Use List for samples

src/hnsw.c

@@ -33,12 +33,6 @@ HnswInit(void)
 					  HNSW_DEFAULT_EF_CONSTRUCTION, HNSW_MIN_EF_CONSTRUCTION, HNSW_MAX_EF_CONSTRUCTION
 #if PG_VERSION_NUM >= 130000
 					  ,AccessExclusiveLock
-#endif
-		);
-	add_int_reloption(hnsw_relopt_kind, "dimensions", "Number of dimensions",
-					  HNSW_DEFAULT_DIMENSIONS, HNSW_MIN_DIMENSIONS, HNSW_MAX_DIMENSIONS
-#if PG_VERSION_NUM >= 130000
-					  ,AccessExclusiveLock
 #endif
 		);
 
@@ -131,7 +125,6 @@ hnswoptions(Datum reloptions, bool validate)
 	static const relopt_parse_elt tab[] = {
 		{"m", RELOPT_TYPE_INT, offsetof(HnswOptions, m)},
 		{"ef_construction", RELOPT_TYPE_INT, offsetof(HnswOptions, efConstruction)},
-		{"dimensions", RELOPT_TYPE_INT, offsetof(HnswOptions, dimensions)},
 	};
 
 #if PG_VERSION_NUM >= 130000

src/hnsw.h

@@ -42,9 +42,6 @@
 #define HNSW_DEFAULT_EF_SEARCH	40
 #define HNSW_MIN_EF_SEARCH		1
 #define HNSW_MAX_EF_SEARCH		1000
-#define HNSW_DEFAULT_DIMENSIONS	-1
-#define HNSW_MIN_DIMENSIONS		1
-#define HNSW_MAX_DIMENSIONS		HNSW_MAX_DIM
 
 /* Tuple types */
 #define HNSW_ELEMENT_TUPLE_TYPE  1
@@ -62,7 +59,7 @@
 
 #define HNSW_MAX_SIZE (BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData)) - sizeof(ItemIdData))
 
-#define HNSW_ELEMENT_TUPLE_SIZE(_datum)	MAXALIGN(offsetof(HnswElementTupleData, value) + VARSIZE_ANY(_datum))
+#define HNSW_ELEMENT_TUPLE_SIZE(_dim)	MAXALIGN(offsetof(HnswElementTupleData, vec) + VECTOR_SIZE(_dim))
 #define HNSW_NEIGHBOR_TUPLE_SIZE(level, m)	MAXALIGN(offsetof(HnswNeighborTupleData, indextids) + ((level) + 2) * (m) * sizeof(ItemPointerData))
 
 #define HnswPageGetOpaque(page)	((HnswPageOpaque) PageGetSpecialPointer(page))
@@ -106,7 +103,7 @@ typedef struct HnswElementData
 	OffsetNumber offno;
 	OffsetNumber neighborOffno;
 	BlockNumber neighborPage;
-	Datum		value;
+	Vector	   *vec;
 }			HnswElementData;
 
 typedef HnswElementData * HnswElement;
@@ -137,7 +134,6 @@ typedef struct HnswOptions
 	int32		vl_len_;		/* varlena header (do not touch directly!) */
 	int			m;				/* number of connections */
 	int			efConstruction; /* size of dynamic candidate list */
-	int			dimensions;
 }			HnswOptions;
 
 typedef struct HnswBuildState
@@ -208,7 +204,7 @@ typedef struct HnswElementTupleData
 	ItemPointerData heaptids[HNSW_HEAPTIDS];
 	ItemPointerData neighbortid;
 	uint16		unused2;
-	char		value[FLEXIBLE_ARRAY_MEMBER];
+	Vector		vec;
 }			HnswElementTupleData;
 
 typedef HnswElementTupleData * HnswElementTuple;
@@ -266,7 +262,6 @@ typedef struct HnswVacuumState
 /* Methods */
 int			HnswGetM(Relation index);
 int			HnswGetEfConstruction(Relation index);
-int			HnswGetDimensions(Relation index);
 FmgrInfo   *HnswOptionalProcInfo(Relation index, uint16 procnum);
 bool		HnswNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
 void		HnswCommitBuffer(Buffer buf, GenericXLogState *state);

src/hnswbuild.c

@@ -8,7 +8,6 @@
 #include "lib/pairingheap.h"
 #include "nodes/pg_list.h"
 #include "storage/bufmgr.h"
-#include "utils/datum.h"
 #include "utils/memutils.h"
 
