Switched to TidStore for HNSW vacuuming for Postgres 17+ [skip ci]

Co-authored-by: Itai Spiegel <itai@mave.com>

CHANGELOG.md

@@ -1,6 +1,12 @@
-## 0.8.4 (unreleased)
+## 0.8.5 (unreleased)
 
+- Reduced memory usage for small tables for IVFFlat index builds
+
+## 0.8.4 (2026-06-30)
+
+- Fixed `hnsw graph not repaired` error with HNSW vacuuming
 - Fixed possible error with inserts during HNSW vacuuming
+- Fixed memory exceeding `maintenance_work_mem` with IVFFlat index builds
 
 ## 0.8.3 (2026-06-17)
 

Dockerfile

@@ -5,7 +5,7 @@ ARG DEBIAN_CODENAME=bookworm
 FROM postgres:$PG_MAJOR-$DEBIAN_CODENAME
 ARG PG_MAJOR
 
-ADD https://github.com/pgvector/pgvector.git#v0.8.3 /tmp/pgvector
+ADD https://github.com/pgvector/pgvector.git#v0.8.4 /tmp/pgvector
 
 RUN apt-get update && \
 		apt-mark hold locales && \

META.json

@@ -2,7 +2,7 @@
 	"name": "vector",
 	"abstract": "Open-source vector similarity search for Postgres",
 	"description": "Supports L2 distance, inner product, and cosine distance",
-	"version": "0.8.3",
+	"version": "0.8.4",
 	"maintainer": [
 		"Andrew Kane <andrew@ankane.org>"
 	],
@@ -20,7 +20,7 @@
 		"vector": {
 			"file": "sql/vector.sql",
 			"docfile": "README.md",
-			"version": "0.8.3",
+			"version": "0.8.4",
 			"abstract": "Open-source vector similarity search for Postgres"
 		}
 	},

Makefile

@@ -1,5 +1,5 @@
 EXTENSION = vector
-EXTVERSION = 0.8.3
+EXTVERSION = 0.8.4
 
 MODULE_big = vector
 DATA = $(wildcard sql/*--*--*.sql)

Makefile.win

@@ -1,5 +1,5 @@
 EXTENSION = vector
-EXTVERSION = 0.8.3
+EXTVERSION = 0.8.4
 
 DATA_built = sql\$(EXTENSION)--$(EXTVERSION).sql
 OBJS = src\bitutils.obj src\bitvec.obj src\halfutils.obj src\halfvec.obj src\hnsw.obj src\hnswbuild.obj src\hnswinsert.obj src\hnswscan.obj src\hnswutils.obj src\hnswvacuum.obj src\ivfbuild.obj src\ivfflat.obj src\ivfinsert.obj src\ivfkmeans.obj src\ivfscan.obj src\ivfutils.obj src\ivfvacuum.obj src\sparsevec.obj src\vector.obj

README.md

@@ -23,7 +23,7 @@ Compile and install the extension (supports Postgres 13+)
 
 ```sh
 cd /tmp
-git clone --branch v0.8.3 https://github.com/pgvector/pgvector.git
+git clone --branch v0.8.4 https://github.com/pgvector/pgvector.git
 cd pgvector
 make
 make install # may need sudo
@@ -40,7 +40,7 @@ Ensure [C++ support in Visual Studio](https://learn.microsoft.com/en-us/cpp/buil
 ```cmd
 set "PGROOT=C:\Program Files\PostgreSQL\18"
 cd %TEMP%
-git clone --branch v0.8.3 https://github.com/pgvector/pgvector.git
+git clone --branch v0.8.4 https://github.com/pgvector/pgvector.git
 cd pgvector
 nmake /F Makefile.win
 nmake /F Makefile.win install
@@ -1151,23 +1151,23 @@ This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (r
 
 Supported tags are:
 
