Fixed Windows build [skip ci]

As the data is aligned, for hnsw, the combined size won't be in the range
(8156 (maxSize exlucding `ItemIdData`), 8160]. So even if the
ItemIdData is not included in the maxSize, it works well now, but I
think it's better to make it correct.

CHANGELOG.md

@@ -1,7 +1,7 @@
 ## 0.5.1 (unreleased)
 
+- Added check for MVCC-compliant snapshot for HNSW index scans
 - Improved performance of index scans for IVFFlat after updates and deletes
-- Fixed locking for index scans for HNSW
 
 ## 0.5.0 (2023-08-28)
 

Makefile

@@ -3,8 +3,8 @@ EXTVERSION = 0.5.0
 
 MODULE_big = vector
 DATA = $(wildcard sql/*--*.sql)
-OBJS = src/hnsw.o src/hnswbuild.o src/hnswinsert.o src/hnswscan.o src/hnswutils.o src/hnswvacuum.o src/ivfbuild.o src/ivfflat.o src/ivfinsert.o src/ivfkmeans.o src/ivfscan.o src/ivfutils.o src/ivfvacuum.o src/vector.o
-HEADERS = src/vector.h
+OBJS = src/hnsw.o src/hnswbuild.o src/hnswinsert.o src/hnswscan.o src/hnswutils.o src/hnswvacuum.o src/ivfbuild.o src/ivfflat.o src/ivfinsert.o src/ivfkmeans.o src/ivfscan.o src/ivfutils.o src/ivfvacuum.o src/tinyint.o src/vector.o
+HEADERS = src/tinyint.h src/vector.h
 
 TESTS = $(wildcard test/sql/*.sql)
 REGRESS = $(patsubst test/sql/%.sql,%,$(TESTS))

Makefile.win

@@ -1,8 +1,8 @@
 EXTENSION = vector
 EXTVERSION = 0.5.0
 
-OBJS = 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\vector.obj
-HEADERS = src\vector.h
+OBJS = 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\tinyint.obj src\vector.obj
+HEADERS = src\tinyint.h 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=$(EXTENSION)
@@ -56,7 +56,7 @@ install:
 	copy $(EXTENSION).control "$(SHAREDIR)\extension"
 	copy sql\$(EXTENSION)--*.sql "$(SHAREDIR)\extension"
 	mkdir "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
-	copy $(HEADERS) "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
+	for %f in ($(HEADERS)) do copy %f "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
 
 installcheck:
 	"$(BINDIR)\pg_regress" --bindir="$(BINDIR)" $(REGRESS_OPTS) $(REGRESS)

sql/vector--0.5.0--0.5.1.sql

@@ -0,0 +1,67 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.5.1'" to load this file. \quit
+
+-- tinyint
+
+CREATE TYPE tinyint;
+
+CREATE FUNCTION tinyint_in(cstring, oid, integer) RETURNS tinyint
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION tinyint_out(tinyint) RETURNS cstring
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION tinyint_recv(internal, oid, integer) RETURNS tinyint
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION tinyint_send(tinyint) RETURNS bytea
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE TYPE tinyint (
+	INPUT     = tinyint_in,
+	OUTPUT    = tinyint_out,
+	RECEIVE   = tinyint_recv,
+	SEND      = tinyint_send,
+	INTERNALLENGTH = 1,
+	PASSEDBYVALUE,
+	ALIGNMENT = char
+);
+
+CREATE FUNCTION integer_to_tinyint(integer, integer, boolean) RETURNS tinyint
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION numeric_to_tinyint(numeric, integer, boolean) RETURNS tinyint
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE CAST (integer AS tinyint)
+	WITH FUNCTION integer_to_tinyint(integer, integer, boolean) AS IMPLICIT;
+
+CREATE CAST (numeric AS tinyint)
+	WITH FUNCTION numeric_to_tinyint(numeric, integer, boolean) AS IMPLICIT;
+
+CREATE FUNCTION l2_distance(tinyint[], tinyint[]) RETURNS float8
+	AS 'MODULE_PATHNAME', 'tinyint_l2_distance' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION inner_product(tinyint[], tinyint[]) RETURNS float8
+	AS 'MODULE_PATHNAME', 'tinyint_inner_product' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION cosine_distance(tinyint[], tinyint[]) RETURNS float8
+	AS 'MODULE_PATHNAME', 'tinyint_cosine_distance' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION tinyint_negative_inner_product(tinyint[], tinyint[]) RETURNS float8
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE OPERATOR <-> (
+	LEFTARG = tinyint[], RIGHTARG = tinyint[], PROCEDURE = l2_distance,
+	COMMUTATOR = '<->'
+);
+
+CREATE OPERATOR <#> (
+	LEFTARG = tinyint[], RIGHTARG = tinyint[], PROCEDURE = tinyint_negative_inner_product,
+	COMMUTATOR = '<#>'
+);
+
+CREATE OPERATOR <=> (
+	LEFTARG = tinyint[], RIGHTARG = tinyint[], PROCEDURE = cosine_distance,
+	COMMUTATOR = '<=>'
+);

