Updated readme [skip ci]

This takes the approach from commit a713e2acaa further. Once we have
remove a candidate from the "closer" set, we still don't need to
recalculate everything that follows. Any candidates that were in the
closer set before still only need to be compared with any new
candidates that we have added.

.github/workflows/build.yml

@@ -8,6 +8,10 @@ jobs:
       fail-fast: false
       matrix:
         include:
+          - postgres: 17
+            os: ubuntu-22.04
+          - postgres: 16
+            os: ubuntu-22.04
           - postgres: 15
             os: ubuntu-22.04
           - postgres: 14
@@ -17,14 +21,16 @@ jobs:
           - postgres: 12
             os: ubuntu-20.04
           - postgres: 11
-            os: ubuntu-18.04
+            os: ubuntu-20.04
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: ankane/setup-postgres@v1
         with:
           postgres-version: ${{ matrix.postgres }}
           dev-files: true
       - run: make
+        env:
+          PG_CFLAGS: -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
       - run: |
           export PG_CONFIG=`which pg_config`
           sudo --preserve-env=PG_CONFIG make install
@@ -39,11 +45,13 @@ jobs:
     runs-on: macos-latest
     if: ${{ !startsWith(github.ref_name, 'windows') }}
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: ankane/setup-postgres@v1
         with:
           postgres-version: 14
       - run: make
+        env:
+          PG_CFLAGS: -Wall -Wextra -Werror -Wno-unused-parameter
       - run: make install
       - run: make installcheck
       - if: ${{ failure() }}
@@ -54,11 +62,12 @@ jobs:
           wget -q https://github.com/postgres/postgres/archive/refs/tags/REL_14_5.tar.gz
           tar xf REL_14_5.tar.gz
       - run: make prove_installcheck PROVE_FLAGS="-I ./postgres-REL_14_5/src/test/perl" PERL5LIB="/Users/runner/perl5/lib/perl5"
+      - run: make clean && /usr/local/opt/llvm@15/bin/scan-build --status-bugs make
   windows:
     runs-on: windows-latest
     if: ${{ !startsWith(github.ref_name, 'mac') }}
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: ankane/setup-postgres@v1
         with:
           postgres-version: 14
@@ -70,3 +79,24 @@ jobs:
           nmake /NOLOGO /F Makefile.win clean && ^
           nmake /NOLOGO /F Makefile.win uninstall
         shell: cmd
+  i386:
+    if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
+    runs-on: ubuntu-latest
+    container:
+      image: debian:11
+      options: --platform linux/386
+    steps:
+      - run: apt-get update && apt-get install -y build-essential git libipc-run-perl postgresql-13 postgresql-server-dev-13 sudo
+      - run: service postgresql start
+      - run: |
+          git clone https://github.com/${{ github.repository }}.git pgvector
+          cd pgvector
+          git fetch origin ${{ github.ref }}
+          git reset --hard FETCH_HEAD
+          make
+          make install
+          chown -R postgres .
+          sudo -u postgres make installcheck
+          sudo -u postgres make prove_installcheck
+        env:
+          PG_CFLAGS: -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare

.gitignore

@@ -1,4 +1,5 @@
 /dist/
+/log/
 /results/
 /tmp_check/
 /sql/vector--?.?.?.sql
@@ -7,6 +8,7 @@ regression.*
 *.so
 *.bc
 *.dll
+*.dylib
 *.obj
 *.lib
 *.exp

CHANGELOG.md

@@ -1,3 +1,51 @@
+## 0.5.2 (unreleased)
+
+- Added support for on-disk parallel index builds for HNSW
+
+## 0.5.1 (2023-10-10)
+
+- Improved performance of HNSW index builds
+- Added check for MVCC-compliant snapshot for index scans
+
+## 0.5.0 (2023-08-28)
+
+- Added HNSW index type
+- Added support for parallel index builds for IVFFlat
+- Added `l1_distance` function
+- Added element-wise multiplication for vectors
+- Added `sum` aggregate
+- Improved performance of distance functions
+- Fixed out of range results for cosine distance
+- Fixed results for NULL and NaN distances for IVFFlat
+
+## 0.4.4 (2023-06-12)
+
+- Improved error message for malformed vector literal
+- Fixed segmentation fault with text input
+- Fixed consecutive delimiters with text input
+
+## 0.4.3 (2023-06-10)
+
+- Improved cost estimation
+- Improved support for spaces with text input
+- Fixed infinite and NaN values with binary input
+- Fixed infinite values with vector addition and subtraction
+- Fixed infinite values with list centers
+- Fixed compilation error when `float8` is pass by reference
+- Fixed compilation error on PowerPC
+- Fixed segmentation fault with index creation on i386
+
+## 0.4.2 (2023-05-13)
+
+- Added notice when index created with little data
+- Fixed dimensions check for some direct function calls
+- Fixed installation error with Postgres 12.0-12.2
+
+## 0.4.1 (2023-03-21)
+
+- Improved performance of cosine distance
+- Fixed index scan count
+
 ## 0.4.0 (2023-01-11)
 
 If upgrading with Postgres < 13, see [this note](https://github.com/pgvector/pgvector#040).

Dockerfile

@@ -1,9 +1,12 @@
-FROM postgres:15
+ARG PG_MAJOR=15
+FROM postgres:$PG_MAJOR
+ARG PG_MAJOR
 
 COPY . /tmp/pgvector
 
 RUN apt-get update && \
-		apt-get install -y --no-install-recommends build-essential postgresql-server-dev-15 && \
+		apt-mark hold locales && \
+		apt-get install -y --no-install-recommends build-essential postgresql-server-dev-$PG_MAJOR && \
 		cd /tmp/pgvector && \
 		make clean && \
 		make OPTFLAGS="" && \
@@ -11,6 +14,7 @@ RUN apt-get update && \
 		mkdir /usr/share/doc/pgvector && \
 		cp LICENSE README.md /usr/share/doc/pgvector && \
 		rm -r /tmp/pgvector && \
-		apt-get remove -y build-essential postgresql-server-dev-15 && \
+		apt-get remove -y build-essential postgresql-server-dev-$PG_MAJOR && \
 		apt-get autoremove -y && \
+		apt-mark unhold locales && \
 		rm -rf /var/lib/apt/lists/*

LICENSE

@@ -1,4 +1,4 @@
-Portions Copyright (c) 1996-2022, PostgreSQL Global Development Group
+Portions Copyright (c) 1996-2023, PostgreSQL Global Development Group
 
 Portions Copyright (c) 1994, The Regents of the University of California
 

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": "0.4.0",
+	"version": "0.5.1",
 	"maintainer": [
 		"Andrew Kane <andrew@ankane.org>"
 	],
@@ -20,7 +20,7 @@
 		"vector": {
 			"file": "sql/vector.sql",
 			"docfile": "README.md",
-			"version": "0.4.0",
+			"version": "0.5.1",
 			"abstract": "Open-source vector similarity search for Postgres"
 		}
 	},

Makefile

@@ -1,23 +1,30 @@
 EXTENSION = vector
-EXTVERSION = 0.4.0
+EXTVERSION = 0.5.1
 
 MODULE_big = vector
 DATA = $(wildcard sql/*--*.sql)
-OBJS = src/ivfbuild.o src/ivfflat.o src/ivfinsert.o src/ivfkmeans.o src/ivfscan.o src/ivfutils.o src/ivfvacuum.o src/vector.o
+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
 
 TESTS = $(wildcard test/sql/*.sql)
 REGRESS = $(patsubst test/sql/%.sql,%,$(TESTS))
-REGRESS_OPTS = --inputdir=test --load-extension=vector
+REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION)
 
 OPTFLAGS = -march=native
 
 # Mac ARM doesn't support -march=native
 ifeq ($(shell uname -s), Darwin)
 	ifeq ($(shell uname -p), arm)
+		# no difference with -march=armv8.5-a
 		OPTFLAGS =
 	endif
 endif
 
+# PowerPC doesn't support -march=native
+ifneq ($(filter ppc64%, $(shell uname -m)), )
+	OPTFLAGS =
+endif
+
 # For auto-vectorization:
 # - GCC (needs -ftree-vectorize OR -O3) - https://gcc.gnu.org/projects/tree-ssa/vectorization.html
 # - Clang (could use pragma instead) - https://llvm.org/docs/Vectorizers.html
@@ -61,4 +68,10 @@ dist:
 .PHONY: docker
 
 docker:
-	docker build --pull --no-cache -t ankane/pgvector:latest .
+	docker build --pull --no-cache --platform linux/amd64 -t ankane/pgvector:latest .
+
+.PHONY: docker-release
+
+docker-release:
+	docker buildx build --push --pull --no-cache --platform linux/amd64,linux/arm64 -t ankane/pgvector:latest .
+	docker buildx build --push --platform linux/amd64,linux/arm64 -t ankane/pgvector:v$(EXTVERSION) .

Makefile.win

@@ -1,10 +1,11 @@
 EXTENSION = vector
-EXTVERSION = 0.4.0
+EXTVERSION = 0.5.1
 
-OBJS = src\ivfbuild.obj src\ivfflat.obj src\ivfinsert.obj src\ivfkmeans.obj src\ivfscan.obj src\ivfutils.obj src\ivfvacuum.obj src\vector.obj
+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
 
 REGRESS = btree cast copy functions input ivfflat_cosine ivfflat_ip ivfflat_l2 ivfflat_options ivfflat_unlogged
-REGRESS_OPTS = --inputdir=test --load-extension=vector
+REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION)
 
 # For /arch flags
 # https://learn.microsoft.com/en-us/cpp/build/reference/arch-minimum-cpu-architecture
@@ -54,6 +55,8 @@ install:
 	copy $(SHLIB) "$(PKGLIBDIR)"
 	copy $(EXTENSION).control "$(SHAREDIR)\extension"
 	copy sql\$(EXTENSION)--*.sql "$(SHAREDIR)\extension"
+	mkdir "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
+	for %f in ($(HEADERS)) do copy %f "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
 
 installcheck:
 	"$(BINDIR)\pg_regress" --bindir="$(BINDIR)" $(REGRESS_OPTS) $(REGRESS)
@@ -61,7 +64,9 @@ installcheck:
 uninstall:
 	del /f "$(PKGLIBDIR)\$(SHLIB)"
 	del /f "$(SHAREDIR)\extension\$(EXTENSION).control"
-	del /f "$(SHAREDIR)\extension\vector--*.sql"
+	del /f "$(SHAREDIR)\extension\$(EXTENSION)--*.sql"
+	del /f "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)\*.h"
+	rmdir "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
 
 clean:
 	del /f $(SHLIB) $(EXTENSION).lib $(EXTENSION).exp

README.md

@@ -2,13 +2,13 @@
 
 Open-source vector similarity search for Postgres
 
-```sql
-CREATE TABLE items (embedding vector(3));
-CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops);
-SELECT * FROM items ORDER BY embedding <-> '[1,2,3]' LIMIT 5;
-```
+Store your vectors with the rest of your data. Supports:
 
-Supports L2 distance, inner product, and cosine distance
+- exact and approximate nearest neighbor search
+- L2 distance, inner product, and cosine distance
+- any [language](#languages) with a Postgres client
+
+Plus [ACID](https://en.wikipedia.org/wiki/ACID) compliance, point-in-time recovery, JOINs, and all of the other [great features](https://www.postgresql.org/about/) of Postgres
 
 [![Build Status](https://github.com/pgvector/pgvector/workflows/build/badge.svg?branch=master)](https://github.com/pgvector/pgvector/actions)
 
@@ -17,93 +17,200 @@ Supports L2 distance, inner product, and cosine distance
 Compile and install the extension (supports Postgres 11+)
 
 ```sh
-git clone --branch v0.4.0 https://github.com/pgvector/pgvector.git
+cd /tmp
+git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
 cd pgvector
 make
 make install # may need sudo
 ```
 
-Then load it in databases where you want to use it
+See the [installation notes](#installation-notes) if you run into issues
 
-```sql
-CREATE EXTENSION vector;
-```
-
-You can also install it with [Docker](#docker), [Homebrew](#homebrew), or [PGXN](#pgxn)
+You can also install it with [Docker](#docker), [Homebrew](#homebrew), [PGXN](#pgxn), [APT](#apt), [Yum](#yum), or [conda-forge](#conda-forge), and it comes preinstalled with [Postgres.app](#postgresapp) and many [hosted providers](#hosted-postgres). There are also instructions for [GitHub Actions](https://github.com/pgvector/setup-pgvector).
 
 ## Getting Started
 
+Enable the extension (do this once in each database where you want to use it)
+
+```tsql
+CREATE EXTENSION vector;
+```
+
 Create a vector column with 3 dimensions
 
 ```sql
-CREATE TABLE items (embedding vector(3));
+CREATE TABLE items (id bigserial PRIMARY KEY, embedding vector(3));
 ```
 
-Insert values
+Insert vectors
 
 ```sql
-INSERT INTO items VALUES ('[1,2,3]'), ('[4,5,6]');
+INSERT INTO items (embedding) VALUES ('[1,2,3]'), ('[4,5,6]');
 ```
 
-Get the nearest neighbor by L2 distance
+Get the nearest neighbors by L2 distance
 
 ```sql
-SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 1;
+SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
 ```
 
 Also supports inner product (`<#>`) and cosine distance (`<=>`)
 
 Note: `<#>` returns the negative inner product since Postgres only supports `ASC` order index scans on operators
 
+## Storing
+
+Create a new table with a vector column
+
+```sql
+CREATE TABLE items (id bigserial PRIMARY KEY, embedding vector(3));
+```
+
+Or add a vector column to an existing table
+
+```sql
+ALTER TABLE items ADD COLUMN embedding vector(3);
+```
+
+Insert vectors
+
+```sql
+INSERT INTO items (embedding) VALUES ('[1,2,3]'), ('[4,5,6]');
+```
+
+Upsert vectors
+
+```sql
+INSERT INTO items (id, embedding) VALUES (1, '[1,2,3]'), (2, '[4,5,6]')
+    ON CONFLICT (id) DO UPDATE SET embedding = EXCLUDED.embedding;
+```
+
+Update vectors
+
+```sql
+UPDATE items SET embedding = '[1,2,3]' WHERE id = 1;
+```
+
+Delete vectors
+
+```sql
+DELETE FROM items WHERE id = 1;
+```
+
+## Querying
+
+Get the nearest neighbors to a vector
+
+```sql
+SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
+```
+
+Get the nearest neighbors to a row
+
+```sql
+SELECT * FROM items WHERE id != 1 ORDER BY embedding <-> (SELECT embedding FROM items WHERE id = 1) LIMIT 5;
+```
+
+Get rows within a certain distance
+
+```sql
+SELECT * FROM items WHERE embedding <-> '[3,1,2]' < 5;
+```
+
+Note: Combine with `ORDER BY` and `LIMIT` to use an index
+
+#### Distances
+
+Get the distance
+
+```sql
+SELECT embedding <-> '[3,1,2]' AS distance FROM items;
+```
+
+For inner product, multiply by -1 (since `<#>` returns the negative inner product)
+
+```tsql
+SELECT (embedding <#> '[3,1,2]') * -1 AS inner_product FROM items;
+```
+
+For cosine similarity, use 1 - cosine distance
+
+```sql
+SELECT 1 - (embedding <=> '[3,1,2]') AS cosine_similarity FROM items;
+```
+
+#### Aggregates
+
+Average vectors
+
+```sql
+SELECT AVG(embedding) FROM items;
+```
+
+Average groups of vectors
+
+```sql
+SELECT category_id, AVG(embedding) FROM items GROUP BY category_id;
+```
+
 ## Indexing
 
-Speed up queries with an approximate index. Add an index for each distance function you want to use.
+By default, pgvector performs exact nearest neighbor search, which provides perfect recall.
+
+You can add an index to use approximate nearest neighbor search, which trades some recall for speed. Unlike typical indexes, you will see different results for queries after adding an approximate index.
+
+Supported index types are:
+
+- [IVFFlat](#ivfflat)
+- [HNSW](#hnsw) - added in 0.5.0
+
+## IVFFlat
+
+An IVFFlat index divides vectors into lists, and then searches a subset of those lists that are closest to the query vector. It has faster build times and uses less memory than HNSW, but has lower query performance (in terms of speed-recall tradeoff).
+
+Three keys to achieving good recall are:
+
+1. Create the index *after* the table has some data
+2. Choose an appropriate number of lists - a good place to start is `rows / 1000` for up to 1M rows and `sqrt(rows)` for over 1M rows
+3. When querying, specify an appropriate number of [probes](#query-options) (higher is better for recall, lower is better for speed) - a good place to start is `sqrt(lists)`
+
+Add an index for each distance function you want to use.
 
 L2 distance
 
-```sql
-CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops);
-```
-
-Inner product
-
-```sql
-CREATE INDEX ON items USING ivfflat (embedding vector_ip_ops);
-```
-
-Cosine distance
-
-```sql
-CREATE INDEX ON items USING ivfflat (embedding vector_cosine_ops);
-```
-
-Indexes should be created after the table has some data for optimal clustering. Also, unlike typical indexes which only affect performance, you may see different results for queries after adding an approximate index. Vectors with up to 2,000 dimensions can be indexed.
-
-### Index Options
-
-Specify the number of inverted lists (100 by default)
-
 ```sql
 CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 100);
 ```
 
-A [good place to start](https://github.com/facebookresearch/faiss/issues/112) is `4 * sqrt(rows)`
+Inner product
+
+```sql
+CREATE INDEX ON items USING ivfflat (embedding vector_ip_ops) WITH (lists = 100);
+```
+
+Cosine distance
+
+```sql
+CREATE INDEX ON items USING ivfflat (embedding vector_cosine_ops) WITH (lists = 100);
+```
+
+Vectors with up to 2,000 dimensions can be indexed.
 
 ### Query Options
 
 Specify the number of probes (1 by default)
 
 ```sql
-SET ivfflat.probes = 1;
+SET ivfflat.probes = 10;
 ```
 
-A higher value improves recall at the cost of speed.
+A higher value provides better recall at the cost of speed, and it can be set to the number of lists for exact nearest neighbor search (at which point the planner won’t use the index)
 
 Use `SET LOCAL` inside a transaction to set it for a single query
 
 ```sql
 BEGIN;
-SET LOCAL ivfflat.probes = 1;
+SET LOCAL ivfflat.probes = 10;
 SELECT ...
 COMMIT;
 ```
@@ -113,97 +220,182 @@ COMMIT;
 Check [indexing progress](https://www.postgresql.org/docs/current/progress-reporting.html#CREATE-INDEX-PROGRESS-REPORTING) with Postgres 12+
 
 ```sql
-SELECT phase, tuples_done, tuples_total FROM pg_stat_progress_create_index;
+SELECT phase, round(100.0 * tuples_done / nullif(tuples_total, 0), 1) AS "%" FROM pg_stat_progress_create_index;
 ```
 
-The phases are:
+The phases for IVFFlat are:
 
 1. `initializing`
 2. `performing k-means`
-3. `sorting tuples`
+3. `assigning tuples`
 4. `loading tuples`
 
-Note: `tuples_done` and `tuples_total` are only populated during the `loading tuples` phase
+Note: `%` is only populated during the `loading tuples` phase
 
-### Partial Indexes
+## HNSW
 
-Consider [partial indexes](https://www.postgresql.org/docs/current/indexes-partial.html) for queries with a `WHERE` clause
+An HNSW index creates a multilayer graph. It has slower build times and uses more memory than IVFFlat, but has better query performance (in terms of speed-recall tradeoff). There’s no training step like IVFFlat, so the index can be created without any data in the table.
+
+Add an index for each distance function you want to use.
+
+L2 distance
+
+```sql
+CREATE INDEX ON items USING hnsw (embedding vector_l2_ops);
+```
+
+Inner product
+
+```sql
+CREATE INDEX ON items USING hnsw (embedding vector_ip_ops);
+```
+
+Cosine distance
+
+```sql
+CREATE INDEX ON items USING hnsw (embedding vector_cosine_ops);
+```
+
+Vectors with up to 2,000 dimensions can be indexed.
+
+### Index Options
+
+Specify HNSW parameters
+
+- `m` - the max number of connections per layer (16 by default)
+- `ef_construction` - the size of the dynamic candidate list for constructing the graph (64 by default)
+
+```sql
+CREATE INDEX ON items USING hnsw (embedding vector_l2_ops) WITH (m = 16, ef_construction = 64);
+```
+
+### Query Options
+
+Specify the size of the dynamic candidate list for search (40 by default)
+
+```sql
+SET hnsw.ef_search = 100;
+```
+
+A higher value provides better recall at the cost of speed.
+
+Use `SET LOCAL` inside a transaction to set it for a single query
+
+```sql
+BEGIN;
+SET LOCAL hnsw.ef_search = 100;
+SELECT ...
+COMMIT;
+```
+
+### Indexing Progress
+
+Check [indexing progress](https://www.postgresql.org/docs/current/progress-reporting.html#CREATE-INDEX-PROGRESS-REPORTING) with Postgres 12+
+
+```sql
+SELECT phase, round(100.0 * blocks_done / nullif(blocks_total, 0), 1) AS "%" FROM pg_stat_progress_create_index;
+```
+
+The phases for HNSW are:
+
+1. `initializing`
+2. `loading tuples`
+
+## Filtering
+
+There are a few ways to index nearest neighbor queries with a `WHERE` clause
 
 ```sql
 SELECT * FROM items WHERE category_id = 123 ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
 ```
 
-can be indexed with:
+Create an index on one [or more](https://www.postgresql.org/docs/current/indexes-multicolumn.html) of the `WHERE` columns for exact search
 
 ```sql
-CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WHERE (category_id = 123);
+CREATE INDEX ON items (category_id);
 ```
 
-To index many different values of `category_id`, consider [partitioning](https://www.postgresql.org/docs/current/ddl-partitioning.html) on `category_id`.
+Or a [partial index](https://www.postgresql.org/docs/current/indexes-partial.html) on the vector column for approximate search
+
+```sql
+CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 100)
+    WHERE (category_id = 123);
+```
+
+Use [partitioning](https://www.postgresql.org/docs/current/ddl-partitioning.html) for approximate search on many different values of the `WHERE` columns
 
 ```sql
 CREATE TABLE items (embedding vector(3), category_id int) PARTITION BY LIST(category_id);
 ```
 
+## Hybrid Search
+
+Use together with Postgres [full-text search](https://www.postgresql.org/docs/current/textsearch-intro.html) for hybrid search.
+
+```sql
+SELECT id, content FROM items, plainto_tsquery('hello search') query
+    WHERE textsearch @@ query ORDER BY ts_rank_cd(textsearch, query) DESC LIMIT 5;
+```
+
+You can use [Reciprocal Rank Fusion](https://github.com/pgvector/pgvector-python/blob/master/examples/hybrid_search_rrf.py) or a [cross-encoder](https://github.com/pgvector/pgvector-python/blob/master/examples/hybrid_search.py) to combine results.
+
 ## Performance
 
+Use `EXPLAIN ANALYZE` to debug performance.
+
+```sql
+EXPLAIN ANALYZE SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
+```
+
+### Exact Search
+
 To speed up queries without an index, increase `max_parallel_workers_per_gather`.
 
 ```sql
 SET max_parallel_workers_per_gather = 4;
 ```
 
-To speed up queries with an index, increase the number of inverted lists (at the expense of recall).
+If vectors are normalized to length 1 (like [OpenAI embeddings](https://platform.openai.com/docs/guides/embeddings/which-distance-function-should-i-use)), use inner product for best performance.
+
+```tsql
+SELECT * FROM items ORDER BY embedding <#> '[3,1,2]' LIMIT 5;
+```
+
+### Approximate Search
+
+To speed up queries with an IVFFlat index, increase the number of inverted lists (at the expense of recall).
 
 ```sql
 CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 1000);
 ```
 
-## Reference
+## Languages
 
-### Vector Type
+Use pgvector from any language with a Postgres client. You can even generate and store vectors in one language and query them in another.
 
-Each vector takes `4 * dimensions + 8` bytes of storage. Each element is a single precision floating-point number (like the `real` type in Postgres), and all elements must be finite (no `NaN`, `Infinity` or `-Infinity`). Vectors can have up to 16,000 dimensions.
-
-### Vector Operators
-
-Operator | Description
+Language | Libraries / Examples
 --- | ---
-\+ | element-wise addition
-\- | element-wise subtraction
-<-> | Euclidean distance
-<#> | negative inner product
-<=> | cosine distance
-
-### Vector Functions
-
-Function | Description
---- | ---
-cosine_distance(vector, vector) → double precision | cosine distance
-inner_product(vector, vector) → double precision | inner product
-l2_distance(vector, vector) → double precision | Euclidean distance
-vector_dims(vector) → integer | number of dimensions
-vector_norm(vector) → double precision | Euclidean norm
-
-### Aggregate Functions
-
-Function | Description | Partial Mode
---- | --- | ---
-avg(vector) → vector | arithmetic mean | Yes
-
-## Libraries
-
-Libraries that use pgvector:
-
-- [pgvector-python](https://github.com/pgvector/pgvector-python) (Python)
-- [Neighbor](https://github.com/ankane/neighbor) (Ruby)
-- [pgvector-ruby](https://github.com/pgvector/pgvector-ruby) (Ruby)
-- [pgvector-node](https://github.com/pgvector/pgvector-node) (Node.js)
-- [pgvector-go](https://github.com/pgvector/pgvector-go) (Go)
-- [pgvector-php](https://github.com/pgvector/pgvector-php) (PHP)
-- [pgvector-rust](https://github.com/pgvector/pgvector-rust) (Rust)
-- [pgvector-cpp](https://github.com/pgvector/pgvector-cpp) (C++)
-- [pgvector-elixir](https://github.com/pgvector/pgvector-elixir) (Elixir)
+C | [pgvector-c](https://github.com/pgvector/pgvector-c)
+C++ | [pgvector-cpp](https://github.com/pgvector/pgvector-cpp)
+C# | [pgvector-dotnet](https://github.com/pgvector/pgvector-dotnet)
+Crystal | [pgvector-crystal](https://github.com/pgvector/pgvector-crystal)
+Dart | [pgvector-dart](https://github.com/pgvector/pgvector-dart)
+Elixir | [pgvector-elixir](https://github.com/pgvector/pgvector-elixir)
+Go | [pgvector-go](https://github.com/pgvector/pgvector-go)
+Haskell | [pgvector-haskell](https://github.com/pgvector/pgvector-haskell)
+Java, Kotlin, Groovy, Scala | [pgvector-java](https://github.com/pgvector/pgvector-java)
+JavaScript, TypeScript | [pgvector-node](https://github.com/pgvector/pgvector-node)
+Julia | [pgvector-julia](https://github.com/pgvector/pgvector-julia)
+Lua | [pgvector-lua](https://github.com/pgvector/pgvector-lua)
+Nim | [pgvector-nim](https://github.com/pgvector/pgvector-nim)
+Perl | [pgvector-perl](https://github.com/pgvector/pgvector-perl)
+PHP | [pgvector-php](https://github.com/pgvector/pgvector-php)
+Python | [pgvector-python](https://github.com/pgvector/pgvector-python)
+R | [pgvector-r](https://github.com/pgvector/pgvector-r)
+Ruby | [pgvector-ruby](https://github.com/pgvector/pgvector-ruby), [Neighbor](https://github.com/ankane/neighbor)
+Rust | [pgvector-rust](https://github.com/pgvector/pgvector-rust)
+Swift | [pgvector-swift](https://github.com/pgvector/pgvector-swift)
+Zig | [pgvector-zig](https://github.com/pgvector/pgvector-zig)
 
 ## Frequently Asked Questions
 
@@ -217,10 +409,180 @@ Yes, pgvector uses the write-ahead log (WAL), which allows for replication and p
 
 #### What if I want to index vectors with more than 2,000 dimensions?
 
-Two things you can try are:
+You’ll need to use [dimensionality reduction](https://en.wikipedia.org/wiki/Dimensionality_reduction) at the moment.
 