-- `pg18-trixie`, `0.8.3-pg18-trixie`
-- `pg18-bookworm`, `0.8.3-pg18-bookworm`, `pg18`, `0.8.3-pg18`
-- `pg17-trixie`, `0.8.3-pg17-trixie`
-- `pg17-bookworm`, `0.8.3-pg17-bookworm`, `pg17`, `0.8.3-pg17`
-- `pg16-trixie`, `0.8.3-pg16-trixie`
-- `pg16-bookworm`, `0.8.3-pg16-bookworm`, `pg16`, `0.8.3-pg16`
-- `pg15-trixie`, `0.8.3-pg15-trixie`
-- `pg15-bookworm`, `0.8.3-pg15-bookworm`, `pg15`, `0.8.3-pg15`
-- `pg14-trixie`, `0.8.3-pg14-trixie`
-- `pg14-bookworm`, `0.8.3-pg14-bookworm`, `pg14`, `0.8.3-pg14`
-- `pg13-trixie`, `0.8.3-pg13-trixie`
-- `pg13-bookworm`, `0.8.3-pg13-bookworm`, `pg13`, `0.8.3-pg13`
+- `pg18-trixie`, `0.8.4-pg18-trixie`
+- `pg18-bookworm`, `0.8.4-pg18-bookworm`, `pg18`, `0.8.4-pg18`
+- `pg17-trixie`, `0.8.4-pg17-trixie`
+- `pg17-bookworm`, `0.8.4-pg17-bookworm`, `pg17`, `0.8.4-pg17`
+- `pg16-trixie`, `0.8.4-pg16-trixie`
+- `pg16-bookworm`, `0.8.4-pg16-bookworm`, `pg16`, `0.8.4-pg16`
+- `pg15-trixie`, `0.8.4-pg15-trixie`
+- `pg15-bookworm`, `0.8.4-pg15-bookworm`, `pg15`, `0.8.4-pg15`
+- `pg14-trixie`, `0.8.4-pg14-trixie`
+- `pg14-bookworm`, `0.8.4-pg14-bookworm`, `pg14`, `0.8.4-pg14`
+- `pg13-trixie`, `0.8.4-pg13-trixie`
+- `pg13-bookworm`, `0.8.4-pg13-bookworm`, `pg13`, `0.8.4-pg13`
 
 You can also build the image manually:
 
 ```sh
-git clone --branch v0.8.3 https://github.com/pgvector/pgvector.git
+git clone --branch v0.8.4 https://github.com/pgvector/pgvector.git
 cd pgvector
 docker build --pull --build-arg PG_MAJOR=18 -t myuser/pgvector .
 ```

sql/vector--0.8.3--0.8.4.sql

@@ -0,0 +1,2 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.8.4'" to load this file. \quit

src/hnsw.c

@@ -290,7 +290,11 @@ hnswhandler(PG_FUNCTION_ARGS)
 		.amcanparallel = false,
 		.amcanbuildparallel = true,
 		.amcaninclude = false,
+#if PG_VERSION_NUM >= 170000
+		.amusemaintenanceworkmem = true,
+#else
 		.amusemaintenanceworkmem = false,
+#endif
 		.amsummarizing = false,
 		.amparallelvacuumoptions = VACUUM_OPTION_PARALLEL_BULKDEL,
 		.amkeytype = InvalidOid,
@@ -351,7 +355,11 @@ hnswhandler(PG_FUNCTION_ARGS)
 	amroutine->amcanbuildparallel = true;
 #endif
 	amroutine->amcaninclude = false;
-	amroutine->amusemaintenanceworkmem = false; /* not used during VACUUM */
+#if PG_VERSION_NUM >= 170000
+	amroutine->amusemaintenanceworkmem = true;
+#else
+	amroutine->amusemaintenanceworkmem = false;
+#endif
 #if PG_VERSION_NUM >= 160000
 	amroutine->amsummarizing = false;
 #endif

src/hnsw.h

@@ -26,6 +26,12 @@
 typedef Pointer Item;
 #endif
 
+#if PG_VERSION_NUM >= 170000
+#define HnswTidStore TidStore
+#else
+#define HnswTidStore tidhash_hash
+#endif
+
 #define HNSW_MAX_DIM 2000
 #define HNSW_MAX_NNZ 1000
 
@@ -427,7 +433,7 @@ typedef struct HnswVacuumState
 	HnswSupport support;
 