sql/vector.sql

@@ -290,3 +290,68 @@ CREATE OPERATOR CLASS vector_cosine_ops
 	OPERATOR 1 <=> (vector, vector) FOR ORDER BY float_ops,
 	FUNCTION 1 vector_negative_inner_product(vector, vector),
 	FUNCTION 2 vector_norm(vector);
+
+-- tinyint
+
+CREATE TYPE tinyint;
+
+CREATE FUNCTION tinyint_in(cstring, oid, integer) RETURNS tinyint
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION tinyint_out(tinyint) RETURNS cstring
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION tinyint_recv(internal, oid, integer) RETURNS tinyint
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION tinyint_send(tinyint) RETURNS bytea
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE TYPE tinyint (
+	INPUT     = tinyint_in,
+	OUTPUT    = tinyint_out,
+	RECEIVE   = tinyint_recv,
+	SEND      = tinyint_send,
+	INTERNALLENGTH = 1,
+	PASSEDBYVALUE,
+	ALIGNMENT = char
+);
+
+CREATE FUNCTION integer_to_tinyint(integer, integer, boolean) RETURNS tinyint
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION numeric_to_tinyint(numeric, integer, boolean) RETURNS tinyint
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE CAST (integer AS tinyint)
+	WITH FUNCTION integer_to_tinyint(integer, integer, boolean) AS IMPLICIT;
+
+CREATE CAST (numeric AS tinyint)
+	WITH FUNCTION numeric_to_tinyint(numeric, integer, boolean) AS IMPLICIT;
+
+CREATE FUNCTION l2_distance(tinyint[], tinyint[]) RETURNS float8
+	AS 'MODULE_PATHNAME', 'tinyint_l2_distance' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION inner_product(tinyint[], tinyint[]) RETURNS float8
+	AS 'MODULE_PATHNAME', 'tinyint_inner_product' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION cosine_distance(tinyint[], tinyint[]) RETURNS float8
+	AS 'MODULE_PATHNAME', 'tinyint_cosine_distance' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION tinyint_negative_inner_product(tinyint[], tinyint[]) RETURNS float8
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE OPERATOR <-> (
+	LEFTARG = tinyint[], RIGHTARG = tinyint[], PROCEDURE = l2_distance,
+	COMMUTATOR = '<->'
+);
+
+CREATE OPERATOR <#> (
+	LEFTARG = tinyint[], RIGHTARG = tinyint[], PROCEDURE = tinyint_negative_inner_product,
+	COMMUTATOR = '<#>'
+);
+
+CREATE OPERATOR <=> (
+	LEFTARG = tinyint[], RIGHTARG = tinyint[], PROCEDURE = cosine_distance,
+	COMMUTATOR = '<=>'
+);

src/hnsw.h

@@ -57,6 +57,8 @@
 /* PROGRESS_CREATEIDX_SUBPHASE_INITIALIZE is 1 */
 #define PROGRESS_HNSW_PHASE_LOAD		2
 
+#define HNSW_MAX_SIZE (BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData)) - sizeof(ItemIdData))
+
 #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))
 
@@ -258,7 +260,7 @@ 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);

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);
 
@@ -118,12 +117,12 @@ CreateElementPages(HnswBuildState * buildstate)
 	ListCell   *lc;
 
 	/* Calculate sizes */
-	maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData));
+	maxSize = HNSW_MAX_SIZE;
 	etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions);
 