 #if PG_VERSION_NUM >= 140000
@@ -106,6 +105,8 @@ CreateElementPages(HnswBuildState * buildstate)
 {
 	Relation	index = buildstate->index;
 	ForkNumber	forkNum = buildstate->forkNum;
+	int			dimensions = buildstate->dimensions;
+	Size		etupSize;
 	Size		maxSize;
 	HnswElementTuple etup;
 	HnswNeighborTuple ntup;
@@ -117,9 +118,10 @@ CreateElementPages(HnswBuildState * buildstate)
 
 	/* Calculate sizes */
 	maxSize = HNSW_MAX_SIZE;
+	etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions);
 
 	/* Allocate once */
-	etup = palloc0(BLCKSZ);
+	etup = palloc0(etupSize);
 	ntup = palloc0(BLCKSZ);
 
 	/* Prepare first page */
@@ -131,14 +133,12 @@ CreateElementPages(HnswBuildState * buildstate)
 	foreach(lc, buildstate->elements)
 	{
 		HnswElement element = lfirst(lc);
-		Size		etupSize;
 		Size		ntupSize;
 		Size		combinedSize;
 
 		HnswSetElementTuple(etup, element);
 
 		/* Calculate sizes */
-		etupSize = HNSW_ELEMENT_TUPLE_SIZE(element->value);
 		ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(element->level, buildstate->m);
 		combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
 
@@ -273,15 +273,18 @@ InsertTuple(Relation index, Datum *values, HnswElement element, HnswBuildState *
 	int			m = buildstate->m;
 
 	/* Detoast once for all calls */
-	element->value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
+	Datum		value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
 
 	/* Normalize if needed */
 	if (buildstate->normprocinfo != NULL)
 	{
-		if (!HnswNormValue(buildstate->normprocinfo, collation, &element->value, buildstate->normvec))
+		if (!HnswNormValue(buildstate->normprocinfo, collation, &value, buildstate->normvec))
 			return false;
 	}
 
+	/* Copy value to element so accessible outside of memory context */
+	memcpy(element->vec, DatumGetVector(value), VECTOR_SIZE(buildstate->dimensions));
+
 	/* Insert element in graph */
 	HnswInsertElement(element, entryPoint, NULL, procinfo, collation, m, efConstruction, false);
 
@@ -357,6 +360,7 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
 
 	/* Allocate necessary memory outside of memory context */
 	element = HnswInitElement(tid, buildstate->m, buildstate->ml, buildstate->maxLevel);
+	element->vec = palloc(VECTOR_SIZE(buildstate->dimensions));
 
 	/* Use memory context since detoast can allocate */
 	oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
@@ -364,8 +368,9 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
 	/* Insert tuple */
 	inserted = InsertTuple(index, values, element, buildstate, &dup);
 
-	/* Switch memory context */
+	/* Reset memory context */
 	MemoryContextSwitchTo(oldCtx);
+	MemoryContextReset(buildstate->tmpCtx);
 
 	/* Add outside memory context */
 	if (dup != NULL)
@@ -373,18 +378,9 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
 
 	/* Add to buildstate or free */
 	if (inserted)
-	{
-		element->value = datumCopy(element->value, false, -1);
 		buildstate->elements = lappend(buildstate->elements, element);
-	}
 	else
-	{
-		element->value = PointerGetDatum(NULL);
 		HnswFreeElement(element);
-	}
-
-	/* Reset memory context */
-	MemoryContextReset(buildstate->tmpCtx);
 }
 