 	/* Variables */
-	struct tidhash_hash *deleted;
+	struct HnswTidStore *deleting;
 	BufferAccessStrategy bas;
 	HnswNeighborTuple ntup;
 	HnswElementData highestPoint;

src/hnswvacuum.c

@@ -10,6 +10,12 @@
 #include "utils/memutils.h"
 #include "utils/rel.h"
 
+#if PG_VERSION_NUM >= 170000
+#include "access/tidstore.h"
+#include "miscadmin.h"
+#include "postmaster/autovacuum.h"
+#endif
+
 #if PG_VERSION_NUM >= 160000
 #include "varatt.h"
 #endif
@@ -19,12 +25,16 @@
 #endif
 
 /*
- * Check if deleted list contains an index TID
+ * Check if deletion list contains an element
  */
 static bool
-DeletedContains(tidhash_hash * deleted, ItemPointer indextid)
+DeletingElement(HnswTidStore * deleting, ItemPointer indextid)
 {
-	return tidhash_lookup(deleted, *indextid) != NULL;
+#if PG_VERSION_NUM >= 170000
+	return TidStoreIsMember(deleting, indextid);
+#else
+	return tidhash_lookup(deleting, *indextid) != NULL;
+#endif
 }
 
 /*
@@ -60,6 +70,10 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
 		OffsetNumber offno;
 		OffsetNumber maxoffno;
 		bool		updated = false;
+#if PG_VERSION_NUM >= 170000
+		OffsetNumber deletedoffs[MaxOffsetNumber];
+		int			ndeletedoffs = 0;
+#endif
 
 		vacuum_delay_point();
 
@@ -80,6 +94,14 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
 			if (!HnswIsElementTuple(etup))
 				continue;
 
+			/*
+			 * Skip deleted tuples. It is important they are not added to the
+			 * deletion list to avoid false positives in NeedsUpdated and
+			 * ConfirmRepaired.
+			 */
+			if (etup->deleted)
+				continue;
+
 			if (ItemPointerIsValid(&etup->heaptids[0]))
 			{
 				for (int i = 0; i < HNSW_HEAPTIDS; i++)
@@ -113,14 +135,18 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
 
 			if (!ItemPointerIsValid(&etup->heaptids[0]))
 			{
-				ItemPointerData ip;
+#if PG_VERSION_NUM >= 170000
+				deletedoffs[ndeletedoffs++] = offno;
+#else
+				ItemPointerData indextid;
 				bool		found;
 
-				/* Add to deleted list */
-				ItemPointerSet(&ip, blkno, offno);
+				/* Add to deletion list */
+				ItemPointerSet(&indextid, blkno, offno);
 
-				tidhash_insert(vacuumstate->deleted, ip, &found);
+				tidhash_insert(vacuumstate->deleting, indextid, &found);
 				Assert(!found);
+#endif
 			}
 			else if (etup->level > highestLevel)
 			{
@@ -149,6 +175,10 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
 			}
 		}
 
+#if PG_VERSION_NUM >= 170000
+		TidStoreSetBlockOffsets(vacuumstate->deleting, blkno, deletedoffs, ndeletedoffs);
+#endif
+
 		blkno = HnswPageGetOpaque(page)->nextblkno;
 
 		if (updated)
@@ -192,8 +222,8 @@ NeedsUpdated(HnswVacuumState * vacuumstate, HnswElement element)
 		if (!ItemPointerIsValid(indextid))
 			continue;
 
-		/* Check if in deleted list */
-		if (DeletedContains(vacuumstate->deleted, indextid))
+		/* Check if in deletion list */
+		if (DeletingElement(vacuumstate->deleting, indextid))
 		{
 			needsUpdated = true;
 			break;
@@ -327,7 +357,7 @@ RepairGraphEntryPoint(HnswVacuumState * vacuumstate)
 
 		ItemPointerSet(&epData, entryPoint->blkno, entryPoint->offno);
 
-		if (DeletedContains(vacuumstate->deleted, &epData))
+		if (DeletingElement(vacuumstate->deleting, &epData))
 		{
 			/*
 			 * Replace the entry point with the highest point. If highest
@@ -413,6 +443,10 @@ RepairGraph(HnswVacuumState * vacuumstate)
 			if (!HnswIsElementTuple(etup))
 				continue;
 
