Improved name [skip ci]

sql/vector.sql

@@ -266,6 +266,9 @@ COMMENT ON ACCESS METHOD hnsw IS 'hnsw index access method';
 CREATE FUNCTION ivfflat_halfvec_support(internal) RETURNS internal
 	AS 'MODULE_PATHNAME' LANGUAGE C;
 
+CREATE FUNCTION ivfflat_minivec_support(internal) RETURNS internal
+	AS 'MODULE_PATHNAME' LANGUAGE C;
+
 CREATE FUNCTION ivfflat_bit_support(internal) RETURNS internal
 	AS 'MODULE_PATHNAME' LANGUAGE C;
 
@@ -748,6 +751,9 @@ CREATE FUNCTION minivec_l2_squared_distance(minivec, minivec) RETURNS float8
 CREATE FUNCTION minivec_negative_inner_product(minivec, minivec) RETURNS float8
 	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
 
+CREATE FUNCTION minivec_spherical_distance(minivec, minivec) RETURNS float8
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
 -- minivec cast functions
 
 CREATE FUNCTION minivec(minivec, integer, boolean) RETURNS minivec
@@ -887,6 +893,30 @@ CREATE OPERATOR CLASS minivec_ops
 	OPERATOR 5 > ,
 	FUNCTION 1 minivec_cmp(minivec, minivec);
 
+CREATE OPERATOR CLASS minivec_l2_ops
+	FOR TYPE minivec USING ivfflat AS
+	OPERATOR 1 <-> (minivec, minivec) FOR ORDER BY float_ops,
+	FUNCTION 1 minivec_l2_squared_distance(minivec, minivec),
+	FUNCTION 3 l2_distance(minivec, minivec),
+	FUNCTION 5 ivfflat_minivec_support(internal);
+
+CREATE OPERATOR CLASS minivec_ip_ops
+	FOR TYPE minivec USING ivfflat AS
+	OPERATOR 1 <#> (minivec, minivec) FOR ORDER BY float_ops,
+	FUNCTION 1 minivec_negative_inner_product(minivec, minivec),
+	FUNCTION 3 minivec_spherical_distance(minivec, minivec),
+	FUNCTION 4 l2_norm(minivec),
+	FUNCTION 5 ivfflat_minivec_support(internal);
+
+CREATE OPERATOR CLASS minivec_cosine_ops
+	FOR TYPE minivec USING ivfflat AS
+	OPERATOR 1 <=> (minivec, minivec) FOR ORDER BY float_ops,
+	FUNCTION 1 cosine_distance(minivec, minivec),
+	FUNCTION 2 l2_norm(minivec),
+	FUNCTION 3 minivec_spherical_distance(minivec, minivec),
+	FUNCTION 4 l2_norm(minivec),
+	FUNCTION 5 ivfflat_minivec_support(internal);
+
 CREATE OPERATOR CLASS minivec_l2_ops
 	FOR TYPE minivec USING hnsw AS
 	OPERATOR 1 <-> (minivec, minivec) FOR ORDER BY float_ops,
@@ -902,7 +932,7 @@ CREATE OPERATOR CLASS minivec_ip_ops
 CREATE OPERATOR CLASS minivec_cosine_ops
 	FOR TYPE minivec USING hnsw AS
 	OPERATOR 1 <=> (minivec, minivec) FOR ORDER BY float_ops,
-	FUNCTION 1 minivec_negative_inner_product(minivec, minivec),
+	FUNCTION 1 cosine_distance(minivec, minivec),
 	FUNCTION 2 l2_norm(minivec),
 	FUNCTION 3 hnsw_minivec_support(internal);
 

src/hnswutils.c

@@ -159,6 +159,9 @@ HnswOptionalProcInfo(Relation index, uint16 procnum)
 Datum
 HnswNormValue(const HnswTypeInfo * typeInfo, Oid collation, Datum value)
 {
+	if (!typeInfo->normalize)
+		return value;
+
 	return DirectFunctionCall1Coll(typeInfo->normalize, collation, value);
 }
 
@@ -1382,7 +1385,8 @@ hnsw_minivec_support(PG_FUNCTION_ARGS)
 {
 	static const HnswTypeInfo typeInfo = {
 		.maxDimensions = HNSW_MAX_DIM * 4,
-		.normalize = minivec_l2_normalize,
+		/* Do not normalize to maximize precision */
+		.normalize = NULL,
 		.checkValue = NULL
 	};
 

src/ivfutils.c

@@ -7,6 +7,7 @@
 #include "halfutils.h"
 #include "halfvec.h"
 #include "ivfflat.h"
+#include "minivec.h"
 #include "storage/bufmgr.h"
 
 /*
@@ -70,6 +71,9 @@ IvfflatOptionalProcInfo(Relation index, uint16 procnum)
 Datum
 IvfflatNormValue(const IvfflatTypeInfo * typeInfo, Oid collation, Datum value)
 {
+	if (!typeInfo->normalize)
+		return value;
+
 	return DirectFunctionCall1Coll(typeInfo->normalize, collation, value);
 }
 
@@ -231,6 +235,7 @@ IvfflatUpdateList(Relation index, ListInfo listInfo,
 
 PGDLLEXPORT Datum l2_normalize(PG_FUNCTION_ARGS);
 PGDLLEXPORT Datum halfvec_l2_normalize(PG_FUNCTION_ARGS);
+PGDLLEXPORT Datum minivec_l2_normalize(PG_FUNCTION_ARGS);
 PGDLLEXPORT Datum sparsevec_l2_normalize(PG_FUNCTION_ARGS);
 
 static Size
@@ -245,6 +250,12 @@ HalfvecItemSize(int dimensions)
 	return HALFVEC_SIZE(dimensions);
 }
 
