+
 				halfcdist[j * numCenters + k] = distance;
 				halfcdist[k * numCenters + j] = distance;
 			}
@@ -291,10 +283,12 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 		/* For all centers c, compute s(c) */
 		for (j = 0; j < numCenters; j++)
 		{
-			minDistance = DBL_MAX;
+			float		minDistance = FLT_MAX;
 
 			for (k = 0; k < numCenters; k++)
 			{
+				float		distance;
+
 				if (j == k)
 					continue;
 
@@ -310,6 +304,8 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 		for (j = 0; j < numSamples; j++)
 		{
+			bool		rj;
+
 			/* Step 2: Identify all points x such that u(x) <= s(c(x)) */
 			if (upperBound[j] <= s[closestCenters[j]])
 				continue;
@@ -318,6 +314,8 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 			for (k = 0; k < numCenters; k++)
 			{
+				float		dxcx;
+
 				/* Step 3: For all remaining points x and centers c */
 				if (k == closestCenters[j])
 					continue;
@@ -347,7 +345,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 				/* Step 3b */
 				if (dxcx > lowerBound[j * numCenters + k] || dxcx > halfcdist[closestCenters[j] * numCenters + k])
 				{
-					dxc = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
+					float		dxc = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
 
 					/* d(x,c) calculated */
 					lowerBound[j * numCenters + k] = dxc;
@@ -361,7 +359,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 						changes++;
 					}
-
 				}
 			}
 		}
@@ -378,6 +375,8 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 		for (j = 0; j < numSamples; j++)
 		{
+			int			closestCenter;
+
 			vec = VectorArrayGet(samples, j);
 			closestCenter = closestCenters[j];
 
@@ -426,7 +425,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 		{
 			for (k = 0; k < numCenters; k++)
 			{
-				distance = lowerBound[j * numCenters + k] - newcdist[k];
+				float		distance = lowerBound[j * numCenters + k] - newcdist[k];
 
 				if (distance < 0)
 					distance = 0;
@@ -442,7 +441,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 		/* Step 7 */
 		for (j = 0; j < numCenters; j++)
-			memcpy(VectorArrayGet(centers, j), VectorArrayGet(newCenters, j), VECTOR_SIZE(dimensions));
+			VectorArraySet(centers, j, VectorArrayGet(newCenters, j));
 
 		if (changes == 0 && iteration != 0)
 			break;
@@ -465,9 +464,6 @@ static void
 CheckCenters(Relation index, VectorArray centers)
 {
 	FmgrInfo   *normprocinfo;
-	Oid			collation;
-	Vector	   *vec;
-	double		norm;
 
 	if (centers->length != centers->maxlen)
 		elog(ERROR, "Not enough centers. Please report a bug.");
@@ -475,7 +471,7 @@ CheckCenters(Relation index, VectorArray centers)
 	/* Ensure no NaN or infinite values */
 	for (int i = 0; i < centers->length; i++)
 	{
-		vec = VectorArrayGet(centers, i);
+		Vector	   *vec = VectorArrayGet(centers, i);
 
 		for (int j = 0; j < vec->dim; j++)
 		{
@@ -501,11 +497,12 @@ CheckCenters(Relation index, VectorArray centers)
 	normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC);
 	if (normprocinfo != NULL)
 	{
-		collation = index->rd_indcollation[0];
+		Oid			collation = index->rd_indcollation[0];
 
 		for (int i = 0; i < centers->length; i++)
 		{
-			norm = DatumGetFloat8(FunctionCall1Coll(normprocinfo, collation, PointerGetDatum(VectorArrayGet(centers, i))));
+			double		norm = DatumGetFloat8(FunctionCall1Coll(normprocinfo, collation, PointerGetDatum(VectorArrayGet(centers, i))));
+
 			if (norm == 0)
 				elog(ERROR, "Zero norm detected. Please report a bug.");
 		}

src/ivfscan.c

@@ -31,36 +31,36 @@ CompareLists(const pairingheap_node *a, const pairingheap_node *b, void *arg)
 static void
 GetScanLists(IndexScanDesc scan, Datum value)
 {
-	Buffer		cbuf;
+	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
-	Page		cpage;
-	IvfflatList list;
-	OffsetNumber offno;
-	OffsetNumber maxoffno;
 	BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
 	int			listCount = 0;
-	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
-	double		distance;
-	IvfflatScanList *scanlist;
 	double		maxDistance = DBL_MAX;
 