+			/* Skip deleted tuples */
+			if (etup->deleted)
+				continue;
+
 			/* Skip updating neighbors if being deleted */
 			if (!ItemPointerIsValid(&etup->heaptids[0]))
 				continue;
@@ -527,6 +561,10 @@ ConfirmRepaired(HnswVacuumState * vacuumstate)
 			if (!HnswIsElementTuple(etup))
 				continue;
 
+			/* Skip deleted tuples */
+			if (etup->deleted)
+				continue;
+
 			/* Skip if being deleted */
 			if (!ItemPointerIsValid(&etup->heaptids[0]))
 				continue;
@@ -557,8 +595,8 @@ ConfirmRepaired(HnswVacuumState * vacuumstate)
 				if (!ItemPointerIsValid(indextid))
 					continue;
 
-				/* Check if in deleted list */
-				if (DeletedContains(vacuumstate->deleted, indextid))
+				/* Check if in deletion list */
+				if (DeletingElement(vacuumstate->deleting, indextid))
 					elog(ERROR, "hnsw graph not repaired");
 			}
 
@@ -740,7 +778,11 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
 	HnswGetMetaPageInfo(index, &vacuumstate->m, NULL);
 
 	/* Create hash table */
-	vacuumstate->deleted = tidhash_create(CurrentMemoryContext, 256, NULL);
+#if PG_VERSION_NUM >= 170000
+	vacuumstate->deleting = TidStoreCreateLocal((AmAutoVacuumWorkerProcess() && autovacuum_work_mem != -1) ? autovacuum_work_mem : maintenance_work_mem, true);
+#else
+	vacuumstate->deleting = tidhash_create(CurrentMemoryContext, 256, NULL);
+#endif
 }
 
 /*
@@ -749,7 +791,11 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
 static void
 FreeVacuumState(HnswVacuumState * vacuumstate)
 {
-	tidhash_destroy(vacuumstate->deleted);
+#if PG_VERSION_NUM >= 170000
+	TidStoreDestroy(vacuumstate->deleting);
+#else
+	tidhash_destroy(vacuumstate->deleting);
+#endif
 	FreeAccessStrategy(vacuumstate->bas);
 	pfree(vacuumstate->ntup);
 	MemoryContextDelete(vacuumstate->tmpCtx);

src/ivfbuild.c

@@ -438,15 +438,22 @@ ComputeCenters(IvfflatBuildState * buildstate)
 
 	pgstat_progress_update_param(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_KMEANS);
 
-	/* Target 50 samples per list, with at least 10000 samples */
-	/* The number of samples has a large effect on index build time */
-	numSamples = buildstate->lists * 50;
-	if (numSamples < 10000)
-		numSamples = 10000;
-
 	/* Skip samples for unlogged table */
 	if (buildstate->heap == NULL)
 		numSamples = 1;
+	else
+	{
+		int64		maxTuples = (int64) RelationGetNumberOfBlocks(buildstate->heap) * MaxHeapTuplesPerPage;
+
+		/* Target 50 samples per list, with at least 10000 samples */
+		/* The number of samples has a large effect on index build time */
+		numSamples = buildstate->lists * 50;
+		if (numSamples < 10000)
+			numSamples = 10000;
+
+		/* Save memory since will not have more than max tuples */
+		numSamples = Max(Min(numSamples, maxTuples), 1);
+	}
 
 	/* Sample rows */
 	buildstate->memoryUsed += VECTOR_ARRAY_SIZE(numSamples, buildstate->itemsize);

src/vector.c

@@ -40,7 +40,7 @@
 #endif
 
 #if PG_VERSION_NUM >= 180000
-PG_MODULE_MAGIC_EXT(.name = "vector", .version = "0.8.3");
+PG_MODULE_MAGIC_EXT(.name = "vector", .version = "0.8.4");
 #else
 PG_MODULE_MAGIC;
 #endif