 /*
@@ -399,7 +395,6 @@ HnswGetMaxInMemoryElements(int m, double ml, int dimensions)
 	elementSize += sizeof(HnswNeighborArray) * (avgLevel + 1);
 	elementSize += sizeof(HnswCandidate) * (m * (avgLevel + 2));
 	elementSize += sizeof(ItemPointerData);
-	/* TODO Handle non-vector types */
 	elementSize += VECTOR_SIZE(dimensions);
 	return (maintenance_work_mem * 1024L) / elementSize;
 }
@@ -417,10 +412,7 @@ InitBuildState(HnswBuildState * buildstate, Relation heap, Relation index, Index
 
 	buildstate->m = HnswGetM(index);
 	buildstate->efConstruction = HnswGetEfConstruction(index);
-	buildstate->dimensions = HnswGetDimensions(index);
-
-	if (buildstate->dimensions < 0)
-		buildstate->dimensions = TupleDescAttr(index->rd_att, 0)->atttypmod;
+	buildstate->dimensions = TupleDescAttr(index->rd_att, 0)->atttypmod;
 
 	/* Require column to have dimensions to be indexed */
 	if (buildstate->dimensions < 0)

src/hnswinsert.c

@@ -123,6 +123,7 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
 	Size		minCombinedSize;
 	HnswElementTuple etup;
 	BlockNumber currentPage = insertPage;
+	int			dimensions = e->vec->dim;
 	HnswNeighborTuple ntup;
 	Buffer		nbuf;
 	Page		npage;
@@ -131,7 +132,7 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
 	BlockNumber newInsertPage = InvalidBlockNumber;
 
 	/* Calculate sizes */
-	etupSize = HNSW_ELEMENT_TUPLE_SIZE(e->value);
+	etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions);
 	ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m);
 	combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
 	maxSize = HNSW_MAX_SIZE;
@@ -404,7 +405,7 @@ HnswAddDuplicate(Relation index, HnswElement element, HnswElement dup)
 	Buffer		buf;
 	Page		page;
 	GenericXLogState *state;
-	Size		etupSize = HNSW_ELEMENT_TUPLE_SIZE(dup->value);
+	Size		etupSize = HNSW_ELEMENT_TUPLE_SIZE(dup->vec->dim);
 	HnswElementTuple etup;
 	int			i;
 
@@ -514,7 +515,7 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
 
 	/* Create an element */
 	element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m));
-	element->value = value;
+	element->vec = DatumGetVector(value);
 
 	/* Prevent concurrent inserts when likely updating entry point */
 	if (entryPoint == NULL || element->level > entryPoint->level)

src/hnswutils.c

@@ -4,7 +4,6 @@
 
 #include "hnsw.h"
 #include "storage/bufmgr.h"
-#include "utils/datum.h"
 #include "vector.h"
 
 /*
@@ -35,20 +34,6 @@ HnswGetEfConstruction(Relation index)
 	return HNSW_DEFAULT_EF_CONSTRUCTION;
 }
 
-/*
- * Get the number of dimensions in the index
- */
-int
-HnswGetDimensions(Relation index)
-{
-	HnswOptions *opts = (HnswOptions *) index->rd_options;
-
-	if (opts)
-		return opts->dimensions;
-
-	return HNSW_DEFAULT_DIMENSIONS;
-}
-
 /*
  * Get proc
  */
@@ -202,8 +187,7 @@ HnswFreeElement(HnswElement element)
 {
 	HnswFreeNeighbors(element);
 	list_free_deep(element->heaptids);
-	if (DatumGetPointer(element->value))
-		pfree(DatumGetPointer(element->value));
+	pfree(element->vec);
 	pfree(element);
 }
 
@@ -230,7 +214,7 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
 	element->blkno = blkno;
 	element->offno = offno;
 	element->neighbors = NULL;
-	element->value = PointerGetDatum(NULL);
+	element->vec = NULL;
 	return element;
 }
 
@@ -340,7 +324,7 @@ HnswSetElementTuple(HnswElementTuple etup, HnswElement element)
 		else
 			ItemPointerSetInvalid(&etup->heaptids[i]);
 	}
-	memcpy(&etup->value, DatumGetPointer(element->value), VARSIZE_ANY(element->value));
+	memcpy(&etup->vec, element->vec, VECTOR_SIZE(element->vec->dim));
 }
 