+static Size
+MinivecItemSize(int dimensions)
+{
+	return MINIVEC_SIZE(dimensions);
+}
+
 static Size
 BitItemSize(int dimensions)
 {
@@ -275,6 +286,18 @@ HalfvecUpdateCenter(Pointer v, int dimensions, float *x)
 		vec->x[k] = Float4ToHalfUnchecked(x[k]);
 }
 
+static void
+MinivecUpdateCenter(Pointer v, int dimensions, float *x)
+{
+	MiniVector *vec = (MiniVector *) v;
+
+	SET_VARSIZE(vec, MINIVEC_SIZE(dimensions));
+	vec->dim = dimensions;
+
+	for (int k = 0; k < dimensions; k++)
+		vec->x[k] = Float4ToFp8Unchecked(x[k]);
+}
+
 static void
 BitUpdateCenter(Pointer v, int dimensions, float *x)
 {
@@ -309,6 +332,15 @@ HalfvecSumCenter(Pointer v, float *x)
 		x[k] += HalfToFloat4(vec->x[k]);
 }
 
+static void
+MinivecSumCenter(Pointer v, float *x)
+{
+	MiniVector *vec = (MiniVector *) v;
+
+	for (int k = 0; k < vec->dim; k++)
+		x[k] += Fp8ToFloat4(vec->x[k]);
+}
+
 static void
 BitSumCenter(Pointer v, float *x)
 {
@@ -357,6 +389,22 @@ ivfflat_halfvec_support(PG_FUNCTION_ARGS)
 	PG_RETURN_POINTER(&typeInfo);
 };
 
+FUNCTION_PREFIX PG_FUNCTION_INFO_V1(ivfflat_minivec_support);
+Datum
+ivfflat_minivec_support(PG_FUNCTION_ARGS)
+{
+	static const IvfflatTypeInfo typeInfo = {
+		.maxDimensions = IVFFLAT_MAX_DIM * 4,
+		/* Do not normalize to maximize precision */
+		.normalize = NULL,
+		.itemSize = MinivecItemSize,
+		.updateCenter = MinivecUpdateCenter,
+		.sumCenter = MinivecSumCenter
+	};
+
+	PG_RETURN_POINTER(&typeInfo);
+};
+
 FUNCTION_PREFIX PG_FUNCTION_INFO_V1(ivfflat_bit_support);
 Datum
 ivfflat_bit_support(PG_FUNCTION_ARGS)

src/minivec.c

@@ -69,6 +69,11 @@ CheckElement(fp8 value)
 		ereport(ERROR,
 				(errcode(ERRCODE_DATA_EXCEPTION),
 				 errmsg("NaN not allowed in minivec")));
+
+	if (Fp8IsInf(value))
+		ereport(ERROR,
+				(errcode(ERRCODE_DATA_EXCEPTION),
+				 errmsg("infinite value not allowed in minivec")));
 }
 
 /*
@@ -169,7 +174,7 @@ minivec_in(PG_FUNCTION_ARGS)
 		x[dim] = Float4ToFp8Unchecked(val);
 
 		/* Check for range error like float4in */
-		if ((errno == ERANGE && isinf(val)) || (Fp8IsNan(x[dim]) && !isnan(val)))
+		if ((errno == ERANGE && isinf(val)) || (Fp8IsInf(x[dim]) && !isinf(val)))
 			ereport(ERROR,
 					(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
 					 errmsg("\"%s\" is out of range for type minivec", pnstrdup(pt, stringEnd - pt))));
@@ -739,7 +744,7 @@ minivec_l2_normalize(PG_FUNCTION_ARGS)
 		/* Check for overflow */
 		for (int i = 0; i < a->dim; i++)
 		{
-			if (Fp8IsNan(rx[i]))
+			if (Fp8IsInf(rx[i]))
 				float_overflow_error();
 		}
 	}
@@ -768,18 +773,12 @@ minivec_add(PG_FUNCTION_ARGS)
 
 	/* Auto-vectorized */
 	for (int i = 0, imax = a->dim; i < imax; i++)
-	{
-#ifdef FLT16_SUPPORT
-		rx[i] = ax[i] + bx[i];
-#else
 		rx[i] = Float4ToFp8Unchecked(Fp8ToFloat4(ax[i]) + Fp8ToFloat4(bx[i]));
-#endif
-	}
 
 	/* Check for overflow */
 	for (int i = 0, imax = a->dim; i < imax; i++)
 	{
-		if (Fp8IsNan(rx[i]))
+		if (Fp8IsInf(rx[i]))
 			float_overflow_error();
 	}
 
@@ -807,18 +806,12 @@ minivec_sub(PG_FUNCTION_ARGS)
 
 	/* Auto-vectorized */
 	for (int i = 0, imax = a->dim; i < imax; i++)
-	{
-#ifdef FLT16_SUPPORT
 		rx[i] = ax[i] - bx[i];
-#else
-		rx[i] = Float4ToFp8Unchecked(Fp8ToFloat4(ax[i]) - Fp8ToFloat4(bx[i]));
-#endif
-	}
 