test/expected/hnsw_bit.out

@@ -49,3 +49,11 @@ CREATE INDEX ON t USING hnsw ((val::bit(3)) bit_hamming_ops);
 CREATE INDEX ON t USING hnsw ((val::bit(64001)) bit_hamming_ops);
 ERROR:  column cannot have more than 64000 dimensions for hnsw index
 DROP TABLE t;
+-- dimensions
+CREATE TABLE t (val bit(64000));
+CREATE INDEX ON t USING hnsw (val bit_hamming_ops);
+DROP TABLE t;
+CREATE TABLE t (val bit(64001));
+CREATE INDEX ON t USING hnsw (val bit_hamming_ops);
+ERROR:  column cannot have more than 64000 dimensions for hnsw index
+DROP TABLE t;

test/expected/hnsw_halfvec.out

@@ -100,3 +100,11 @@ SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <+> (SELECT NULL::halfvec)) t
 (1 row)
 
 DROP TABLE t;
+-- dimensions
+CREATE TABLE t (val halfvec(4000));
+CREATE INDEX ON t USING hnsw (val halfvec_l2_ops);
+DROP TABLE t;
+CREATE TABLE t (val halfvec(4001));
+CREATE INDEX ON t USING hnsw (val halfvec_l2_ops);
+ERROR:  column cannot have more than 4000 dimensions for hnsw index
+DROP TABLE t;

test/expected/hnsw_vector.out

@@ -161,6 +161,7 @@ ERROR:  value 1001 out of bounds for option "ef_construction"
 DETAIL:  Valid values are between "4" and "1000".
 CREATE INDEX ON t USING hnsw (val vector_l2_ops) WITH (m = 16, ef_construction = 31);
 ERROR:  ef_construction must be greater than or equal to 2 * m
+DROP TABLE t;
 SHOW hnsw.ef_search;
  hnsw.ef_search 
 ----------------
@@ -198,4 +199,11 @@ SET hnsw.scan_mem_multiplier = 0;
 ERROR:  0 is outside the valid range for parameter "hnsw.scan_mem_multiplier" (1 .. 1000)
 SET hnsw.scan_mem_multiplier = 1001;
 ERROR:  1001 is outside the valid range for parameter "hnsw.scan_mem_multiplier" (1 .. 1000)
+-- dimensions
+CREATE TABLE t (val vector(2000));
+CREATE INDEX ON t USING hnsw (val vector_l2_ops);
+DROP TABLE t;
+CREATE TABLE t (val vector(2001));
+CREATE INDEX ON t USING hnsw (val vector_l2_ops);
+ERROR:  column cannot have more than 2000 dimensions for hnsw index
 DROP TABLE t;

test/expected/ivfflat_bit.out

@@ -35,3 +35,32 @@ NOTICE:  ivfflat index created with little data
 DETAIL:  This will cause low recall.
 HINT:  Drop the index until the table has more data.
 DROP TABLE t;
+-- dimensions
+CREATE TABLE t (val bit(64000));
+CREATE INDEX ON t USING ivfflat (val bit_hamming_ops);
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
+DROP TABLE t;
+CREATE TABLE t (val bit(64001));
+CREATE INDEX ON t USING ivfflat (val bit_hamming_ops);
+ERROR:  column cannot have more than 64000 dimensions for ivfflat index
+DROP TABLE t;
+-- memory
+SET maintenance_work_mem = '1MB';
+CREATE TABLE t (val bit(64000));
+CREATE INDEX ON t USING ivfflat (val bit_hamming_ops);
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
+DROP TABLE t;
+RESET maintenance_work_mem;
+SET maintenance_work_mem = '29MB';
+CREATE TABLE t (val bit(64000));
+INSERT INTO t (val) VALUES (B'0'::bit(64000));
+CREATE INDEX ON t USING ivfflat (val bit_hamming_ops);
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
+DROP TABLE t;
+RESET maintenance_work_mem;

test/expected/ivfflat_halfvec.out

@@ -82,3 +82,32 @@ SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> (SELECT NULL::halfvec)) t
 (1 row)
 