 	/* Search all list pages */
 	while (BlockNumberIsValid(nextblkno))
 	{
+		Buffer		cbuf;
+		Page		cpage;
+		OffsetNumber maxoffno;
+
 		cbuf = ReadBuffer(scan->indexRelation, nextblkno);
 		LockBuffer(cbuf, BUFFER_LOCK_SHARE);
 		cpage = BufferGetPage(cbuf);
 
 		maxoffno = PageGetMaxOffsetNumber(cpage);
 
-		for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+		for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
 		{
-			list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
+			IvfflatList list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
+			double		distance;
 
 			/* Use procinfo from the index instead of scan key for performance */
 			distance = DatumGetFloat8(FunctionCall2Coll(so->procinfo, so->collation, PointerGetDatum(&list->center), value));
 
 			if (listCount < so->probes)
 			{
+				IvfflatScanList *scanlist;
+
 				scanlist = &so->lists[listCount];
 				scanlist->startPage = list->startPage;
 				scanlist->distance = distance;
@@ -75,6 +75,8 @@ GetScanLists(IndexScanDesc scan, Datum value)
 			}
 			else if (distance < maxDistance)
 			{
+				IvfflatScanList *scanlist;
+
 				/* Remove */
 				scanlist = (IvfflatScanList *) pairingheap_remove_first(so->listQueue);
 
@@ -101,14 +103,6 @@ static void
 GetScanItems(IndexScanDesc scan, Datum value)
 {
 	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);
 	double		tuples = 0;
 
@@ -128,19 +122,32 @@ GetScanItems(IndexScanDesc scan, Datum value)
 	/* Search closest probes lists */
 	while (!pairingheap_is_empty(so->listQueue))
 	{
-		searchPage = ((IvfflatScanList *) pairingheap_remove_first(so->listQueue))->startPage;
+		BlockNumber searchPage = ((IvfflatScanList *) pairingheap_remove_first(so->listQueue))->startPage;
 
 		/* Search all entry pages for list */
 		while (BlockNumberIsValid(searchPage))
 		{
+			Buffer		buf;
+			Page		page;
+			OffsetNumber maxoffno;
+
 			buf = ReadBufferExtended(scan->indexRelation, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas);
 			LockBuffer(buf, BUFFER_LOCK_SHARE);
 			page = BufferGetPage(buf);
 			maxoffno = PageGetMaxOffsetNumber(page);
 
-			for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+			for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
 			{
-				itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offno));
+				IndexTuple	itup;
+				Datum		datum;
+				bool		isnull;
+				ItemId		itemid = PageGetItemId(page, offno);
+
+				/* Skip dead tuples */
+				if (scan->ignore_killed_tuples && ItemIdIsDead(itemid))
+					continue;
+
+				itup = (IndexTuple) PageGetItem(page, itemid);
 				datum = index_getattr(itup, 1, tupdesc, &isnull);
 
 				/*
@@ -181,26 +188,52 @@ GetScanItems(IndexScanDesc scan, Datum value)
 }
 
 /*
- * Get dimensions from metapage
+ * Mark prior tuple as dead
  */
-static int
+static void
-GetDimensions(Relation index)
+MarkPriorTupleDead(IndexScanDesc scan)
 {
-	Buffer		buf;
+	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
+	Buffer		buf = so->buf;
 	Page		page;
-	IvfflatMetaPage metap;
+	OffsetNumber maxoffno;
-	int			dimensions;
 