 	/* Check for overflow */
 	for (int i = 0, imax = a->dim; i < imax; i++)
 	{
-		if (Fp8IsNan(rx[i]))
+		if (Fp8IsInf(rx[i]))
 			float_overflow_error();
 	}
 
@@ -846,18 +839,12 @@ minivec_mul(PG_FUNCTION_ARGS)
 
 	/* Auto-vectorized */
 	for (int i = 0, imax = a->dim; i < imax; i++)
-	{
-#ifdef FLT16_SUPPORT
-		rx[i] = ax[i] * bx[i];
-#else
 		rx[i] = Float4ToFp8Unchecked(Fp8ToFloat4(ax[i]) * Fp8ToFloat4(bx[i]));
-#endif
-	}
 
 	/* Check for overflow and underflow */
 	for (int i = 0, imax = a->dim; i < imax; i++)
 	{
-		if (Fp8IsNan(rx[i]))
+		if (Fp8IsInf(rx[i]))
 			float_overflow_error();
 
 		if (Fp8IsZero(rx[i]) && !(Fp8IsZero(ax[i]) || Fp8IsZero(bx[i])))

src/minivec.h

@@ -1,8 +1,6 @@
 #ifndef MINIVEC_H
 #define MINIVEC_H
 
-#include <float.h>
-
 #define MINIVEC_MAX_DIM 16000
 
 #define fp8 uint8
@@ -28,7 +26,16 @@ MiniVector *InitMiniVector(int dim);
 static inline bool
 Fp8IsNan(fp8 num)
 {
-	return (num & 0x7F) == 0x7F;
+	return (num & 0x7C) == 0x7C && (num & 0x7F) != 0x7C;
+}
+
+/*
+ * Check if fp8 is infinite
+ */
+static inline bool
+Fp8IsInf(fp8 num)
+{
+	return (num & 0x7F) == 0x7C;
 }
 
 /*
@@ -46,10 +53,19 @@ Fp8IsZero(fp8 num)
 static inline float
 Fp8ToFloat4(fp8 num)
 {
-	float		lookup[128] = {0, 0.00195312, 0.00390625, 0.00585938, 0.0078125, 0.00976562, 0.0117188, 0.0136719, 0.015625, 0.0175781, 0.0195312, 0.0214844, 0.0234375, 0.0253906, 0.0273438, 0.0292969, 0.03125, 0.0351562, 0.0390625, 0.0429688, 0.046875, 0.0507812, 0.0546875, 0.0585938, 0.0625, 0.0703125, 0.078125, 0.0859375, 0.09375, 0.101562, 0.109375, 0.117188, 0.125, 0.140625, 0.15625, 0.171875, 0.1875, 0.203125, 0.21875, 0.234375, 0.25, 0.28125, 0.3125, 0.34375, 0.375, 0.40625, 0.4375, 0.46875, 0.5, 0.5625, 0.625, 0.6875, 0.75, 0.8125, 0.875, 0.9375, 1, 1.125, 1.25, 1.375, 1.5, 1.625, 1.75, 1.875, 2, 2.25, 2.5, 2.75, 3, 3.25, 3.5, 3.75, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 20, 22, 24, 26, 28, 30, 32, 36, 40, 44, 48, 52, 56, 60, 64, 72, 80, 88, 96, 104, 112, 120, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, NAN};
+	/* Lookup table for non-sign bits */
-	float		v = lookup[num & 0x7F];
+	/* Uses uint32 for correctness */
+	uint32		lookup[128] = {0, 931135488, 939524096, 943718400, 947912704, 950009856, 952107008, 954204160, 956301312, 958398464, 960495616, 962592768, 964689920, 966787072, 968884224, 970981376, 973078528, 975175680, 977272832, 979369984, 981467136, 983564288, 985661440, 987758592, 989855744, 991952896, 994050048, 996147200, 998244352, 1000341504, 1002438656, 1004535808, 1006632960, 1008730112, 1010827264, 1012924416, 1015021568, 1017118720, 1019215872, 1021313024, 1023410176, 1025507328, 1027604480, 1029701632, 1031798784, 1033895936, 1035993088, 1038090240, 1040187392, 1042284544, 1044381696, 1046478848, 1048576000, 1050673152, 1052770304, 1054867456, 1056964608, 1059061760, 1061158912, 1063256064, 1065353216, 1067450368, 1069547520, 1071644672, 1073741824, 1075838976, 1077936128, 1080033280, 1082130432, 1084227584, 1086324736, 1088421888, 1090519040, 1092616192, 1094713344, 1096810496, 1098907648, 1101004800, 1103101952, 1105199104, 1107296256, 1109393408, 1111490560, 1113587712, 1115684864, 1117782016, 1119879168, 1121976320, 1124073472, 1126170624, 1128267776, 1130364928, 1132462080, 1134559232, 1136656384, 1138753536, 1140850688, 1142947840, 1145044992, 1147142144, 1149239296, 1151336448, 1153433600, 1155530752, 1157627904, 1159725056, 1161822208, 1163919360, 1166016512, 1168113664, 1170210816, 1172307968, 1174405120, 1176502272, 1178599424, 1180696576, 1182793728, 1184890880, 1186988032, 1189085184, 1191182336, 1193279488, 1195376640, 1197473792, 2139095040, 2145386496, 2143289344, 2145386496};
 
-	return (num & 0x80) == 0x80 ? -v : v;
+	union
+	{
+		float		f;
+		uint32		i;
+	}			swap;
+
+	swap.i = lookup[num & 0x7F];
+
+	return (num & 0x80) == 0x80 ? -swap.f : swap.f;
 }
 