 DROP TABLE t;
+-- dimensions
+CREATE TABLE t (val halfvec(4000));
+CREATE INDEX ON t USING ivfflat (val halfvec_l2_ops);
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
+DROP TABLE t;
+CREATE TABLE t (val halfvec(4001));
+CREATE INDEX ON t USING ivfflat (val halfvec_l2_ops);
+ERROR:  column cannot have more than 4000 dimensions for ivfflat index
+DROP TABLE t;
+-- memory
+SET maintenance_work_mem = '1MB';
+CREATE TABLE t (val halfvec(4000));
+CREATE INDEX ON t USING ivfflat (val halfvec_l2_ops);
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
+DROP TABLE t;
+RESET maintenance_work_mem;
+SET maintenance_work_mem = '6MB';
+CREATE TABLE t (val halfvec(4000));
+INSERT INTO t (val) VALUES (array_fill(0, ARRAY[4000]));
+CREATE INDEX ON t USING ivfflat (val halfvec_l2_ops);
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
+DROP TABLE t;
+RESET maintenance_work_mem;

test/expected/ivfflat_vector.out

@@ -143,6 +143,7 @@ DETAIL:  Valid values are between "1" and "32768".
 CREATE INDEX ON t USING ivfflat (val vector_l2_ops) WITH (lists = 32769);
 ERROR:  value 32769 out of bounds for option "lists"
 DETAIL:  Valid values are between "1" and "32768".
+DROP TABLE t;
 SHOW ivfflat.probes;
  ivfflat.probes 
 ----------------
@@ -172,4 +173,32 @@ SET ivfflat.max_probes = 0;
 ERROR:  0 is outside the valid range for parameter "ivfflat.max_probes" (1 .. 32768)
 SET ivfflat.max_probes = 32769;
 ERROR:  32769 is outside the valid range for parameter "ivfflat.max_probes" (1 .. 32768)
+-- dimensions
+CREATE TABLE t (val vector(2000));
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops);
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
 DROP TABLE t;
+CREATE TABLE t (val vector(2001));
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops);
+ERROR:  column cannot have more than 2000 dimensions for ivfflat index
+DROP TABLE t;
+-- memory
+SET maintenance_work_mem = '1MB';
+CREATE TABLE t (val vector(2000));
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops);
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
+DROP TABLE t;
+RESET maintenance_work_mem;
+SET maintenance_work_mem = '5MB';
+CREATE TABLE t (val vector(2000));
+INSERT INTO t (val) VALUES (array_fill(0, ARRAY[2000]));
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops);
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
+DROP TABLE t;
+RESET maintenance_work_mem;

test/sql/hnsw_bit.sql

@@ -33,3 +33,13 @@ CREATE INDEX ON t USING hnsw (val bit_hamming_ops);
 CREATE INDEX ON t USING hnsw ((val::bit(3)) bit_hamming_ops);
 CREATE INDEX ON t USING hnsw ((val::bit(64001)) bit_hamming_ops);
 DROP TABLE t;
+
+-- dimensions
+
+CREATE TABLE t (val bit(64000));
+CREATE INDEX ON t USING hnsw (val bit_hamming_ops);
+DROP TABLE t;
+
+CREATE TABLE t (val bit(64001));
+CREATE INDEX ON t USING hnsw (val bit_hamming_ops);
+DROP TABLE t;

test/sql/hnsw_halfvec.sql

@@ -56,3 +56,13 @@ SELECT * FROM t ORDER BY val <+> '[3,3,3]';
 SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <+> (SELECT NULL::halfvec)) t2;
 
 DROP TABLE t;
+
+-- dimensions
+
+CREATE TABLE t (val halfvec(4000));
+CREATE INDEX ON t USING hnsw (val halfvec_l2_ops);
+DROP TABLE t;
+
+CREATE TABLE t (val halfvec(4001));
+CREATE INDEX ON t USING hnsw (val halfvec_l2_ops);
+DROP TABLE t;

test/sql/hnsw_vector.sql

@@ -95,23 +95,28 @@ CREATE INDEX ON t USING hnsw (val vector_l2_ops) WITH (m = 101);
 CREATE INDEX ON t USING hnsw (val vector_l2_ops) WITH (ef_construction = 3);
 CREATE INDEX ON t USING hnsw (val vector_l2_ops) WITH (ef_construction = 1001);
 CREATE INDEX ON t USING hnsw (val vector_l2_ops) WITH (m = 16, ef_construction = 31);
+DROP TABLE t;
 