-1. use dimensionality reduction
-2. compile Postgres with a larger block size (`./configure --with-blocksize=32`) and edit the limit in `src/ivfflat.h`
+#### Can I store vectors with different dimensions in the same column?
+
+You can use `vector` as the type (instead of `vector(3)`).
+
+```sql
+CREATE TABLE embeddings (model_id bigint, item_id bigint, embedding vector, PRIMARY KEY (model_id, item_id));
+```
+
+However, you can only create indexes on rows with the same number of dimensions (using [expression](https://www.postgresql.org/docs/current/indexes-expressional.html) and [partial](https://www.postgresql.org/docs/current/indexes-partial.html) indexing):
+
+```sql
+CREATE INDEX ON embeddings USING hnsw ((embedding::vector(3)) vector_l2_ops) WHERE (model_id = 123);
+```
+
+and query with:
+
+```sql
+SELECT * FROM embeddings WHERE model_id = 123 ORDER BY embedding::vector(3) <-> '[3,1,2]' LIMIT 5;
+```
+
+#### Can I store vectors with more precision?
+
+You can use the `double precision[]` or `numeric[]` type to store vectors with more precision.
+
+```sql
+CREATE TABLE items (id bigserial PRIMARY KEY, embedding double precision[]);
+
+-- use {} instead of [] for Postgres arrays
+INSERT INTO items (embedding) VALUES ('{1,2,3}'), ('{4,5,6}');
+```
+
+Optionally, add a [check constraint](https://www.postgresql.org/docs/current/ddl-constraints.html) to ensure data can be converted to the `vector` type and has the expected dimensions.
+
+```sql
+ALTER TABLE items ADD CHECK (vector_dims(embedding::vector) = 3);
+```
+
+Use [expression indexing](https://www.postgresql.org/docs/current/indexes-expressional.html) to index (at a lower precision):
+
+```sql
+CREATE INDEX ON items USING hnsw ((embedding::vector(3)) vector_l2_ops);
+```
+
+and query with:
+
+```sql
+SELECT * FROM items ORDER BY embedding::vector(3) <-> '[3,1,2]' LIMIT 5;
+```
+
+## Troubleshooting
+
+#### Why isn’t a query using an index?
+
+The cost estimation in pgvector < 0.4.3 does not always work well with the planner. You can encourage the planner to use an index for a query with:
+
+```sql
+BEGIN;
+SET LOCAL enable_seqscan = off;
+SELECT ...
+COMMIT;
+```
+
+Also, if the table is small, a table scan may be faster.
+
+#### Why isn’t a query using a parallel table scan?
+
+The planner doesn’t consider [out-of-line storage](https://www.postgresql.org/docs/current/storage-toast.html) in cost estimates, which can make a serial scan look cheaper. You can reduce the cost of a parallel scan for a query with:
+
+```sql
+BEGIN;
+SET LOCAL min_parallel_table_scan_size = 1;
+SET LOCAL parallel_setup_cost = 1;
+SELECT ...
+COMMIT;
+```
+
+or choose to store vectors inline:
+
+```sql
+ALTER TABLE items ALTER COLUMN embedding SET STORAGE PLAIN;
+```
+
+#### Why are there less results for a query after adding an IVFFlat index?
+
+The index was likely created with too little data for the number of lists. Drop the index until the table has more data.
+
+```sql
+DROP INDEX index_name;
+```
+
+## Reference
+
+### Vector Type
+
+Each vector takes `4 * dimensions + 8` bytes of storage. Each element is a single precision floating-point number (like the `real` type in Postgres), and all elements must be finite (no `NaN`, `Infinity` or `-Infinity`). Vectors can have up to 16,000 dimensions.
+
+### Vector Operators
+
+Operator | Description | Added
+--- | --- | ---
+\+ | element-wise addition |
+\- | element-wise subtraction |
+\* | element-wise multiplication | 0.5.0
+<-> | Euclidean distance |
+<#> | negative inner product |
+<=> | cosine distance |
+
+### Vector Functions
+
+Function | Description | Added
+--- | --- | ---
+cosine_distance(vector, vector) → double precision | cosine distance |
+inner_product(vector, vector) → double precision | inner product |
+l2_distance(vector, vector) → double precision | Euclidean distance |
+l1_distance(vector, vector) → double precision | taxicab distance | 0.5.0
+vector_dims(vector) → integer | number of dimensions |
+vector_norm(vector) → double precision | Euclidean norm |
+
+### Aggregate Functions
+
+Function | Description | Added
+--- | --- | ---
+avg(vector) → vector | average |
+sum(vector) → vector | sum | 0.5.0
+
+## Installation Notes
+
+### Postgres Location
+
+If your machine has multiple Postgres installations, specify the path to [pg_config](https://www.postgresql.org/docs/current/app-pgconfig.html) with:
+
+```sh
+export PG_CONFIG=/Applications/Postgres.app/Contents/Versions/latest/bin/pg_config
+```
+
+Then re-run the installation instructions (run `make clean` before `make` if needed). If `sudo` is needed for `make install`, use:
+
+```sh
+sudo --preserve-env=PG_CONFIG make install
+```
+
+### Missing Header
+
+If compilation fails with `fatal error: postgres.h: No such file or directory`, make sure Postgres development files are installed on the server.
+
+For Ubuntu and Debian, use:
+
+```sh
+sudo apt install postgresql-server-dev-15
+```
+
+Note: Replace `15` with your Postgres server version
+
+### Windows
+
+Support for Windows is currently experimental. Ensure [C++ support in Visual Studio](https://learn.microsoft.com/en-us/cpp/build/building-on-the-command-line?view=msvc-170#download-and-install-the-tools) is installed, and run:
+
+```cmd
+call "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvars64.bat"
+```
+
+Note: The exact path will vary depending on your Visual Studio version and edition
+
+Then use `nmake` to build:
+
+```cmd
+set "PGROOT=C:\Program Files\PostgreSQL\15"
+git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
+cd pgvector
+nmake /F Makefile.win
+nmake /F Makefile.win install
+```
 
 ## Additional Installation Methods
 
@@ -232,14 +594,14 @@ Get the [Docker image](https://hub.docker.com/r/ankane/pgvector) with:
 docker pull ankane/pgvector
 ```
 
-This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres).
+This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (run it the same way).
 
-You can also build the image manually
+You can also build the image manually:
 
 ```sh
-git clone --branch v0.4.0 https://github.com/pgvector/pgvector.git
+git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
 cd pgvector
-docker build -t pgvector .
+docker build --build-arg PG_MAJOR=15 -t myuser/pgvector .
 ```
 
 ### Homebrew
@@ -247,9 +609,11 @@ docker build -t pgvector .
 With Homebrew Postgres, you can use:
 
 ```sh
-brew install pgvector/brew/pgvector
+brew install pgvector
 ```
 
+Note: This only adds it to the `postgresql@14` formula
+
 ### PGXN
 
 Install from the [PostgreSQL Extension Network](https://pgxn.org/dist/vector) with:
@@ -258,28 +622,75 @@ Install from the [PostgreSQL Extension Network](https://pgxn.org/dist/vector) wi
 pgxn install vector
 ```
 
+### APT
+
+Debian and Ubuntu packages are available from the [PostgreSQL APT Repository](https://wiki.postgresql.org/wiki/Apt). Follow the [setup instructions](https://wiki.postgresql.org/wiki/Apt#Quickstart) and run:
+
+```sh
+sudo apt install postgresql-15-pgvector
+```
+
+Note: Replace `15` with your Postgres server version
+
+### Yum
+
+RPM packages are available from the [PostgreSQL Yum Repository](https://yum.postgresql.org/). Follow the [setup instructions](https://www.postgresql.org/download/linux/redhat/) for your distribution and run:
+
+```sh
+sudo yum install pgvector_15
+# or
+sudo dnf install pgvector_15
+```
+
+Note: Replace `15` with your Postgres server version
+
+### conda-forge
+
+With Conda Postgres, install from [conda-forge](https://anaconda.org/conda-forge/pgvector) with:
+
+```sh
+conda install -c conda-forge pgvector
+```
+
+This method is [community-maintained](https://github.com/conda-forge/pgvector-feedstock) by [@mmcauliffe](https://github.com/mmcauliffe)
+
+### Postgres.app
+
+Download the [latest release](https://postgresapp.com/downloads.html) with Postgres 15+.
+
 ## Hosted Postgres
 
-Some Postgres providers only support specific extensions. To request a new extension:
-
-- Amazon RDS - follow the instructions on [this page](https://aws.amazon.com/rds/postgresql/faqs/)
-- Google Cloud SQL - follow the instructions on [this page](https://cloud.google.com/sql/docs/postgres/extensions#requesting-support-for-a-new-extension)
-- DigitalOcean Managed Databases - vote or comment on [this page](https://ideas.digitalocean.com/app-framework-services/p/pgvector-extension-for-postgresql)
-- Azure Database for PostgreSQL - vote or comment on [this page](https://feedback.azure.com/d365community/idea/7b423322-6189-ed11-a81b-000d3ae49307)
+pgvector is available on [these providers](https://github.com/pgvector/pgvector/issues/54).
 
 ## Upgrading
 
-Install the latest version and run:
+[Install](#installation) the latest version (use the same method as the original installation). Then in each database you want to upgrade, run:
 
 ```sql
 ALTER EXTENSION vector UPDATE;
 ```
 
+You can check the version in the current database with:
+
+```sql
+SELECT extversion FROM pg_extension WHERE extname = 'vector';
+```
+
 ## Upgrade Notes
 
+### 0.6.0
+
+If upgrading with Postgres < 13, remove this line from `sql/vector--0.5.1--0.6.0.sql`:
+
+```sql
+ALTER TYPE vector SET (STORAGE = external);
+```
+
+Then run `make install` and `ALTER EXTENSION vector UPDATE;`.
+
 ### 0.4.0
 