 /*
@@ -62,27 +78,32 @@ Float4ToFp8Unchecked(float num)
 	{
 		float		f;
 		uint32		i;
-	}			swapfloat;
+	}			swap;
 
 	uint32		bin;
 	int			exponent;
 	int			mantissa;
 	uint8		result;
 
-	swapfloat.f = num;
+	swap.f = num;
-	bin = swapfloat.i;
+	bin = swap.i;
 	exponent = (bin & 0x7F800000) >> 23;
 	mantissa = bin & 0x007FFFFF;
 
 	/* Sign */
 	result = (bin & 0x80000000) >> 24;
 
-	if (isinf(num) || isnan(num))
+	if (isinf(num))
+	{
+		/* Infinite */
+		result |= 0x7C;
+	}
+	else if (isnan(num))
 	{
 		/* NaN */
 		result |= 0x7F;
 	}
-	else if (exponent > 116)
+	else if (exponent > 98)
 	{
 		int			m;
 		int			gr;
@@ -92,37 +113,37 @@ Float4ToFp8Unchecked(float num)
 		s = mantissa & 0x000FFFFF;
 
 		/* Subnormal */
-		if (exponent < -6)
+		if (exponent < -14)
 		{
-			int			diff = -exponent - 6;
+			int			diff = -exponent - 14;
 
 			mantissa >>= diff;
 			mantissa += 1 << (23 - diff);
 			s |= mantissa & 0x000FFFFF;
 		}
 
-		m = mantissa >> 20;
+		m = mantissa >> 21;
 
 		/* Round */
-		gr = (mantissa >> 19) % 4;
+		gr = (mantissa >> 20) % 4;
 		if (gr == 3 || (gr == 1 && s != 0))
 			m += 1;
 
-		if (m == 8)
+		if (m == 4)
 		{
 			m = 0;
 			exponent += 1;
 		}
 
-		if (exponent > 8)
+		if (exponent > 15)
 		{
-			/* Infinite, which is NaN */
+			/* Infinite */
-			result |= 0x7F;
+			result |= 0x7C;
 		}
 		else
 		{
-			if (exponent >= -7)
+			if (exponent >= -14)
-				result |= (exponent + 7) << 3;
+				result |= (exponent + 15) << 2;
 
 			result |= m;
 		}
@@ -139,7 +160,7 @@ Float4ToFp8(float num)
 {
 	fp8			result = Float4ToFp8Unchecked(num);
 
-	if (unlikely(Fp8IsNan(result)) && !isnan(num))
+	if (unlikely(Fp8IsInf(result)) && !isinf(num))
 	{
 		char	   *buf = palloc(FLOAT_SHORTEST_DECIMAL_LEN);
 

test/expected/cast.out

@@ -154,8 +154,8 @@ SELECT '[1,2,3]'::vector::minivec(3);
 
 SELECT '[1,2,3]'::vector::minivec(2);
 ERROR:  expected 2 dimensions, not 3
-SELECT '[465]'::vector::minivec;
+SELECT '[61440]'::vector::minivec;
-ERROR:  "465" is out of range for type minivec
+ERROR:  "61440" is out of range for type minivec
 SELECT '[1e-8]'::vector::minivec;
  minivec 
 ---------
@@ -190,8 +190,8 @@ SELECT '{1,2,3}'::real[]::minivec(3);
 
 SELECT '{1,2,3}'::real[]::minivec(2);
 ERROR:  expected 2 dimensions, not 3
-SELECT '{465,-465}'::real[]::minivec;
+SELECT '{61440,-61440}'::real[]::minivec;
-ERROR:  "465" is out of range for type minivec
+ERROR:  "61440" is out of range for type minivec
 SELECT '{1e-8,-1e-8}'::real[]::minivec;
  minivec 
 ---------

test/expected/ivfflat_minivec.out

@@ -0,0 +1,84 @@
+SET enable_seqscan = off;
+-- L2
+CREATE TABLE t (val minivec(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING ivfflat (val minivec_l2_ops) WITH (lists = 1);
+INSERT INTO t (val) VALUES ('[1,2,4]');
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+   val   
+---------
+ [1,2,3]
+ [1,2,4]
+ [1,1,1]
+ [0,0,0]
+(4 rows)
+
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <-> (SELECT NULL::minivec)) t2;
+ count 
+-------
+     4
+(1 row)
+
+SELECT COUNT(*) FROM t;
+ count 
+-------
+     5
+(1 row)
+
+TRUNCATE t;
+NOTICE:  ivfflat index created with little data
+DETAIL:  This will cause low recall.
+HINT:  Drop the index until the table has more data.
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+ val 
+-----
+(0 rows)
+
+DROP TABLE t;
+-- inner product
+CREATE TABLE t (val minivec(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING ivfflat (val minivec_ip_ops) WITH (lists = 1);
+INSERT INTO t (val) VALUES ('[1,2,4]');
+SELECT * FROM t ORDER BY val <#> '[3,3,3]';
+   val   
+---------
+ [1,2,4]
+ [1,2,3]
+ [1,1,1]
+ [0,0,0]
+(4 rows)
+
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <#> (SELECT NULL::minivec)) t2;
+ count 
+-------
+     4
+(1 row)
+
+DROP TABLE t;
+-- cosine
+CREATE TABLE t (val minivec(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING ivfflat (val minivec_cosine_ops) WITH (lists = 1);
+INSERT INTO t (val) VALUES ('[1,2,4]');
+SELECT * FROM t ORDER BY val <=> '[3,3,3]';
+   val   
+---------
+ [1,1,1]
+ [1,2,3]
+ [1,2,4]
+(3 rows)
+
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> '[0,0,0]') t2;
+ count 
+-------
+     3
+(1 row)
+
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> (SELECT NULL::minivec)) t2;
+ count 
+-------
+     3
+(1 row)
+
+DROP TABLE t;