 	/* Allocate once */
 	etup = palloc0(etupSize);
-	ntup = palloc0(maxSize);
+	ntup = palloc0(BLCKSZ);
 
 	/* Prepare first page */
 	buf = HnswNewBuffer(index, forkNum);
@@ -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

@@ -135,7 +135,7 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
 	etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions);
 	ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m);
 	combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
-	maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData));
+	maxSize = HNSW_MAX_SIZE;
 	minCombinedSize = etupSize + HNSW_NEIGHBOR_TUPLE_SIZE(0, m) + sizeof(ItemIdData);
 
 	/* Prepare element tuple */
@@ -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)
@@ -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);
 

src/hnswscan.c

@@ -113,12 +113,6 @@ hnswbeginscan(Relation index, int nkeys, int norderbys)
 
 	scan->opaque = so;
 
-	/*
-	 * Get a shared lock. This allows vacuum to ensure no in-flight scans
-	 * before marking tuples as deleted.
-	 */
-	LockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
-
 	return scan;
 }
 
@@ -166,11 +160,25 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 		if (scan->orderByData == NULL)
 			elog(ERROR, "cannot scan hnsw index without order");
 
+		/* Requires MVCC-compliant snapshot as not able to maintain a pin */
+		/* https://www.postgresql.org/docs/current/index-locking.html */
+		if (!IsMVCCSnapshot(scan->xs_snapshot))
+			elog(ERROR, "non-MVCC snapshots are not supported with hnsw");
+
 		/* Get scan value */
 		value = GetScanValue(scan);
 
+		/*
+		 * Get a shared lock. This allows vacuum to ensure no in-flight scans
+		 * before marking tuples as deleted.
+		 */
+		LockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
+
 		so->w = GetScanItems(scan, value);
 
+		/* Release shared lock */
+		UnlockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
+
 		so->first = false;
 	}
 
@@ -198,15 +206,6 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 		scan->xs_ctup.t_self = *heaptid;
 #endif
 
-		/*
-		 * Typically, an index scan must maintain a pin on the index page
-		 * 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
-		 */
-
 		scan->xs_recheckorderby = false;
 		return true;
 	}
@@ -223,9 +222,6 @@ hnswendscan(IndexScanDesc scan)
 {
 	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
 
-	/* Release shared lock */
-	UnlockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
-
 	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);
 }

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);
 
@@ -547,9 +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);

src/ivfbuild.c

@@ -506,29 +506,30 @@ CreateListPages(Relation index, VectorArray centers, int dimensions,
 	Buffer		buf;
 	Page		page;
 	GenericXLogState *state;
-	OffsetNumber offno;
-	Size		itemsz;
+	Size		listSize;
 	IvfflatList list;
 
-	itemsz = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
-	list = palloc(itemsz);
+	listSize = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
+	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.h

@@ -277,7 +277,7 @@ 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);

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);
-	collation = rel->rd_indcollation[0];
+	procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC);
+	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;
@@ -80,33 +81,33 @@ 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 */
 	itemsz = MAXALIGN(IndexTupleSize(itup));
-	Assert(itemsz <= BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(IvfflatPageOpaqueData)));
+	Assert(itemsz <= BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(IvfflatPageOpaqueData)) - sizeof(ItemIdData));
 
 	/* 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);
-			newbuf = IvfflatNewBuffer(rel, MAIN_FORKNUM);
-			UnlockRelationForExtension(rel, ExclusiveLock);
+			LockRelationForExtension(index, ExclusiveLock);
+			newbuf = IvfflatNewBuffer(index, MAIN_FORKNUM);
+			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;
-		closestCenter = 0;
+		float		minDistance = FLT_MAX;
+		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/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 */