 /*
@@ -463,9 +447,8 @@ HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHe
 
 	if (loadVec)
 	{
-		Datum		value = PointerGetDatum(&etup->value);
-
-		element->value = datumCopy(value, false, -1);
+		element->vec = palloc(VECTOR_SIZE(etup->vec.dim));
+		memcpy(element->vec, &etup->vec, VECTOR_SIZE(etup->vec.dim));
 	}
 }
 
@@ -493,7 +476,7 @@ HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index,
 
 	/* Calculate distance */
 	if (distance != NULL)
-		*distance = (float) DatumGetFloat8(FunctionCall2Coll(procinfo, collation, *q, PointerGetDatum(&etup->value)));
+		*distance = (float) DatumGetFloat8(FunctionCall2Coll(procinfo, collation, *q, PointerGetDatum(&etup->vec)));
 
 	UnlockReleaseBuffer(buf);
 }
@@ -504,7 +487,7 @@ HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index,
 static float
 GetCandidateDistance(HnswCandidate * hc, Datum q, FmgrInfo *procinfo, Oid collation)
 {
-	return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, q, hc->element->value));
+	return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, q, PointerGetDatum(hc->element->vec)));
 }
 
 /*
@@ -767,7 +750,7 @@ HnswGetDistance(HnswElement a, HnswElement b, int lc, FmgrInfo *procinfo, Oid co
 		}
 	}
 
-	return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, a->value, b->value));
+	return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(a->vec), PointerGetDatum(b->vec)));
 }
 
 /*
@@ -894,7 +877,7 @@ HnswFindDuplicate(HnswElement e)
 		HnswCandidate *neighbor = &neighbors->items[i];
 
 		/* Exit early since ordered by distance */
-		if (!datumIsEqual(e->value, neighbor->element->value, false, -1))
+		if (vector_cmp_internal(e->vec, neighbor->element->vec) != 0)
 			break;
 
 		/* Check for space */
@@ -947,13 +930,13 @@ HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int
 		/* Load elements on insert */
 		if (index != NULL)
 		{
-			Datum		q = hc->element->value;
+			Datum		q = PointerGetDatum(hc->element->vec);
 
 			for (int i = 0; i < currentNeighbors->length; i++)
 			{
 				HnswCandidate *hc3 = &currentNeighbors->items[i];
 
-				if (!DatumGetPointer(hc3->element->value))
+				if (hc3->element->vec == NULL)
 					HnswLoadElement(hc3->element, &hc3->distance, &q, index, procinfo, collation, true);
 				else
 					hc3->distance = GetCandidateDistance(hc3, q, procinfo, collation);
@@ -1034,7 +1017,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
 	List	   *w;
 	int			level = element->level;
 	int			entryLevel;
-	Datum		q = element->value;
+	Datum		q = PointerGetDatum(element->vec);
 	HnswElement skipElement = existing ? element : NULL;
 
 	/* No neighbors if no entry point */

src/hnswvacuum.c

@@ -93,7 +93,7 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
 
 				if (itemUpdated)
 				{
-					Size		etupSize = HNSW_ELEMENT_TUPLE_SIZE(PointerGetDatum(&etup->value));
+					Size		etupSize = HNSW_ELEMENT_TUPLE_SIZE(etup->vec.dim);
 
 					/* Mark rest as invalid */
 					for (int i = idx; i < HNSW_HEAPTIDS; i++)
@@ -481,7 +481,6 @@ MarkDeleted(HnswVacuumState * vacuumstate)
 			HnswNeighborTuple ntup;
 			Size		etupSize;
 			Size		ntupSize;
-			Datum		value;
 			Buffer		nbuf;
 			Page		npage;
 			BlockNumber neighborPage;
@@ -505,11 +504,8 @@ MarkDeleted(HnswVacuumState * vacuumstate)
 			if (ItemPointerIsValid(&etup->heaptids[0]))
 				continue;
 