test/expected/minivec.out

@@ -37,20 +37,22 @@ ERROR:  NaN not allowed in minivec
 LINE 1: SELECT '[NaN,1]'::minivec;
                ^
 SELECT '[Infinity,1]'::minivec;
-ERROR:  "Infinity" is out of range for type minivec
+ERROR:  infinite value not allowed in minivec
 LINE 1: SELECT '[Infinity,1]'::minivec;
                ^
 SELECT '[-Infinity,1]'::minivec;
-ERROR:  "-Infinity" is out of range for type minivec
+ERROR:  infinite value not allowed in minivec
 LINE 1: SELECT '[-Infinity,1]'::minivec;
                ^
-SELECT '[65519,-65519]'::minivec;
+SELECT '[61439,-61439]'::minivec;
-ERROR:  "65519" is out of range for type minivec
+    minivec     
-LINE 1: SELECT '[65519,-65519]'::minivec;
+----------------
-               ^
+ [57344,-57344]
-SELECT '[65520,-65520]'::minivec;
+(1 row)
-ERROR:  "65520" is out of range for type minivec
+
-LINE 1: SELECT '[65520,-65520]'::minivec;
+SELECT '[61440,-61440]'::minivec;
+ERROR:  "61440" is out of range for type minivec
+LINE 1: SELECT '[61440,-61440]'::minivec;
                ^
 SELECT '[1e-8,-1e-8]'::minivec;
  minivec 
@@ -163,30 +165,30 @@ ERROR:  expected 2 dimensions, not 3
 SELECT '[1,2,3]'::minivec + '[4,5,6]';
  ?column? 
 ----------
- [5,7,9]
+ [5,7,8]
 (1 row)
 
-SELECT '[448]'::minivec + '[448]';
+SELECT '[61439]'::minivec + '[61439]';
 ERROR:  value out of range: overflow
 SELECT '[1,2]'::minivec + '[3]';
 ERROR:  different minivec dimensions 2 and 1
 SELECT '[1,2,3]'::minivec - '[4,5,6]';
-  ?column?  
+ERROR:  value out of range: overflow
-------------
+SELECT '[-61439]'::minivec - '[61439]';
- [-3,-3,-3]
+ ?column? 
+----------
+ [-0]
 (1 row)
 
-SELECT '[-448]'::minivec - '[448]';
-ERROR:  value out of range: overflow
 SELECT '[1,2]'::minivec - '[3]';
 ERROR:  different minivec dimensions 2 and 1
 SELECT '[1,2,3]'::minivec * '[4,5,6]';
  ?column?  
 -----------
- [4,10,18]
+ [4,10,16]
 (1 row)
 
-SELECT '[448]'::minivec * '[448]';
+SELECT '[61439]'::minivec * '[61439]';
 ERROR:  value out of range: overflow
 SELECT '[1e-7]'::minivec * '[1e-7]';
  ?column? 
@@ -397,7 +399,7 @@ SELECT inner_product('[448]'::minivec, '[448]');
 SELECT inner_product('[1,1,1,1,1,1,1,1,1]'::minivec, '[1,2,3,4,5,6,7,8,9]');
  inner_product 
 ---------------
-            45
+            44
 (1 row)
 
 SELECT '[1,2]'::minivec <#> '[3,4]';
@@ -491,7 +493,7 @@ SELECT l1_distance('[1,2,3,4,5,6,7,8,9]'::minivec, '[1,2,3,4,5,6,7,8,9]');
 SELECT l1_distance('[1,2,3,4,5,6,7,8,9]'::minivec, '[0,3,2,5,4,7,6,9,8]');
  l1_distance 
 -------------
-           9
+           7
 (1 row)
 
 SELECT '[0,0]'::minivec <+> '[3,4]';
@@ -501,9 +503,9 @@ SELECT '[0,0]'::minivec <+> '[3,4]';
 (1 row)
 
 SELECT l2_normalize('[3,4]'::minivec);
-  l2_normalize  
+ l2_normalize 
-----------------
+--------------
- [0.625,0.8125]
+ [0.625,0.75]
 (1 row)
 
 SELECT l2_normalize('[3,0]'::minivec);

test/sql/cast.sql

@@ -41,7 +41,7 @@ SELECT '{1e-8,-1e-8}'::real[]::halfvec;
 SELECT '[1,2,3]'::vector::minivec;
 SELECT '[1,2,3]'::vector::minivec(3);
 SELECT '[1,2,3]'::vector::minivec(2);
-SELECT '[465]'::vector::minivec;
+SELECT '[61440]'::vector::minivec;
 SELECT '[1e-8]'::vector::minivec;
 