src/ivfvacuum.c

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

src/tinyint.c

@@ -0,0 +1,294 @@
+#include "postgres.h"
+
+#include <math.h>
+#include <stdint.h>
+
+#include "fmgr.h"
+#include "lib/stringinfo.h"
+#include "libpq/pqformat.h"
+#include "tinyint.h"
+#include "utils/array.h"
+#include "utils/builtins.h"
+#include "utils/numeric.h"
+
+/*
+ * Check if array is a vector
+ */
+static bool
+ArrayIsVector(ArrayType *a)
+{
+	return ARR_NDIM(a) == 1 && !array_contains_nulls(a);
+}
+
+/*
+ * Check if dimensions are the same
+ */
+static int
+CheckDims(ArrayType *a, ArrayType *b)
+{
+	int			dima;
+	int			dimb;
+
+	if (!ArrayIsVector(a) || !ArrayIsVector(b))
+		return 0;
+
+	dima = ARR_DIMS(a)[0];
+	dimb = ARR_DIMS(b)[0];
+
+	if (dima != dimb)
+		return 0;
+
+	return dima;
+}
+
+/*
+ * Check range
+ */
+static void
+CheckRange(long i)
+{
+	if (i < INT8_MIN || i > INT8_MAX)
+		ereport(ERROR,
+				(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
+				 errmsg("value \"%ld\" is out of range for type tinyint", i)));
+}
+
+/*
+ * Convert textual representation to internal representation
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(tinyint_in);
+Datum
+tinyint_in(PG_FUNCTION_ARGS)
+{
+	char	   *s = PG_GETARG_CSTRING(0);
+	const char *ptr = s;
+	long		i;
+	char	   *end;
+
+	/* skip leading spaces */
+	while (*ptr != '\0' && isspace((unsigned char) *ptr))
+		ptr++;
+
+	if (*ptr == '\0')
+		ereport(ERROR,
+				(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+				 errmsg("invalid input syntax for type tinyint: \"%s\"", s)));
+
+	i = strtol(ptr, &end, 10);
+	ptr = end;
+
+	if (i < INT8_MIN || i > INT8_MAX)
+		ereport(ERROR,
+				(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
+				 errmsg("value \"%s\" is out of range for type tinyint", s)));
+
+	/* allow trailing whitespace, but not other trailing chars */
+	while (*ptr != '\0' && isspace((unsigned char) *ptr))
+		ptr++;
+
+	if (*ptr != '\0')
+		ereport(ERROR,
+				(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+				 errmsg("invalid input syntax for type tinyint: \"%s\"", s)));
+
+	PG_RETURN_INT8(i);
+}
+
+/*
+ * Convert internal representation to textual representation
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(tinyint_out);
+Datum
+tinyint_out(PG_FUNCTION_ARGS)
+{
+	int8		num = PG_GETARG_INT8(0);
+	char	   *result = (char *) palloc(5);	/* sign, 3 digits, '\0' */
+
+	pg_ltoa((int32) num, result);
+	PG_RETURN_CSTRING(result);
+}
+
+/*
+ * Convert external binary representation to internal representation
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(tinyint_recv);
+Datum
+tinyint_recv(PG_FUNCTION_ARGS)
+{
+	StringInfo	buf = (StringInfo) PG_GETARG_POINTER(0);
+
+	PG_RETURN_INT8((int8) pq_getmsgint(buf, sizeof(int8)));
+}
+
+/*
+ * Convert internal representation to the external binary representation
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(tinyint_send);
+Datum
+tinyint_send(PG_FUNCTION_ARGS)
+{
+	int8		arg1 = PG_GETARG_INT8(0);
+	StringInfoData buf;
+
+	pq_begintypsend(&buf);
+	pq_sendint8(&buf, arg1);
+	PG_RETURN_BYTEA_P(pq_endtypsend(&buf));
+}
+
+/*
+ * Convert integer to tinyint
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(integer_to_tinyint);
+Datum
+integer_to_tinyint(PG_FUNCTION_ARGS)
+{
+	int32		i = PG_GETARG_INT32(0);
+
+	CheckRange(i);