-			/* Get datum */
-			value = PointerGetDatum(&etup->value);
-
 			/* Calculate sizes */
-			etupSize = HNSW_ELEMENT_TUPLE_SIZE(value);
+			etupSize = HNSW_ELEMENT_TUPLE_SIZE(etup->vec.dim);
 			ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(etup->level, vacuumstate->m);
 
 			/* Get neighbor page */
@@ -532,7 +528,7 @@ MarkDeleted(HnswVacuumState * vacuumstate)
 
 			/* Overwrite element */
 			etup->deleted = 1;
-			MemSet(&etup->value, 0, VARSIZE_ANY(value));
+			MemSet(&etup->vec.x, 0, etup->vec.dim * sizeof(float));
 
 			/* Overwrite neighbors */
 			for (int i = 0; i < ntup->count; i++)

src/ivfbuild.c

@@ -11,6 +11,7 @@
 #include "miscadmin.h"
 #include "storage/bufmgr.h"
 #include "tcop/tcopprot.h"
+#include "utils/datum.h"
 #include "utils/memutils.h"
 
 #if PG_VERSION_NUM >= 140000
@@ -65,11 +66,18 @@
 static void
 AddSample(Datum *values, IvfflatBuildState * buildstate)
 {
-	VectorArray samples = buildstate->samples;
-	int			targsamples = samples->maxlen;
+	MemoryContext oldCtx;
+	Datum		value;
+	int			targsamples = buildstate->targsamples;
+
+	/* Use memory context since detoast can allocate */
+	oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
 
 	/* Detoast once for all calls */
-	Datum		value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
+	value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
+
+	/* Restore memory context */
+	MemoryContextSwitchTo(oldCtx);
 
 	/*
 	 * Normalize with KMEANS_NORM_PROC since spherical distance function
@@ -81,18 +89,23 @@ AddSample(Datum *values, IvfflatBuildState * buildstate)
 			return;
 	}
 
-	if (samples->length < targsamples)
-	{
-		VectorArraySet(samples, samples->length, DatumGetVector(value));
-		samples->length++;
-	}
+	/* Copy datum */
+	value = datumCopy(value, false, -1);
+
+	/* Reset memory context */
+	MemoryContextReset(buildstate->tmpCtx);
+
+	if (list_length(buildstate->samples) < targsamples)
+		buildstate->samples = lappend(buildstate->samples, DatumGetVector(value));
 	else
 	{
 		if (buildstate->rowstoskip < 0)
-			buildstate->rowstoskip = reservoir_get_next_S(&buildstate->rstate, samples->length, targsamples);
+			buildstate->rowstoskip = reservoir_get_next_S(&buildstate->rstate, list_length(buildstate->samples), targsamples);
 
 		if (buildstate->rowstoskip <= 0)
 		{
+			ListCell   *lc;
+
 #if PG_VERSION_NUM >= 150000
 			int			k = (int) (targsamples * sampler_random_fract(&buildstate->rstate.randstate));
 #else
@@ -100,7 +113,8 @@ AddSample(Datum *values, IvfflatBuildState * buildstate)
 #endif
 
 			Assert(k >= 0 && k < targsamples);
-			VectorArraySet(samples, k, DatumGetVector(value));
+			lc = list_nth_cell(buildstate->samples, k);
+			lfirst(lc) = DatumGetVector(value);
 		}
 
 		buildstate->rowstoskip -= 1;
@@ -115,21 +129,13 @@ SampleCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
 			   bool *isnull, bool tupleIsAlive, void *state)
 {
 	IvfflatBuildState *buildstate = (IvfflatBuildState *) state;
-	MemoryContext oldCtx;
 
 	/* Skip nulls */
 	if (isnull[0])
 		return;
 
-	/* Use memory context since detoast can allocate */
-	oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
-
 	/* Add sample */
-	AddSample(values, state);
-
-	/* Reset memory context */
-	MemoryContextSwitchTo(oldCtx);
-	MemoryContextReset(buildstate->tmpCtx);
+	AddSample(values, buildstate);
 }
 