-For Postgres < 13, remove this line from `sql/vector--0.3.2--0.4.0.sql`:
+If upgrading with Postgres < 13, remove this line from `sql/vector--0.3.2--0.4.0.sql`:
 
 ```sql
 ALTER TYPE vector SET (STORAGE = extended);
@@ -307,9 +718,10 @@ Thanks to:
 
 - [PASE: PostgreSQL Ultra-High-Dimensional Approximate Nearest Neighbor Search Extension](https://dl.acm.org/doi/pdf/10.1145/3318464.3386131)
 - [Faiss: A Library for Efficient Similarity Search and Clustering of Dense Vectors](https://github.com/facebookresearch/faiss)
-- [Using the Triangle Inequality to Accelerate k-means](https://www.aaai.org/Papers/ICML/2003/ICML03-022.pdf)
+- [Using the Triangle Inequality to Accelerate k-means](https://cdn.aaai.org/ICML/2003/ICML03-022.pdf)
 - [k-means++: The Advantage of Careful Seeding](https://theory.stanford.edu/~sergei/papers/kMeansPP-soda.pdf)
 - [Concept Decompositions for Large Sparse Text Data using Clustering](https://www.cs.utexas.edu/users/inderjit/public_papers/concept_mlj.pdf)
+- [Efficient and Robust Approximate Nearest Neighbor Search using Hierarchical Navigable Small World Graphs](https://arxiv.org/ftp/arxiv/papers/1603/1603.09320.pdf)
 
 ## History
 
@@ -357,4 +769,4 @@ Resources for contributors
 
 - [Extension Building Infrastructure](https://www.postgresql.org/docs/current/extend-pgxs.html)
 - [Index Access Method Interface Definition](https://www.postgresql.org/docs/current/indexam.html)
-- [Generic WAL Records](https://www.postgresql.org/docs/13/generic-wal.html)
+- [Generic WAL Records](https://www.postgresql.org/docs/current/generic-wal.html)

sql/vector--0.4.0--0.4.1.sql

@@ -0,0 +1,2 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.4.1'" to load this file. \quit

sql/vector--0.4.1--0.4.2.sql

@@ -0,0 +1,2 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.4.2'" to load this file. \quit

sql/vector--0.4.2--0.4.3.sql

@@ -0,0 +1,2 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.4.3'" to load this file. \quit

sql/vector--0.4.3--0.4.4.sql

@@ -0,0 +1,2 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.4.4'" to load this file. \quit

sql/vector--0.4.4--0.5.0.sql

@@ -0,0 +1,43 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.5.0'" to load this file. \quit
+
+CREATE FUNCTION l1_distance(vector, vector) RETURNS float8
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE FUNCTION vector_mul(vector, vector) RETURNS vector
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
+CREATE OPERATOR * (
+	LEFTARG = vector, RIGHTARG = vector, PROCEDURE = vector_mul,
+	COMMUTATOR = *
+);
+
+CREATE AGGREGATE sum(vector) (
+	SFUNC = vector_add,
+	STYPE = vector,
+	COMBINEFUNC = vector_add,
+	PARALLEL = SAFE
+);
+
+CREATE FUNCTION hnswhandler(internal) RETURNS index_am_handler
+	AS 'MODULE_PATHNAME' LANGUAGE C;
+
+CREATE ACCESS METHOD hnsw TYPE INDEX HANDLER hnswhandler;
+
+COMMENT ON ACCESS METHOD hnsw IS 'hnsw index access method';
+
+CREATE OPERATOR CLASS vector_l2_ops
+	FOR TYPE vector USING hnsw AS
+	OPERATOR 1 <-> (vector, vector) FOR ORDER BY float_ops,
+	FUNCTION 1 vector_l2_squared_distance(vector, vector);
+
+CREATE OPERATOR CLASS vector_ip_ops
+	FOR TYPE vector USING hnsw AS
+	OPERATOR 1 <#> (vector, vector) FOR ORDER BY float_ops,
+	FUNCTION 1 vector_negative_inner_product(vector, vector);
+
+CREATE OPERATOR CLASS vector_cosine_ops
+	FOR TYPE vector USING hnsw AS
+	OPERATOR 1 <=> (vector, vector) FOR ORDER BY float_ops,
+	FUNCTION 1 vector_negative_inner_product(vector, vector),
+	FUNCTION 2 vector_norm(vector);

sql/vector--0.5.0--0.5.1.sql

@@ -0,0 +1,2 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.5.1'" to load this file. \quit

sql/vector--0.5.1--0.6.0.sql

@@ -0,0 +1,5 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.6.0'" to load this file. \quit
+
+-- remove this single line for Postgres < 13
+ALTER TYPE vector SET (STORAGE = external);

sql/vector.sql

@@ -26,7 +26,7 @@ CREATE TYPE vector (
 	TYPMOD_IN = vector_typmod_in,
 	RECEIVE   = vector_recv,
 	SEND      = vector_send,
-	STORAGE   = extended
+	STORAGE   = external
 );
 
 -- functions
@@ -40,6 +40,9 @@ CREATE FUNCTION inner_product(vector, vector) RETURNS float8
 CREATE FUNCTION cosine_distance(vector, vector) RETURNS float8
 	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
 
+CREATE FUNCTION l1_distance(vector, vector) RETURNS float8
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
 CREATE FUNCTION vector_dims(vector) RETURNS integer
 	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
 
@@ -52,6 +55,9 @@ CREATE FUNCTION vector_add(vector, vector) RETURNS vector
 CREATE FUNCTION vector_sub(vector, vector) RETURNS vector
 	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
 
+CREATE FUNCTION vector_mul(vector, vector) RETURNS vector
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
 -- private functions
 
 CREATE FUNCTION vector_lt(vector, vector) RETURNS bool
@@ -104,6 +110,13 @@ CREATE AGGREGATE avg(vector) (
 	PARALLEL = SAFE
 );
 
+CREATE AGGREGATE sum(vector) (
+	SFUNC = vector_add,
+	STYPE = vector,
+	COMBINEFUNC = vector_add,
+	PARALLEL = SAFE
+);
+
 -- cast functions
 
 CREATE FUNCTION vector(vector, integer, boolean) RETURNS vector
@@ -171,6 +184,11 @@ CREATE OPERATOR - (
 	COMMUTATOR = -
 );
 
+CREATE OPERATOR * (
+	LEFTARG = vector, RIGHTARG = vector, PROCEDURE = vector_mul,
+	COMMUTATOR = *
+);
+
 CREATE OPERATOR < (
 	LEFTARG = vector, RIGHTARG = vector, PROCEDURE = vector_lt,
 	COMMUTATOR = > , NEGATOR = >= ,
@@ -209,7 +227,7 @@ CREATE OPERATOR > (
 	RESTRICT = scalargtsel, JOIN = scalargtjoinsel
 );
 
--- access method
+-- access methods
 
 CREATE FUNCTION ivfflathandler(internal) RETURNS index_am_handler
 	AS 'MODULE_PATHNAME' LANGUAGE C;
@@ -218,6 +236,13 @@ CREATE ACCESS METHOD ivfflat TYPE INDEX HANDLER ivfflathandler;
 
 COMMENT ON ACCESS METHOD ivfflat IS 'ivfflat index access method';
 
+CREATE FUNCTION hnswhandler(internal) RETURNS index_am_handler
+	AS 'MODULE_PATHNAME' LANGUAGE C;
+
+CREATE ACCESS METHOD hnsw TYPE INDEX HANDLER hnswhandler;
+
+COMMENT ON ACCESS METHOD hnsw IS 'hnsw index access method';
+
 -- opclasses
 
 CREATE OPERATOR CLASS vector_ops
@@ -249,3 +274,19 @@ CREATE OPERATOR CLASS vector_cosine_ops
 	FUNCTION 2 vector_norm(vector),
 	FUNCTION 3 vector_spherical_distance(vector, vector),
 	FUNCTION 4 vector_norm(vector);
+
+CREATE OPERATOR CLASS vector_l2_ops
+	FOR TYPE vector USING hnsw AS
+	OPERATOR 1 <-> (vector, vector) FOR ORDER BY float_ops,
+	FUNCTION 1 vector_l2_squared_distance(vector, vector);
+
+CREATE OPERATOR CLASS vector_ip_ops
+	FOR TYPE vector USING hnsw AS
+	OPERATOR 1 <#> (vector, vector) FOR ORDER BY float_ops,
+	FUNCTION 1 vector_negative_inner_product(vector, vector);
+
+CREATE OPERATOR CLASS vector_cosine_ops
+	FOR TYPE vector USING hnsw AS
+	OPERATOR 1 <=> (vector, vector) FOR ORDER BY float_ops,
+	FUNCTION 1 vector_negative_inner_product(vector, vector),
+	FUNCTION 2 vector_norm(vector);

src/hnsw.c

@@ -0,0 +1,230 @@
+#include "postgres.h"
+
+#include <float.h>
+#include <math.h>
+
+#include "access/amapi.h"
+#include "commands/vacuum.h"
+#include "hnsw.h"
+#include "utils/guc.h"
+#include "utils/selfuncs.h"
+
+#if PG_VERSION_NUM >= 120000
+#include "commands/progress.h"
+#endif
+
+int			hnsw_ef_search;
+bool		hnsw_enable_parallel_build;
+static relopt_kind hnsw_relopt_kind;
+
+/*
+ * Initialize index options and variables
+ */
+void
+HnswInit(void)
+{
+	hnsw_relopt_kind = add_reloption_kind();
+	add_int_reloption(hnsw_relopt_kind, "m", "Max number of connections",
+					  HNSW_DEFAULT_M, HNSW_MIN_M, HNSW_MAX_M
+#if PG_VERSION_NUM >= 130000
+					  ,AccessExclusiveLock
+#endif
+		);
+	add_int_reloption(hnsw_relopt_kind, "ef_construction", "Size of the dynamic candidate list for construction",
+					  HNSW_DEFAULT_EF_CONSTRUCTION, HNSW_MIN_EF_CONSTRUCTION, HNSW_MAX_EF_CONSTRUCTION
+#if PG_VERSION_NUM >= 130000
+					  ,AccessExclusiveLock
+#endif
+		);
+
+	DefineCustomIntVariable("hnsw.ef_search", "Sets the size of the dynamic candidate list for search",
+							"Valid range is 1..1000.", &hnsw_ef_search,
+							HNSW_DEFAULT_EF_SEARCH, HNSW_MIN_EF_SEARCH, HNSW_MAX_EF_SEARCH, PGC_USERSET, 0, NULL, NULL, NULL);
+
+	/* Behind a variable for now since can be slower than building in memory */
+	DefineCustomBoolVariable("hnsw.enable_parallel_build", "Enables or disables building indexes in parallel",
+							 NULL, &hnsw_enable_parallel_build,
+							 false, PGC_USERSET, 0, NULL, NULL, NULL);
+}
+
+/*
+ * Get the name of index build phase
+ */
+#if PG_VERSION_NUM >= 120000
+static char *
+hnswbuildphasename(int64 phasenum)
+{
+	switch (phasenum)
+	{
+		case PROGRESS_CREATEIDX_SUBPHASE_INITIALIZE:
+			return "initializing";
+		case PROGRESS_HNSW_PHASE_LOAD:
+			return "loading tuples";
+		default:
+			return NULL;
+	}
+}
+#endif
+
+/*
+ * Estimate the cost of an index scan
+ */
+static void
+hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
+				 Cost *indexStartupCost, Cost *indexTotalCost,
+				 Selectivity *indexSelectivity, double *indexCorrelation,
+				 double *indexPages)
+{
+	GenericCosts costs;
+	int			m;
+	int			entryLevel;
+	Relation	index;
+#if PG_VERSION_NUM < 120000
+	List	   *qinfos;
+#endif
+
+	/* Never use index without order */
+	if (path->indexorderbys == NULL)
+	{
+		*indexStartupCost = DBL_MAX;
+		*indexTotalCost = DBL_MAX;
+		*indexSelectivity = 0;
+		*indexCorrelation = 0;
+		*indexPages = 0;
+		return;
+	}
+
+	MemSet(&costs, 0, sizeof(costs));
+
+	index = index_open(path->indexinfo->indexoid, NoLock);
+	HnswGetMetaPageInfo(index, &m, NULL);
+	index_close(index, NoLock);
+
+	/* Approximate entry level */
+	entryLevel = (int) -log(1.0 / path->indexinfo->tuples) * HnswGetMl(m);
+
+	/* TODO Improve estimate of visited tuples (currently underestimates) */
+	/* Account for number of tuples (or entry level), m, and ef_search */
+	costs.numIndexTuples = (entryLevel + 2) * m;
+
+#if PG_VERSION_NUM >= 120000
+	genericcostestimate(root, path, loop_count, &costs);
+#else
+	qinfos = deconstruct_indexquals(path);
+	genericcostestimate(root, path, loop_count, qinfos, &costs);
+#endif
+
+	/* Use total cost since most work happens before first tuple is returned */
+	*indexStartupCost = costs.indexTotalCost;
+	*indexTotalCost = costs.indexTotalCost;
+	*indexSelectivity = costs.indexSelectivity;
+	*indexCorrelation = costs.indexCorrelation;
+	*indexPages = costs.numIndexPages;
+}
+
+/*
+ * Parse and validate the reloptions
+ */
+static bytea *
+hnswoptions(Datum reloptions, bool validate)
+{
+	static const relopt_parse_elt tab[] = {
+		{"m", RELOPT_TYPE_INT, offsetof(HnswOptions, m)},
+		{"ef_construction", RELOPT_TYPE_INT, offsetof(HnswOptions, efConstruction)},
+	};
+
+#if PG_VERSION_NUM >= 130000
+	return (bytea *) build_reloptions(reloptions, validate,
+									  hnsw_relopt_kind,
+									  sizeof(HnswOptions),
+									  tab, lengthof(tab));
+#else
+	relopt_value *options;
+	int			numoptions;
+	HnswOptions *rdopts;
+
+	options = parseRelOptions(reloptions, validate, hnsw_relopt_kind, &numoptions);
+	rdopts = allocateReloptStruct(sizeof(HnswOptions), options, numoptions);
+	fillRelOptions((void *) rdopts, sizeof(HnswOptions), options, numoptions,
+				   validate, tab, lengthof(tab));
+
+	return (bytea *) rdopts;
+#endif
+}
+
+/*
+ * Validate catalog entries for the specified operator class
+ */
+static bool
+hnswvalidate(Oid opclassoid)
+{
+	return true;
+}
+
+/*
+ * Define index handler
+ *
+ * See https://www.postgresql.org/docs/current/index-api.html
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(hnswhandler);
+Datum
+hnswhandler(PG_FUNCTION_ARGS)
+{
+	IndexAmRoutine *amroutine = makeNode(IndexAmRoutine);
+
+	amroutine->amstrategies = 0;
+	amroutine->amsupport = 2;
+#if PG_VERSION_NUM >= 130000
+	amroutine->amoptsprocnum = 0;
+#endif
+	amroutine->amcanorder = false;
+	amroutine->amcanorderbyop = true;
+	amroutine->amcanbackward = false;	/* can change direction mid-scan */
+	amroutine->amcanunique = false;
+	amroutine->amcanmulticol = false;
+	amroutine->amoptionalkey = true;
+	amroutine->amsearcharray = false;
+	amroutine->amsearchnulls = false;
+	amroutine->amstorage = false;
+	amroutine->amclusterable = false;
+	amroutine->ampredlocks = false;
+	amroutine->amcanparallel = false;
+	amroutine->amcaninclude = false;
+#if PG_VERSION_NUM >= 130000
+	amroutine->amusemaintenanceworkmem = false; /* not used during VACUUM */
+	amroutine->amparallelvacuumoptions = VACUUM_OPTION_PARALLEL_BULKDEL;
+#endif
+	amroutine->amkeytype = InvalidOid;
+
+	/* Interface functions */
+	amroutine->ambuild = hnswbuild;
+	amroutine->ambuildempty = hnswbuildempty;
+	amroutine->aminsert = hnswinsert;
+	amroutine->ambulkdelete = hnswbulkdelete;
+	amroutine->amvacuumcleanup = hnswvacuumcleanup;
+	amroutine->amcanreturn = NULL;
+	amroutine->amcostestimate = hnswcostestimate;
+	amroutine->amoptions = hnswoptions;
+	amroutine->amproperty = NULL;	/* TODO AMPROP_DISTANCE_ORDERABLE */
+#if PG_VERSION_NUM >= 120000
+	amroutine->ambuildphasename = hnswbuildphasename;
+#endif
+	amroutine->amvalidate = hnswvalidate;
+#if PG_VERSION_NUM >= 140000
+	amroutine->amadjustmembers = NULL;
+#endif
+	amroutine->ambeginscan = hnswbeginscan;
+	amroutine->amrescan = hnswrescan;
+	amroutine->amgettuple = hnswgettuple;
+	amroutine->amgetbitmap = NULL;
+	amroutine->amendscan = hnswendscan;
+	amroutine->ammarkpos = NULL;
+	amroutine->amrestrpos = NULL;
+
+	/* Interface functions to support parallel index scans */
+	amroutine->amestimateparallelscan = NULL;
+	amroutine->aminitparallelscan = NULL;
+	amroutine->amparallelrescan = NULL;
+
+	PG_RETURN_POINTER(amroutine);
+}

src/hnsw.h

@@ -0,0 +1,363 @@
+#ifndef HNSW_H
+#define HNSW_H
+
+#include "postgres.h"
+
+#include "access/generic_xlog.h"
+#include "access/parallel.h"
+#include "access/reloptions.h"
+#include "nodes/execnodes.h"
+#include "port.h"				/* for random() */
+#include "utils/sampling.h"
+#include "vector.h"
+
+#if PG_VERSION_NUM < 110000
+#error "Requires PostgreSQL 11+"
+#endif
+
+#if PG_VERSION_NUM < 120000
+#include "access/relscan.h"
+#endif
+
+#define HNSW_MAX_DIM 2000
+
+/* Support functions */
+#define HNSW_DISTANCE_PROC 1
+#define HNSW_NORM_PROC 2
+
+#define HNSW_VERSION	1
+#define HNSW_MAGIC_NUMBER 0xA953A953
+#define HNSW_PAGE_ID	0xFF90
+
+/* Preserved page numbers */
+#define HNSW_METAPAGE_BLKNO	0
+#define HNSW_HEAD_BLKNO		1	/* first element page */
+
+/* Must correspond to page numbers since page lock is used */
+#define HNSW_UPDATE_LOCK 	0
+#define HNSW_SCAN_LOCK		1
+
+/* HNSW parameters */
+#define HNSW_DEFAULT_M	16
+#define HNSW_MIN_M	2
+#define HNSW_MAX_M		100
+#define HNSW_DEFAULT_EF_CONSTRUCTION	64
+#define HNSW_MIN_EF_CONSTRUCTION	4
+#define HNSW_MAX_EF_CONSTRUCTION		1000
+#define HNSW_DEFAULT_EF_SEARCH	40
+#define HNSW_MIN_EF_SEARCH		1
+#define HNSW_MAX_EF_SEARCH		1000
+
+/* Tuple types */
+#define HNSW_ELEMENT_TUPLE_TYPE  1
+#define HNSW_NEIGHBOR_TUPLE_TYPE 2
+
+/* Make graph robust against non-HOT updates */
+#define HNSW_HEAPTIDS 10
+
+#define HNSW_UPDATE_ENTRY_GREATER 1
+#define HNSW_UPDATE_ENTRY_ALWAYS 2
+
+/* Build phases */
+/* 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(size)	MAXALIGN(offsetof(HnswElementTupleData, data) + (size))
+#define HNSW_NEIGHBOR_TUPLE_SIZE(level, m)	MAXALIGN(offsetof(HnswNeighborTupleData, indextids) + ((level) + 2) * (m) * sizeof(ItemPointerData))
+
+#define HnswPageGetOpaque(page)	((HnswPageOpaque) PageGetSpecialPointer(page))
+#define HnswPageGetMeta(page)	((HnswMetaPageData *) PageGetContents(page))
+
+#if PG_VERSION_NUM >= 150000
+#define RandomDouble() pg_prng_double(&pg_global_prng_state)
+#else
+#define RandomDouble() (((double) random()) / MAX_RANDOM_VALUE)
+#endif
+
+#if PG_VERSION_NUM < 130000
+#define list_delete_last(list) list_truncate(list, list_length(list) - 1)
+#define list_sort(list, cmp) list_qsort(list, cmp)
+#endif
+
+#define HnswIsElementTuple(tup) ((tup)->type == HNSW_ELEMENT_TUPLE_TYPE)
+#define HnswIsNeighborTuple(tup) ((tup)->type == HNSW_NEIGHBOR_TUPLE_TYPE)
+
+/* 2 * M connections for ground layer */
+#define HnswGetLayerM(m, layer) (layer == 0 ? (m) * 2 : (m))
+
+/* Optimal ML from paper */
+#define HnswGetMl(m) (1 / log(m))
+
+/* Ensure fits on page and in uint8 */
+#define HnswGetMaxLevel(m) Min(((BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData)) - offsetof(HnswNeighborTupleData, indextids) - sizeof(ItemIdData)) / (sizeof(ItemPointerData)) / m) - 2, 255)
+
+/* Variables */
+extern int	hnsw_ef_search;
+extern bool hnsw_enable_parallel_build;
+
+typedef struct HnswNeighborArray HnswNeighborArray;
+
+typedef struct HnswElementData
+{
+	List	   *heaptids;
+	uint8		level;
+	uint8		deleted;
+	HnswNeighborArray *neighbors;
+	BlockNumber blkno;
+	OffsetNumber offno;
+	OffsetNumber neighborOffno;
+	BlockNumber neighborPage;
+	Datum		value;
+}			HnswElementData;
+
+typedef HnswElementData * HnswElement;
+
+typedef struct HnswCandidate
+{
+	HnswElement element;
+	float		distance;
+	bool		closer;
+}			HnswCandidate;
+
+typedef struct HnswNeighborArray
+{
+	int			length;
+	bool		closerSet;
+	HnswCandidate *items;
+}			HnswNeighborArray;
+
+typedef struct HnswPairingHeapNode
+{
+	pairingheap_node ph_node;
+	HnswCandidate *inner;
+}			HnswPairingHeapNode;
+
+/* HNSW index options */
+typedef struct HnswOptions
+{
+	int32		vl_len_;		/* varlena header (do not touch directly!) */
+	int			m;				/* number of connections */
+	int			efConstruction; /* size of dynamic candidate list */
+}			HnswOptions;
+
+typedef struct HnswSpool
+{
+	Relation	heap;
+	Relation	index;
+}			HnswSpool;
+
+typedef struct HnswShared
+{
+	/* Immutable state */
+	Oid			heaprelid;
+	Oid			indexrelid;
+	bool		isconcurrent;
+	int			scantuplesortstates;
+
+	/* Worker progress */
+	ConditionVariable workersdonecv;
+
+	/* Mutex for mutable state */
+	slock_t		mutex;
+
+	/* Mutable state */
+	int			nparticipantsdone;
+	double		reltuples;
+	double		indtuples;
+
+#if PG_VERSION_NUM < 120000
+	ParallelHeapScanDescData heapdesc;	/* must come last */
+#endif
+}			HnswShared;
+
+#if PG_VERSION_NUM >= 120000
+#define ParallelTableScanFromHnswShared(shared) \
+	(ParallelTableScanDesc) ((char *) (shared) + BUFFERALIGN(sizeof(HnswShared)))
+#endif
+
+typedef struct HnswLeader
+{
+	ParallelContext *pcxt;
+	int			nparticipanttuplesorts;
+	HnswShared *hnswshared;
+	Snapshot	snapshot;
+}			HnswLeader;
+
+typedef struct HnswBuildState
+{
+	/* Info */
+	Relation	heap;
+	Relation	index;
+	IndexInfo  *indexInfo;
+	ForkNumber	forkNum;
+
+	/* Settings */
+	int			dimensions;
+	int			m;
+	int			efConstruction;
+
+	/* Statistics */
+	double		indtuples;
+	double		reltuples;
+
+	/* Support functions */
+	FmgrInfo   *procinfo;
+	FmgrInfo   *normprocinfo;
+	Oid			collation;
+
+	/* Variables */
+	List	   *elements;
+	HnswElement entryPoint;
+	double		ml;
+	int			maxLevel;
+	long		memoryLeft;
+	bool		flushed;
+	Vector	   *normvec;
+
+	/* Memory */
+	MemoryContext tmpCtx;
+
+	/* Parallel builds */
+	HnswLeader *hnswleader;
+	HnswShared *hnswshared;
+}			HnswBuildState;
+
+typedef struct HnswMetaPageData
+{
+	uint32		magicNumber;
+	uint32		version;
+	uint32		dimensions;
+	uint16		m;
+	uint16		efConstruction;
+	BlockNumber entryBlkno;
+	OffsetNumber entryOffno;
+	int16		entryLevel;
+	BlockNumber insertPage;
+}			HnswMetaPageData;
+
+typedef HnswMetaPageData * HnswMetaPage;
+
+typedef struct HnswPageOpaqueData
+{
+	BlockNumber nextblkno;
+	uint16		unused;
+	uint16		page_id;		/* for identification of HNSW indexes */
+}			HnswPageOpaqueData;
+
+typedef HnswPageOpaqueData * HnswPageOpaque;
+
+typedef struct HnswElementTupleData
+{
+	uint8		type;
+	uint8		level;
+	uint8		deleted;
+	uint8		unused;
+	ItemPointerData heaptids[HNSW_HEAPTIDS];
+	ItemPointerData neighbortid;
+	uint16		unused2;
+	Vector		data;
+}			HnswElementTupleData;
+
+typedef HnswElementTupleData * HnswElementTuple;
+
+typedef struct HnswNeighborTupleData
+{
+	uint8		type;
+	uint8		unused;
+	uint16		count;
+	ItemPointerData indextids[FLEXIBLE_ARRAY_MEMBER];
+}			HnswNeighborTupleData;
+
+typedef HnswNeighborTupleData * HnswNeighborTuple;
+
+typedef struct HnswScanOpaqueData
+{
+	bool		first;
+	List	   *w;
+	MemoryContext tmpCtx;
+
+	/* Support functions */
+	FmgrInfo   *procinfo;
+	FmgrInfo   *normprocinfo;
+	Oid			collation;
+}			HnswScanOpaqueData;
+
+typedef HnswScanOpaqueData * HnswScanOpaque;
+
+typedef struct HnswVacuumState
+{
+	/* Info */
+	Relation	index;
+	IndexBulkDeleteResult *stats;
+	IndexBulkDeleteCallback callback;
+	void	   *callback_state;
+
+	/* Settings */
+	int			m;
+	int			efConstruction;
+
+	/* Support functions */
+	FmgrInfo   *procinfo;
+	Oid			collation;
+
+	/* Variables */
+	HTAB	   *deleted;
+	BufferAccessStrategy bas;
+	HnswNeighborTuple ntup;
+	HnswElementData highestPoint;
+
+	/* Memory */
+	MemoryContext tmpCtx;
+}			HnswVacuumState;
+
+/* Methods */
+int			HnswGetM(Relation index);
+int			HnswGetEfConstruction(Relation index);
+FmgrInfo   *HnswOptionalProcInfo(Relation index, uint16 procnum);
+bool		HnswNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
+void		HnswCommitBuffer(Buffer buf, GenericXLogState *state);
+Buffer		HnswNewBuffer(Relation index, ForkNumber forkNum);
+void		HnswInitPage(Buffer buf, Page page);
+void		HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state);
+void		HnswInit(void);
+List	   *HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, int m, bool inserting, HnswElement skipElement);
+HnswElement HnswGetEntryPoint(Relation index);
+void		HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint);
+HnswElement HnswInitElement(ItemPointer tid, int m, double ml, int maxLevel);
+void		HnswFreeElement(HnswElement element);
+HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno);
+void		HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, FmgrInfo *procinfo, Oid collation, int m, int efConstruction, bool existing);
+HnswElement HnswFindDuplicate(HnswElement e);
+HnswCandidate *HnswEntryCandidate(HnswElement em, Datum q, Relation rel, FmgrInfo *procinfo, Oid collation, bool loadVec);
+void		HnswUpdateMetaPage(Relation index, int updateEntry, HnswElement entryPoint, BlockNumber insertPage, ForkNumber forkNum);
+void		HnswSetNeighborTuple(HnswNeighborTuple ntup, HnswElement e, int m);
+void		HnswAddHeapTid(HnswElement element, ItemPointer heaptid);
+void		HnswInitNeighbors(HnswElement element, int m);
+bool		HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel);
+void		HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement e, int m, bool checkExisting);
+void		HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHeaptids, bool loadVec);
+void		HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec);
+void		HnswSetElementTuple(HnswElementTuple etup, HnswElement element);
+void		HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation);
+void		HnswLoadNeighbors(HnswElement element, Relation index, int m);
+PGDLLEXPORT void HnswParallelBuildMain(dsm_segment *seg, shm_toc *toc);
+
+/* Index access methods */
+IndexBuildResult *hnswbuild(Relation heap, Relation index, IndexInfo *indexInfo);
+void		hnswbuildempty(Relation index);
+bool		hnswinsert(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heap, IndexUniqueCheck checkUnique
+#if PG_VERSION_NUM >= 140000
+					   ,bool indexUnchanged
+#endif
+					   ,IndexInfo *indexInfo
+);
+IndexBulkDeleteResult *hnswbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats, IndexBulkDeleteCallback callback, void *callback_state);
+IndexBulkDeleteResult *hnswvacuumcleanup(IndexVacuumInfo *info, IndexBulkDeleteResult *stats);
+IndexScanDesc hnswbeginscan(Relation index, int nkeys, int norderbys);
+void		hnswrescan(IndexScanDesc scan, ScanKey keys, int nkeys, ScanKey orderbys, int norderbys);
+bool		hnswgettuple(IndexScanDesc scan, ScanDirection dir);
+void		hnswendscan(IndexScanDesc scan);
+
+#endif

src/hnswbuild.c

@@ -0,0 +1,926 @@
+#include "postgres.h"
+
+#include <math.h>
+
+#include "access/parallel.h"
+#include "access/xact.h"
+#include "catalog/index.h"
+#include "hnsw.h"
+#include "miscadmin.h"
+#include "lib/pairingheap.h"
+#include "nodes/pg_list.h"
+#include "storage/bufmgr.h"
+#include "tcop/tcopprot.h"
+#include "utils/datum.h"