+
+	PG_RETURN_INT8(i);
+}
+
+/*
+ * Convert numeric to tinyint
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(numeric_to_tinyint);
+Datum
+numeric_to_tinyint(PG_FUNCTION_ARGS)
+{
+	Numeric		num = PG_GETARG_NUMERIC(0);
+	int32		i = numeric_int4_opt_error(num, NULL);
+
+	CheckRange(i);
+
+	PG_RETURN_INT8(i);
+}
+
+/*
+ * Get the L2 distance between tinyint arrays
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(tinyint_l2_distance);
+Datum
+tinyint_l2_distance(PG_FUNCTION_ARGS)
+{
+	ArrayType  *a = PG_GETARG_ARRAYTYPE_P(0);
+	ArrayType  *b = PG_GETARG_ARRAYTYPE_P(1);
+	int8    *ax = (int8 *) ARR_DATA_PTR(a);
+	int8    *bx = (int8 *) ARR_DATA_PTR(b);
+	double		distance = 0.0;
+	int			dim = CheckDims(a, b);
+
+	/* TODO Decide on error or NULL */
+	if (!dim)
+		PG_RETURN_NULL();
+
+	/* Auto-vectorized */
+	for (int i = 0; i < dim; i++)
+	{
+		double		diff = ax[i] - bx[i];
+
+		distance += diff * diff;
+	}
+
+	PG_RETURN_FLOAT8(sqrt(distance));
+}
+
+/*
+ * Get the inner product of two tinyint arrays
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(tinyint_inner_product);
+Datum
+tinyint_inner_product(PG_FUNCTION_ARGS)
+{
+	ArrayType  *a = PG_GETARG_ARRAYTYPE_P(0);
+	ArrayType  *b = PG_GETARG_ARRAYTYPE_P(1);
+	int8	   *ax = (int8 *) ARR_DATA_PTR(a);
+	int8	   *bx = (int8 *) ARR_DATA_PTR(b);
+	double		distance = 0.0;
+	int			dim = CheckDims(a, b);
+
+	/* TODO Decide on error or NULL */
+	if (!dim)
+		PG_RETURN_NULL();
+
+	/* Auto-vectorized */
+	for (int i = 0; i < dim; i++)
+		distance += ax[i] * bx[i];
+
+	PG_RETURN_FLOAT8(distance);
+}
+
+/*
+ * Get the negative inner product of two tinyint arrays
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(tinyint_negative_inner_product);
+Datum
+tinyint_negative_inner_product(PG_FUNCTION_ARGS)
+{
+	ArrayType  *a = PG_GETARG_ARRAYTYPE_P(0);
+	ArrayType  *b = PG_GETARG_ARRAYTYPE_P(1);
+	int8	   *ax = (int8 *) ARR_DATA_PTR(a);
+	int8	   *bx = (int8 *) ARR_DATA_PTR(b);
+	double		distance = 0.0;
+	int			dim = CheckDims(a, b);
+
+	/* TODO Decide on error or NULL */
+	if (!dim)
+		PG_RETURN_NULL();
+
+	/* Auto-vectorized */
+	for (int i = 0; i < dim; i++)
+		distance += ax[i] * bx[i];
+
+	PG_RETURN_FLOAT8(distance * -1);
+}
+
+/*
+ * Get the cosine distance between two float2 arrays
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(tinyint_cosine_distance);
+Datum
+tinyint_cosine_distance(PG_FUNCTION_ARGS)
+{
+	ArrayType  *a = PG_GETARG_ARRAYTYPE_P(0);
+	ArrayType  *b = PG_GETARG_ARRAYTYPE_P(1);
+	int8   *ax = (int8 *) ARR_DATA_PTR(a);
+	int8   *bx = (int8 *) ARR_DATA_PTR(b);
+	double		distance = 0.0;
+	double		norma = 0.0;
+	double		normb = 0.0;
+	double		similarity;
+	int			dim = CheckDims(a, b);
+
+	/* TODO Decide on error or NULL */
+	if (!dim)
+		PG_RETURN_NULL();
+
+	/* Auto-vectorized */
+	for (int i = 0; i < dim; i++)
+	{
+		float		axi = ax[i];
+		float		bxi = bx[i];
+
+		distance += axi * bxi;
+		norma += axi * axi;
+		normb += bxi * bxi;
+	}
+
+	/* Use sqrt(a * b) over sqrt(a) * sqrt(b) */
+	similarity = distance / sqrt(norma * normb);
+
+#ifdef _MSC_VER
+	/* /fp:fast may not propagate NaN */
+	if (isnan(similarity))
+		PG_RETURN_FLOAT8(NAN);
+#endif
+
+	/* Keep in range */
+	if (similarity > 1)
+		similarity = 1;
+	else if (similarity < -1)
+		similarity = -1;
+
+	PG_RETURN_FLOAT8(1 - similarity);
+}