 /*
@@ -138,7 +144,7 @@ SampleCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
 static void
 SampleRows(IvfflatBuildState * buildstate)
 {
-	int			targsamples = buildstate->samples->maxlen;
+	int			targsamples = buildstate->targsamples;
 	BlockNumber totalblocks = RelationGetNumberOfBlocks(buildstate->heap);
 
 	buildstate->rowstoskip = -1;
@@ -449,12 +455,13 @@ ComputeCenters(IvfflatBuildState * buildstate)
 
 	/* Sample rows */
 	/* TODO Ensure within maintenance_work_mem */
-	buildstate->samples = VectorArrayInit(numSamples, buildstate->dimensions);
+	buildstate->samples = NIL;
+	buildstate->targsamples = numSamples;
 	if (buildstate->heap != NULL)
 	{
 		SampleRows(buildstate);
 
-		if (buildstate->samples->length < buildstate->lists)
+		if (list_length(buildstate->samples) < buildstate->lists)
 		{
 			ereport(NOTICE,
 					(errmsg("ivfflat index created with little data"),
@@ -467,7 +474,7 @@ ComputeCenters(IvfflatBuildState * buildstate)
 	IvfflatBench("k-means", IvfflatKmeans(buildstate->index, buildstate->samples, buildstate->centers));
 
 	/* Free samples before we allocate more memory */
-	VectorArrayFree(buildstate->samples);
+	list_free_deep(buildstate->samples);
 }
 
 /*

src/ivfflat.h

@@ -80,6 +80,10 @@
 #define RandomInt() random()
 #endif
 
+#if PG_VERSION_NUM < 130000
+#define list_sort(list, cmp) list_qsort(list, cmp)
+#endif
+
 /* Variables */
 extern int	ivfflat_probes;
 
@@ -178,7 +182,8 @@ typedef struct IvfflatBuildState
 	Oid			collation;
 
 	/* Variables */
-	VectorArray samples;
+	List	   *samples;
+	int			targsamples;
 	VectorArray centers;
 	ListInfo   *listInfo;
 	Vector	   *normvec;
@@ -274,7 +279,7 @@ typedef IvfflatScanOpaqueData * IvfflatScanOpaque;
 VectorArray VectorArrayInit(int maxlen, int dimensions);
 void		VectorArrayFree(VectorArray arr);
 void		PrintVectorArray(char *msg, VectorArray arr);
-void		IvfflatKmeans(Relation index, VectorArray samples, VectorArray centers);
+void		IvfflatKmeans(Relation index, List *samples, VectorArray centers);
 FmgrInfo   *IvfflatOptionalProcInfo(Relation index, uint16 procnum);
 bool		IvfflatNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
 int			IvfflatGetLists(Relation index);

src/ivfkmeans.c

@@ -12,20 +12,20 @@
  * https://theory.stanford.edu/~sergei/papers/kMeansPP-soda.pdf
  */
 static void
-InitCenters(Relation index, VectorArray samples, VectorArray centers, float *lowerBound)
+InitCenters(Relation index, List *samples, VectorArray centers, float *lowerBound)
 {
 	FmgrInfo   *procinfo;
 	Oid			collation;
 	int64		j;
-	float	   *weight = palloc(samples->length * sizeof(float));
+	float	   *weight = palloc(list_length(samples) * sizeof(float));
 	int			numCenters = centers->maxlen;
-	int			numSamples = samples->length;
+	int			numSamples = list_length(samples);
 
 	procinfo = index_getprocinfo(index, 1, IVFFLAT_KMEANS_DISTANCE_PROC);
 	collation = index->rd_indcollation[0];
 
 	/* Choose an initial center uniformly at random */
-	VectorArraySet(centers, 0, VectorArrayGet(samples, RandomInt() % samples->length));
+	VectorArraySet(centers, 0, list_nth(samples, RandomInt() % list_length(samples)));
 	centers->length++;
 
 	for (j = 0; j < numSamples; j++)
@@ -42,7 +42,7 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
 
 		for (j = 0; j < numSamples; j++)
 		{
-			Vector	   *vec = VectorArrayGet(samples, j);
+			Vector	   *vec = list_nth(samples, j);
 			double		distance;
 
 			/* Only need to compute distance for new center */
@@ -74,7 +74,7 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
 				break;
 		}
 