+#include "utils/memutils.h"
+
+#if PG_VERSION_NUM >= 140000
+#include "utils/backend_progress.h"
+#elif PG_VERSION_NUM >= 120000
+#include "pgstat.h"
+#endif
+
+#if PG_VERSION_NUM >= 120000
+#include "access/tableam.h"
+#include "commands/progress.h"
+#else
+#define PROGRESS_CREATEIDX_TUPLES_DONE 0
+#endif
+
+#if PG_VERSION_NUM >= 130000
+#define CALLBACK_ITEM_POINTER ItemPointer tid
+#else
+#define CALLBACK_ITEM_POINTER HeapTuple hup
+#endif
+
+#if PG_VERSION_NUM >= 120000
+#define UpdateProgress(index, val) pgstat_progress_update_param(index, val)
+#else
+#define UpdateProgress(index, val) ((void)val)
+#endif
+
+#if PG_VERSION_NUM >= 140000
+#include "utils/backend_status.h"
+#include "utils/wait_event.h"
+#endif
+
+#if PG_VERSION_NUM >= 120000
+#include "access/table.h"
+#include "optimizer/optimizer.h"
+#else
+#include "access/heapam.h"
+#include "optimizer/planner.h"
+#include "pgstat.h"
+#endif
+
+#define PARALLEL_KEY_HNSW_SHARED		UINT64CONST(0xA000000000000001)
+#define PARALLEL_KEY_QUERY_TEXT			UINT64CONST(0xA000000000000002)
+
+/*
+ * Create the metapage
+ */
+static void
+CreateMetaPage(HnswBuildState * buildstate)
+{
+	Relation	index = buildstate->index;
+	ForkNumber	forkNum = buildstate->forkNum;
+	Buffer		buf;
+	Page		page;
+	GenericXLogState *state;
+	HnswMetaPage metap;
+
+	buf = HnswNewBuffer(index, forkNum);
+	HnswInitRegisterPage(index, &buf, &page, &state);
+
+	/* Set metapage data */
+	metap = HnswPageGetMeta(page);
+	metap->magicNumber = HNSW_MAGIC_NUMBER;
+	metap->version = HNSW_VERSION;
+	metap->dimensions = buildstate->dimensions;
+	metap->m = buildstate->m;
+	metap->efConstruction = buildstate->efConstruction;
+	metap->entryBlkno = InvalidBlockNumber;
+	metap->entryOffno = InvalidOffsetNumber;
+	metap->entryLevel = -1;
+	metap->insertPage = InvalidBlockNumber;
+	((PageHeader) page)->pd_lower =
+		((char *) metap + sizeof(HnswMetaPageData)) - (char *) page;
+
+	HnswCommitBuffer(buf, state);
+}
+
+/*
+ * Add a new page
+ */
+static void
+HnswBuildAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state, ForkNumber forkNum)
+{
+	/* Add a new page */
+	Buffer		newbuf = HnswNewBuffer(index, forkNum);
+
+	/* Update previous page */
+	HnswPageGetOpaque(*page)->nextblkno = BufferGetBlockNumber(newbuf);
+
+	/* Commit */
+	GenericXLogFinish(*state);
+	UnlockReleaseBuffer(*buf);
+
+	/* Can take a while, so ensure we can interrupt */
+	/* Needs to be called when no buffer locks are held */
+	LockBuffer(newbuf, BUFFER_LOCK_UNLOCK);
+	CHECK_FOR_INTERRUPTS();
+	LockBuffer(newbuf, BUFFER_LOCK_EXCLUSIVE);
+
+	/* Prepare new page */
+	*buf = newbuf;
+	*state = GenericXLogStart(index);
+	*page = GenericXLogRegisterBuffer(*state, *buf, GENERIC_XLOG_FULL_IMAGE);
+	HnswInitPage(*buf, *page);
+}
+
+/*
+ * Create element pages
+ */
+static void
+CreateElementPages(HnswBuildState * buildstate)
+{
+	Relation	index = buildstate->index;
+	ForkNumber	forkNum = buildstate->forkNum;
+	Size		etupAllocSize;
+	Size		maxSize;
+	HnswElementTuple etup;
+	HnswNeighborTuple ntup;
+	BlockNumber insertPage;
+	Buffer		buf;
+	Page		page;
+	GenericXLogState *state;
+	ListCell   *lc;
+
+	/* Calculate sizes */
+	etupAllocSize = BLCKSZ;
+	maxSize = HNSW_MAX_SIZE;
+
+	/* Allocate once */
+	etup = palloc0(etupAllocSize);
+	ntup = palloc0(BLCKSZ);
+
+	/* Prepare first page */
+	buf = HnswNewBuffer(index, forkNum);
+	state = GenericXLogStart(index);
+	page = GenericXLogRegisterBuffer(state, buf, GENERIC_XLOG_FULL_IMAGE);
+	HnswInitPage(buf, page);
+
+	foreach(lc, buildstate->elements)
+	{
+		HnswElement element = lfirst(lc);
+		Size		etupSize;
+		Size		ntupSize;
+		Size		combinedSize;
+
+		/* Zero memory for each element */
+		MemSet(etup, 0, etupAllocSize);
+
+		/* Calculate sizes */
+		etupSize = HNSW_ELEMENT_TUPLE_SIZE(VARSIZE_ANY(DatumGetPointer(element->value)));
+		ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(element->level, buildstate->m);
+		combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
+
+		/* Initial size check */
+		if (etupSize > etupAllocSize)
+			elog(ERROR, "index tuple too large");
+
+		HnswSetElementTuple(etup, element);
+
+		/* Keep element and neighbors on the same page if possible */
+		if (PageGetFreeSpace(page) < etupSize || (combinedSize <= maxSize && PageGetFreeSpace(page) < combinedSize))
+			HnswBuildAppendPage(index, &buf, &page, &state, forkNum);
+
+		/* Calculate offsets */
+		element->blkno = BufferGetBlockNumber(buf);
+		element->offno = OffsetNumberNext(PageGetMaxOffsetNumber(page));
+		if (combinedSize <= maxSize)
+		{
+			element->neighborPage = element->blkno;
+			element->neighborOffno = OffsetNumberNext(element->offno);
+		}
+		else
+		{
+			element->neighborPage = element->blkno + 1;
+			element->neighborOffno = FirstOffsetNumber;
+		}
+
+		ItemPointerSet(&etup->neighbortid, element->neighborPage, element->neighborOffno);
+
+		/* Add element */
+		if (PageAddItem(page, (Item) etup, etupSize, InvalidOffsetNumber, false, false) != element->offno)
+			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+		/* Add new page if needed */
+		if (PageGetFreeSpace(page) < ntupSize)
+			HnswBuildAppendPage(index, &buf, &page, &state, forkNum);
+
+		/* Add placeholder for neighbors */
+		if (PageAddItem(page, (Item) ntup, ntupSize, InvalidOffsetNumber, false, false) != element->neighborOffno)
+			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+	}
+
+	insertPage = BufferGetBlockNumber(buf);
+
+	/* Commit */
+	GenericXLogFinish(state);
+	UnlockReleaseBuffer(buf);
+
+	HnswUpdateMetaPage(index, HNSW_UPDATE_ENTRY_ALWAYS, buildstate->entryPoint, insertPage, forkNum);
+
+	pfree(etup);
+	pfree(ntup);
+}
+
+/*
+ * Create neighbor pages
+ */
+static void
+CreateNeighborPages(HnswBuildState * buildstate)
+{
+	Relation	index = buildstate->index;
+	ForkNumber	forkNum = buildstate->forkNum;
+	int			m = buildstate->m;
+	ListCell   *lc;
+	HnswNeighborTuple ntup;
+
+	/* Allocate once */
+	ntup = palloc0(BLCKSZ);
+
+	foreach(lc, buildstate->elements)
+	{
+		HnswElement e = lfirst(lc);
+		Buffer		buf;
+		Page		page;
+		GenericXLogState *state;
+		Size		ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m);
+
+		/* Can take a while, so ensure we can interrupt */
+		/* Needs to be called when no buffer locks are held */
+		CHECK_FOR_INTERRUPTS();
+
+		buf = ReadBufferExtended(index, forkNum, e->neighborPage, RBM_NORMAL, NULL);
+		LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
+		state = GenericXLogStart(index);
+		page = GenericXLogRegisterBuffer(state, buf, 0);
+
+		HnswSetNeighborTuple(ntup, e, m);
+
+		if (!PageIndexTupleOverwrite(page, e->neighborOffno, (Item) ntup, ntupSize))
+			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+		/* Commit */
+		GenericXLogFinish(state);
+		UnlockReleaseBuffer(buf);
+	}
+
+	pfree(ntup);
+}
+
+/*
+ * Free elements
+ */
+static void
+FreeElements(HnswBuildState * buildstate)
+{
+	ListCell   *lc;
+
+	foreach(lc, buildstate->elements)
+		HnswFreeElement(lfirst(lc));
+
+	list_free(buildstate->elements);
+}
+
+/*
+ * Flush pages
+ */
+static void
+FlushPages(HnswBuildState * buildstate)
+{
+	CreateMetaPage(buildstate);
+	CreateElementPages(buildstate);
+	CreateNeighborPages(buildstate);
+
+	buildstate->flushed = true;
+	FreeElements(buildstate);
+}
+
+/*
+ * Insert tuple
+ */
+static bool
+InsertTuple(Relation index, Datum *values, HnswElement element, HnswBuildState * buildstate, HnswElement * dup, MemoryContext outerCtx)
+{
+	FmgrInfo   *procinfo = buildstate->procinfo;
+	Oid			collation = buildstate->collation;
+	HnswElement entryPoint = buildstate->entryPoint;
+	int			efConstruction = buildstate->efConstruction;
+	int			m = buildstate->m;
+	MemoryContext oldCtx;
+
+	/* Detoast once for all calls */
+	Datum		value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
+
+	/* Normalize if needed */
+	if (buildstate->normprocinfo != NULL)
+	{
+		if (!HnswNormValue(buildstate->normprocinfo, collation, &value, buildstate->normvec))
+			return false;
+	}
+
+	/* Copy value to element so accessible outside of memory context */
+	oldCtx = MemoryContextSwitchTo(outerCtx);
+	element->value = datumCopy(value, false, -1);
+	MemoryContextSwitchTo(oldCtx);
+
+	/* Insert element in graph */
+	HnswInsertElement(element, entryPoint, NULL, procinfo, collation, m, efConstruction, false);
+
+	/* Look for duplicate */
+	*dup = HnswFindDuplicate(element);
+
+	/* Update neighbors if needed */
+	if (*dup == NULL)
+	{
+		for (int lc = element->level; lc >= 0; lc--)
+		{
+			int			lm = HnswGetLayerM(m, lc);
+			HnswNeighborArray *neighbors = &element->neighbors[lc];
+
+			for (int i = 0; i < neighbors->length; i++)
+				HnswUpdateConnection(element, &neighbors->items[i], lm, lc, NULL, NULL, procinfo, collation);
+		}
+	}
+
+	/* Update entry point if needed */
+	if (*dup == NULL && (entryPoint == NULL || element->level > entryPoint->level))
+		buildstate->entryPoint = element;
+
+	UpdateProgress(PROGRESS_CREATEIDX_TUPLES_DONE, ++buildstate->indtuples);
+
+	return *dup == NULL;
+}
+
+/*
+ * Get the memory used by an element
+ */
+static long
+HnswElementMemory(HnswElement e, int m)
+{
+	long		elementSize = sizeof(HnswElementData);
+
+	elementSize += sizeof(HnswNeighborArray) * (e->level + 1);
+	elementSize += sizeof(HnswCandidate) * (m * (e->level + 2));
+	elementSize += sizeof(ItemPointerData);
+	elementSize += VARSIZE_ANY(DatumGetPointer(e->value));
+	return elementSize;
+}
+
+/*
+ * Callback for table_index_build_scan
+ */
+static void
+BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
+			  bool *isnull, bool tupleIsAlive, void *state)
+{
+	HnswBuildState *buildstate = (HnswBuildState *) state;
+	MemoryContext oldCtx;
+	HnswElement element;
+	HnswElement dup = NULL;
+	bool		inserted;
+
+#if PG_VERSION_NUM < 130000
+	ItemPointer tid = &hup->t_self;
+#endif
+
+	/* Skip nulls */
+	if (isnull[0])
+		return;
+
+	if (buildstate->memoryLeft <= 0)
+	{
+		if (!buildstate->flushed)
+		{
+			ereport(NOTICE,
+					(errmsg("hnsw graph no longer fits into maintenance_work_mem after " INT64_FORMAT " tuples", (int64) buildstate->indtuples),
+					 errdetail("Building will take significantly more time."),
+					 errhint("Increase maintenance_work_mem to speed up builds.")));
+
+			FlushPages(buildstate);
+		}
+
+		oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
+
+		if (HnswInsertTuple(buildstate->index, values, isnull, tid, buildstate->heap))
+		{
+			if (buildstate->hnswshared)
+			{
+				HnswShared *hnswshared = buildstate->hnswshared;
+
+				SpinLockAcquire(&hnswshared->mutex);
+				UpdateProgress(PROGRESS_CREATEIDX_TUPLES_DONE, ++hnswshared->indtuples);
+				SpinLockRelease(&hnswshared->mutex);
+			}
+			else
+				UpdateProgress(PROGRESS_CREATEIDX_TUPLES_DONE, ++buildstate->indtuples);
+		}
+
+		/* Reset memory context */
+		MemoryContextSwitchTo(oldCtx);
+		MemoryContextReset(buildstate->tmpCtx);
+
+		return;
+	}
+
+	/* Allocate necessary memory outside of memory context */
+	element = HnswInitElement(tid, buildstate->m, buildstate->ml, buildstate->maxLevel);
+
+	/* Use memory context since detoast can allocate */
+	oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
+
+	/* Insert tuple */
+	inserted = InsertTuple(index, values, element, buildstate, &dup, oldCtx);
+
+	/* Reset memory context */
+	MemoryContextSwitchTo(oldCtx);
+	MemoryContextReset(buildstate->tmpCtx);
+
+	/* Add outside memory context */
+	if (dup != NULL)
+	{
+		HnswAddHeapTid(dup, tid);
+		buildstate->memoryLeft -= sizeof(ItemPointerData);
+	}
+
+	/* Add to buildstate or free */
+	if (inserted)
+	{
+		buildstate->elements = lappend(buildstate->elements, element);
+		buildstate->memoryLeft -= HnswElementMemory(element, buildstate->m);
+	}
+	else
+		HnswFreeElement(element);
+}
+
+/*
+ * Initialize the build state
+ */
+static void
+InitBuildState(HnswBuildState * buildstate, Relation heap, Relation index, IndexInfo *indexInfo, ForkNumber forkNum)
+{
+	buildstate->heap = heap;
+	buildstate->index = index;
+	buildstate->indexInfo = indexInfo;
+	buildstate->forkNum = forkNum;
+
+	buildstate->m = HnswGetM(index);
+	buildstate->efConstruction = HnswGetEfConstruction(index);
+	buildstate->dimensions = TupleDescAttr(index->rd_att, 0)->atttypmod;
+
+	/* Require column to have dimensions to be indexed */
+	if (buildstate->dimensions < 0)
+		elog(ERROR, "column does not have dimensions");
+
+	if (buildstate->dimensions > HNSW_MAX_DIM)
+		elog(ERROR, "column cannot have more than %d dimensions for hnsw index", HNSW_MAX_DIM);
+
+	if (buildstate->efConstruction < 2 * buildstate->m)
+		elog(ERROR, "ef_construction must be greater than or equal to 2 * m");
+
+	buildstate->reltuples = 0;
+	buildstate->indtuples = 0;
+
+	/* Get support functions */
+	buildstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
+	buildstate->normprocinfo = HnswOptionalProcInfo(index, HNSW_NORM_PROC);
+	buildstate->collation = index->rd_indcollation[0];
+
+	buildstate->elements = NIL;
+	buildstate->entryPoint = NULL;
+	buildstate->ml = HnswGetMl(buildstate->m);
+	buildstate->maxLevel = HnswGetMaxLevel(buildstate->m);
+	buildstate->memoryLeft = maintenance_work_mem * 1024L;
+	buildstate->flushed = false;
+
+	/* Reuse for each tuple */
+	buildstate->normvec = InitVector(buildstate->dimensions);
+
+	buildstate->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
+											   "Hnsw build temporary context",
+											   ALLOCSET_DEFAULT_SIZES);
+
+	buildstate->hnswleader = NULL;
+	buildstate->hnswshared = NULL;
+}
+
+/*
+ * Free resources
+ */
+static void
+FreeBuildState(HnswBuildState * buildstate)
+{
+	pfree(buildstate->normvec);
+	MemoryContextDelete(buildstate->tmpCtx);
+}
+
+/*
+ * Within leader, wait for end of heap scan
+ */
+static double
+ParallelHeapScan(HnswBuildState * buildstate)
+{
+	HnswShared *hnswshared = buildstate->hnswleader->hnswshared;
+	int			nparticipanttuplesorts;
+	double		reltuples;
+
+	nparticipanttuplesorts = buildstate->hnswleader->nparticipanttuplesorts;
+	for (;;)
+	{
+		SpinLockAcquire(&hnswshared->mutex);
+		if (hnswshared->nparticipantsdone == nparticipanttuplesorts)
+		{
+			buildstate->indtuples = hnswshared->indtuples;
+			reltuples = hnswshared->reltuples;
+			SpinLockRelease(&hnswshared->mutex);
+			break;
+		}
+		SpinLockRelease(&hnswshared->mutex);
+
+		ConditionVariableSleep(&hnswshared->workersdonecv,
+							   WAIT_EVENT_PARALLEL_CREATE_INDEX_SCAN);
+	}
+
+	ConditionVariableCancelSleep();
+
+	return reltuples;
+}
+
+/*
+ * Perform a worker's portion of a parallel insert
+ */
+static void
+HnswParallelScanAndInsert(HnswSpool * hnswspool, HnswShared * hnswshared, bool progress)
+{
+	HnswBuildState buildstate;
+#if PG_VERSION_NUM >= 120000
+	TableScanDesc scan;
+#else
+	HeapScanDesc scan;
+#endif
+	double		reltuples;
+	IndexInfo  *indexInfo;
+
+	/* Join parallel scan */
+	indexInfo = BuildIndexInfo(hnswspool->index);
+	indexInfo->ii_Concurrent = hnswshared->isconcurrent;
+	InitBuildState(&buildstate, hnswspool->heap, hnswspool->index, indexInfo, MAIN_FORKNUM);
+	/* TODO Support in-memory builds */
+	buildstate.memoryLeft = 0;
+	buildstate.flushed = true;
+	buildstate.hnswshared = hnswshared;
+#if PG_VERSION_NUM >= 120000
+	scan = table_beginscan_parallel(hnswspool->heap,
+									ParallelTableScanFromHnswShared(hnswshared));
+	reltuples = table_index_build_scan(hnswspool->heap, hnswspool->index, indexInfo,
+									   true, progress, BuildCallback,
+									   (void *) &buildstate, scan);
+#else
+	scan = heap_beginscan_parallel(hnswspool->heap, &hnswshared->heapdesc);
+	reltuples = IndexBuildHeapScan(hnswspool->heap, hnswspool->index, indexInfo,
+								   true, BuildCallback,
+								   (void *) &buildstate, scan);
+#endif
+
+	/* Record statistics */
+	SpinLockAcquire(&hnswshared->mutex);
+	hnswshared->nparticipantsdone++;
+	hnswshared->reltuples += reltuples;
+	SpinLockRelease(&hnswshared->mutex);
+
+	/* Log statistics */
+	if (progress)
+		ereport(DEBUG1, (errmsg("leader processed " INT64_FORMAT " tuples", (int64) reltuples)));
+	else
+		ereport(DEBUG1, (errmsg("worker processed " INT64_FORMAT " tuples", (int64) reltuples)));
+
+	/* Notify leader */
+	ConditionVariableSignal(&hnswshared->workersdonecv);
+
+	FreeBuildState(&buildstate);
+}
+
+/*
+ * Perform work within a launched parallel process
+ */
+void
+HnswParallelBuildMain(dsm_segment *seg, shm_toc *toc)
+{
+	char	   *sharedquery;
+	HnswSpool  *hnswspool;
+	HnswShared *hnswshared;
+	Relation	heapRel;
+	Relation	indexRel;
+	LOCKMODE	heapLockmode;
+	LOCKMODE	indexLockmode;
+
+	/* Set debug_query_string for individual workers first */
+	sharedquery = shm_toc_lookup(toc, PARALLEL_KEY_QUERY_TEXT, true);
+	debug_query_string = sharedquery;
+
+	/* Report the query string from leader */
+	pgstat_report_activity(STATE_RUNNING, debug_query_string);
+
+	/* Look up shared state */
+	hnswshared = shm_toc_lookup(toc, PARALLEL_KEY_HNSW_SHARED, false);
+
+	/* Open relations using lock modes known to be obtained by index.c */
+	if (!hnswshared->isconcurrent)
+	{
+		heapLockmode = ShareLock;
+		indexLockmode = AccessExclusiveLock;
+	}
+	else
+	{
+		heapLockmode = ShareUpdateExclusiveLock;
+		indexLockmode = RowExclusiveLock;
+	}
+
+	/* Open relations within worker */
+#if PG_VERSION_NUM >= 120000
+	heapRel = table_open(hnswshared->heaprelid, heapLockmode);
+#else
+	heapRel = heap_open(hnswshared->heaprelid, heapLockmode);
+#endif
+	indexRel = index_open(hnswshared->indexrelid, indexLockmode);
+
+	/* Initialize worker's own spool */
+	hnswspool = (HnswSpool *) palloc0(sizeof(HnswSpool));
+	hnswspool->heap = heapRel;
+	hnswspool->index = indexRel;
+
+	/* Perform inserts */
+	HnswParallelScanAndInsert(hnswspool, hnswshared, false);
+
+	/* Close relations within worker */
+	index_close(indexRel, indexLockmode);
+#if PG_VERSION_NUM >= 120000
+	table_close(heapRel, heapLockmode);
+#else
+	heap_close(heapRel, heapLockmode);
+#endif
+}
+
+/*
+ * End parallel build
+ */
+static void
+HnswEndParallel(HnswLeader * hnswleader)
+{
+	/* Shutdown worker processes */
+	WaitForParallelWorkersToFinish(hnswleader->pcxt);
+
+	/* Free last reference to MVCC snapshot, if one was used */
+	if (IsMVCCSnapshot(hnswleader->snapshot))
+		UnregisterSnapshot(hnswleader->snapshot);
+	DestroyParallelContext(hnswleader->pcxt);
+	ExitParallelMode();
+}
+
+/*
+ * Return size of shared memory required for parallel index build
+ */
+static Size
+ParallelEstimateShared(Relation heap, Snapshot snapshot)
+{
+#if PG_VERSION_NUM >= 120000
+	return add_size(BUFFERALIGN(sizeof(HnswShared)), table_parallelscan_estimate(heap, snapshot));
+#else
+	if (!IsMVCCSnapshot(snapshot))
+	{
+		Assert(snapshot == SnapshotAny);
+		return sizeof(HnswShared);
+	}
+
+	return add_size(offsetof(HnswShared, heapdesc) +
+					offsetof(ParallelHeapScanDescData, phs_snapshot_data),
+					EstimateSnapshotSpace(snapshot));
+#endif
+}
+
+/*
+ * Within leader, participate as a parallel worker
+ */
+static void
+HnswLeaderParticipateAsWorker(HnswBuildState * buildstate)
+{
+	HnswLeader *hnswleader = buildstate->hnswleader;
+	HnswSpool  *leaderworker;
+
+	/* Allocate memory and initialize private spool */
+	leaderworker = (HnswSpool *) palloc0(sizeof(HnswSpool));
+	leaderworker->heap = buildstate->heap;
+	leaderworker->index = buildstate->index;
+
+	/* Perform work common to all participants */
+	HnswParallelScanAndInsert(leaderworker, hnswleader->hnswshared, true);
+}
+
+/*
+ * Begin parallel build
+ */
+static void
+HnswBeginParallel(HnswBuildState * buildstate, bool isconcurrent, int request)
+{
+	ParallelContext *pcxt;
+	int			scantuplesortstates;
+	Snapshot	snapshot;
+	Size		esthnswshared;
+	HnswShared *hnswshared;
+	HnswLeader *hnswleader = (HnswLeader *) palloc0(sizeof(HnswLeader));
+	bool		leaderparticipates = true;
+	int			querylen;
+
+#ifdef DISABLE_LEADER_PARTICIPATION
+	leaderparticipates = false;
+#endif
+
+	/* Enter parallel mode and create context */
+	EnterParallelMode();
+	Assert(request > 0);
+#if PG_VERSION_NUM >= 120000
+	pcxt = CreateParallelContext("vector", "HnswParallelBuildMain", request);
+#else
+	pcxt = CreateParallelContext("vector", "HnswParallelBuildMain", request, true);
+#endif
+
+	scantuplesortstates = leaderparticipates ? request + 1 : request;
+
+	/* Get snapshot for table scan */
+	if (!isconcurrent)
+		snapshot = SnapshotAny;
+	else
+		snapshot = RegisterSnapshot(GetTransactionSnapshot());
+
+	/* Estimate size of workspaces */
+	esthnswshared = ParallelEstimateShared(buildstate->heap, snapshot);
+	shm_toc_estimate_chunk(&pcxt->estimator, esthnswshared);
+	shm_toc_estimate_keys(&pcxt->estimator, 1);
+
+	/* Finally, estimate PARALLEL_KEY_QUERY_TEXT space */
+	if (debug_query_string)
+	{
+		querylen = strlen(debug_query_string);
+		shm_toc_estimate_chunk(&pcxt->estimator, querylen + 1);
+		shm_toc_estimate_keys(&pcxt->estimator, 1);
+	}
+	else
+		querylen = 0;			/* keep compiler quiet */
+
+	/* Everyone's had a chance to ask for space, so now create the DSM */
+	InitializeParallelDSM(pcxt);
+
+	/* If no DSM segment was available, back out (do serial build) */
+	if (pcxt->seg == NULL)
+	{
+		if (IsMVCCSnapshot(snapshot))
+			UnregisterSnapshot(snapshot);
+		DestroyParallelContext(pcxt);
+		ExitParallelMode();
+		return;
+	}
+
+	/* Store shared build state, for which we reserved space */
+	hnswshared = (HnswShared *) shm_toc_allocate(pcxt->toc, esthnswshared);
+	/* Initialize immutable state */
+	hnswshared->heaprelid = RelationGetRelid(buildstate->heap);
+	hnswshared->indexrelid = RelationGetRelid(buildstate->index);
+	hnswshared->isconcurrent = isconcurrent;
+	hnswshared->scantuplesortstates = scantuplesortstates;
+	ConditionVariableInit(&hnswshared->workersdonecv);
+	SpinLockInit(&hnswshared->mutex);
+	/* Initialize mutable state */
+	hnswshared->nparticipantsdone = 0;
+	hnswshared->reltuples = 0;
+	hnswshared->indtuples = 0;
+#if PG_VERSION_NUM >= 120000
+	table_parallelscan_initialize(buildstate->heap,
+								  ParallelTableScanFromHnswShared(hnswshared),
+								  snapshot);
+#else
+	heap_parallelscan_initialize(&hnswshared->heapdesc, buildstate->heap, snapshot);
+#endif
+
+	shm_toc_insert(pcxt->toc, PARALLEL_KEY_HNSW_SHARED, hnswshared);
+
+	/* Store query string for workers */
+	if (debug_query_string)
+	{
+		char	   *sharedquery;
+
+		sharedquery = (char *) shm_toc_allocate(pcxt->toc, querylen + 1);
+		memcpy(sharedquery, debug_query_string, querylen + 1);
+		shm_toc_insert(pcxt->toc, PARALLEL_KEY_QUERY_TEXT, sharedquery);
+	}
+
+	/* Launch workers, saving status for leader/caller */
+	LaunchParallelWorkers(pcxt);
+	hnswleader->pcxt = pcxt;
+	hnswleader->nparticipanttuplesorts = pcxt->nworkers_launched;
+	if (leaderparticipates)
+		hnswleader->nparticipanttuplesorts++;
+	hnswleader->hnswshared = hnswshared;
+	hnswleader->snapshot = snapshot;
+
+	/* If no workers were successfully launched, back out (do serial build) */
+	if (pcxt->nworkers_launched == 0)
+	{
+		HnswEndParallel(hnswleader);
+		return;
+	}
+
+	/* Log participants */
+	ereport(DEBUG1, (errmsg("using %d parallel workers", pcxt->nworkers_launched)));
+
+	/* Save leader state now that it's clear build will be parallel */
+	buildstate->hnswleader = hnswleader;
+
+	/* Join heap scan ourselves */
+	if (leaderparticipates)
+		HnswLeaderParticipateAsWorker(buildstate);
+
+	/* Wait for all launched workers */
+	WaitForParallelWorkersToAttach(pcxt);
+}
+
+/*
+ * Build graph
+ */
+static void
+BuildGraph(HnswBuildState * buildstate, ForkNumber forkNum)
+{
+	int			parallel_workers = 0;
+
+	UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_HNSW_PHASE_LOAD);
+
+	/* Calculate parallel workers */
+	if (hnsw_enable_parallel_build)
+		parallel_workers = plan_create_index_workers(RelationGetRelid(buildstate->heap), RelationGetRelid(buildstate->index));
+
+	/* Attempt to launch parallel worker scan when required */
+	if (parallel_workers > 0)
+	{
+		/* TODO Support in-memory builds */
+		FlushPages(buildstate);
+		HnswBeginParallel(buildstate, buildstate->indexInfo->ii_Concurrent, parallel_workers);
+	}
+
+	/* Add tuples to sort */
+	if (buildstate->hnswleader)
+		buildstate->reltuples = ParallelHeapScan(buildstate);
+	else
+	{
+#if PG_VERSION_NUM >= 120000