src/tinyint.h

@@ -0,0 +1,8 @@
+#ifndef TINYINT_H
+#define TINYINT_H
+
+#define DatumGetInt8(X) ((int8) (X))
+#define PG_GETARG_INT8(n)	 DatumGetInt8(PG_GETARG_DATUM(n))
+#define PG_RETURN_INT8(x)	 return Int8GetDatum(x)
+
+#endif

test/expected/functions.out

@@ -54,85 +54,85 @@ SELECT vector_norm('[3e37,4e37]')::real;
        5e+37
 (1 row)
 
-SELECT l2_distance('[0,0]', '[3,4]');
+SELECT l2_distance('[0,0]'::vector, '[3,4]');
  l2_distance 
 -------------
            5
 (1 row)
 
-SELECT l2_distance('[0,0]', '[0,1]');
+SELECT l2_distance('[0,0]'::vector, '[0,1]');
  l2_distance 
 -------------
            1
 (1 row)
 
-SELECT l2_distance('[1,2]', '[3]');
+SELECT l2_distance('[1,2]'::vector, '[3]');
 ERROR:  different vector dimensions 2 and 1
-SELECT l2_distance('[3e38]', '[-3e38]');
+SELECT l2_distance('[3e38]'::vector, '[-3e38]');
  l2_distance 
 -------------
     Infinity
 (1 row)
 
-SELECT inner_product('[1,2]', '[3,4]');
+SELECT inner_product('[1,2]'::vector, '[3,4]');
  inner_product 
 ---------------
             11
 (1 row)
 
-SELECT inner_product('[1,2]', '[3]');
+SELECT inner_product('[1,2]'::vector, '[3]');
 ERROR:  different vector dimensions 2 and 1
-SELECT inner_product('[3e38]', '[3e38]');
+SELECT inner_product('[3e38]'::vector, '[3e38]');
  inner_product 
 ---------------
       Infinity
 (1 row)
 
-SELECT cosine_distance('[1,2]', '[2,4]');
+SELECT cosine_distance('[1,2]'::vector, '[2,4]');
  cosine_distance 
 -----------------
                0
 (1 row)
 
-SELECT cosine_distance('[1,2]', '[0,0]');
+SELECT cosine_distance('[1,2]'::vector, '[0,0]');
  cosine_distance 
 -----------------
              NaN
 (1 row)
 
-SELECT cosine_distance('[1,1]', '[1,1]');
+SELECT cosine_distance('[1,1]'::vector, '[1,1]');
  cosine_distance 
 -----------------
                0
 (1 row)
 
-SELECT cosine_distance('[1,0]', '[0,2]');
+SELECT cosine_distance('[1,0]'::vector, '[0,2]');
  cosine_distance 
 -----------------
                1
 (1 row)
 
-SELECT cosine_distance('[1,1]', '[-1,-1]');
+SELECT cosine_distance('[1,1]'::vector, '[-1,-1]');
  cosine_distance 
 -----------------
                2
 (1 row)
 
-SELECT cosine_distance('[1,2]', '[3]');
+SELECT cosine_distance('[1,2]'::vector, '[3]');
 ERROR:  different vector dimensions 2 and 1
-SELECT cosine_distance('[1,1]', '[1.1,1.1]');
+SELECT cosine_distance('[1,1]'::vector, '[1.1,1.1]');
  cosine_distance 
 -----------------
                0
 (1 row)
 
-SELECT cosine_distance('[1,1]', '[-1.1,-1.1]');
+SELECT cosine_distance('[1,1]'::vector, '[-1.1,-1.1]');
  cosine_distance 
 -----------------
                2
 (1 row)
 
-SELECT cosine_distance('[3e38]', '[3e38]');
+SELECT cosine_distance('[3e38]'::vector, '[3e38]');
  cosine_distance 
 -----------------
              NaN