-		VectorArraySet(centers, i + 1, VectorArrayGet(samples, j));
+		VectorArraySet(centers, i + 1, list_nth(samples, j));
 		centers->length++;
 	}
 
@@ -106,25 +106,41 @@ CompareVectors(const void *a, const void *b)
 	return vector_cmp_internal((Vector *) a, (Vector *) b);
 }
 
+/*
+ * Compare list vectors
+ */
+static int
+#if PG_VERSION_NUM >= 130000
+CompareListVectors(const ListCell *a, const ListCell *b)
+#else
+CompareListVectors(const void *a, const void *b)
+#endif
+{
+	Vector	   *va = lfirst((ListCell *) a);
+	Vector	   *vb = lfirst((ListCell *) b);
+
+	return CompareVectors(va, vb);
+}
+
 /*
  * Quick approach if we have little data
  */
 static void
-QuickCenters(Relation index, VectorArray samples, VectorArray centers)
+QuickCenters(Relation index, List *samples, VectorArray centers)
 {
 	int			dimensions = centers->dim;
 	Oid			collation = index->rd_indcollation[0];
 	FmgrInfo   *normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_KMEANS_NORM_PROC);
 
 	/* Copy existing vectors while avoiding duplicates */
-	if (samples->length > 0)
+	if (list_length(samples) > 0)
 	{
-		qsort(samples->items, samples->length, VECTOR_SIZE(samples->dim), CompareVectors);
-		for (int i = 0; i < samples->length; i++)
+		list_sort(samples, CompareListVectors);
+		for (int i = 0; i < list_length(samples); i++)
 		{
-			Vector	   *vec = VectorArrayGet(samples, i);
+			Vector	   *vec = list_nth(samples, i);
 
-			if (i == 0 || CompareVectors(vec, VectorArrayGet(samples, i - 1)) != 0)
+			if (i == 0 || CompareVectors(vec, list_nth(samples, i - 1)) != 0)
 			{
 				VectorArraySet(centers, centers->length, vec);
 				centers->length++;
@@ -160,7 +176,7 @@ QuickCenters(Relation index, VectorArray samples, VectorArray centers)
  * https://www.aaai.org/Papers/ICML/2003/ICML03-022.pdf
  */
 static void
-ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
+ElkanKmeans(Relation index, List *samples, VectorArray centers)
 {
 	FmgrInfo   *procinfo;
 	FmgrInfo   *normprocinfo;
@@ -171,7 +187,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 	int64		k;
 	int			dimensions = centers->dim;
 	int			numCenters = centers->maxlen;
-	int			numSamples = samples->length;
+	int			numSamples = list_length(samples);
 	VectorArray newCenters;
 	int		   *centerCounts;
 	int		   *closestCenters;
@@ -182,7 +198,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 	float	   *newcdist;
 
 	/* Calculate allocation sizes */
-	Size		samplesSize = VECTOR_ARRAY_SIZE(samples->maxlen, samples->dim);
+	Size		samplesSize = 0;
 	Size		centersSize = VECTOR_ARRAY_SIZE(centers->maxlen, centers->dim);
 	Size		newCentersSize = VECTOR_ARRAY_SIZE(numCenters, dimensions);
 	Size		centerCountsSize = sizeof(int) * numCenters;
@@ -326,7 +342,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 				if (upperBound[j] <= halfcdist[closestCenters[j] * numCenters + k])
 					continue;
 
-				vec = VectorArrayGet(samples, j);
+				vec = list_nth(samples, j);
 
 				/* Step 3a */
 				if (rj)
@@ -377,7 +393,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 		{
 			int			closestCenter;
 
-			vec = VectorArrayGet(samples, j);
+			vec = list_nth(samples, j);
 			closestCenter = closestCenters[j];
 