+		buildstate->reltuples = table_index_build_scan(buildstate->heap, buildstate->index, buildstate->indexInfo,
+													   true, true, BuildCallback, (void *) buildstate, NULL);
+#else
+		buildstate->reltuples = IndexBuildHeapScan(buildstate->heap, buildstate->index, buildstate->indexInfo,
+												   true, BuildCallback, (void *) buildstate, NULL);
+#endif
+	}
+
+	/* End parallel build */
+	if (buildstate->hnswleader)
+		HnswEndParallel(buildstate->hnswleader);
+}
+
+/*
+ * Build the index
+ */
+static void
+BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
+		   HnswBuildState * buildstate, ForkNumber forkNum)
+{
+	InitBuildState(buildstate, heap, index, indexInfo, forkNum);
+
+	if (buildstate->heap != NULL)
+		BuildGraph(buildstate, forkNum);
+
+	if (!buildstate->flushed)
+		FlushPages(buildstate);
+
+	FreeBuildState(buildstate);
+}
+
+/*
+ * Build the index for a logged table
+ */
+IndexBuildResult *
+hnswbuild(Relation heap, Relation index, IndexInfo *indexInfo)
+{
+	IndexBuildResult *result;
+	HnswBuildState buildstate;
+
+	BuildIndex(heap, index, indexInfo, &buildstate, MAIN_FORKNUM);
+
+	result = (IndexBuildResult *) palloc(sizeof(IndexBuildResult));
+	result->heap_tuples = buildstate.reltuples;
+	result->index_tuples = buildstate.indtuples;
+
+	return result;
+}
+
+/*
+ * Build the index for an unlogged table
+ */
+void
+hnswbuildempty(Relation index)
+{
+	IndexInfo  *indexInfo = BuildIndexInfo(index);
+	HnswBuildState buildstate;
+
+	BuildIndex(NULL, index, indexInfo, &buildstate, INIT_FORKNUM);
+}

src/hnswinsert.c

@@ -0,0 +1,584 @@
+#include "postgres.h"
+
+#include <math.h>
+
+#include "hnsw.h"
+#include "storage/bufmgr.h"
+#include "storage/lmgr.h"
+#include "utils/memutils.h"
+
+/*
+ * Get the insert page
+ */
+static BlockNumber
+GetInsertPage(Relation index)
+{
+	Buffer		buf;
+	Page		page;
+	HnswMetaPage metap;
+	BlockNumber insertPage;
+
+	buf = ReadBuffer(index, HNSW_METAPAGE_BLKNO);
+	LockBuffer(buf, BUFFER_LOCK_SHARE);
+	page = BufferGetPage(buf);
+	metap = HnswPageGetMeta(page);
+
+	insertPage = metap->insertPage;
+
+	UnlockReleaseBuffer(buf);
+
+	return insertPage;
+}
+
+/*
+ * Check for a free offset
+ */
+static bool
+HnswFreeOffset(Relation index, Buffer buf, Page page, HnswElement element, Size ntupSize, Buffer *nbuf, Page *npage, OffsetNumber *freeOffno, OffsetNumber *freeNeighborOffno, BlockNumber *newInsertPage)
+{
+	OffsetNumber offno;
+	OffsetNumber maxoffno = PageGetMaxOffsetNumber(page);
+
+	for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+	{
+		HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, PageGetItemId(page, offno));
+
+		/* Skip neighbor tuples */
+		if (!HnswIsElementTuple(etup))
+			continue;
+
+		if (etup->deleted)
+		{
+			BlockNumber elementPage = BufferGetBlockNumber(buf);
+			BlockNumber neighborPage = ItemPointerGetBlockNumber(&etup->neighbortid);
+			OffsetNumber neighborOffno = ItemPointerGetOffsetNumber(&etup->neighbortid);
+			ItemId		itemid;
+
+			if (!BlockNumberIsValid(*newInsertPage))
+				*newInsertPage = elementPage;
+
+			if (neighborPage == elementPage)
+			{
+				*nbuf = buf;
+				*npage = page;
+			}
+			else
+			{
+				*nbuf = ReadBuffer(index, neighborPage);
+				LockBuffer(*nbuf, BUFFER_LOCK_EXCLUSIVE);
+
+				/* Skip WAL for now */
+				*npage = BufferGetPage(*nbuf);
+			}
+
+			itemid = PageGetItemId(*npage, neighborOffno);
+
+			/* Check for space on neighbor tuple page */
+			if (PageGetFreeSpace(*npage) + ItemIdGetLength(itemid) - sizeof(ItemIdData) >= ntupSize)
+			{
+				*freeOffno = offno;
+				*freeNeighborOffno = neighborOffno;
+				return true;
+			}
+			else if (*nbuf != buf)
+				UnlockReleaseBuffer(*nbuf);
+		}
+	}
+
+	return false;
+}
+
+/*
+ * Add a new page
+ */
+static void
+HnswInsertAppendPage(Relation index, Buffer *nbuf, Page *npage, GenericXLogState *state, Page page)
+{
+	/* Add a new page */
+	LockRelationForExtension(index, ExclusiveLock);
+	*nbuf = HnswNewBuffer(index, MAIN_FORKNUM);
+	UnlockRelationForExtension(index, ExclusiveLock);
+
+	/* Init new page */
+	*npage = GenericXLogRegisterBuffer(state, *nbuf, GENERIC_XLOG_FULL_IMAGE);
+	HnswInitPage(*nbuf, *npage);
+
+	/* Update previous buffer */
+	HnswPageGetOpaque(page)->nextblkno = BufferGetBlockNumber(*nbuf);
+}
+
+/*
+ * Add to element and neighbor pages
+ */
+static void
+WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPage, BlockNumber *updatedInsertPage)
+{
+	Buffer		buf;
+	Page		page;
+	GenericXLogState *state;
+	Size		etupSize;
+	Size		ntupSize;
+	Size		combinedSize;
+	Size		maxSize;
+	Size		minCombinedSize;
+	HnswElementTuple etup;
+	BlockNumber currentPage = insertPage;
+	HnswNeighborTuple ntup;
+	Buffer		nbuf;
+	Page		npage;
+	OffsetNumber freeOffno = InvalidOffsetNumber;
+	OffsetNumber freeNeighborOffno = InvalidOffsetNumber;
+	BlockNumber newInsertPage = InvalidBlockNumber;
+
+	/* Calculate sizes */
+	etupSize = HNSW_ELEMENT_TUPLE_SIZE(VARSIZE_ANY(DatumGetPointer(e->value)));
+	ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m);
+	combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
+	maxSize = HNSW_MAX_SIZE;
+	minCombinedSize = etupSize + HNSW_NEIGHBOR_TUPLE_SIZE(0, m) + sizeof(ItemIdData);
+
+	/* Prepare element tuple */
+	etup = palloc0(etupSize);
+	HnswSetElementTuple(etup, e);
+
+	/* Prepare neighbor tuple */
+	ntup = palloc0(ntupSize);
+	HnswSetNeighborTuple(ntup, e, m);
+
+	/* Find a page (or two if needed) to insert the tuples */
+	for (;;)
+	{
+		buf = ReadBuffer(index, currentPage);
+		LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
+
+		state = GenericXLogStart(index);
+		page = GenericXLogRegisterBuffer(state, buf, 0);
+
+		/* Keep track of first page where element at level 0 can fit */
+		if (!BlockNumberIsValid(newInsertPage) && PageGetFreeSpace(page) >= minCombinedSize)
+			newInsertPage = currentPage;
+
+		/* First, try the fastest path */
+		/* Space for both tuples on the current page */
+		/* This can split existing tuples in rare cases */
+		if (PageGetFreeSpace(page) >= combinedSize)
+		{
+			nbuf = buf;
+			npage = page;
+			break;
+		}
+
+		/* Next, try space from a deleted element */
+		if (HnswFreeOffset(index, buf, page, e, ntupSize, &nbuf, &npage, &freeOffno, &freeNeighborOffno, &newInsertPage))
+		{
+			if (nbuf != buf)
+				npage = GenericXLogRegisterBuffer(state, nbuf, 0);
+
+			break;
+		}
+
+		/* Finally, try space for element only if last page */
+		/* Skip if both tuples can fit on the same page */
+		if (combinedSize > maxSize && PageGetFreeSpace(page) >= etupSize && !BlockNumberIsValid(HnswPageGetOpaque(page)->nextblkno))
+		{
+			HnswInsertAppendPage(index, &nbuf, &npage, state, page);
+			break;
+		}
+
+		currentPage = HnswPageGetOpaque(page)->nextblkno;
+
+		if (BlockNumberIsValid(currentPage))
+		{
+			/* Move to next page */
+			GenericXLogAbort(state);
+			UnlockReleaseBuffer(buf);
+		}
+		else
+		{
+			Buffer		newbuf;
+			Page		newpage;
+
+			HnswInsertAppendPage(index, &newbuf, &newpage, state, page);
+
+			/* Commit */
+			GenericXLogFinish(state);
+
+			/* Unlock previous buffer */
+			UnlockReleaseBuffer(buf);
+
+			/* Prepare new buffer */
+			state = GenericXLogStart(index);
+			buf = newbuf;
+			page = GenericXLogRegisterBuffer(state, buf, 0);
+
+			/* Create new page for neighbors if needed */
+			if (PageGetFreeSpace(page) < combinedSize)
+				HnswInsertAppendPage(index, &nbuf, &npage, state, page);
+			else
+			{
+				nbuf = buf;
+				npage = page;
+			}
+
+			break;
+		}
+	}
+
+	e->blkno = BufferGetBlockNumber(buf);
+	e->neighborPage = BufferGetBlockNumber(nbuf);
+
+	/* Added tuple to new page if newInsertPage is not set */
+	/* So can set to neighbor page instead of element page */
+	if (!BlockNumberIsValid(newInsertPage))
+		newInsertPage = e->neighborPage;
+
+	if (OffsetNumberIsValid(freeOffno))
+	{
+		e->offno = freeOffno;
+		e->neighborOffno = freeNeighborOffno;
+	}
+	else
+	{
+		e->offno = OffsetNumberNext(PageGetMaxOffsetNumber(page));
+		if (nbuf == buf)
+			e->neighborOffno = OffsetNumberNext(e->offno);
+		else
+			e->neighborOffno = FirstOffsetNumber;
+	}
+
+	ItemPointerSet(&etup->neighbortid, e->neighborPage, e->neighborOffno);
+
+	/* Add element and neighbors */
+	if (OffsetNumberIsValid(freeOffno))
+	{
+		if (!PageIndexTupleOverwrite(page, e->offno, (Item) etup, etupSize))
+			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+		if (!PageIndexTupleOverwrite(npage, e->neighborOffno, (Item) ntup, ntupSize))
+			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+	}
+	else
+	{
+		if (PageAddItem(page, (Item) etup, etupSize, InvalidOffsetNumber, false, false) != e->offno)
+			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+		if (PageAddItem(npage, (Item) ntup, ntupSize, InvalidOffsetNumber, false, false) != e->neighborOffno)
+			elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+	}
+
+	/* Commit */
+	GenericXLogFinish(state);
+	UnlockReleaseBuffer(buf);
+	if (nbuf != buf)
+		UnlockReleaseBuffer(nbuf);
+
+	/* Update the insert page */
+	if (BlockNumberIsValid(newInsertPage) && newInsertPage != insertPage)
+		*updatedInsertPage = newInsertPage;
+}
+
+/*
+ * Check if connection already exists
+ */
+static bool
+ConnectionExists(HnswElement e, HnswNeighborTuple ntup, int startIdx, int lm)
+{
+	for (int i = 0; i < lm; i++)
+	{
+		ItemPointer indextid = &ntup->indextids[startIdx + i];
+
+		if (!ItemPointerIsValid(indextid))
+			break;
+
+		if (ItemPointerGetBlockNumber(indextid) == e->blkno && ItemPointerGetOffsetNumber(indextid) == e->offno)
+			return true;
+	}
+
+	return false;
+}
+
+/*
+ * Update neighbors
+ */
+void
+HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement e, int m, bool checkExisting)
+{
+	for (int lc = e->level; lc >= 0; lc--)
+	{
+		int			lm = HnswGetLayerM(m, lc);
+		HnswNeighborArray *neighbors = &e->neighbors[lc];
+
+		for (int i = 0; i < neighbors->length; i++)
+		{
+			HnswCandidate *hc = &neighbors->items[i];
+			Buffer		buf;
+			Page		page;
+			GenericXLogState *state;
+			ItemId		itemid;
+			HnswNeighborTuple ntup;
+			Size		ntupSize;
+			int			idx = -1;
+			int			startIdx;
+			OffsetNumber offno = hc->element->neighborOffno;
+
+			/* Get latest neighbors since they may have changed */
+			/* Do not lock yet since selecting neighbors can take time */
+			HnswLoadNeighbors(hc->element, index, m);
+
+			/*
+			 * Could improve performance for vacuuming by checking neighbors
+			 * against list of elements being deleted to find index. It's
+			 * important to exclude already deleted elements for this since
+			 * they can be replaced at any time.
+			 */
+
+			/* Select neighbors */
+			HnswUpdateConnection(e, hc, lm, lc, &idx, index, procinfo, collation);
+
+			/* New element was not selected as a neighbor */
+			if (idx == -1)
+				continue;
+
+			/* Register page */
+			buf = ReadBuffer(index, hc->element->neighborPage);
+			LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
+			state = GenericXLogStart(index);
+			page = GenericXLogRegisterBuffer(state, buf, 0);
+
+			/* Get tuple */
+			itemid = PageGetItemId(page, offno);
+			ntup = (HnswNeighborTuple) PageGetItem(page, itemid);
+			ntupSize = ItemIdGetLength(itemid);
+
+			/* Calculate index for update */
+			startIdx = (hc->element->level - lc) * m;
+
+			/* Check for existing connection */
+			if (checkExisting && ConnectionExists(e, ntup, startIdx, lm))
+				idx = -1;
+			else if (idx == -2)
+			{
+				/* Find free offset if still exists */
+				/* TODO Retry updating connections if not */
+				for (int j = 0; j < lm; j++)
+				{
+					if (!ItemPointerIsValid(&ntup->indextids[startIdx + j]))
+					{
+						idx = startIdx + j;
+						break;
+					}
+				}
+			}
+			else
+				idx += startIdx;
+
+			/* Make robust to issues */
+			if (idx >= 0 && idx < ntup->count)
+			{
+				ItemPointer indextid = &ntup->indextids[idx];
+
+				/* Update neighbor */
+				ItemPointerSet(indextid, e->blkno, e->offno);
+
+				/* Overwrite tuple */
+				if (!PageIndexTupleOverwrite(page, offno, (Item) ntup, ntupSize))
+					elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+				/* Commit */
+				GenericXLogFinish(state);
+			}
+			else
+				GenericXLogAbort(state);
+
+			UnlockReleaseBuffer(buf);
+		}
+	}
+}
+
+/*
+ * Add a heap TID to an existing element
+ */
+static bool
+HnswAddDuplicate(Relation index, HnswElement element, HnswElement dup)
+{
+	Buffer		buf;
+	Page		page;
+	GenericXLogState *state;
+	ItemId		itemid;
+	HnswElementTuple etup;
+	Size		etupSize;
+	int			i;
+
+	/* Read page */
+	buf = ReadBuffer(index, dup->blkno);
+	LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
+	state = GenericXLogStart(index);
+	page = GenericXLogRegisterBuffer(state, buf, 0);
+
+	/* Find space */
+	itemid = PageGetItemId(page, dup->offno);
+	etup = (HnswElementTuple) PageGetItem(page, itemid);
+	etupSize = ItemIdGetLength(itemid);
+	for (i = 0; i < HNSW_HEAPTIDS; i++)
+	{
+		if (!ItemPointerIsValid(&etup->heaptids[i]))
+			break;
+	}
+
+	/* Either being deleted or we lost our chance to another backend */
+	if (i == 0 || i == HNSW_HEAPTIDS)
+	{
+		GenericXLogAbort(state);
+		UnlockReleaseBuffer(buf);
+		return false;
+	}
+
+	/* Add heap TID */
+	etup->heaptids[i] = *((ItemPointer) linitial(element->heaptids));
+
+	/* Overwrite tuple */
+	if (!PageIndexTupleOverwrite(page, dup->offno, (Item) etup, etupSize))
+		elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+	/* Commit */
+	GenericXLogFinish(state);
+	UnlockReleaseBuffer(buf);
+
+	return true;
+}
+
+/*
+ * Write changes to disk
+ */
+static void
+WriteElement(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement element, int m, int efConstruction, HnswElement dup, HnswElement entryPoint)
+{
+	BlockNumber newInsertPage = InvalidBlockNumber;
+
+	/* Try to add to existing page */
+	if (dup != NULL)
+	{
+		if (HnswAddDuplicate(index, element, dup))
+			return;
+	}
+
+	/* Write element and neighbor tuples */
+	WriteNewElementPages(index, element, m, GetInsertPage(index), &newInsertPage);
+
+	/* Update insert page if needed */
+	if (BlockNumberIsValid(newInsertPage))
+		HnswUpdateMetaPage(index, 0, NULL, newInsertPage, MAIN_FORKNUM);
+
+	/* Update neighbors */
+	HnswUpdateNeighborPages(index, procinfo, collation, element, m, false);
+
+	/* Update metapage if needed */
+	if (entryPoint == NULL || element->level > entryPoint->level)
+		HnswUpdateMetaPage(index, HNSW_UPDATE_ENTRY_GREATER, element, InvalidBlockNumber, MAIN_FORKNUM);
+}
+
+/*
+ * Insert a tuple into the index
+ */
+bool
+HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
+{
+	Datum		value;
+	FmgrInfo   *normprocinfo;
+	HnswElement entryPoint;
+	HnswElement element;
+	int			m;
+	int			efConstruction = HnswGetEfConstruction(index);
+	FmgrInfo   *procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
+	Oid			collation = index->rd_indcollation[0];
+	HnswElement dup;
+	LOCKMODE	lockmode = ShareLock;
+
+	/* Detoast once for all calls */
+	value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
+
+	/* Normalize if needed */
+	normprocinfo = HnswOptionalProcInfo(index, HNSW_NORM_PROC);
+	if (normprocinfo != NULL)
+	{
+		if (!HnswNormValue(normprocinfo, collation, &value, NULL))
+			return false;
+	}
+
+	/*
+	 * Get a shared lock. This allows vacuum to ensure no in-flight inserts
+	 * before repairing graph. Use a page lock so it does not interfere with
+	 * buffer lock (or reads when vacuuming).
+	 */
+	LockPage(index, HNSW_UPDATE_LOCK, lockmode);
+
+	/* Get m and entry point */
+	HnswGetMetaPageInfo(index, &m, &entryPoint);
+
+	/* Create an element */
+	element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m));
+	element->value = value;
+
+	/* Prevent concurrent inserts when likely updating entry point */
+	if (entryPoint == NULL || element->level > entryPoint->level)
+	{
+		/* Release shared lock */
+		UnlockPage(index, HNSW_UPDATE_LOCK, lockmode);
+
+		/* Get exclusive lock */
+		lockmode = ExclusiveLock;
+		LockPage(index, HNSW_UPDATE_LOCK, lockmode);
+
+		/* Get latest entry point after lock is acquired */
+		entryPoint = HnswGetEntryPoint(index);
+	}
+
+	/* Insert element in graph */
+	HnswInsertElement(element, entryPoint, index, procinfo, collation, m, efConstruction, false);
+
+	/* Look for duplicate */
+	dup = HnswFindDuplicate(element);
+
+	/* Write to disk */
+	WriteElement(index, procinfo, collation, element, m, efConstruction, dup, entryPoint);
+
+	/* Release lock */
+	UnlockPage(index, HNSW_UPDATE_LOCK, lockmode);
+
+	return true;
+}
+
+/*
+ * Insert a tuple into the index
+ */
+bool
+hnswinsert(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid,
+		   Relation heap, IndexUniqueCheck checkUnique
+#if PG_VERSION_NUM >= 140000
+		   ,bool indexUnchanged
+#endif
+		   ,IndexInfo *indexInfo
+)
+{
+	MemoryContext oldCtx;
+	MemoryContext insertCtx;
+
+	/* Skip nulls */
+	if (isnull[0])
+		return false;
+
+	/* Create memory context */
+	insertCtx = AllocSetContextCreate(CurrentMemoryContext,
+									  "Hnsw insert temporary context",
+									  ALLOCSET_DEFAULT_SIZES);
+	oldCtx = MemoryContextSwitchTo(insertCtx);
+
+	/* Insert tuple */
+	HnswInsertTuple(index, values, isnull, heap_tid, heap);
+
+	/* Delete memory context */
+	MemoryContextSwitchTo(oldCtx);
+	MemoryContextDelete(insertCtx);
+
+	return false;
+}

src/hnswscan.c

@@ -0,0 +1,229 @@
+#include "postgres.h"
+
+#include "access/relscan.h"
+#include "hnsw.h"
+#include "pgstat.h"
+#include "storage/bufmgr.h"
+#include "storage/lmgr.h"
+#include "utils/memutils.h"
+
+/*
+ * Algorithm 5 from paper
+ */
+static List *
+GetScanItems(IndexScanDesc scan, Datum q)
+{
+	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
+	Relation	index = scan->indexRelation;
+	FmgrInfo   *procinfo = so->procinfo;
+	Oid			collation = so->collation;
+	List	   *ep;
+	List	   *w;
+	int			m;
+	HnswElement entryPoint;
+
+	/* Get m and entry point */
+	HnswGetMetaPageInfo(index, &m, &entryPoint);
+
+	if (entryPoint == NULL)
+		return NIL;
+
+	ep = list_make1(HnswEntryCandidate(entryPoint, q, index, procinfo, collation, false));
+
+	for (int lc = entryPoint->level; lc >= 1; lc--)
+	{
+		w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, m, false, NULL);
+		ep = w;
+	}
+
+	return HnswSearchLayer(q, ep, hnsw_ef_search, 0, index, procinfo, collation, m, false, NULL);
+}
+
+/*
+ * Get dimensions from metapage
+ */
+static int
+GetDimensions(Relation index)
+{
+	Buffer		buf;
+	Page		page;
+	HnswMetaPage metap;
+	int			dimensions;
+
+	buf = ReadBuffer(index, HNSW_METAPAGE_BLKNO);
+	LockBuffer(buf, BUFFER_LOCK_SHARE);
+	page = BufferGetPage(buf);
+	metap = HnswPageGetMeta(page);
+
+	dimensions = metap->dimensions;
+
+	UnlockReleaseBuffer(buf);
+
+	return dimensions;
+}
+
+/*
+ * Get scan value
+ */
+static Datum
+GetScanValue(IndexScanDesc scan)
+{
+	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
+	Datum		value;
+
+	if (scan->orderByData->sk_flags & SK_ISNULL)
+		value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
+	else
+	{
+		value = scan->orderByData->sk_argument;
+
+		/* Value should not be compressed or toasted */
+		Assert(!VARATT_IS_COMPRESSED(DatumGetPointer(value)));
+		Assert(!VARATT_IS_EXTENDED(DatumGetPointer(value)));
+
+		/* Fine if normalization fails */
+		if (so->normprocinfo != NULL)
+			HnswNormValue(so->normprocinfo, so->collation, &value, NULL);
+	}
+
+	return value;
+}
+
+/*
+ * Prepare for an index scan
+ */
+IndexScanDesc
+hnswbeginscan(Relation index, int nkeys, int norderbys)
+{
+	IndexScanDesc scan;
+	HnswScanOpaque so;
+
+	scan = RelationGetIndexScan(index, nkeys, norderbys);
+
+	so = (HnswScanOpaque) palloc(sizeof(HnswScanOpaqueData));
+	so->first = true;
+	so->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
+									   "Hnsw scan temporary context",
+									   ALLOCSET_DEFAULT_SIZES);
+
+	/* Set support functions */
+	so->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
+	so->normprocinfo = HnswOptionalProcInfo(index, HNSW_NORM_PROC);
+	so->collation = index->rd_indcollation[0];
+
+	scan->opaque = so;
+
+	return scan;
+}
+
+/*
+ * Start or restart an index scan
+ */
+void
+hnswrescan(IndexScanDesc scan, ScanKey keys, int nkeys, ScanKey orderbys, int norderbys)
+{
+	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
+
+	so->first = true;
+	MemoryContextReset(so->tmpCtx);
+
+	if (keys && scan->numberOfKeys > 0)
+		memmove(scan->keyData, keys, scan->numberOfKeys * sizeof(ScanKeyData));
+
+	if (orderbys && scan->numberOfOrderBys > 0)
+		memmove(scan->orderByData, orderbys, scan->numberOfOrderBys * sizeof(ScanKeyData));
+}
+
+/*
+ * Fetch the next tuple in the given scan
+ */
+bool
+hnswgettuple(IndexScanDesc scan, ScanDirection dir)
+{
+	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
+	MemoryContext oldCtx = MemoryContextSwitchTo(so->tmpCtx);
+
+	/*
+	 * Index can be used to scan backward, but Postgres doesn't support
+	 * backward scan on operators
+	 */
+	Assert(ScanDirectionIsForward(dir));
+
+	if (so->first)
+	{
+		Datum		value;
+
+		/* Count index scan for stats */
+		pgstat_count_index_scan(scan->indexRelation);
+
+		/* Safety check */
+		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;
+	}
+
+	while (list_length(so->w) > 0)
+	{
+		HnswCandidate *hc = llast(so->w);
+		ItemPointer heaptid;
+
+		/* Move to next element if no valid heap TIDs */
+		if (list_length(hc->element->heaptids) == 0)
+		{
+			so->w = list_delete_last(so->w);
+			continue;
+		}
+
+		heaptid = llast(hc->element->heaptids);
+
+		hc->element->heaptids = list_delete_last(hc->element->heaptids);
+
+		MemoryContextSwitchTo(oldCtx);
+
+#if PG_VERSION_NUM >= 120000
+		scan->xs_heaptid = *heaptid;
+#else
+		scan->xs_ctup.t_self = *heaptid;
+#endif
+
+		scan->xs_recheckorderby = false;
+		return true;
+	}
+
+	MemoryContextSwitchTo(oldCtx);
+	return false;
+}
+
+/*
+ * End a scan and release resources
+ */
+void
+hnswendscan(IndexScanDesc scan)
+{
+	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
+
+	MemoryContextDelete(so->tmpCtx);
+
+	pfree(so);
+	scan->opaque = NULL;
+}

src/hnswvacuum.c

@@ -0,0 +1,662 @@
+#include "postgres.h"
+
+#include <math.h>
+
+#include "commands/vacuum.h"
+#include "hnsw.h"
+#include "storage/bufmgr.h"
+#include "storage/lmgr.h"
+#include "utils/memutils.h"
+
+/*
+ * Check if deleted list contains an index TID
+ */
+static bool
+DeletedContains(HTAB *deleted, ItemPointer indextid)
+{
+	bool		found;
+
+	hash_search(deleted, indextid, HASH_FIND, &found);
+	return found;
+}
+
+/*
+ * Remove deleted heap TIDs
+ *
+ * OK to remove for entry point, since always considered for searches and inserts
+ */
+static void
+RemoveHeapTids(HnswVacuumState * vacuumstate)
+{
+	BlockNumber blkno = HNSW_HEAD_BLKNO;
+	HnswElement highestPoint = &vacuumstate->highestPoint;
+	Relation	index = vacuumstate->index;
+	BufferAccessStrategy bas = vacuumstate->bas;
+	HnswElement entryPoint = HnswGetEntryPoint(vacuumstate->index);
+	IndexBulkDeleteResult *stats = vacuumstate->stats;
+
+	/* Store separately since highestPoint.level is uint8 */
+	int			highestLevel = -1;
+
+	/* Initialize highest point */
+	highestPoint->blkno = InvalidBlockNumber;
+	highestPoint->offno = InvalidOffsetNumber;
+
+	while (BlockNumberIsValid(blkno))
+	{
+		Buffer		buf;
+		Page		page;
+		GenericXLogState *state;
+		OffsetNumber offno;
+		OffsetNumber maxoffno;
+		bool		updated = false;
+
+		vacuum_delay_point();
+
+		buf = ReadBufferExtended(index, MAIN_FORKNUM, blkno, RBM_NORMAL, bas);
+		LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
+		state = GenericXLogStart(index);
+		page = GenericXLogRegisterBuffer(state, buf, 0);
+		maxoffno = PageGetMaxOffsetNumber(page);
+
+		/* Iterate over nodes */
+		for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+		{
+			ItemId		itemid = PageGetItemId(page, offno);
+			HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, itemid);
+			int			idx = 0;
+			bool		itemUpdated = false;
+
+			/* Skip neighbor tuples */
+			if (!HnswIsElementTuple(etup))
+				continue;
+
+			if (ItemPointerIsValid(&etup->heaptids[0]))
+			{
+				for (int i = 0; i < HNSW_HEAPTIDS; i++)
+				{
+					/* Stop at first unused */
+					if (!ItemPointerIsValid(&etup->heaptids[i]))
+						break;
+
+					if (vacuumstate->callback(&etup->heaptids[i], vacuumstate->callback_state))