test/expected/tinyint.out

@@ -0,0 +1,148 @@
+SELECT '127'::tinyint;
+ tinyint 
+---------
+ 127
+(1 row)
+
+SELECT '128'::tinyint;
+ERROR:  value "128" is out of range for type tinyint
+LINE 1: SELECT '128'::tinyint;
+               ^
+SELECT '-128'::tinyint;
+ tinyint 
+---------
+ -128
+(1 row)
+
+SELECT '-129'::tinyint;
+ERROR:  value "-129" is out of range for type tinyint
+LINE 1: SELECT '-129'::tinyint;
+               ^
+SELECT ''::tinyint;
+ERROR:  invalid input syntax for type tinyint: ""
+LINE 1: SELECT ''::tinyint;
+               ^
+SELECT ' 1'::tinyint;
+ tinyint 
+---------
+ 1
+(1 row)
+
+SELECT '1 '::tinyint;
+ tinyint 
+---------
+ 1
+(1 row)
+
+SELECT '1a'::tinyint;
+ERROR:  invalid input syntax for type tinyint: "1a"
+LINE 1: SELECT '1a'::tinyint;
+               ^
+SELECT '{1,2,3}'::tinyint[];
+ tinyint 
+---------
+ {1,2,3}
+(1 row)
+
+SELECT '128'::numeric::tinyint;
+ERROR:  value "128" is out of range for type tinyint
+SELECT 'NaN'::numeric::tinyint;
+ERROR:  cannot convert NaN to integer
+SELECT l2_distance('{0,0}'::tinyint[], '{3,4}'::tinyint[]);
+ l2_distance 
+-------------
+           5
+(1 row)
+
+SELECT l2_distance('{0,0}'::tinyint[], '{0,1}'::tinyint[]);
+ l2_distance 
+-------------
+           1
+(1 row)
+
+SELECT l2_distance('{1,2}'::tinyint[], '{3}'::tinyint[]);
+ l2_distance 
+-------------
+            
+(1 row)
+
+SELECT l2_distance('{3e38}'::tinyint[], '{-3e38}'::tinyint[]);
+ERROR:  invalid input syntax for type tinyint: "3e38"
+LINE 1: SELECT l2_distance('{3e38}'::tinyint[], '{-3e38}'::tinyint[]...
+                           ^
+SELECT '{0,0}'::tinyint[] <-> '{3,4}'::tinyint[];
+ ?column? 
+----------
+        5
+(1 row)
+
+SELECT inner_product('{1,2}'::tinyint[], '{3,4}'::tinyint[]);
+ inner_product 
+---------------
+            11
+(1 row)
+
+SELECT inner_product('{1,2}'::tinyint[], '{3}'::tinyint[]);
+ inner_product 
+---------------
+              
+(1 row)
+
+SELECT inner_product('{127}'::tinyint[], '{127}'::tinyint[]);
+ inner_product 
+---------------
+         16129
+(1 row)
+
+SELECT '{1,2}'::tinyint[] <#> '{3,4}'::tinyint[];
+ ?column? 
+----------
+      -11
+(1 row)
+
+SELECT cosine_distance('{1,2}'::tinyint[], '{2,4}'::tinyint[]);
+ cosine_distance 
+-----------------
+               0
+(1 row)
+
+SELECT cosine_distance('{1,2}'::tinyint[], '{0,0}'::tinyint[]);
+ cosine_distance 
+-----------------
+             NaN
+(1 row)
+
+SELECT cosine_distance('{1,1}'::tinyint[], '{1,1}'::tinyint[]);
+ cosine_distance 
+-----------------
+               0
+(1 row)
+
+SELECT cosine_distance('{1,0}'::tinyint[], '{0,2}'::tinyint[]);
+ cosine_distance 
+-----------------
+               1
+(1 row)
+
+SELECT cosine_distance('{1,1}'::tinyint[], '{-1,-1}'::tinyint[]);
+ cosine_distance 
+-----------------
+               2
+(1 row)
+
+SELECT cosine_distance('{1,2}'::tinyint[], '{3}'::tinyint[]);
+ cosine_distance 
+-----------------
+                
+(1 row)
+
+SELECT cosine_distance('{3e38}'::tinyint[], '{3e38}'::tinyint[]);
+ERROR:  invalid input syntax for type tinyint: "3e38"
+LINE 1: SELECT cosine_distance('{3e38}'::tinyint[], '{3e38}'::tinyin...
+                               ^
+SELECT '{1,2}'::tinyint[] <=> '{2,4}'::tinyint[];
+ ?column? 
+----------
+        0
+(1 row)
+

test/sql/functions.sql

@@ -13,24 +13,24 @@ SELECT vector_norm('[3,4]');
 SELECT vector_norm('[0,1]');
 SELECT vector_norm('[3e37,4e37]')::real;
 