 SELECT '[1,2,3]'::minivec::vector;
@@ -51,7 +51,7 @@ SELECT '[1,2,3]'::minivec::vector(2);
 SELECT '{1,2,3}'::real[]::minivec;
 SELECT '{1,2,3}'::real[]::minivec(3);
 SELECT '{1,2,3}'::real[]::minivec(2);
-SELECT '{465,-465}'::real[]::minivec;
+SELECT '{61440,-61440}'::real[]::minivec;
 SELECT '{1e-8,-1e-8}'::real[]::minivec;
 
 SELECT '[0,1.5,0,3.5,0]'::vector::sparsevec;

test/sql/ivfflat_minivec.sql

@@ -0,0 +1,45 @@
+SET enable_seqscan = off;
+
+-- L2
+
+CREATE TABLE t (val minivec(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING ivfflat (val minivec_l2_ops) WITH (lists = 1);
+
+INSERT INTO t (val) VALUES ('[1,2,4]');
+
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <-> (SELECT NULL::minivec)) t2;
+SELECT COUNT(*) FROM t;
+
+TRUNCATE t;
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+
+DROP TABLE t;
+
+-- inner product
+
+CREATE TABLE t (val minivec(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING ivfflat (val minivec_ip_ops) WITH (lists = 1);
+
+INSERT INTO t (val) VALUES ('[1,2,4]');
+
+SELECT * FROM t ORDER BY val <#> '[3,3,3]';
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <#> (SELECT NULL::minivec)) t2;
+
+DROP TABLE t;
+
+-- cosine
+
+CREATE TABLE t (val minivec(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING ivfflat (val minivec_cosine_ops) WITH (lists = 1);
+
+INSERT INTO t (val) VALUES ('[1,2,4]');
+
+SELECT * FROM t ORDER BY val <=> '[3,3,3]';
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> '[0,0,0]') t2;
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> (SELECT NULL::minivec)) t2;
+
+DROP TABLE t;

test/sql/minivec.sql

@@ -7,8 +7,8 @@ SELECT '[hello,1]'::minivec;
 SELECT '[NaN,1]'::minivec;
 SELECT '[Infinity,1]'::minivec;
 SELECT '[-Infinity,1]'::minivec;
-SELECT '[65519,-65519]'::minivec;
+SELECT '[61439,-61439]'::minivec;
-SELECT '[65520,-65520]'::minivec;
+SELECT '[61440,-61440]'::minivec;
 SELECT '[1e-8,-1e-8]'::minivec;
 SELECT '[4e38,1]'::minivec;
 SELECT '[1e-46,1]'::minivec;
@@ -38,15 +38,15 @@ SELECT unnest('{"[1,2,3]", "[4,5,6]"}'::minivec[]);
 SELECT '{"[1,2,3]"}'::minivec(2)[];
 
 SELECT '[1,2,3]'::minivec + '[4,5,6]';
-SELECT '[448]'::minivec + '[448]';
+SELECT '[61439]'::minivec + '[61439]';
 SELECT '[1,2]'::minivec + '[3]';
 
 SELECT '[1,2,3]'::minivec - '[4,5,6]';
-SELECT '[-448]'::minivec - '[448]';
+SELECT '[-61439]'::minivec - '[61439]';
 SELECT '[1,2]'::minivec - '[3]';
 
 SELECT '[1,2,3]'::minivec * '[4,5,6]';
-SELECT '[448]'::minivec * '[448]';
+SELECT '[61439]'::minivec * '[61439]';
 SELECT '[1e-7]'::minivec * '[1e-7]';
 SELECT '[1,2]'::minivec * '[3]';
 

test/t/033_comparison.pl

@@ -40,6 +40,10 @@ for (1 .. 50)
 	$actual = $node->safe_psql("postgres", "SELECT halfvec_cmp(v::halfvec, '$query'::real[]::halfvec) FROM tst");
 	is($expected, $actual);
 
+	# Test minivec
+	$actual = $node->safe_psql("postgres", "SELECT minivec_cmp(v::minivec, '$query'::real[]::minivec) FROM tst");
+	is($expected, $actual);
+
 	# Test sparsevec
 	$actual = $node->safe_psql("postgres", "SELECT sparsevec_cmp(v::vector::sparsevec, '$query'::real[]::vector::sparsevec) FROM tst");
 	is($expected, $actual);

test/t/034_distance_functions.pl

@@ -45,6 +45,10 @@ for my $function (@functions)
 		my $actual = $node->safe_psql("postgres", "SELECT $function(v::halfvec, '$query'::vector::halfvec) FROM tst");
 		is($expected, $actual, "halfvec $function");
 
+		# Test minivec
+		$actual = $node->safe_psql("postgres", "SELECT $function(v::minivec, '$query'::vector::minivec) FROM tst");
+		is($expected, $actual, "minivec $function");
+
 		# Test sparsevec
 		$actual = $node->safe_psql("postgres", "SELECT $function(v::sparsevec, '$query'::vector::sparsevec) FROM tst");
 		is($expected, $actual, "sparsevec $function");

test/t/037_inputs.pl

@@ -12,8 +12,8 @@ $node->start;
 # Create extension
 $node->safe_psql("postgres", "CREATE EXTENSION vector;");
 
-my @types = ("vector", "halfvec", "sparsevec");
+my @types = ("vector", "halfvec", "minivec", "sparsevec");
-my @inputs = ("[1.23,4.56,7.89]", "[1.23,4.56,7.89]", "{1:1.23,2:4.56,3:7.89}/3");
+my @inputs = ("[1.23,4.56,7.89]", "[1.23,4.56,7.89]", "[1.23,4.56,7.89]", "{1:1.23,2:4.56,3:7.89}/3");
 my @subs = (" ", " ", ",", ":", "-", "1", "9", "\0", "2147483648", "-2147483649");
 
 for my $i (0 .. $#types)

test/t/039_hnsw_minivec_build_recall.pl

@@ -95,10 +95,6 @@ for my $i (0 .. $#operators)
 