+					{
+						itemUpdated = true;
+						stats->tuples_removed++;
+					}
+					else
+					{
+						/* Move to front of list */
+						etup->heaptids[idx++] = etup->heaptids[i];
+						stats->num_index_tuples++;
+					}
+				}
+
+				if (itemUpdated)
+				{
+					Size		etupSize = ItemIdGetLength(itemid);
+
+					/* Mark rest as invalid */
+					for (int i = idx; i < HNSW_HEAPTIDS; i++)
+						ItemPointerSetInvalid(&etup->heaptids[i]);
+
+					if (!PageIndexTupleOverwrite(page, offno, (Item) etup, etupSize))
+						elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+					updated = true;
+				}
+			}
+
+			if (!ItemPointerIsValid(&etup->heaptids[0]))
+			{
+				ItemPointerData ip;
+
+				/* Add to deleted list */
+				ItemPointerSet(&ip, blkno, offno);
+
+				(void) hash_search(vacuumstate->deleted, &ip, HASH_ENTER, NULL);
+			}
+			else if (etup->level > highestLevel && !(entryPoint != NULL && blkno == entryPoint->blkno && offno == entryPoint->offno))
+			{
+				/* Keep track of highest non-entry point */
+				highestPoint->blkno = blkno;
+				highestPoint->offno = offno;
+				highestPoint->level = etup->level;
+				highestLevel = etup->level;
+			}
+		}
+
+		blkno = HnswPageGetOpaque(page)->nextblkno;
+
+		if (updated)
+			GenericXLogFinish(state);
+		else
+			GenericXLogAbort(state);
+
+		UnlockReleaseBuffer(buf);
+	}
+}
+
+/*
+ * Check for deleted neighbors
+ */
+static bool
+NeedsUpdated(HnswVacuumState * vacuumstate, HnswElement element)
+{
+	Relation	index = vacuumstate->index;
+	BufferAccessStrategy bas = vacuumstate->bas;
+	Buffer		buf;
+	Page		page;
+	HnswNeighborTuple ntup;
+	bool		needsUpdated = false;
+
+	buf = ReadBufferExtended(index, MAIN_FORKNUM, element->neighborPage, RBM_NORMAL, bas);
+	LockBuffer(buf, BUFFER_LOCK_SHARE);
+	page = BufferGetPage(buf);
+	ntup = (HnswNeighborTuple) PageGetItem(page, PageGetItemId(page, element->neighborOffno));
+
+	Assert(HnswIsNeighborTuple(ntup));
+
+	/* Check neighbors */
+	for (int i = 0; i < ntup->count; i++)
+	{
+		ItemPointer indextid = &ntup->indextids[i];
+
+		if (!ItemPointerIsValid(indextid))
+			continue;
+
+		/* Check if in deleted list */
+		if (DeletedContains(vacuumstate->deleted, indextid))
+		{
+			needsUpdated = true;
+			break;
+		}
+	}
+
+	/* Also update if layer 0 is not full */
+	/* This could indicate too many candidates being deleted during insert */
+	if (!needsUpdated)
+		needsUpdated = !ItemPointerIsValid(&ntup->indextids[ntup->count - 1]);
+
+	UnlockReleaseBuffer(buf);
+
+	return needsUpdated;
+}
+
+/*
+ * Repair graph for a single element
+ */
+static void
+RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswElement entryPoint)
+{
+	Relation	index = vacuumstate->index;
+	Buffer		buf;
+	Page		page;
+	GenericXLogState *state;
+	int			m = vacuumstate->m;
+	int			efConstruction = vacuumstate->efConstruction;
+	FmgrInfo   *procinfo = vacuumstate->procinfo;
+	Oid			collation = vacuumstate->collation;
+	BufferAccessStrategy bas = vacuumstate->bas;
+	HnswNeighborTuple ntup = vacuumstate->ntup;
+	Size		ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(element->level, m);
+
+	/* Skip if element is entry point */
+	if (entryPoint != NULL && element->blkno == entryPoint->blkno && element->offno == entryPoint->offno)
+		return;
+
+	/* Init fields */
+	HnswInitNeighbors(element, m);
+	element->heaptids = NIL;
+
+	/* Add element to graph, skipping itself */
+	HnswInsertElement(element, entryPoint, index, procinfo, collation, m, efConstruction, true);
+
+	/* Update neighbor tuple */
+	/* Do this before getting page to minimize locking */
+	HnswSetNeighborTuple(ntup, element, m);
+
+	/* Get neighbor page */
+	buf = ReadBufferExtended(index, MAIN_FORKNUM, element->neighborPage, RBM_NORMAL, bas);
+	LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
+	state = GenericXLogStart(index);
+	page = GenericXLogRegisterBuffer(state, buf, 0);
+
+	/* Overwrite tuple */
+	if (!PageIndexTupleOverwrite(page, element->neighborOffno, (Item) ntup, ntupSize))
+		elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+	/* Commit */
+	GenericXLogFinish(state);
+	UnlockReleaseBuffer(buf);
+
+	/* Update neighbors */
+	HnswUpdateNeighborPages(index, procinfo, collation, element, m, true);
+}
+
+/*
+ * Repair graph entry point
+ */
+static void
+RepairGraphEntryPoint(HnswVacuumState * vacuumstate)
+{
+	Relation	index = vacuumstate->index;
+	HnswElement highestPoint = &vacuumstate->highestPoint;
+	HnswElement entryPoint;
+	MemoryContext oldCtx = MemoryContextSwitchTo(vacuumstate->tmpCtx);
+
+	if (!BlockNumberIsValid(highestPoint->blkno))
+		highestPoint = NULL;
+
+	/*
+	 * Repair graph for highest non-entry point. Highest point may be outdated
+	 * due to inserts that happen during and after RemoveHeapTids.
+	 */
+	if (highestPoint != NULL)
+	{
+		/* Get a shared lock */
+		LockPage(index, HNSW_UPDATE_LOCK, ShareLock);
+
+		/* Load element */
+		HnswLoadElement(highestPoint, NULL, NULL, index, vacuumstate->procinfo, vacuumstate->collation, true);
+
+		/* Repair if needed */
+		if (NeedsUpdated(vacuumstate, highestPoint))
+			RepairGraphElement(vacuumstate, highestPoint, HnswGetEntryPoint(index));
+
+		/* Release lock */
+		UnlockPage(index, HNSW_UPDATE_LOCK, ShareLock);
+	}
+
+	/* Prevent concurrent inserts when possibly updating entry point */
+	LockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
+
+	/* Get latest entry point */
+	entryPoint = HnswGetEntryPoint(index);
+
+	if (entryPoint != NULL)
+	{
+		ItemPointerData epData;
+
+		ItemPointerSet(&epData, entryPoint->blkno, entryPoint->offno);
+
+		if (DeletedContains(vacuumstate->deleted, &epData))
+		{
+			/*
+			 * Replace the entry point with the highest point. If highest
+			 * point is outdated and empty, the entry point will be empty
+			 * until an element is repaired.
+			 */
+			HnswUpdateMetaPage(index, HNSW_UPDATE_ENTRY_ALWAYS, highestPoint, InvalidBlockNumber, MAIN_FORKNUM);
+		}
+		else
+		{
+			/*
+			 * Repair the entry point with the highest point. If highest point
+			 * is outdated, this can remove connections at higher levels in
+			 * the graph until they are repaired, but this should be fine.
+			 */
+			HnswLoadElement(entryPoint, NULL, NULL, index, vacuumstate->procinfo, vacuumstate->collation, true);
+
+			if (NeedsUpdated(vacuumstate, entryPoint))
+			{
+				/* Reset neighbors from previous update */
+				if (highestPoint != NULL)
+					highestPoint->neighbors = NULL;
+
+				RepairGraphElement(vacuumstate, entryPoint, highestPoint);
+			}
+		}
+	}
+
+	/* Release lock */
+	UnlockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
+
+	/* Reset memory context */
+	MemoryContextSwitchTo(oldCtx);
+	MemoryContextReset(vacuumstate->tmpCtx);
+}
+
+/*
+ * Repair graph for all elements
+ */
+static void
+RepairGraph(HnswVacuumState * vacuumstate)
+{
+	Relation	index = vacuumstate->index;
+	BufferAccessStrategy bas = vacuumstate->bas;
+	BlockNumber blkno = HNSW_HEAD_BLKNO;
+
+	/*
+	 * Wait for inserts to complete. Inserts before this point may have
+	 * neighbors about to be deleted. Inserts after this point will not.
+	 */
+	LockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
+	UnlockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
+
+	/* Repair entry point first */
+	RepairGraphEntryPoint(vacuumstate);
+
+	while (BlockNumberIsValid(blkno))
+	{
+		Buffer		buf;
+		Page		page;
+		OffsetNumber offno;
+		OffsetNumber maxoffno;
+		List	   *elements = NIL;
+		ListCell   *lc2;
+		MemoryContext oldCtx;
+
+		vacuum_delay_point();
+
+		oldCtx = MemoryContextSwitchTo(vacuumstate->tmpCtx);
+
+		buf = ReadBufferExtended(index, MAIN_FORKNUM, blkno, RBM_NORMAL, bas);
+		LockBuffer(buf, BUFFER_LOCK_SHARE);
+		page = BufferGetPage(buf);
+		maxoffno = PageGetMaxOffsetNumber(page);
+
+		/* Load items into memory to minimize locking */
+		for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+		{
+			HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, PageGetItemId(page, offno));
+			HnswElement element;
+
+			/* Skip neighbor tuples */
+			if (!HnswIsElementTuple(etup))
+				continue;
+
+			/* Skip updating neighbors if being deleted */
+			if (!ItemPointerIsValid(&etup->heaptids[0]))
+				continue;
+
+			/* Create an element */
+			element = HnswInitElementFromBlock(blkno, offno);
+			HnswLoadElementFromTuple(element, etup, false, true);
+
+			elements = lappend(elements, element);
+		}
+
+		blkno = HnswPageGetOpaque(page)->nextblkno;
+
+		UnlockReleaseBuffer(buf);
+
+		/* Update neighbor pages */
+		foreach(lc2, elements)
+		{
+			HnswElement element = (HnswElement) lfirst(lc2);
+			HnswElement entryPoint;
+			LOCKMODE	lockmode = ShareLock;
+
+			/* Check if any neighbors point to deleted values */
+			if (!NeedsUpdated(vacuumstate, element))
+				continue;
+
+			/* Get a shared lock */
+			LockPage(index, HNSW_UPDATE_LOCK, lockmode);
+
+			/* Refresh entry point for each element */
+			entryPoint = HnswGetEntryPoint(index);
+
+			/* Prevent concurrent inserts when likely updating entry point */
+			if (entryPoint == NULL || element->level > entryPoint->level)
+			{
+				/* Release shared lock */
+				UnlockPage(index, HNSW_UPDATE_LOCK, lockmode);
+
+				/* Get exclusive lock */
+				lockmode = ExclusiveLock;
+				LockPage(index, HNSW_UPDATE_LOCK, lockmode);
+
+				/* Get latest entry point after lock is acquired */
+				entryPoint = HnswGetEntryPoint(index);
+			}
+
+			/* Repair connections */
+			RepairGraphElement(vacuumstate, element, entryPoint);
+
+			/*
+			 * Update metapage if needed. Should only happen if entry point
+			 * was replaced and highest point was outdated.
+			 */
+			if (entryPoint == NULL || element->level > entryPoint->level)
+				HnswUpdateMetaPage(index, HNSW_UPDATE_ENTRY_GREATER, element, InvalidBlockNumber, MAIN_FORKNUM);
+
+			/* Release lock */
+			UnlockPage(index, HNSW_UPDATE_LOCK, lockmode);
+		}
+
+		/* Reset memory context */
+		MemoryContextSwitchTo(oldCtx);
+		MemoryContextReset(vacuumstate->tmpCtx);
+	}
+}
+
+/*
+ * Mark items as deleted
+ */
+static void
+MarkDeleted(HnswVacuumState * vacuumstate)
+{
+	BlockNumber blkno = HNSW_HEAD_BLKNO;
+	BlockNumber insertPage = InvalidBlockNumber;
+	Relation	index = vacuumstate->index;
+	BufferAccessStrategy bas = vacuumstate->bas;
+
+	/*
+	 * Wait for index scans to complete. Scans before this point may contain
+	 * tuples about to be deleted. Scans after this point will not, since the
+	 * graph has been repaired.
+	 */
+	LockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
+	UnlockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
+
+	while (BlockNumberIsValid(blkno))
+	{
+		Buffer		buf;
+		Page		page;
+		GenericXLogState *state;
+		OffsetNumber offno;
+		OffsetNumber maxoffno;
+
+		vacuum_delay_point();
+
+		buf = ReadBufferExtended(index, MAIN_FORKNUM, blkno, RBM_NORMAL, bas);
+
+		/*
+		 * ambulkdelete cannot delete entries from pages that are pinned by
+		 * other backends
+		 *
+		 * https://www.postgresql.org/docs/current/index-locking.html
+		 */
+		LockBufferForCleanup(buf);
+
+		state = GenericXLogStart(index);
+		page = GenericXLogRegisterBuffer(state, buf, 0);
+		maxoffno = PageGetMaxOffsetNumber(page);
+
+		/* Update element and neighbors together */
+		for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+		{
+			ItemId		itemid = PageGetItemId(page, offno);
+			HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, itemid);
+			HnswNeighborTuple ntup;
+			Size		etupSize;
+			Size		ntupSize;
+			Buffer		nbuf;
+			Page		npage;
+			BlockNumber neighborPage;
+			OffsetNumber neighborOffno;
+
+			/* Skip neighbor tuples */
+			if (!HnswIsElementTuple(etup))
+				continue;
+
+			/* Skip deleted tuples */
+			if (etup->deleted)
+			{
+				/* Set to first free page */
+				if (!BlockNumberIsValid(insertPage))
+					insertPage = blkno;
+
+				continue;
+			}
+
+			/* Skip live tuples */
+			if (ItemPointerIsValid(&etup->heaptids[0]))
+				continue;
+
+			/* Calculate sizes */
+			etupSize = ItemIdGetLength(itemid);
+			ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(etup->level, vacuumstate->m);
+
+			/* Get neighbor page */
+			neighborPage = ItemPointerGetBlockNumber(&etup->neighbortid);
+			neighborOffno = ItemPointerGetOffsetNumber(&etup->neighbortid);
+
+			if (neighborPage == blkno)
+			{
+				nbuf = buf;
+				npage = page;
+			}
+			else
+			{
+				nbuf = ReadBufferExtended(index, MAIN_FORKNUM, neighborPage, RBM_NORMAL, bas);
+				LockBuffer(nbuf, BUFFER_LOCK_EXCLUSIVE);
+				npage = GenericXLogRegisterBuffer(state, nbuf, 0);
+			}
+
+			ntup = (HnswNeighborTuple) PageGetItem(npage, PageGetItemId(npage, neighborOffno));
+
+			/* Overwrite element */
+			etup->deleted = 1;
+			MemSet(&etup->data, 0, VARSIZE_ANY(&etup->data));
+
+			/* Overwrite neighbors */
+			for (int i = 0; i < ntup->count; i++)
+				ItemPointerSetInvalid(&ntup->indextids[i]);
+
+			/* Overwrite element tuple */
+			if (!PageIndexTupleOverwrite(page, offno, (Item) etup, etupSize))
+				elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+			/* Overwrite neighbor tuple */
+			if (!PageIndexTupleOverwrite(npage, neighborOffno, (Item) ntup, ntupSize))
+				elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
+
+			/* Commit */
+			GenericXLogFinish(state);
+			if (nbuf != buf)
+				UnlockReleaseBuffer(nbuf);
+
+			/* Set to first free page */
+			if (!BlockNumberIsValid(insertPage))
+				insertPage = blkno;
+
+			/* Prepare new xlog */
+			state = GenericXLogStart(index);
+			page = GenericXLogRegisterBuffer(state, buf, 0);
+		}
+
+		blkno = HnswPageGetOpaque(page)->nextblkno;
+
+		GenericXLogAbort(state);
+		UnlockReleaseBuffer(buf);
+	}
+
+	/* Update insert page last, after everything has been marked as deleted */
+	HnswUpdateMetaPage(index, 0, NULL, insertPage, MAIN_FORKNUM);
+}
+
+/*
+ * Initialize the vacuum state
+ */
+static void
+InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkDeleteResult *stats, IndexBulkDeleteCallback callback, void *callback_state)
+{
+	Relation	index = info->index;
+	HASHCTL		hash_ctl;
+
+	if (stats == NULL)
+		stats = (IndexBulkDeleteResult *) palloc0(sizeof(IndexBulkDeleteResult));
+
+	vacuumstate->index = index;
+	vacuumstate->stats = stats;
+	vacuumstate->callback = callback;
+	vacuumstate->callback_state = callback_state;
+	vacuumstate->efConstruction = HnswGetEfConstruction(index);
+	vacuumstate->bas = GetAccessStrategy(BAS_BULKREAD);
+	vacuumstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
+	vacuumstate->collation = index->rd_indcollation[0];
+	vacuumstate->ntup = palloc0(BLCKSZ);
+	vacuumstate->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
+												"Hnsw vacuum temporary context",
+												ALLOCSET_DEFAULT_SIZES);
+
+	/* Get m from metapage */
+	HnswGetMetaPageInfo(index, &vacuumstate->m, NULL);
+
+	/* Create hash table */
+	hash_ctl.keysize = sizeof(ItemPointerData);
+	hash_ctl.entrysize = sizeof(ItemPointerData);
+	hash_ctl.hcxt = CurrentMemoryContext;
+	vacuumstate->deleted = hash_create("hnswbulkdelete indextids", 256, &hash_ctl, HASH_ELEM | HASH_BLOBS | HASH_CONTEXT);
+}
+
+/*
+ * Free resources
+ */
+static void
+FreeVacuumState(HnswVacuumState * vacuumstate)
+{
+	hash_destroy(vacuumstate->deleted);
+	FreeAccessStrategy(vacuumstate->bas);
+	pfree(vacuumstate->ntup);
+	MemoryContextDelete(vacuumstate->tmpCtx);
+}
+
+/*
+ * Bulk delete tuples from the index
+ */
+IndexBulkDeleteResult *
+hnswbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
+			   IndexBulkDeleteCallback callback, void *callback_state)
+{
+	HnswVacuumState vacuumstate;
+
+	InitVacuumState(&vacuumstate, info, stats, callback, callback_state);
+
+	/* Pass 1: Remove heap TIDs */
+	RemoveHeapTids(&vacuumstate);
+
+	/* Pass 2: Repair graph */
+	RepairGraph(&vacuumstate);
+
+	/* Pass 3: Mark as deleted */
+	MarkDeleted(&vacuumstate);
+
+	FreeVacuumState(&vacuumstate);
+
+	return vacuumstate.stats;
+}
+
+/*
+ * Clean up after a VACUUM operation
+ */
+IndexBulkDeleteResult *
+hnswvacuumcleanup(IndexVacuumInfo *info, IndexBulkDeleteResult *stats)
+{
+	Relation	rel = info->index;
+
+	if (info->analyze_only)
+		return stats;
+
+	/* stats is NULL if ambulkdelete not called */
+	/* OK to return NULL if index not changed */
+	if (stats == NULL)
+		return NULL;
+
+	stats->num_pages = RelationGetNumberOfBlocks(rel);
+
+	return stats;
+}

src/ivfbuild.c

@@ -2,10 +2,15 @@
 
 #include <float.h>
 
+#include "access/parallel.h"
+#include "access/xact.h"
 #include "catalog/index.h"
+#include "catalog/pg_operator_d.h"
+#include "catalog/pg_type_d.h"
 #include "ivfflat.h"
 #include "miscadmin.h"
 #include "storage/bufmgr.h"
+#include "tcop/tcopprot.h"
 #include "utils/memutils.h"
 
 #if PG_VERSION_NUM >= 140000
@@ -23,9 +28,6 @@
 #define PROGRESS_CREATEIDX_TUPLES_DONE 0
 #endif
 
-#include "catalog/pg_operator_d.h"
-#include "catalog/pg_type_d.h"
-
 #if PG_VERSION_NUM >= 130000
 #define CALLBACK_ITEM_POINTER ItemPointer tid
 #else
@@ -38,6 +40,25 @@
 #define UpdateProgress(index, val) ((void)val)
 #endif
 
+#if PG_VERSION_NUM >= 140000
+#include "utils/backend_status.h"
+#include "utils/wait_event.h"
+#endif
+
+#if PG_VERSION_NUM >= 120000
+#include "access/table.h"
+#include "optimizer/optimizer.h"
+#else
+#include "access/heapam.h"
+#include "optimizer/planner.h"
+#include "pgstat.h"
+#endif
+
+#define PARALLEL_KEY_IVFFLAT_SHARED		UINT64CONST(0xA000000000000001)
+#define PARALLEL_KEY_TUPLESORT			UINT64CONST(0xA000000000000002)
+#define PARALLEL_KEY_IVFFLAT_CENTERS	UINT64CONST(0xA000000000000003)
+#define PARALLEL_KEY_QUERY_TEXT			UINT64CONST(0xA000000000000004)
+
 /*
  * Add sample
  */
@@ -147,10 +168,9 @@ AddTupleToSort(Relation index, ItemPointer tid, Datum *values, IvfflatBuildState
 {
 	double		distance;
 	double		minDistance = DBL_MAX;
-	int			closestCenter = -1;
+	int			closestCenter = 0;
 	VectorArray centers = buildstate->centers;
 	TupleTableSlot *slot = buildstate->slot;
-	int			i;
 
 	/* Detoast once for all calls */
 	Datum		value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
@@ -163,7 +183,7 @@ AddTupleToSort(Relation index, ItemPointer tid, Datum *values, IvfflatBuildState
 	}
 
 	/* Find the list that minimizes the distance */
-	for (i = 0; i < centers->length; i++)
+	for (int i = 0; i < centers->length; i++)
 	{
 		distance = DatumGetFloat8(FunctionCall2Coll(buildstate->procinfo, buildstate->collation, value, PointerGetDatum(VectorArrayGet(centers, i))));
 
@@ -258,15 +278,8 @@ GetNextTuple(Tuplesortstate *sortstate, TupleDesc tupdesc, TupleTableSlot *slot,
 static void
 InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
 {
-	Buffer		buf;
-	Page		page;
-	GenericXLogState *state;
 	int			list;
 	IndexTuple	itup = NULL;	/* silence compiler warning */
-	BlockNumber startPage;
-	BlockNumber insertPage;
-	Size		itemsz;
-	int			i;
 	int64		inserted = 0;
 
 #if PG_VERSION_NUM >= 120000
@@ -282,8 +295,14 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
 
 	GetNextTuple(buildstate->sortstate, tupdesc, slot, &itup, &list);
 
-	for (i = 0; i < buildstate->centers->length; i++)
+	for (int i = 0; i < buildstate->centers->length; i++)
 	{
+		Buffer		buf;
+		Page		page;
+		GenericXLogState *state;
+		BlockNumber startPage;
+		BlockNumber insertPage;
+
 		/* Can take a while, so ensure we can interrupt */
 		/* Needs to be called when no buffer locks are held */
 		CHECK_FOR_INTERRUPTS();
@@ -297,7 +316,8 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
 		while (list == i)
 		{
 			/* Check for free space */
-			itemsz = MAXALIGN(IndexTupleSize(itup));
+			Size		itemsz = MAXALIGN(IndexTupleSize(itup));
+
 			if (PageGetFreeSpace(page) < itemsz)
 				IvfflatAppendPage(index, &buf, &page, &state, forkNum);
 
@@ -317,7 +337,7 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
 		IvfflatCommitBuffer(buf, state);
 
 		/* Set the start and insert pages */
-		IvfflatUpdateList(index, state, buildstate->listInfo[i], insertPage, InvalidBlockNumber, startPage, forkNum);
+		IvfflatUpdateList(index, buildstate->listInfo[i], insertPage, InvalidBlockNumber, startPage, forkNum);
 	}
 }
 
@@ -351,9 +371,7 @@ InitBuildState(IvfflatBuildState * buildstate, Relation heap, Relation index, In
 	buildstate->collation = index->rd_indcollation[0];
 
 	/* Require more than one dimension for spherical k-means */
-	/* Lists check for backwards compatibility */
-	/* TODO Remove lists check in 0.3.0 */
-	if (buildstate->kmeansnormprocinfo != NULL && buildstate->dimensions == 1 && buildstate->lists > 1)
+	if (buildstate->kmeansnormprocinfo != NULL && buildstate->dimensions == 1)
 		elog(ERROR, "dimensions must be greater than one for this opclass");
 
 	/* Create tuple description for sorting */
@@ -387,6 +405,8 @@ InitBuildState(IvfflatBuildState * buildstate, Relation heap, Relation index, In
 	buildstate->listSums = palloc0(sizeof(double) * buildstate->lists);
 	buildstate->listCounts = palloc0(sizeof(int) * buildstate->lists);
 #endif
+
+	buildstate->ivfleader = NULL;
 }
 
 /*
@@ -431,8 +451,18 @@ ComputeCenters(IvfflatBuildState * buildstate)
 	/* TODO Ensure within maintenance_work_mem */
 	buildstate->samples = VectorArrayInit(numSamples, buildstate->dimensions);
 	if (buildstate->heap != NULL)
+	{
 		SampleRows(buildstate);
 
+		if (buildstate->samples->length < buildstate->lists)
+		{
+			ereport(NOTICE,
+					(errmsg("ivfflat index created with little data"),
+					 errdetail("This will cause low recall."),
+					 errhint("Drop the index until the table has more data.")));
+		}
+	}
+
 	/* Calculate centers */
 	IvfflatBench("k-means", IvfflatKmeans(buildstate->index, buildstate->samples, buildstate->centers));
 
@@ -473,33 +503,33 @@ static void
 CreateListPages(Relation index, VectorArray centers, int dimensions,
 				int lists, ForkNumber forkNum, ListInfo * *listInfo)
 {
-	int			i;
 	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 (i = 0; i < lists; i++)
+	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));
 
@@ -523,7 +553,7 @@ PrintKmeansMetrics(IvfflatBuildState * buildstate)
 	elog(INFO, "inertia: %.3e", buildstate->inertia);
 
 	/* Calculate Davies-Bouldin index */
-	if (buildstate->lists > 1)
+	if (buildstate->lists > 1 && !buildstate->ivfleader)
 	{
 		double		db = 0.0;
 
@@ -558,43 +588,478 @@ PrintKmeansMetrics(IvfflatBuildState * buildstate)
 }
 #endif
 
+/*
+ * Within leader, wait for end of heap scan
+ */
+static double
+ParallelHeapScan(IvfflatBuildState * buildstate)
+{
+	IvfflatShared *ivfshared = buildstate->ivfleader->ivfshared;
+	int			nparticipanttuplesorts;
+	double		reltuples;
+
+	nparticipanttuplesorts = buildstate->ivfleader->nparticipanttuplesorts;
+	for (;;)
+	{
+		SpinLockAcquire(&ivfshared->mutex);
+		if (ivfshared->nparticipantsdone == nparticipanttuplesorts)
+		{
+			buildstate->indtuples = ivfshared->indtuples;
+			reltuples = ivfshared->reltuples;
+#ifdef IVFFLAT_KMEANS_DEBUG
+			buildstate->inertia = ivfshared->inertia;
+#endif
+			SpinLockRelease(&ivfshared->mutex);
+			break;
+		}
+		SpinLockRelease(&ivfshared->mutex);
+
+		ConditionVariableSleep(&ivfshared->workersdonecv,
+							   WAIT_EVENT_PARALLEL_CREATE_INDEX_SCAN);
+	}
+
+	ConditionVariableCancelSleep();
+
+	return reltuples;
+}
+
+/*
+ * Perform a worker's portion of a parallel sort
+ */
+static void
+IvfflatParallelScanAndSort(IvfflatSpool * ivfspool, IvfflatShared * ivfshared, Sharedsort *sharedsort, Vector * ivfcenters, int sortmem, bool progress)
+{
+	SortCoordinate coordinate;
+	IvfflatBuildState buildstate;
+#if PG_VERSION_NUM >= 120000
+	TableScanDesc scan;
+#else
+	HeapScanDesc scan;
+#endif
+	double		reltuples;
+	IndexInfo  *indexInfo;
+
+	/* Sort options, which must match AssignTuples */
+	AttrNumber	attNums[] = {1};
+	Oid			sortOperators[] = {Int4LessOperator};
+	Oid			sortCollations[] = {InvalidOid};
+	bool		nullsFirstFlags[] = {false};
+
+	/* Initialize local tuplesort coordination state */
+	coordinate = palloc0(sizeof(SortCoordinateData));
+	coordinate->isWorker = true;
+	coordinate->nParticipants = -1;
+	coordinate->sharedsort = sharedsort;
+
+	/* Join parallel scan */
+	indexInfo = BuildIndexInfo(ivfspool->index);
+	indexInfo->ii_Concurrent = ivfshared->isconcurrent;
+	InitBuildState(&buildstate, ivfspool->heap, ivfspool->index, indexInfo);