-	buf = ReadBuffer(index, IVFFLAT_METAPAGE_BLKNO);
+	/* Safety check */
+	if (!BufferIsValid(so->buf) || !ItemPointerIsValid(&so->heaptid))
+		return;
+
+	/* Only a shared locked is needed for ItemIdMarkDead */
 	LockBuffer(buf, BUFFER_LOCK_SHARE);
 	page = BufferGetPage(buf);
-	metap = IvfflatPageGetMeta(page);
+	maxoffno = PageGetMaxOffsetNumber(page);
 
-	dimensions = metap->dimensions;
+	for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+	{
+		ItemId		itemid = PageGetItemId(page, offno);
+		IndexTuple	itup = (IndexTuple) PageGetItem(page, itemid);
 
-	UnlockReleaseBuffer(buf);
+		/*
+		 * Find tuple. Since buffer has been pinned, tuple cannot have been
+		 * vacuumed (and heap TID reused).
+		 */
+		if (ItemPointerEquals(&itup->t_tid, &so->heaptid))
+		{
+			/*
+			 * Make sure tuple has not already been marked dead to avoid extra
+			 * WAL if wal_log_hints or data checksums enabled
+			 */
+			if (!ItemIdIsDead(itemid))
+			{
+				ItemIdMarkDead(itemid);
+				MarkBufferDirtyHint(buf, true);
+			}
 
-	return dimensions;
+			break;
+		}
+	}
+
+	/* Unlock buffer */
+	LockBuffer(buf, BUFFER_LOCK_UNLOCK);
 }
 
 /*
@@ -212,6 +245,7 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 	IndexScanDesc scan;
 	IvfflatScanOpaque so;
 	int			lists;
+	int			dimensions;
 	AttrNumber	attNums[] = {1};
 	Oid			sortOperators[] = {Float8LessOperator};
 	Oid			sortCollations[] = {InvalidOid};
@@ -219,7 +253,9 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 	int			probes = ivfflat_probes;
 
 	scan = RelationGetIndexScan(index, nkeys, norderbys);
-	lists = IvfflatGetLists(scan->indexRelation);
+
+	/* Get lists and dimensions from metapage */
+	IvfflatGetMetaPageInfo(index, &lists, &dimensions);
 
 	if (probes > lists)
 		probes = lists;
@@ -227,7 +263,9 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 	so = (IvfflatScanOpaque) palloc(offsetof(IvfflatScanOpaqueData, lists) + probes * sizeof(IvfflatScanList));
 	so->buf = InvalidBuffer;
 	so->first = true;
+	ItemPointerSetInvalid(&so->heaptid);
 	so->probes = probes;
+	so->dimensions = dimensions;
 
 	/* Set support functions */
 	so->procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC);
@@ -241,7 +279,7 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 	so->tupdesc = CreateTemplateTupleDesc(3, false);
 #endif
 	TupleDescInitEntry(so->tupdesc, (AttrNumber) 1, "distance", FLOAT8OID, -1, 0);
-	TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "tid", TIDOID, -1, 0);
+	TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "heaptid", TIDOID, -1, 0);
 	TupleDescInitEntry(so->tupdesc, (AttrNumber) 3, "indexblkno", INT4OID, -1, 0);
 
 	/* Prep sort */
@@ -274,6 +312,7 @@ ivfflatrescan(IndexScanDesc scan, ScanKey keys, int nkeys, ScanKey orderbys, int
 #endif
 
 	so->first = true;
+	ItemPointerSetInvalid(&so->heaptid);
 	pairingheap_reset(so->listQueue);
 
 	if (keys && scan->numberOfKeys > 0)
@@ -309,7 +348,7 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
 			elog(ERROR, "cannot scan ivfflat index without order");
 