 SHOW hnsw.ef_search;
-
 SET hnsw.ef_search = 0;
 SET hnsw.ef_search = 1001;
 
 SHOW hnsw.iterative_scan;
-
 SET hnsw.iterative_scan = on;
 
 SHOW hnsw.max_scan_tuples;
-
 SET hnsw.max_scan_tuples = 0;
 
 SHOW hnsw.scan_mem_multiplier;
-
 SET hnsw.scan_mem_multiplier = 0;
 SET hnsw.scan_mem_multiplier = 1001;
 
+-- dimensions
+
+CREATE TABLE t (val vector(2000));
+CREATE INDEX ON t USING hnsw (val vector_l2_ops);
+DROP TABLE t;
+
+CREATE TABLE t (val vector(2001));
+CREATE INDEX ON t USING hnsw (val vector_l2_ops);
 DROP TABLE t;

test/sql/ivfflat_bit.sql

@@ -21,3 +21,28 @@ CREATE INDEX ON t USING ivfflat ((val::bit(3)) bit_hamming_ops) WITH (lists = 1)
 CREATE INDEX ON t USING ivfflat ((val::bit(64001)) bit_hamming_ops) WITH (lists = 1);
 CREATE INDEX ON t USING ivfflat ((val::bit(2)) bit_hamming_ops) WITH (lists = 5);
 DROP TABLE t;
+
+-- dimensions
+
+CREATE TABLE t (val bit(64000));
+CREATE INDEX ON t USING ivfflat (val bit_hamming_ops);
+DROP TABLE t;
+
+CREATE TABLE t (val bit(64001));
+CREATE INDEX ON t USING ivfflat (val bit_hamming_ops);
+DROP TABLE t;
+
+-- memory
+
+SET maintenance_work_mem = '1MB';
+CREATE TABLE t (val bit(64000));
+CREATE INDEX ON t USING ivfflat (val bit_hamming_ops);
+DROP TABLE t;
+RESET maintenance_work_mem;
+
+SET maintenance_work_mem = '29MB';
+CREATE TABLE t (val bit(64000));
+INSERT INTO t (val) VALUES (B'0'::bit(64000));
+CREATE INDEX ON t USING ivfflat (val bit_hamming_ops);
+DROP TABLE t;
+RESET maintenance_work_mem;

test/sql/ivfflat_halfvec.sql

@@ -43,3 +43,28 @@ SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> '[0,0,0]') t2;
 SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> (SELECT NULL::halfvec)) t2;
 
 DROP TABLE t;
+
+-- dimensions
+
+CREATE TABLE t (val halfvec(4000));
+CREATE INDEX ON t USING ivfflat (val halfvec_l2_ops);
+DROP TABLE t;
+
+CREATE TABLE t (val halfvec(4001));
+CREATE INDEX ON t USING ivfflat (val halfvec_l2_ops);
+DROP TABLE t;
+
+-- memory
+
+SET maintenance_work_mem = '1MB';
+CREATE TABLE t (val halfvec(4000));
+CREATE INDEX ON t USING ivfflat (val halfvec_l2_ops);
+DROP TABLE t;
+RESET maintenance_work_mem;
+
+SET maintenance_work_mem = '6MB';
+CREATE TABLE t (val halfvec(4000));
+INSERT INTO t (val) VALUES (array_fill(0, ARRAY[4000]));
+CREATE INDEX ON t USING ivfflat (val halfvec_l2_ops);
+DROP TABLE t;
+RESET maintenance_work_mem;

test/sql/ivfflat_vector.sql

@@ -81,19 +81,40 @@ DROP TABLE t;
 CREATE TABLE t (val vector(3));
 CREATE INDEX ON t USING ivfflat (val vector_l2_ops) WITH (lists = 0);
 CREATE INDEX ON t USING ivfflat (val vector_l2_ops) WITH (lists = 32769);
+DROP TABLE t;
 
 SHOW ivfflat.probes;
-
 SET ivfflat.probes = 0;
 SET ivfflat.probes = 32769;
 