 	# Test approximate results
 	my $min = 0.98;
-	if ($operator eq '<=>')
-	{
-		$min = 0.65;
-	}
 	test_recall($min, $operator);
 
 	$node->safe_psql("postgres", "DROP INDEX idx;");

test/t/040_hnsw_minivec_insert_recall.pl

@@ -0,0 +1,113 @@
+use strict;
+use warnings FATAL => 'all';
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+use Test::More;
+
+my $node;
+my @queries = ();
+my @expected;
+my $limit = 20;
+my $dim = 10;
+my $array_sql = join(",", ('2 * random() * random()') x $dim);
+
+sub test_recall
+{
+	my ($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 = PostgreSQL::Test::Cluster->new('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i serial, v minivec($dim));");
+
+# Generate queries
+for (1 .. 20)
+{
+	my @r = ();
+	for (1 .. $dim)
+	{
+		push(@r, rand());
+	}
+	push(@queries, "[" . join(",", @r) . "]");
+}
+
+# Check each index type
+my @operators = ("<->", "<#>", "<=>", "<+>");
+my @opclasses = ("minivec_l2_ops", "minivec_ip_ops", "minivec_cosine_ops", "minivec_l1_ops");
+
+for my $i (0 .. $#operators)
+{
+	my $operator = $operators[$i];
+	my $opclass = $opclasses[$i];
+
+	# Add index
+	$node->safe_psql("postgres", "CREATE INDEX idx ON tst USING hnsw (v $opclass);");
+
+	# Use concurrent inserts
+	$node->pgbench(
+		"--no-vacuum --client=10 --transactions=1000",
+		0,
+		[qr{actually processed}],
+		[qr{^$}],
+		"concurrent INSERTs",
+		{
+			"040_hnsw_minivec_insert_recall_$opclass" => "INSERT INTO tst (v) VALUES (ARRAY[$array_sql]);"
+		}
+	);
+
+	# Get exact results
+	@expected = ();
+	foreach (@queries)
+	{
+		my $res = $node->safe_psql("postgres", qq(
+			SET enable_indexscan = off;
+			SELECT i FROM tst ORDER BY v $operator '$_' LIMIT $limit;
+		));
+		push(@expected, $res);
+	}
+
+	# Test approximate results
+	my $min = 0.98;
+	test_recall($min, $operator);
+
+	$node->safe_psql("postgres", "DROP INDEX idx;");
+	$node->safe_psql("postgres", "TRUNCATE tst;");
+}
+
+done_testing();

test/t/041_hnsw_minivec_vacuum_recall.pl

@@ -0,0 +1,97 @@
+use strict;
+use warnings FATAL => 'all';
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+use Test::More;
+
+my $node;
+my @queries = ();
+my @expected;
+my $limit = 20;
+
+sub test_recall
+{
+	my ($min, $ef_search, $test_name) = @_;
+	my $correct = 0;
+	my $total = 0;
+
+	my $explain = $node->safe_psql("postgres", qq(
+		SET enable_seqscan = off;
+		SET hnsw.ef_search = $ef_search;
+		EXPLAIN ANALYZE SELECT i FROM tst ORDER BY v <-> '$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;
+			SET hnsw.ef_search = $ef_search;
+			SELECT i FROM tst ORDER BY v <-> '$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, $test_name);
+}
+
+# Initialize node
+$node = PostgreSQL::Test::Cluster->new('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v minivec(3));");
+$node->safe_psql("postgres", "ALTER TABLE tst SET (autovacuum_enabled = false);");
+$node->safe_psql("postgres",
+	"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 10000) i;"
+);
+
+# Add index
+$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v minivec_l2_ops) WITH (m = 4, ef_construction = 8);");
+
+# Delete data
+$node->safe_psql("postgres", "DELETE FROM tst WHERE i > 2500;");
+
+# Generate queries
+for (1 .. 20)
+{
+	my $r1 = rand();
+	my $r2 = rand();
+	my $r3 = rand();
+	push(@queries, "[$r1,$r2,$r3]");
+}
+
+# Get exact results
+@expected = ();
+foreach (@queries)
+{
+	my $res = $node->safe_psql("postgres", qq(
+		SET enable_indexscan = off;
+		SELECT i FROM tst ORDER BY v <-> '$_' LIMIT $limit;
+	));
+	push(@expected, $res);
+}
+
+test_recall(0.18, $limit, "before vacuum");
+test_recall(0.93, 100, "before vacuum");
+
+# TODO Test concurrent inserts with vacuum
+$node->safe_psql("postgres", "VACUUM tst;");
+
+test_recall(0.95, $limit, "after vacuum");
+
+done_testing();