+	memcpy(buildstate.centers->items, ivfcenters, VECTOR_SIZE(buildstate.centers->dim) * buildstate.centers->maxlen);
+	buildstate.centers->length = buildstate.centers->maxlen;
+	ivfspool->sortstate = tuplesort_begin_heap(buildstate.tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, sortmem, coordinate, false);
+	buildstate.sortstate = ivfspool->sortstate;
+#if PG_VERSION_NUM >= 120000
+	scan = table_beginscan_parallel(ivfspool->heap,
+									ParallelTableScanFromIvfflatShared(ivfshared));
+	reltuples = table_index_build_scan(ivfspool->heap, ivfspool->index, indexInfo,
+									   true, progress, BuildCallback,
+									   (void *) &buildstate, scan);
+#else
+	scan = heap_beginscan_parallel(ivfspool->heap, &ivfshared->heapdesc);
+	reltuples = IndexBuildHeapScan(ivfspool->heap, ivfspool->index, indexInfo,
+								   true, BuildCallback,
+								   (void *) &buildstate, scan);
+#endif
+
+	/* Execute this worker's part of the sort */
+	tuplesort_performsort(ivfspool->sortstate);
+
+	/* Record statistics */
+	SpinLockAcquire(&ivfshared->mutex);
+	ivfshared->nparticipantsdone++;
+	ivfshared->reltuples += reltuples;
+	ivfshared->indtuples += buildstate.indtuples;
+#ifdef IVFFLAT_KMEANS_DEBUG
+	ivfshared->inertia += buildstate.inertia;
+#endif
+	SpinLockRelease(&ivfshared->mutex);
+
+	/* Log statistics */
+	if (progress)
+		ereport(DEBUG1, (errmsg("leader processed " INT64_FORMAT " tuples", (int64) reltuples)));
+	else
+		ereport(DEBUG1, (errmsg("worker processed " INT64_FORMAT " tuples", (int64) reltuples)));
+
+	/* Notify leader */
+	ConditionVariableSignal(&ivfshared->workersdonecv);
+
+	/* We can end tuplesorts immediately */
+	tuplesort_end(ivfspool->sortstate);
+
+	FreeBuildState(&buildstate);
+}
+
+/*
+ * Perform work within a launched parallel process
+ */
+void
+IvfflatParallelBuildMain(dsm_segment *seg, shm_toc *toc)
+{
+	char	   *sharedquery;
+	IvfflatSpool *ivfspool;
+	IvfflatShared *ivfshared;
+	Sharedsort *sharedsort;
+	Vector	   *ivfcenters;
+	Relation	heapRel;
+	Relation	indexRel;
+	LOCKMODE	heapLockmode;
+	LOCKMODE	indexLockmode;
+	int			sortmem;
+
+	/* Set debug_query_string for individual workers first */
+	sharedquery = shm_toc_lookup(toc, PARALLEL_KEY_QUERY_TEXT, true);
+	debug_query_string = sharedquery;
+
+	/* Report the query string from leader */
+	pgstat_report_activity(STATE_RUNNING, debug_query_string);
+
+	/* Look up shared state */
+	ivfshared = shm_toc_lookup(toc, PARALLEL_KEY_IVFFLAT_SHARED, false);
+
+	/* Open relations using lock modes known to be obtained by index.c */
+	if (!ivfshared->isconcurrent)
+	{
+		heapLockmode = ShareLock;
+		indexLockmode = AccessExclusiveLock;
+	}
+	else
+	{
+		heapLockmode = ShareUpdateExclusiveLock;
+		indexLockmode = RowExclusiveLock;
+	}
+
+	/* Open relations within worker */
+#if PG_VERSION_NUM >= 120000
+	heapRel = table_open(ivfshared->heaprelid, heapLockmode);
+#else
+	heapRel = heap_open(ivfshared->heaprelid, heapLockmode);
+#endif
+	indexRel = index_open(ivfshared->indexrelid, indexLockmode);
+
+	/* Initialize worker's own spool */
+	ivfspool = (IvfflatSpool *) palloc0(sizeof(IvfflatSpool));
+	ivfspool->heap = heapRel;
+	ivfspool->index = indexRel;
+
+	/* Look up shared state private to tuplesort.c */
+	sharedsort = shm_toc_lookup(toc, PARALLEL_KEY_TUPLESORT, false);
+	tuplesort_attach_shared(sharedsort, seg);
+
+	ivfcenters = shm_toc_lookup(toc, PARALLEL_KEY_IVFFLAT_CENTERS, false);
+
+	/* Perform sorting */
+	sortmem = maintenance_work_mem / ivfshared->scantuplesortstates;
+	IvfflatParallelScanAndSort(ivfspool, ivfshared, sharedsort, ivfcenters, sortmem, false);
+
+	/* Close relations within worker */
+	index_close(indexRel, indexLockmode);
+#if PG_VERSION_NUM >= 120000
+	table_close(heapRel, heapLockmode);
+#else
+	heap_close(heapRel, heapLockmode);
+#endif
+}
+
+/*
+ * End parallel build
+ */
+static void
+IvfflatEndParallel(IvfflatLeader * ivfleader)
+{
+	/* Shutdown worker processes */
+	WaitForParallelWorkersToFinish(ivfleader->pcxt);
+
+	/* Free last reference to MVCC snapshot, if one was used */
+	if (IsMVCCSnapshot(ivfleader->snapshot))
+		UnregisterSnapshot(ivfleader->snapshot);
+	DestroyParallelContext(ivfleader->pcxt);
+	ExitParallelMode();
+}
+
+/*
+ * Return size of shared memory required for parallel index build
+ */
+static Size
+ParallelEstimateShared(Relation heap, Snapshot snapshot)
+{
+#if PG_VERSION_NUM >= 120000
+	return add_size(BUFFERALIGN(sizeof(IvfflatShared)), table_parallelscan_estimate(heap, snapshot));
+#else
+	if (!IsMVCCSnapshot(snapshot))
+	{
+		Assert(snapshot == SnapshotAny);
+		return sizeof(IvfflatShared);
+	}
+
+	return add_size(offsetof(IvfflatShared, heapdesc) +
+					offsetof(ParallelHeapScanDescData, phs_snapshot_data),
+					EstimateSnapshotSpace(snapshot));
+#endif
+}
+
+/*
+ * Within leader, participate as a parallel worker
+ */
+static void
+IvfflatLeaderParticipateAsWorker(IvfflatBuildState * buildstate)
+{
+	IvfflatLeader *ivfleader = buildstate->ivfleader;
+	IvfflatSpool *leaderworker;
+	int			sortmem;
+
+	/* Allocate memory and initialize private spool */
+	leaderworker = (IvfflatSpool *) palloc0(sizeof(IvfflatSpool));
+	leaderworker->heap = buildstate->heap;
+	leaderworker->index = buildstate->index;
+
+	/* Perform work common to all participants */
+	sortmem = maintenance_work_mem / ivfleader->nparticipanttuplesorts;
+	IvfflatParallelScanAndSort(leaderworker, ivfleader->ivfshared,
+							   ivfleader->sharedsort, ivfleader->ivfcenters,
+							   sortmem, true);
+}
+
+/*
+ * Begin parallel build
+ */
+static void
+IvfflatBeginParallel(IvfflatBuildState * buildstate, bool isconcurrent, int request)
+{
+	ParallelContext *pcxt;
+	int			scantuplesortstates;
+	Snapshot	snapshot;
+	Size		estivfshared;
+	Size		estsort;
+	Size		estcenters;
+	IvfflatShared *ivfshared;
+	Sharedsort *sharedsort;
+	Vector	   *ivfcenters;
+	IvfflatLeader *ivfleader = (IvfflatLeader *) palloc0(sizeof(IvfflatLeader));
+	bool		leaderparticipates = true;
+	int			querylen;
+
+#ifdef DISABLE_LEADER_PARTICIPATION
+	leaderparticipates = false;
+#endif
+
+	/* Enter parallel mode and create context */
+	EnterParallelMode();
+	Assert(request > 0);
+#if PG_VERSION_NUM >= 120000
+	pcxt = CreateParallelContext("vector", "IvfflatParallelBuildMain", request);
+#else
+	pcxt = CreateParallelContext("vector", "IvfflatParallelBuildMain", request, true);
+#endif
+
+	scantuplesortstates = leaderparticipates ? request + 1 : request;
+
+	/* Get snapshot for table scan */
+	if (!isconcurrent)
+		snapshot = SnapshotAny;
+	else
+		snapshot = RegisterSnapshot(GetTransactionSnapshot());
+
+	/* Estimate size of workspaces */
+	estivfshared = ParallelEstimateShared(buildstate->heap, snapshot);
+	shm_toc_estimate_chunk(&pcxt->estimator, estivfshared);
+	estsort = tuplesort_estimate_shared(scantuplesortstates);
+	shm_toc_estimate_chunk(&pcxt->estimator, estsort);
+	estcenters = VECTOR_SIZE(buildstate->dimensions) * buildstate->lists;
+	shm_toc_estimate_chunk(&pcxt->estimator, estcenters);
+	shm_toc_estimate_keys(&pcxt->estimator, 3);
+
+	/* Finally, estimate PARALLEL_KEY_QUERY_TEXT space */
+	if (debug_query_string)
+	{
+		querylen = strlen(debug_query_string);
+		shm_toc_estimate_chunk(&pcxt->estimator, querylen + 1);
+		shm_toc_estimate_keys(&pcxt->estimator, 1);
+	}
+	else
+		querylen = 0;			/* keep compiler quiet */
+
+	/* Everyone's had a chance to ask for space, so now create the DSM */
+	InitializeParallelDSM(pcxt);
+
+	/* If no DSM segment was available, back out (do serial build) */
+	if (pcxt->seg == NULL)
+	{
+		if (IsMVCCSnapshot(snapshot))
+			UnregisterSnapshot(snapshot);
+		DestroyParallelContext(pcxt);
+		ExitParallelMode();
+		return;
+	}
+
+	/* Store shared build state, for which we reserved space */
+	ivfshared = (IvfflatShared *) shm_toc_allocate(pcxt->toc, estivfshared);
+	/* Initialize immutable state */
+	ivfshared->heaprelid = RelationGetRelid(buildstate->heap);
+	ivfshared->indexrelid = RelationGetRelid(buildstate->index);
+	ivfshared->isconcurrent = isconcurrent;
+	ivfshared->scantuplesortstates = scantuplesortstates;
+	ConditionVariableInit(&ivfshared->workersdonecv);
+	SpinLockInit(&ivfshared->mutex);
+	/* Initialize mutable state */
+	ivfshared->nparticipantsdone = 0;
+	ivfshared->reltuples = 0;
+	ivfshared->indtuples = 0;
+#ifdef IVFFLAT_KMEANS_DEBUG
+	ivfshared->inertia = 0;
+#endif
+#if PG_VERSION_NUM >= 120000
+	table_parallelscan_initialize(buildstate->heap,
+								  ParallelTableScanFromIvfflatShared(ivfshared),
+								  snapshot);
+#else
+	heap_parallelscan_initialize(&ivfshared->heapdesc, buildstate->heap, snapshot);
+#endif
+
+	/* Store shared tuplesort-private state, for which we reserved space */
+	sharedsort = (Sharedsort *) shm_toc_allocate(pcxt->toc, estsort);
+	tuplesort_initialize_shared(sharedsort, scantuplesortstates,
+								pcxt->seg);
+
+	ivfcenters = (Vector *) shm_toc_allocate(pcxt->toc, estcenters);
+	memcpy(ivfcenters, buildstate->centers->items, estcenters);
+
+	shm_toc_insert(pcxt->toc, PARALLEL_KEY_IVFFLAT_SHARED, ivfshared);
+	shm_toc_insert(pcxt->toc, PARALLEL_KEY_TUPLESORT, sharedsort);
+	shm_toc_insert(pcxt->toc, PARALLEL_KEY_IVFFLAT_CENTERS, ivfcenters);
+
+	/* Store query string for workers */
+	if (debug_query_string)
+	{
+		char	   *sharedquery;
+
+		sharedquery = (char *) shm_toc_allocate(pcxt->toc, querylen + 1);
+		memcpy(sharedquery, debug_query_string, querylen + 1);
+		shm_toc_insert(pcxt->toc, PARALLEL_KEY_QUERY_TEXT, sharedquery);
+	}
+
+	/* Launch workers, saving status for leader/caller */
+	LaunchParallelWorkers(pcxt);
+	ivfleader->pcxt = pcxt;
+	ivfleader->nparticipanttuplesorts = pcxt->nworkers_launched;
+	if (leaderparticipates)
+		ivfleader->nparticipanttuplesorts++;
+	ivfleader->ivfshared = ivfshared;
+	ivfleader->sharedsort = sharedsort;
+	ivfleader->snapshot = snapshot;
+	ivfleader->ivfcenters = ivfcenters;
+
+	/* If no workers were successfully launched, back out (do serial build) */
+	if (pcxt->nworkers_launched == 0)
+	{
+		IvfflatEndParallel(ivfleader);
+		return;
+	}
+
+	/* Log participants */
+	ereport(DEBUG1, (errmsg("using %d parallel workers", pcxt->nworkers_launched)));
+
+	/* Save leader state now that it's clear build will be parallel */
+	buildstate->ivfleader = ivfleader;
+
+	/* Join heap scan ourselves */
+	if (leaderparticipates)
+		IvfflatLeaderParticipateAsWorker(buildstate);
+
+	/* Wait for all launched workers */
+	WaitForParallelWorkersToAttach(pcxt);
+}
+
+/*
+ * Scan table for tuples to index
+ */
+static void
+AssignTuples(IvfflatBuildState * buildstate)
+{
+	int			parallel_workers = 0;
+	SortCoordinate coordinate = NULL;
+
+	/* Sort options, which must match IvfflatParallelScanAndSort */
+	AttrNumber	attNums[] = {1};
+	Oid			sortOperators[] = {Int4LessOperator};
+	Oid			sortCollations[] = {InvalidOid};
+	bool		nullsFirstFlags[] = {false};
+
+	UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_ASSIGN);
+
+	/* Calculate parallel workers */
+	if (buildstate->heap != NULL)
+		parallel_workers = plan_create_index_workers(RelationGetRelid(buildstate->heap), RelationGetRelid(buildstate->index));
+
+	/* Attempt to launch parallel worker scan when required */
+	if (parallel_workers > 0)
+		IvfflatBeginParallel(buildstate, buildstate->indexInfo->ii_Concurrent, parallel_workers);
+
+	/* Set up coordination state if at least one worker launched */
+	if (buildstate->ivfleader)
+	{
+		coordinate = (SortCoordinate) palloc0(sizeof(SortCoordinateData));
+		coordinate->isWorker = false;
+		coordinate->nParticipants = buildstate->ivfleader->nparticipanttuplesorts;
+		coordinate->sharedsort = buildstate->ivfleader->sharedsort;
+	}
+
+	/* Begin serial/leader tuplesort */
+	buildstate->sortstate = tuplesort_begin_heap(buildstate->tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, maintenance_work_mem, coordinate, false);
+
+	/* Add tuples to sort */
+	if (buildstate->heap != NULL)
+	{
+		if (buildstate->ivfleader)
+			buildstate->reltuples = ParallelHeapScan(buildstate);
+		else
+		{
+#if PG_VERSION_NUM >= 120000
+			buildstate->reltuples = table_index_build_scan(buildstate->heap, buildstate->index, buildstate->indexInfo,
+														   true, true, BuildCallback, (void *) buildstate, NULL);
+#else
+			buildstate->reltuples = IndexBuildHeapScan(buildstate->heap, buildstate->index, buildstate->indexInfo,
+													   true, BuildCallback, (void *) buildstate, NULL);
+#endif
+		}
+
+#ifdef IVFFLAT_KMEANS_DEBUG
+		PrintKmeansMetrics(buildstate);
+#endif
+	}
+}
+
 /*
  * Create entry pages
  */
 static void
 CreateEntryPages(IvfflatBuildState * buildstate, ForkNumber forkNum)
 {
-	AttrNumber	attNums[] = {1};
-	Oid			sortOperators[] = {Float8LessOperator};
-	Oid			sortCollations[] = {InvalidOid};
-	bool		nullsFirstFlags[] = {false};
-
-	UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_SORT);
-
-	buildstate->sortstate = tuplesort_begin_heap(buildstate->tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, maintenance_work_mem, NULL, false);
-
-	/* Add tuples to sort */
-	if (buildstate->heap != NULL)
-	{
-#if PG_VERSION_NUM >= 120000
-		buildstate->reltuples = table_index_build_scan(buildstate->heap, buildstate->index, buildstate->indexInfo,
-													   true, true, BuildCallback, (void *) buildstate, NULL);
-#else
-		buildstate->reltuples = IndexBuildHeapScan(buildstate->heap, buildstate->index, buildstate->indexInfo,
-												   true, BuildCallback, (void *) buildstate, NULL);
-#endif
-	}
+	/* Assign */
+	IvfflatBench("assign tuples", AssignTuples(buildstate));
 
 	/* Sort */
-	tuplesort_performsort(buildstate->sortstate);
+	IvfflatBench("sort tuples", tuplesort_performsort(buildstate->sortstate));
 
-#ifdef IVFFLAT_KMEANS_DEBUG
-	PrintKmeansMetrics(buildstate);
-#endif
+	/* Load */
+	IvfflatBench("load tuples", InsertTuples(buildstate->index, buildstate, forkNum));
 
-	/* Insert */
-	InsertTuples(buildstate->index, buildstate, forkNum);
+	/* End sort */
 	tuplesort_end(buildstate->sortstate);
+
+	/* End parallel build */
+	if (buildstate->ivfleader)
+		IvfflatEndParallel(buildstate->ivfleader);
 }
 
 /*
@@ -611,7 +1076,7 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
 	/* Create pages */
 	CreateMetaPage(index, buildstate->dimensions, buildstate->lists, forkNum);
 	CreateListPages(index, buildstate->centers, buildstate->dimensions, buildstate->lists, forkNum, &buildstate->listInfo);
-	IvfflatBench("CreateEntryPages", CreateEntryPages(buildstate, forkNum));
+	CreateEntryPages(buildstate, forkNum);
 
 	FreeBuildState(buildstate);
 }

src/ivfflat.c

@@ -7,6 +7,7 @@
 #include "ivfflat.h"
 #include "utils/guc.h"
 #include "utils/selfuncs.h"
+#include "utils/spccache.h"
 
 #if PG_VERSION_NUM >= 120000
 #include "commands/progress.h"
@@ -19,11 +20,11 @@ static relopt_kind ivfflat_relopt_kind;
  * Initialize index options and variables
  */
 void
-_PG_init(void)
+IvfflatInit(void)
 {
 	ivfflat_relopt_kind = add_reloption_kind();
 	add_int_reloption(ivfflat_relopt_kind, "lists", "Number of inverted lists",
-					  IVFFLAT_DEFAULT_LISTS, 1, IVFFLAT_MAX_LISTS
+					  IVFFLAT_DEFAULT_LISTS, IVFFLAT_MIN_LISTS, IVFFLAT_MAX_LISTS
 #if PG_VERSION_NUM >= 130000
 					  ,AccessExclusiveLock
 #endif
@@ -31,7 +32,7 @@ _PG_init(void)
 
 	DefineCustomIntVariable("ivfflat.probes", "Sets the number of probes",
 							"Valid range is 1..lists.", &ivfflat_probes,
-							1, 1, IVFFLAT_MAX_LISTS, PGC_USERSET, 0, NULL, NULL, NULL);
+							IVFFLAT_DEFAULT_PROBES, IVFFLAT_MIN_LISTS, IVFFLAT_MAX_LISTS, PGC_USERSET, 0, NULL, NULL, NULL);
 }
 
 /*
@@ -47,8 +48,8 @@ ivfflatbuildphasename(int64 phasenum)
 			return "initializing";
 		case PROGRESS_IVFFLAT_PHASE_KMEANS:
 			return "performing k-means";
-		case PROGRESS_IVFFLAT_PHASE_SORT:
-			return "sorting tuples";
+		case PROGRESS_IVFFLAT_PHASE_ASSIGN:
+			return "assigning tuples";
 		case PROGRESS_IVFFLAT_PHASE_LOAD:
 			return "loading tuples";
 		default:
@@ -63,14 +64,14 @@ ivfflatbuildphasename(int64 phasenum)
 static void
 ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
 					Cost *indexStartupCost, Cost *indexTotalCost,
-					Selectivity *indexSelectivity, double *indexCorrelation
-					,double *indexPages
-)
+					Selectivity *indexSelectivity, double *indexCorrelation,
+					double *indexPages)
 {
 	GenericCosts costs;
 	int			lists;
 	double		ratio;
-	Relation	indexRel;
+	double		spc_seq_page_cost;
+	Relation	index;
 #if PG_VERSION_NUM < 120000
 	List	   *qinfos;
 #endif
@@ -88,6 +89,22 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
 
 	MemSet(&costs, 0, sizeof(costs));
 
+	index = index_open(path->indexinfo->indexoid, NoLock);
+	IvfflatGetMetaPageInfo(index, &lists, NULL);
+	index_close(index, NoLock);
+
+	/* Get the ratio of lists that we need to visit */
+	ratio = ((double) ivfflat_probes) / lists;
+	if (ratio > 1.0)
+		ratio = 1.0;
+
+	/*
+	 * This gives us the subset of tuples to visit. This value is passed into
+	 * the generic cost estimator to determine the number of pages to visit
+	 * during the index scan.
+	 */
+	costs.numIndexTuples = path->indexinfo->tuples * ratio;
+
 #if PG_VERSION_NUM >= 120000
 	genericcostestimate(root, path, loop_count, &costs);
 #else
@@ -95,17 +112,31 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
 	genericcostestimate(root, path, loop_count, qinfos, &costs);
 #endif
 
-	indexRel = index_open(path->indexinfo->indexoid, NoLock);
-	lists = IvfflatGetLists(indexRel);
-	index_close(indexRel, NoLock);
+	get_tablespace_page_costs(path->indexinfo->reltablespace, NULL, &spc_seq_page_cost);
 
-	ratio = ((double) ivfflat_probes) / lists;
-	if (ratio > 1)
-		ratio = 1;
+	/* Adjust cost if needed since TOAST not included in seq scan cost */
+	if (costs.numIndexPages > path->indexinfo->rel->pages && ratio < 0.5)
+	{
+		/* Change all page cost from random to sequential */
+		costs.indexTotalCost -= costs.numIndexPages * (costs.spc_random_page_cost - spc_seq_page_cost);
 
-	costs.indexTotalCost *= ratio;
+		/* Remove cost of extra pages */
+		costs.indexTotalCost -= (costs.numIndexPages - path->indexinfo->rel->pages) * spc_seq_page_cost;
+	}
+	else
+	{
+		/* Change some page cost from random to sequential */
+		costs.indexTotalCost -= 0.5 * costs.numIndexPages * (costs.spc_random_page_cost - spc_seq_page_cost);
+	}
 
-	/* Startup cost and total cost are same */
+	/*
+	 * If the list selectivity is lower than what is returned from the generic
+	 * cost estimator, use that.
+	 */
+	if (ratio < costs.indexSelectivity)
+		costs.indexSelectivity = ratio;
+
+	/* Use total cost since most work happens before first tuple is returned */
 	*indexStartupCost = costs.indexTotalCost;
 	*indexTotalCost = costs.indexTotalCost;
 	*indexSelectivity = costs.indexSelectivity;

src/ivfflat.h

@@ -3,14 +3,11 @@
 
 #include "postgres.h"
 
-#if PG_VERSION_NUM < 110000
-#error "Requires PostgreSQL 11+"
-#endif
-
 #include "access/generic_xlog.h"
+#include "access/parallel.h"
 #include "access/reloptions.h"
 #include "nodes/execnodes.h"
-#include "port.h"				/* for strtof() and random() */
+#include "port.h"				/* for random() */
 #include "utils/sampling.h"
 #include "utils/tuplesort.h"
 #include "vector.h"
@@ -19,6 +16,10 @@
 #include "common/pg_prng.h"
 #endif
 
+#if PG_VERSION_NUM < 120000
+#include "access/relscan.h"
+#endif
+
 #ifdef IVFFLAT_BENCH
 #include "portability/instr_time.h"
 #endif
@@ -39,13 +40,16 @@
 #define IVFFLAT_METAPAGE_BLKNO	0
 #define IVFFLAT_HEAD_BLKNO		1	/* first list page */
 
+/* IVFFlat parameters */
 #define IVFFLAT_DEFAULT_LISTS	100
+#define IVFFLAT_MIN_LISTS		1
 #define IVFFLAT_MAX_LISTS		32768
+#define IVFFLAT_DEFAULT_PROBES	1
 
 /* Build phases */
 /* PROGRESS_CREATEIDX_SUBPHASE_INITIALIZE is 1 */
 #define PROGRESS_IVFFLAT_PHASE_KMEANS	2
-#define PROGRESS_IVFFLAT_PHASE_SORT		3
+#define PROGRESS_IVFFLAT_PHASE_ASSIGN	3
 #define PROGRESS_IVFFLAT_PHASE_LOAD		4
 
 #define IVFFLAT_LIST_SIZE(_dim)	(offsetof(IvfflatListData, center) + VECTOR_SIZE(_dim))
@@ -79,9 +83,6 @@
 /* Variables */
 extern int	ivfflat_probes;
 
-/* Exported functions */
-PGDLLEXPORT void _PG_init(void);
-
 typedef struct VectorArrayData
 {
 	int			length;
@@ -105,6 +106,56 @@ typedef struct IvfflatOptions
 	int			lists;			/* number of lists */
 }			IvfflatOptions;
 
+typedef struct IvfflatSpool
+{
+	Tuplesortstate *sortstate;
+	Relation	heap;
+	Relation	index;
+}			IvfflatSpool;
+
+typedef struct IvfflatShared
+{
+	/* Immutable state */
+	Oid			heaprelid;
+	Oid			indexrelid;
+	bool		isconcurrent;
+	int			scantuplesortstates;
+
+	/* Worker progress */
+	ConditionVariable workersdonecv;
+
+	/* Mutex for mutable state */
+	slock_t		mutex;
+
+	/* Mutable state */
+	int			nparticipantsdone;
+	double		reltuples;
+	double		indtuples;
+
+#ifdef IVFFLAT_KMEANS_DEBUG
+	double		inertia;
+#endif
+
+#if PG_VERSION_NUM < 120000
+	ParallelHeapScanDescData heapdesc;	/* must come last */
+#endif
+}			IvfflatShared;
+
+#if PG_VERSION_NUM >= 120000
+#define ParallelTableScanFromIvfflatShared(shared) \
+	(ParallelTableScanDesc) ((char *) (shared) + BUFFERALIGN(sizeof(IvfflatShared)))
+#endif
+
+typedef struct IvfflatLeader
+{
+	ParallelContext *pcxt;
+	int			nparticipanttuplesorts;
+	IvfflatShared *ivfshared;
+	Sharedsort *sharedsort;
+	Snapshot	snapshot;
+	Vector	   *ivfcenters;
+}			IvfflatLeader;
+
 typedef struct IvfflatBuildState
 {
 	/* Info */
@@ -150,6 +201,9 @@ typedef struct IvfflatBuildState
 
 	/* Memory */
 	MemoryContext tmpCtx;
+
+	/* Parallel builds */
+	IvfflatLeader *ivfleader;
 }			IvfflatBuildState;
 
 typedef struct IvfflatMetaPageData
@@ -190,8 +244,8 @@ typedef struct IvfflatScanList
 typedef struct IvfflatScanOpaqueData
 {
 	int			probes;
+	int			dimensions;
 	bool		first;
-	Buffer		buf;
 
 	/* Sorting */
 	Tuplesortstate *sortstate;
@@ -221,15 +275,18 @@ 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		IvfflatUpdateList(Relation index, GenericXLogState *state, ListInfo listInfo, BlockNumber insertPage, BlockNumber originalInsertPage, BlockNumber startPage, ForkNumber forkNum);
+void		IvfflatGetMetaPageInfo(Relation index, int *lists, int *dimensions);
+void		IvfflatUpdateList(Relation index, ListInfo listInfo, BlockNumber insertPage, BlockNumber originalInsertPage, BlockNumber startPage, ForkNumber forkNum);
 void		IvfflatCommitBuffer(Buffer buf, GenericXLogState *state);
 void		IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state, ForkNumber forkNum);
 Buffer		IvfflatNewBuffer(Relation index, ForkNumber forkNum);
 void		IvfflatInitPage(Buffer buf, Page page);
 void		IvfflatInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state);
+void		IvfflatInit(void);
+PGDLLEXPORT void IvfflatParallelBuildMain(dsm_segment *seg, shm_toc *toc);
 
 /* Index access methods */
 IndexBuildResult *ivfflatbuild(Relation heap, Relation index, IndexInfo *indexInfo);

src/ivfinsert.c

@@ -4,42 +4,48 @@
 
 #include "ivfflat.h"
 #include "storage/bufmgr.h"
+#include "storage/lmgr.h"
 #include "utils/memutils.h"
 
 /*
  * 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;
 
-	procinfo = index_getprocinfo(rel, 1, IVFFLAT_DISTANCE_PROC);
-	collation = rel->rd_indcollation[0];
+	/* Avoid compiler warning */
+	listInfo->blkno = nextblkno;
+	listInfo->offno = FirstOffsetNumber;
+
+	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)));
 
-			if (distance < minDistance)
+			if (distance < minDistance || !BlockNumberIsValid(*insertPage))
 			{
 				*insertPage = list->insertPage;
 				listInfo->blkno = nextblkno;
@@ -58,7 +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;
@@ -75,33 +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)
@@ -117,23 +123,16 @@ InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Rel
 		}
 		else
 		{
-			Buffer		metabuf;
 			Buffer		newbuf;
 			Page		newpage;
 
-			/*
-			 * From ReadBufferExtended: Caller is responsible for ensuring
-			 * that only one backend tries to extend a relation at the same
-			 * time!
-			 */
-			metabuf = ReadBuffer(rel, IVFFLAT_METAPAGE_BLKNO);
-			LockBuffer(metabuf, BUFFER_LOCK_EXCLUSIVE);
-
 			/* Add a new page */
-			newbuf = IvfflatNewBuffer(rel, MAIN_FORKNUM);
-			newpage = GenericXLogRegisterBuffer(state, newbuf, GENERIC_XLOG_FULL_IMAGE);