 		if (scan->orderByData->sk_flags & SK_ISNULL)
-			value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
+			value = PointerGetDatum(InitVector(so->dimensions));
 		else
 		{
 			value = scan->orderByData->sk_argument;
@@ -331,18 +370,27 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
 		if (value != scan->orderByData->sk_argument)
 			pfree(DatumGetPointer(value));
 	}
+	else
+	{
+		/* Mark prior tuple as dead */
+		if (scan->kill_prior_tuple)
+			MarkPriorTupleDead(scan);
+	}
 
 	if (tuplesort_gettupleslot(so->sortstate, true, false, so->slot, NULL))
 	{
-		ItemPointer tid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
+		ItemPointer heaptid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
 		BlockNumber indexblkno = DatumGetInt32(slot_getattr(so->slot, 3, &so->isnull));
 
 #if PG_VERSION_NUM >= 120000
-		scan->xs_heaptid = *tid;
+		scan->xs_heaptid = *heaptid;
 #else
-		scan->xs_ctup.t_self = *tid;
+		scan->xs_ctup.t_self = *heaptid;
 #endif
 
+		/* Keep track of info needed to mark tuple as dead */
+		so->heaptid = *heaptid;
+
 		/* Unpin buffer */
 		if (BufferIsValid(so->buf))
 			ReleaseBuffer(so->buf);

src/ivfutils.c

@@ -57,12 +57,12 @@ IvfflatGetLists(Relation index)
  * Get proc
  */
 FmgrInfo *
-IvfflatOptionalProcInfo(Relation rel, uint16 procnum)
+IvfflatOptionalProcInfo(Relation index, uint16 procnum)
 {
-	if (!OidIsValid(index_getprocid(rel, 1, procnum)))
+	if (!OidIsValid(index_getprocid(index, 1, procnum)))
 		return NULL;
 
-	return index_getprocinfo(rel, 1, procnum);
+	return index_getprocinfo(index, 1, procnum);
 }
 
 /*
@@ -136,7 +136,6 @@ IvfflatInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogStat
 void
 IvfflatCommitBuffer(Buffer buf, GenericXLogState *state)
 {
-	MarkBufferDirty(buf);
 	GenericXLogFinish(state);
 	UnlockReleaseBuffer(buf);
 }
@@ -160,8 +159,6 @@ IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **st
 	IvfflatInitPage(newbuf, newpage);
 
 	/* Commit */
-	MarkBufferDirty(*buf);
-	MarkBufferDirty(newbuf);
 	GenericXLogFinish(*state);
 
 	/* Unlock */
@@ -172,6 +169,29 @@ IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **st
 	*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
  */

src/ivfvacuum.c

@@ -107,7 +107,6 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
 				{
 					/* Delete tuples */
 					PageIndexMultiDelete(page, deletable, ndeletable);
-					MarkBufferDirty(buf);
 					GenericXLogFinish(state);
 				}
 				else

src/vector.c

@@ -695,6 +695,8 @@ vector_spherical_distance(PG_FUNCTION_ARGS)
 {
 	Vector	   *a = PG_GETARG_VECTOR_P(0);
 	Vector	   *b = PG_GETARG_VECTOR_P(1);
+	float	   *ax = a->x;
+	float	   *bx = b->x;
 	float		dp = 0.0;
 	double		distance;
 
@@ -702,7 +704,7 @@ vector_spherical_distance(PG_FUNCTION_ARGS)
 
 	/* Auto-vectorized */
 	for (int i = 0; i < a->dim; i++)
-		dp += a->x[i] * b->x[i];
+		dp += ax[i] * bx[i];
 
 	distance = (double) dp;
 

test/expected/functions.out

@@ -106,6 +106,12 @@ SELECT cosine_distance('[1,1]', '[1,1]');
                0
 (1 row)
 