-SELECT l2_distance('[0,0]', '[3,4]');
-SELECT l2_distance('[0,0]', '[0,1]');
-SELECT l2_distance('[1,2]', '[3]');
-SELECT l2_distance('[3e38]', '[-3e38]');
+SELECT l2_distance('[0,0]'::vector, '[3,4]');
+SELECT l2_distance('[0,0]'::vector, '[0,1]');
+SELECT l2_distance('[1,2]'::vector, '[3]');
+SELECT l2_distance('[3e38]'::vector, '[-3e38]');
 
-SELECT inner_product('[1,2]', '[3,4]');
-SELECT inner_product('[1,2]', '[3]');
-SELECT inner_product('[3e38]', '[3e38]');
+SELECT inner_product('[1,2]'::vector, '[3,4]');
+SELECT inner_product('[1,2]'::vector, '[3]');
+SELECT inner_product('[3e38]'::vector, '[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]');
-SELECT cosine_distance('[1,1]', '[-1.1,-1.1]');
-SELECT cosine_distance('[3e38]', '[3e38]');
+SELECT cosine_distance('[1,2]'::vector, '[2,4]');
+SELECT cosine_distance('[1,2]'::vector, '[0,0]');
+SELECT cosine_distance('[1,1]'::vector, '[1,1]');
+SELECT cosine_distance('[1,0]'::vector, '[0,2]');
+SELECT cosine_distance('[1,1]'::vector, '[-1,-1]');
+SELECT cosine_distance('[1,2]'::vector, '[3]');
+SELECT cosine_distance('[1,1]'::vector, '[1.1,1.1]');
+SELECT cosine_distance('[1,1]'::vector, '[-1.1,-1.1]');
+SELECT cosine_distance('[3e38]'::vector, '[3e38]');
 
 SELECT l1_distance('[0,0]', '[3,4]');
 SELECT l1_distance('[0,0]', '[0,1]');

test/sql/tinyint.sql

@@ -0,0 +1,34 @@
+SELECT '127'::tinyint;
+SELECT '128'::tinyint;
+SELECT '-128'::tinyint;
+SELECT '-129'::tinyint;
+
+SELECT ''::tinyint;
+SELECT ' 1'::tinyint;
+SELECT '1 '::tinyint;
+SELECT '1a'::tinyint;
+
+SELECT '{1,2,3}'::tinyint[];
+
+SELECT '128'::numeric::tinyint;
+SELECT 'NaN'::numeric::tinyint;
+
+SELECT l2_distance('{0,0}'::tinyint[], '{3,4}'::tinyint[]);
+SELECT l2_distance('{0,0}'::tinyint[], '{0,1}'::tinyint[]);
+SELECT l2_distance('{1,2}'::tinyint[], '{3}'::tinyint[]);
+SELECT l2_distance('{3e38}'::tinyint[], '{-3e38}'::tinyint[]);
+SELECT '{0,0}'::tinyint[] <-> '{3,4}'::tinyint[];
+
+SELECT inner_product('{1,2}'::tinyint[], '{3,4}'::tinyint[]);
+SELECT inner_product('{1,2}'::tinyint[], '{3}'::tinyint[]);
+SELECT inner_product('{127}'::tinyint[], '{127}'::tinyint[]);
+SELECT '{1,2}'::tinyint[] <#> '{3,4}'::tinyint[];
+
+SELECT cosine_distance('{1,2}'::tinyint[], '{2,4}'::tinyint[]);
+SELECT cosine_distance('{1,2}'::tinyint[], '{0,0}'::tinyint[]);
+SELECT cosine_distance('{1,1}'::tinyint[], '{1,1}'::tinyint[]);
+SELECT cosine_distance('{1,0}'::tinyint[], '{0,2}'::tinyint[]);
+SELECT cosine_distance('{1,1}'::tinyint[], '{-1,-1}'::tinyint[]);
+SELECT cosine_distance('{1,2}'::tinyint[], '{3}'::tinyint[]);
+SELECT cosine_distance('{3e38}'::tinyint[], '{3e38}'::tinyint[]);
+SELECT '{1,2}'::tinyint[] <=> '{2,4}'::tinyint[];