test/t/042_hnsw_minivec_duplicates.pl

@@ -0,0 +1,58 @@
+use strict;
+use warnings FATAL => 'all';
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+use Test::More;
+
+# Initialize node
+my $node = PostgreSQL::Test::Cluster->new('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (v minivec(3));");
+
+sub insert_vectors
+{
+	for my $i (1 .. 20)
+	{
+		$node->safe_psql("postgres", "INSERT INTO tst VALUES ('[1,1,1]');");
+	}
+}
+
+sub test_duplicates
+{
+	my $res = $node->safe_psql("postgres", qq(
+		SET enable_seqscan = off;
+		SET hnsw.ef_search = 1;
+		SELECT COUNT(*) FROM (SELECT * FROM tst ORDER BY v <-> '[1,1,1]') t;
+	));
+	is($res, 10);
+}
+
+# Test duplicates with build
+insert_vectors();
+$node->safe_psql("postgres", "CREATE INDEX idx ON tst USING hnsw (v minivec_l2_ops);");
+test_duplicates();
+
+# Reset
+$node->safe_psql("postgres", "TRUNCATE tst;");
+
+# Test duplicates with inserts
+insert_vectors();
+test_duplicates();
+
+# Test fallback path for inserts
+$node->pgbench(
+	"--no-vacuum --client=5 --transactions=100",
+	0,
+	[qr{actually processed}],
+	[qr{^$}],
+	"concurrent INSERTs",
+	{
+		"042_hnsw_minivec_duplicates" => "INSERT INTO tst VALUES ('[1,1,1]');"
+	}
+);
+
+done_testing();

test/t/043_ivfflat_minivec_build_recall.pl

@@ -0,0 +1,154 @@
+use strict;
+use warnings FATAL => 'all';
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+use Test::More;
+
+my $node;
+my @queries = ();
+my @expected;
+my $limit = 20;
+my $dim = 10;
+my $array_sql = join(",", ('random()') x $dim);
+
+sub test_recall
+{
+	my ($probes, $min, $operator) = @_;
+	my $correct = 0;
+	my $total = 0;
+
+	my $explain = $node->safe_psql("postgres", qq(
+		SET enable_seqscan = off;
+		SET ivfflat.probes = $probes;
+		EXPLAIN ANALYZE SELECT i FROM tst ORDER BY v $operator '$queries[0]' LIMIT $limit;
+	));
+	like($explain, qr/Index Scan using idx on tst/);
+
+	for my $i (0 .. $#queries)
+	{
+		my $actual = $node->safe_psql("postgres", qq(
+			SET enable_seqscan = off;
+			SET ivfflat.probes = $probes;
+			SELECT i FROM tst ORDER BY v $operator '$queries[$i]' LIMIT $limit;
+		));
+		my @actual_ids = split("\n", $actual);
+
+		my @expected_ids = split("\n", $expected[$i]);
+		my %expected_set = map { $_ => 1 } @expected_ids;
+
+		foreach (@actual_ids)
+		{
+			if (exists($expected_set{$_}))
+			{
+				$correct++;
+			}
+		}
+
+		$total += $limit;
+	}
+
+	cmp_ok($correct / $total, ">=", $min, $operator);
+}
+
+# Initialize node
+$node = PostgreSQL::Test::Cluster->new('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v minivec($dim));");
+$node->safe_psql("postgres",
+	"INSERT INTO tst SELECT i, ARRAY[$array_sql] FROM generate_series(1, 100000) i;"
+);
+
+# Generate queries
+for (1 .. 20)
+{
+	my @r = ();
+	for (1 .. $dim)
+	{
+		push(@r, rand());
+	}
+	push(@queries, "[" . join(",", @r) . "]");
+}
+
+# Check each index type
+my @operators = ("<->", "<#>", "<=>");
+my @opclasses = ("minivec_l2_ops", "minivec_ip_ops", "minivec_cosine_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", qq(
+			WITH top AS (
+				SELECT v $operator '$_' AS distance FROM tst ORDER BY distance LIMIT $limit
+			)
+			SELECT i FROM tst WHERE (v $operator '$_') <= (SELECT MAX(distance) FROM top)
+		));
+		push(@expected, $res);
+	}
+
+	# Build index serially
+	$node->safe_psql("postgres", qq(
+		SET max_parallel_maintenance_workers = 0;
+		CREATE INDEX idx ON tst USING ivfflat (v $opclass);
+	));
+
+	# Test approximate results
+	if ($operator ne "<#>")
+	{
+		# TODO Fix test (uniform random vectors all have similar inner product)
+		test_recall(1, 0.33, $operator);
+		test_recall(10, 0.93, $operator);
+	}
+
+	# Test probes equals lists
+	if ($operator eq "<=>")
+	{
+		test_recall(100, 0.98, $operator);
+	}
+	else
+	{
+		test_recall(100, 1.00, $operator);
+	}
+
+	$node->safe_psql("postgres", "DROP INDEX idx;");
+
+	# Build index in parallel
+	my ($ret, $stdout, $stderr) = $node->psql("postgres", qq(
+		SET client_min_messages = DEBUG;
+		SET min_parallel_table_scan_size = 1;
+		CREATE INDEX idx ON tst USING ivfflat (v $opclass);
+	));
+	is($ret, 0, $stderr);
+	like($stderr, qr/using \d+ parallel workers/);
+
+	# Test approximate results
+	if ($operator ne "<#>")
+	{
+		# TODO Fix test (uniform random vectors all have similar inner product)
+		test_recall(1, 0.33, $operator);
+		test_recall(10, 0.93, $operator);
+	}
+
+	# Test probes equals lists
+	if ($operator eq "<=>")
+	{
+		test_recall(100, 0.98, $operator);
+	}
+	else
+	{
+		test_recall(100, 1.00, $operator);
+	}
+
+	$node->safe_psql("postgres", "DROP INDEX idx;");
+}
+
+done_testing();