 			/* Increment sum and count of closest center */
@@ -514,9 +530,9 @@ CheckCenters(Relation index, VectorArray centers)
  * We use spherical k-means for inner product and cosine
  */
 void
-IvfflatKmeans(Relation index, VectorArray samples, VectorArray centers)
+IvfflatKmeans(Relation index, List *samples, VectorArray centers)
 {
-	if (samples->length <= centers->maxlen)
+	if (list_length(samples) <= centers->maxlen)
 		QuickCenters(index, samples, centers);
 	else
 		ElkanKmeans(index, samples, centers);

test/t/019_hnsw_array.pl

@@ -1,113 +0,0 @@
-use strict;
-use warnings;
-use PostgresNode;
-use TestLib;
-use Test::More;
-
-my $node;
-my @queries = ();
-my @expected;
-my $limit = 20;
-
-sub test_recall
-{
-	my ($min, $operator) = @_;
-	my $correct = 0;
-	my $total = 0;
-
-	my $explain = $node->safe_psql("postgres", qq(
-		SET enable_seqscan = off;
-		EXPLAIN ANALYZE SELECT i FROM tst ORDER BY v $operator '$queries[0]' LIMIT $limit;
-	));
-	like($explain, qr/Index Scan/);
-
-	for my $i (0 .. $#queries)
-	{
-		my $actual = $node->safe_psql("postgres", qq(
-			SET enable_seqscan = off;
-			SELECT i FROM tst ORDER BY v $operator '$queries[$i]' LIMIT $limit;
-		));
-		my @actual_ids = split("\n", $actual);
-		my %actual_set = map { $_ => 1 } @actual_ids;
-
-		my @expected_ids = split("\n", $expected[$i]);
-
-		foreach (@expected_ids)
-		{
-			if (exists($actual_set{$_}))
-			{
-				$correct++;
-			}
-			$total++;
-		}
-	}
-
-	cmp_ok($correct / $total, ">=", $min, $operator);
-}
-
-# Initialize node
-$node = get_new_node('node');
-$node->init;
-$node->start;
-
-# Create table
-$node->safe_psql("postgres", "CREATE EXTENSION vector;");
-$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v float4[3]);");
-$node->safe_psql("postgres",
-	"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 10000) i;"
-);
-
-$node->safe_psql("postgres", qq(
-	CREATE FUNCTION float4_l2_distance(float4[], float4[]) RETURNS float8
-		AS 'BEGIN RETURN l2_distance(\$1::vector, \$2::vector); END;'
-		LANGUAGE plpgsql IMMUTABLE STRICT PARALLEL SAFE;
-
-	CREATE FUNCTION float4_l2_squared_distance(float4[], float4[]) RETURNS float8
-		AS 'BEGIN RETURN vector_l2_squared_distance(\$1::vector, \$2::vector); END;'
-		LANGUAGE plpgsql IMMUTABLE STRICT PARALLEL SAFE;
-
-	CREATE OPERATOR <-> (
-		LEFTARG = float4[], RIGHTARG = float4[], PROCEDURE = float4_l2_distance,
-		COMMUTATOR = '<->'
-	);
-
-	CREATE OPERATOR CLASS float4_l2_ops
-		FOR TYPE float4[] USING hnsw AS
-		OPERATOR 1 <-> (float4[], float4[]) FOR ORDER BY float_ops,
-		FUNCTION 1 float4_l2_squared_distance(float4[], float4[]);
-));
-
-# Generate queries
-for (1 .. 20)
-{
-	my $r1 = rand();
-	my $r2 = rand();
-	my $r3 = rand();
-	push(@queries, "{$r1,$r2,$r3}");
-}
-
-# Check each index type
-my @operators = ("<->");
-my @opclasses = ("float4_l2_ops");
-
-for my $i (0 .. $#operators)
-{
-	my $operator = $operators[$i];
-	my $opclass = $opclasses[$i];
-
-	# Get exact results
-	@expected = ();
-	foreach (@queries)
-	{
-		my $res = $node->safe_psql("postgres", "SELECT i FROM tst ORDER BY v $operator '$_' LIMIT $limit;");
-		push(@expected, $res);
-	}
-
-	# Add index
-	$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v $opclass) WITH (dimensions = 3);");
-
-	my $min = $operator eq "<#>" ? 0.80 : 0.99;
-	test_recall($min, $operator);
-}
-
-done_testing();