+SELECT cosine_distance('[1,0]', '[0,2]');
+ cosine_distance 
+-----------------
+               1
+(1 row)
+
 SELECT cosine_distance('[1,1]', '[-1,-1]');
  cosine_distance 
 -----------------

test/sql/functions.sql

@@ -25,6 +25,7 @@ SELECT inner_product('[3e38]', '[3e38]');
 SELECT cosine_distance('[1,2]', '[2,4]');
 SELECT cosine_distance('[1,2]', '[0,0]');
 SELECT cosine_distance('[1,1]', '[1,1]');
+SELECT cosine_distance('[1,0]', '[0,2]');
 SELECT cosine_distance('[1,1]', '[-1,-1]');
 SELECT cosine_distance('[1,2]', '[3]');
 SELECT cosine_distance('[1,1]', '[1.1,1.1]');

test/t/001_ivfflat_wal.pl

@@ -19,8 +19,6 @@ sub test_index_replay
 
 	# Wait for replica to catch up
 	my $applname = $node_replica->name;
-
-	my $server_version_num = $node_primary->safe_psql("postgres", "SHOW server_version_num");
 	my $caughtup_query = "SELECT pg_current_wal_lsn() <= replay_lsn FROM pg_stat_replication WHERE application_name = '$applname';";
 	$node_primary->poll_query_until('postgres', $caughtup_query)
 	  or die "Timed out while waiting for replica 1 to catch up";

test/t/010_hnsw_wal.pl

@@ -19,8 +19,6 @@ sub test_index_replay
 
 	# Wait for replica to catch up
 	my $applname = $node_replica->name;
-
-	my $server_version_num = $node_primary->safe_psql("postgres", "SHOW server_version_num");
 	my $caughtup_query = "SELECT pg_current_wal_lsn() <= replay_lsn FROM pg_stat_replication WHERE application_name = '$applname';";
 	$node_primary->poll_query_until('postgres', $caughtup_query)
 	  or die "Timed out while waiting for replica 1 to catch up";
@@ -38,9 +36,8 @@ sub test_index_replay
 	);
 
 	# Run test queries and compare their result
-	# Query replica first since index scan on primary can generate WAL removing tuples
-	my $replica_result = $node_replica->safe_psql("postgres", $queries);
 	my $primary_result = $node_primary->safe_psql("postgres", $queries);
+	my $replica_result = $node_replica->safe_psql("postgres", $queries);
 
 	is($primary_result, $replica_result, "$test_name: query result matches");
 	return;

test/t/015_hnsw_duplicates.pl

@@ -23,27 +23,25 @@ sub insert_vectors
 
 sub test_duplicates
 {
-	my ($exp) = @_;
-
 	my $res = $node->safe_psql("postgres", qq(
 		SET enable_seqscan = off;
 		SET hnsw.ef_search = 1;
 		SELECT COUNT(*) FROM (SELECT * FROM tst ORDER BY v <-> '[1,1,1]') t;
 	));
-	is($res, $exp);
+	is($res, 10);
 }
 
 # Test duplicates with build
 insert_vectors();
 $node->safe_psql("postgres", "CREATE INDEX idx ON tst USING hnsw (v vector_l2_ops);");
-test_duplicates(10);
+test_duplicates();
 
 # Reset
 $node->safe_psql("postgres", "TRUNCATE tst;");
 
 # Test duplicates with inserts
 insert_vectors();
-test_duplicates(10);
+test_duplicates();
 
 # Test fallback path for inserts
 $node->pgbench(
@@ -57,15 +55,4 @@ $node->pgbench(
 	}
 );
 
-# Reset
-$node->safe_psql("postgres", "TRUNCATE tst;");
-
-# Test deletes with index scan
-$node->safe_psql("postgres", "INSERT INTO tst SELECT '[1,1,1]' FROM generate_series(1, 10) i;");
-$node->safe_psql("postgres", "DELETE FROM tst WHERE ctid IN (SELECT ctid FROM tst ORDER BY random() LIMIT 5);");
-for (1 .. 3)
-{
-	test_duplicates(5);
-}
-
 done_testing();

test/t/018_ivfflat_deletes.pl

@@ -0,0 +1,43 @@
+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();

vector.control

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