 SHOW ivfflat.iterative_scan;
-
 SET ivfflat.iterative_scan = on;
 
 SHOW ivfflat.max_probes;
-
 SET ivfflat.max_probes = 0;
 SET ivfflat.max_probes = 32769;
 
+-- dimensions
+
+CREATE TABLE t (val vector(2000));
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops);
 DROP TABLE t;
+
+CREATE TABLE t (val vector(2001));
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops);
+DROP TABLE t;
+
+-- memory
+
+SET maintenance_work_mem = '1MB';
+CREATE TABLE t (val vector(2000));
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops);
+DROP TABLE t;
+RESET maintenance_work_mem;
+
+SET maintenance_work_mem = '5MB';
+CREATE TABLE t (val vector(2000));
+INSERT INTO t (val) VALUES (array_fill(0, ARRAY[2000]));
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops);
+DROP TABLE t;
+RESET maintenance_work_mem;

test/t/046_hnsw_vacuum_scan.pl

@@ -0,0 +1,38 @@
+use strict;
+use warnings FATAL => 'all';
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+use Test::More;
+
+my $dim = 3;
+my $array_sql = join(",", ('random()') x $dim);
+
+# Initialize node
+my $node = PostgreSQL::Test::Cluster->new('node');
+$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));");
+$node->safe_psql("postgres", "ALTER TABLE tst SET (autovacuum_enabled = false);");
+$node->safe_psql("postgres",
+    "INSERT INTO tst (v) SELECT ARRAY[$array_sql] FROM generate_series(1, 1000) i;"
+);
+$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v vector_l2_ops);");
+$node->safe_psql("postgres", "DELETE FROM tst");
+
+# Test HNSW_SCAN_LOCK at the beginning of MarkDeleted is effective
+$node->pgbench(
+    "--no-vacuum --client=5 --transactions=1000",
+    0,
+    [qr{actually processed}],
+    [qr{^$}],
+    "concurrent SELECTs and VACUUM",
+    {
+        "046_hnsw_vacuum_scan_select\@1000" => "SELECT i FROM tst ORDER BY v <-> '[0,0,0]' LIMIT 10;",
+        "046_hnsw_vacuum_scan_vacuum\@1" => "VACUUM tst;"
+    }
+);
+
+done_testing();

test/t/047_hnsw_vacuum_insert.pl

@@ -0,0 +1,39 @@
+use strict;
+use warnings FATAL => 'all';
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+use Test::More;
+
+my $dim = 3;
+my $array_sql = join(",", ('random()') x $dim);
+
+# Initialize node
+my $node = PostgreSQL::Test::Cluster->new('node');
+$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));");
+$node->safe_psql("postgres", "ALTER TABLE tst SET (autovacuum_enabled = false);");
+$node->safe_psql("postgres",
+    "INSERT INTO tst (v) SELECT ARRAY[$array_sql] FROM generate_series(1, 1000) i;"
+);
+$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v vector_l2_ops);");
+
+# Test no "hnsw graph not repaired" errors
+$node->pgbench(
+    "--no-vacuum --client=5 --transactions=1000",
+    0,
+    [qr{actually processed}],
+    [qr{^$}],
+    "concurrent INSERTs, DELETEs, SELECTs, and VACUUM",
+    {
+        "047_hnsw_vacuum_insert_insert\@500" => "INSERT INTO tst (v) VALUES (ARRAY[$array_sql]);",
+        "047_hnsw_vacuum_insert_delete\@500" => "DELETE FROM tst WHERE i = (SELECT i FROM tst LIMIT 1);",
+        "047_hnsw_vacuum_insert_select\@20" => "SELECT i FROM tst ORDER BY v <-> '[0,0,0]' LIMIT 10;",
+        "047_hnsw_vacuum_insert_vacuum\@1" => "VACUUM tst;"
+    }
+);
+
+done_testing();

vector.control

@@ -1,4 +1,4 @@
 comment = 'vector data type and ivfflat and hnsw access methods'
-default_version = '0.8.3'
+default_version = '0.8.4'
 module_pathname = '$libdir/vector'
 relocatable = true