+			LockRelationForExtension(index, ExclusiveLock);
+			newbuf = IvfflatNewBuffer(index, MAIN_FORKNUM);
+			UnlockRelationForExtension(index, ExclusiveLock);
 
 			/* Init new page */
+			newpage = GenericXLogRegisterBuffer(state, newbuf, GENERIC_XLOG_FULL_IMAGE);
 			IvfflatInitPage(newbuf, newpage);
 
 			/* Update insert page */
@@ -143,18 +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 extend relation lock as early as possible */
-			UnlockReleaseBuffer(metabuf);
-
 			/* Unlock previous buffer */
 			UnlockReleaseBuffer(buf);
 
 			/* Prepare new buffer */
-			state = GenericXLogStart(rel);
+			state = GenericXLogStart(index);
 			buf = newbuf;
 			page = GenericXLogRegisterBuffer(state, buf, 0);
 			break;
@@ -163,13 +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, state, listInfo, insertPage, originalInsertPage, InvalidBlockNumber, MAIN_FORKNUM);
+		IvfflatUpdateList(index, listInfo, insertPage, originalInsertPage, InvalidBlockNumber, MAIN_FORKNUM);
 }
 
 /*

src/ivfkmeans.c

@@ -1,6 +1,7 @@
 #include "postgres.h"
 
 #include <float.h>
+#include <math.h>
 
 #include "ivfflat.h"
 #include "miscadmin.h"
@@ -15,12 +16,7 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
 {
 	FmgrInfo   *procinfo;
 	Oid			collation;
-	int			i;
 	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 (i = 0; i < numCenters; i++)
+	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 */
@@ -87,13 +87,12 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
 static inline void
 ApplyNorm(FmgrInfo *normprocinfo, Oid collation, Vector * vec)
 {
-	int			i;
 	double		norm = DatumGetFloat8(FunctionCall1Coll(normprocinfo, collation, PointerGetDatum(vec)));
 
 	/* TODO Handle zero norm */
 	if (norm > 0)
 	{
-		for (i = 0; i < vec->dim; i++)
+		for (int i = 0; i < vec->dim; i++)
 			vec->x[i] /= norm;
 	}
 }
@@ -113,9 +112,6 @@ CompareVectors(const void *a, const void *b)
 static void
 QuickCenters(Relation index, VectorArray samples, VectorArray centers)
 {
-	int			i;
-	int			j;
-	Vector	   *vec;
 	int			dimensions = centers->dim;
 	Oid			collation = index->rd_indcollation[0];
 	FmgrInfo   *normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_KMEANS_NORM_PROC);
@@ -124,9 +120,9 @@ QuickCenters(Relation index, VectorArray samples, VectorArray centers)
 	if (samples->length > 0)
 	{
 		qsort(samples->items, samples->length, VECTOR_SIZE(samples->dim), CompareVectors);
-		for (i = 0; i < samples->length; i++)
+		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)
 			{
@@ -139,12 +135,12 @@ 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;
 
-		for (j = 0; j < dimensions; j++)
+		for (int j = 0; j < dimensions; j++)
 			vec->x[j] = RandomDouble();
 
 		/* Normalize if needed (only needed for random centers) */
@@ -171,7 +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;
@@ -185,14 +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);
@@ -211,7 +198,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 	/* Check memory requirements */
 	/* Add one to error message to ceil */
-	if (totalSize / 1024 > maintenance_work_mem)
+	if (totalSize > (Size) maintenance_work_mem * 1024L)
 		ereport(ERROR,
 				(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
 				 errmsg("memory required is %zu MB, maintenance_work_mem is %d MB",
@@ -250,14 +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 = -1;
+		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)
 			{
@@ -271,13 +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++)
 		{
@@ -285,7 +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;
 			}
@@ -294,10 +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;
 
@@ -313,6 +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;
@@ -321,6 +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;
@@ -350,7 +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;
@@ -364,7 +359,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 						changes++;
 					}
-
 				}
 			}
 		}
@@ -381,6 +375,8 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 		for (j = 0; j < numSamples; j++)
 		{
+			int			closestCenter;
+
 			vec = VectorArrayGet(samples, j);
 			closestCenter = closestCenters[j];
 
@@ -398,6 +394,14 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
 
 			if (centerCounts[j] > 0)
 			{
+				/* Double avoids overflow, but requires more memory */
+				/* TODO Update bounds */
+				for (k = 0; k < dimensions; k++)
+				{
+					if (isinf(vec->x[k]))
+						vec->x[k] = vec->x[k] > 0 ? FLT_MAX : -FLT_MAX;
+				}
+
 				for (k = 0; k < dimensions; k++)
 					vec->x[k] /= centerCounts[j];
 			}
@@ -421,7 +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;
@@ -437,7 +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;
@@ -460,17 +464,29 @@ static void
 CheckCenters(Relation index, VectorArray centers)
 {
 	FmgrInfo   *normprocinfo;
-	Oid			collation;
-	int			i;
-	double		norm;
 
 	if (centers->length != centers->maxlen)
 		elog(ERROR, "Not enough centers. Please report a bug.");
 
+	/* Ensure no NaN or infinite values */
+	for (int i = 0; i < centers->length; i++)
+	{
+		Vector	   *vec = VectorArrayGet(centers, i);
+
+		for (int j = 0; j < vec->dim; j++)
+		{
+			if (isnan(vec->x[j]))
+				elog(ERROR, "NaN detected. Please report a bug.");
+
+			if (isinf(vec->x[j]))
+				elog(ERROR, "Infinite value detected. Please report a bug.");
+		}
+	}
+
 	/* Ensure no duplicate centers */
 	/* Fine to sort in-place */
 	qsort(centers->items, centers->length, VECTOR_SIZE(centers->dim), CompareVectors);
-	for (i = 1; i < centers->length; i++)
+	for (int i = 1; i < centers->length; i++)
 	{
 		if (CompareVectors(VectorArrayGet(centers, i), VectorArrayGet(centers, i - 1)) == 0)
 			elog(ERROR, "Duplicate centers detected. Please report a bug.");
@@ -481,11 +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 (i = 0; i < centers->length; i++)
+		for (int i = 0; i < centers->length; i++)
 		{
-			norm = DatumGetFloat8(FunctionCall1Coll(normprocinfo, collation, PointerGetDatum(VectorArrayGet(centers, i))));
+			double		norm = DatumGetFloat8(FunctionCall1Coll(normprocinfo, collation, PointerGetDatum(VectorArrayGet(centers, i))));
+
 			if (norm == 0)
 				elog(ERROR, "Zero norm detected. Please report a bug.");
 		}

src/ivfscan.c

@@ -3,12 +3,12 @@
 #include <float.h>
 
 #include "access/relscan.h"
-#include "ivfflat.h"
-#include "miscadmin.h"
-#include "storage/bufmgr.h"
-
 #include "catalog/pg_operator_d.h"
 #include "catalog/pg_type_d.h"
+#include "ivfflat.h"
+#include "miscadmin.h"
+#include "pgstat.h"
+#include "storage/bufmgr.h"
 
 /*
  * Compare list distances
@@ -31,36 +31,36 @@ CompareLists(const pairingheap_node *a, const pairingheap_node *b, void *arg)
 static void
 GetScanLists(IndexScanDesc scan, Datum value)
 {
-	Buffer		cbuf;
-	Page		cpage;
-	IvfflatList list;
-	OffsetNumber offno;
-	OffsetNumber maxoffno;
+	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
 	BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
 	int			listCount = 0;
-	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
-	double		distance;
-	IvfflatScanList *scanlist;
 	double		maxDistance = DBL_MAX;
 
 	/* Search all list pages */
 	while (BlockNumberIsValid(nextblkno))
 	{
+		Buffer		cbuf;
+		Page		cpage;
+		OffsetNumber maxoffno;
+
 		cbuf = ReadBuffer(scan->indexRelation, nextblkno);
 		LockBuffer(cbuf, BUFFER_LOCK_SHARE);
 		cpage = BufferGetPage(cbuf);
 
 		maxoffno = PageGetMaxOffsetNumber(cpage);
 
-		for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+		for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
 		{
-			list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
+			IvfflatList list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
+			double		distance;
 
 			/* Use procinfo from the index instead of scan key for performance */
 			distance = DatumGetFloat8(FunctionCall2Coll(so->procinfo, so->collation, PointerGetDatum(&list->center), value));
 
 			if (listCount < so->probes)
 			{
+				IvfflatScanList *scanlist;
+
 				scanlist = &so->lists[listCount];
 				scanlist->startPage = list->startPage;
 				scanlist->distance = distance;
@@ -75,6 +75,8 @@ GetScanLists(IndexScanDesc scan, Datum value)
 			}
 			else if (distance < maxDistance)
 			{
+				IvfflatScanList *scanlist;
+
 				/* Remove */
 				scanlist = (IvfflatScanList *) pairingheap_remove_first(so->listQueue);
 
@@ -101,15 +103,8 @@ static void
 GetScanItems(IndexScanDesc scan, Datum value)
 {
 	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
-	Buffer		buf;
-	Page		page;
-	IndexTuple	itup;
-	BlockNumber searchPage;
-	OffsetNumber offno;
-	OffsetNumber maxoffno;
-	Datum		datum;
-	bool		isnull;
 	TupleDesc	tupdesc = RelationGetDescr(scan->indexRelation);
+	double		tuples = 0;
 
 #if PG_VERSION_NUM >= 120000
 	TupleTableSlot *slot = MakeSingleTupleTableSlot(so->tupdesc, &TTSOpsVirtual);
@@ -127,19 +122,28 @@ GetScanItems(IndexScanDesc scan, Datum value)
 	/* Search closest probes lists */
 	while (!pairingheap_is_empty(so->listQueue))
 	{
-		searchPage = ((IvfflatScanList *) pairingheap_remove_first(so->listQueue))->startPage;
+		BlockNumber searchPage = ((IvfflatScanList *) pairingheap_remove_first(so->listQueue))->startPage;
 
 		/* Search all entry pages for list */
 		while (BlockNumberIsValid(searchPage))
 		{
+			Buffer		buf;
+			Page		page;
+			OffsetNumber maxoffno;
+
 			buf = ReadBufferExtended(scan->indexRelation, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas);
 			LockBuffer(buf, BUFFER_LOCK_SHARE);
 			page = BufferGetPage(buf);
 			maxoffno = PageGetMaxOffsetNumber(page);
 
-			for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
+			for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
 			{
-				itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offno));
+				IndexTuple	itup;
+				Datum		datum;
+				bool		isnull;
+				ItemId		itemid = PageGetItemId(page, offno);
+
+				itup = (IndexTuple) PageGetItem(page, itemid);
 				datum = index_getattr(itup, 1, tupdesc, &isnull);
 
 				/*
@@ -153,11 +157,11 @@ GetScanItems(IndexScanDesc scan, Datum value)
 				slot->tts_isnull[0] = false;
 				slot->tts_values[1] = PointerGetDatum(&itup->t_tid);
 				slot->tts_isnull[1] = false;
-				slot->tts_values[2] = Int32GetDatum((int) searchPage);
-				slot->tts_isnull[2] = false;
 				ExecStoreVirtualTuple(slot);
 
 				tuplesort_puttupleslot(so->sortstate, slot);
+
+				tuples++;
 			}
 
 			searchPage = IvfflatPageGetOpaque(page)->nextblkno;
@@ -166,6 +170,14 @@ GetScanItems(IndexScanDesc scan, Datum value)
 		}
 	}
 
+	FreeAccessStrategy(bas);
+
+	if (tuples < 100)
+		ereport(DEBUG1,
+				(errmsg("index scan found few tuples"),
+				 errdetail("Index may have been created with little data."),
+				 errhint("Recreate the index and possibly decrease lists.")));
+
 	tuplesort_performsort(so->sortstate);
 }
 
@@ -178,6 +190,7 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 	IndexScanDesc scan;
 	IvfflatScanOpaque so;
 	int			lists;
+	int			dimensions;
 	AttrNumber	attNums[] = {1};
 	Oid			sortOperators[] = {Float8LessOperator};
 	Oid			sortCollations[] = {InvalidOid};
@@ -185,15 +198,17 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 	int			probes = ivfflat_probes;
 
 	scan = RelationGetIndexScan(index, nkeys, norderbys);
-	lists = IvfflatGetLists(scan->indexRelation);
+
+	/* Get lists and dimensions from metapage */
+	IvfflatGetMetaPageInfo(index, &lists, &dimensions);
 
 	if (probes > lists)
 		probes = lists;
 
 	so = (IvfflatScanOpaque) palloc(offsetof(IvfflatScanOpaqueData, lists) + probes * sizeof(IvfflatScanList));
-	so->buf = InvalidBuffer;
 	so->first = true;
 	so->probes = probes;
+	so->dimensions = dimensions;
 
 	/* Set support functions */
 	so->procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC);
@@ -202,13 +217,12 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 
 	/* Create tuple description for sorting */
 #if PG_VERSION_NUM >= 120000
-	so->tupdesc = CreateTemplateTupleDesc(3);
+	so->tupdesc = CreateTemplateTupleDesc(2);
 #else
-	so->tupdesc = CreateTemplateTupleDesc(3, false);
+	so->tupdesc = CreateTemplateTupleDesc(2, false);
 #endif
 	TupleDescInitEntry(so->tupdesc, (AttrNumber) 1, "distance", FLOAT8OID, -1, 0);
-	TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "tid", TIDOID, -1, 0);
-	TupleDescInitEntry(so->tupdesc, (AttrNumber) 3, "indexblkno", INT4OID, -1, 0);
+	TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "heaptid", TIDOID, -1, 0);
 
 	/* Prep sort */
 	so->sortstate = tuplesort_begin_heap(so->tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, work_mem, NULL, false);
@@ -267,25 +281,31 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
 	{
 		Datum		value;
 
+		/* Count index scan for stats */
+		pgstat_count_index_scan(scan->indexRelation);
+
 		/* Safety check */
 		if (scan->orderByData == NULL)
 			elog(ERROR, "cannot scan ivfflat index without order");
 
-		/* No items will match if null */
+		/* Requires MVCC-compliant snapshot as not able to pin during sorting */
+		/* https://www.postgresql.org/docs/current/index-locking.html */
+		if (!IsMVCCSnapshot(scan->xs_snapshot))
+			elog(ERROR, "non-MVCC snapshots are not supported with ivfflat");
+
 		if (scan->orderByData->sk_flags & SK_ISNULL)
-			return false;
-
-		value = scan->orderByData->sk_argument;
-
-		/* Value should not be compressed or toasted */
-		Assert(!VARATT_IS_COMPRESSED(DatumGetPointer(value)));
-		Assert(!VARATT_IS_EXTENDED(DatumGetPointer(value)));
-
-		if (so->normprocinfo != NULL)
+			value = PointerGetDatum(InitVector(so->dimensions));
+		else
 		{
-			/* No items will match if normalization fails */
-			if (!IvfflatNormValue(so->normprocinfo, so->collation, &value, NULL))
-				return false;
+			value = scan->orderByData->sk_argument;
+
+			/* Value should not be compressed or toasted */
+			Assert(!VARATT_IS_COMPRESSED(DatumGetPointer(value)));
+			Assert(!VARATT_IS_EXTENDED(DatumGetPointer(value)));
+
+			/* Fine if normalization fails */
+			if (so->normprocinfo != NULL)
+				IvfflatNormValue(so->normprocinfo, so->collation, &value, NULL);
 		}
 
 		IvfflatBench("GetScanLists", GetScanLists(scan, value));
@@ -299,26 +319,14 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
 
 	if (tuplesort_gettupleslot(so->sortstate, true, false, so->slot, NULL))
 	{
-		ItemPointer tid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
-		BlockNumber indexblkno = DatumGetInt32(slot_getattr(so->slot, 3, &so->isnull));
+		ItemPointer heaptid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
 
 #if PG_VERSION_NUM >= 120000
-		scan->xs_heaptid = *tid;
+		scan->xs_heaptid = *heaptid;
 #else
-		scan->xs_ctup.t_self = *tid;
+		scan->xs_ctup.t_self = *heaptid;
 #endif
 
-		if (BufferIsValid(so->buf))
-			ReleaseBuffer(so->buf);
-
-		/*
-		 * An index scan must maintain a pin on the index page holding the
-		 * item last returned by amgettuple
-		 *
-		 * https://www.postgresql.org/docs/current/index-locking.html
-		 */
-		so->buf = ReadBuffer(scan->indexRelation, indexblkno);
-
 		scan->xs_recheckorderby = false;
 		return true;
 	}
@@ -334,10 +342,6 @@ ivfflatendscan(IndexScanDesc scan)
 {
 	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
 
-	/* Release pin */
-	if (BufferIsValid(so->buf))
-		ReleaseBuffer(so->buf);
-
 	pairingheap_free(so->listQueue);
 	tuplesort_end(so->sortstate);
 

src/ivfutils.c

@@ -35,9 +35,7 @@ VectorArrayFree(VectorArray arr)
 void
 PrintVectorArray(char *msg, VectorArray arr)
 {
-	int			i;
-
-	for (i = 0; i < arr->length; i++)
+	for (int i = 0; i < arr->length; i++)
 		PrintVector(msg, VectorArrayGet(arr, i));
 }
 
@@ -59,12 +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);
 }
 
 /*
@@ -78,20 +76,16 @@ IvfflatOptionalProcInfo(Relation rel, uint16 procnum)
 bool
 IvfflatNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result)
 {
-	Vector	   *v;
-	int			i;
-	double		norm;
-
-	norm = DatumGetFloat8(FunctionCall1Coll(procinfo, collation, *value));
+	double		norm = DatumGetFloat8(FunctionCall1Coll(procinfo, collation, *value));
 
 	if (norm > 0)
 	{
-		v = DatumGetVector(*value);
+		Vector	   *v = DatumGetVector(*value);
 
 		if (result == NULL)
 			result = InitVector(v->dim);
 
-		for (i = 0; i < v->dim; i++)
+		for (int i = 0; i < v->dim; i++)
 			result->x[i] = v->x[i] / norm;
 
 		*value = PointerGetDatum(result);
@@ -142,7 +136,6 @@ IvfflatInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogStat
 void
 IvfflatCommitBuffer(Buffer buf, GenericXLogState *state)
 {
-	MarkBufferDirty(buf);
 	GenericXLogFinish(state);
 	UnlockReleaseBuffer(buf);
 }
@@ -166,8 +159,6 @@ IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **st
 	IvfflatInitPage(newbuf, newpage);
 
 	/* Commit */
-	MarkBufferDirty(*buf);
-	MarkBufferDirty(newbuf);
 	GenericXLogFinish(*state);
 
 	/* Unlock */
@@ -178,16 +169,40 @@ IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **st
 	*buf = newbuf;
 }
 
+/*
+ * Get the metapage info
+ */
+void
+IvfflatGetMetaPageInfo(Relation index, int *lists, int *dimensions)
+{
+	Buffer		buf;
+	Page		page;
+	IvfflatMetaPage metap;
+
+	buf = ReadBuffer(index, IVFFLAT_METAPAGE_BLKNO);
+	LockBuffer(buf, BUFFER_LOCK_SHARE);
+	page = BufferGetPage(buf);
+	metap = IvfflatPageGetMeta(page);
+
+	*lists = metap->lists;
+
+	if (dimensions != NULL)
+		*dimensions = metap->dimensions;
+
+	UnlockReleaseBuffer(buf);
+}
+
 /*
  * Update the start or insert page of a list
  */
 void
-IvfflatUpdateList(Relation index, GenericXLogState *state, ListInfo listInfo,
+IvfflatUpdateList(Relation index, ListInfo listInfo,
 				  BlockNumber insertPage, BlockNumber originalInsertPage,
 				  BlockNumber startPage, ForkNumber forkNum)
 {
 	Buffer		buf;
 	Page		page;
+	GenericXLogState *state;
 	IvfflatList list;
 	bool		changed = false;
 

src/ivfvacuum.c

@@ -12,34 +12,23 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
 				  IndexBulkDeleteCallback callback, void *callback_state)
 {
 	Relation	index = info->index;
-	Buffer		cbuf;
-	Page		cpage;
-	Buffer		buf;
-	Page		page;
-	IvfflatList list;
-	IndexTuple	itup;
-	ItemPointer htup;
-	OffsetNumber deletable[MaxOffsetNumber];
-	int			ndeletable;
-	BlockNumber startPages[MaxOffsetNumber];
-	BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
-	BlockNumber searchPage;
-	BlockNumber insertPage;
-	GenericXLogState *state;
-	OffsetNumber coffno;
-	OffsetNumber cmaxoffno;
-	OffsetNumber offno;
-	OffsetNumber maxoffno;
-	ListInfo	listInfo;
+	BlockNumber blkno = IVFFLAT_HEAD_BLKNO;
 	BufferAccessStrategy bas = GetAccessStrategy(BAS_BULKREAD);
 
 	if (stats == NULL)
 		stats = (IndexBulkDeleteResult *) palloc0(sizeof(IndexBulkDeleteResult));
 
 	/* Iterate over list pages */
-	while (BlockNumberIsValid(nextblkno))
+	while (BlockNumberIsValid(blkno))
 	{
-		cbuf = ReadBuffer(index, nextblkno);
+		Buffer		cbuf;
+		Page		cpage;
+		OffsetNumber coffno;
+		OffsetNumber cmaxoffno;
+		BlockNumber startPages[MaxOffsetNumber];
+		ListInfo	listInfo;
+
+		cbuf = ReadBuffer(index, blkno);
 		LockBuffer(cbuf, BUFFER_LOCK_SHARE);
 		cpage = BufferGetPage(cbuf);
 
@@ -48,23 +37,32 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
 		/* Iterate over lists */
 		for (coffno = FirstOffsetNumber; coffno <= cmaxoffno; coffno = OffsetNumberNext(coffno))
 		{
-			list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, coffno));
+			IvfflatList list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, coffno));
+
 			startPages[coffno - FirstOffsetNumber] = list->startPage;
 		}
 
-		listInfo.blkno = nextblkno;
-		nextblkno = IvfflatPageGetOpaque(cpage)->nextblkno;
+		listInfo.blkno = blkno;
+		blkno = IvfflatPageGetOpaque(cpage)->nextblkno;
 
 		UnlockReleaseBuffer(cbuf);
 
 		for (coffno = FirstOffsetNumber; coffno <= cmaxoffno; coffno = OffsetNumberNext(coffno))
 		{
-			searchPage = startPages[coffno - FirstOffsetNumber];
-			insertPage = InvalidBlockNumber;
+			BlockNumber searchPage = startPages[coffno - FirstOffsetNumber];
+			BlockNumber insertPage = InvalidBlockNumber;
 
 			/* Iterate over entry pages */
 			while (BlockNumberIsValid(searchPage))
 			{
+				Buffer		buf;
+				Page		page;
+				GenericXLogState *state;
+				OffsetNumber offno;
+				OffsetNumber maxoffno;
+				OffsetNumber deletable[MaxOffsetNumber];
+				int			ndeletable;
+
 				vacuum_delay_point();
 
 				buf = ReadBufferExtended(index, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas);
@@ -86,8 +84,8 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
 				/* Find deleted tuples */
 				for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
 				{
-					itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offno));
-					htup = &(itup->t_tid);
+					IndexTuple	itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offno));
+					ItemPointer htup = &(itup->t_tid);
 
 					if (callback(htup, callback_state))
 					{
@@ -109,7 +107,6 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
 				{
 					/* Delete tuples */
 					PageIndexMultiDelete(page, deletable, ndeletable);
-					MarkBufferDirty(buf);
 					GenericXLogFinish(state);
 				}
 				else
@@ -127,11 +124,13 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
 			if (BlockNumberIsValid(insertPage))
 			{
 				listInfo.offno = coffno;
-				IvfflatUpdateList(index, state, listInfo, insertPage, InvalidBlockNumber, InvalidBlockNumber, MAIN_FORKNUM);
+				IvfflatUpdateList(index, listInfo, insertPage, InvalidBlockNumber, InvalidBlockNumber, MAIN_FORKNUM);
 			}
 		}
 	}
 
+	FreeAccessStrategy(bas);
+
 	return stats;
 }
 
@@ -143,6 +142,11 @@ ivfflatvacuumcleanup(IndexVacuumInfo *info, IndexBulkDeleteResult *stats)
 {
 	Relation	rel = info->index;
 
+	if (info->analyze_only)
+		return stats;
+
+	/* stats is NULL if ambulkdelete not called */
+	/* OK to return NULL if index not changed */
 	if (stats == NULL)
 		return NULL;
 

src/vector.c

@@ -2,15 +2,22 @@
 
 #include <math.h>
 
-#include "vector.h"
-#include "fmgr.h"
 #include "catalog/pg_type.h"
+#include "fmgr.h"
+#include "hnsw.h"
+#include "ivfflat.h"
 #include "lib/stringinfo.h"
 #include "libpq/pqformat.h"
+#include "port.h"				/* for strtof() */
 #include "utils/array.h"
 #include "utils/builtins.h"
 #include "utils/lsyscache.h"
 #include "utils/numeric.h"
+#include "vector.h"
+
+#if PG_VERSION_NUM >= 160000
+#include "varatt.h"
+#endif
 
 #if PG_VERSION_NUM >= 120000
 #include "common/shortest_dec.h"
@@ -29,6 +36,17 @@
 
 PG_MODULE_MAGIC;
 
+/*
+ * Initialize index options and variables
+ */
+PGDLLEXPORT void _PG_init(void);
+void
+_PG_init(void)
+{
+	HnswInit();
+	IvfflatInit();
+}
+
 /*
  * Ensure same dimensions
  */
@@ -42,7 +60,7 @@ CheckDims(Vector * a, Vector * b)
 }
 
 /*
- * Ensure expected dimension
+ * Ensure expected dimensions
  */
 static inline void
 CheckExpectedDim(int32 typmod, int dim)
@@ -53,7 +71,9 @@ CheckExpectedDim(int32 typmod, int dim)
 				 errmsg("expected %d dimensions, not %d", typmod, dim)));
 }
 
-
+/*
+ * Ensure valid dimensions
+ */
 static inline void
 CheckDim(int dim)
 {
@@ -69,7 +89,7 @@ CheckDim(int dim)
 }
 
 /*
- * Ensure finite elements
+ * Ensure finite element
  */
 static inline void
 CheckElement(float value)
@@ -79,13 +99,45 @@ CheckElement(float value)
 				(errcode(ERRCODE_DATA_EXCEPTION),
 				 errmsg("NaN not allowed in vector")));
 
-
 	if (isinf(value))
 		ereport(ERROR,
 				(errcode(ERRCODE_DATA_EXCEPTION),
 				 errmsg("infinite value not allowed in vector")));
 }
 
+/*
+ * Allocate and initialize a new vector
+ */
+Vector *
+InitVector(int dim)
+{
+	Vector	   *result;
+	int			size;
+
+	size = VECTOR_SIZE(dim);
+	result = (Vector *) palloc0(size);
+	SET_VARSIZE(result, size);
+	result->dim = dim;
+
+	return result;
+}
+
+/*
+ * Check for whitespace, since array_isspace() is static
+ */
+static inline bool
+vector_isspace(char ch)
+{
+	if (ch == ' ' ||
+		ch == '\t' ||
+		ch == '\n' ||
+		ch == '\r' ||
+		ch == '\v' ||
+		ch == '\f')
+		return true;
+	return false;
+}
+
 /*
  * Check state array
  */
@@ -100,7 +152,7 @@ CheckStateArray(ArrayType *statearray, const char *caller)
 	return (float8 *) ARR_DATA_PTR(statearray);
 }
 
-#if PG_VERSION_NUM < 120000
+#if PG_VERSION_NUM < 120003
 static pg_noinline void
 float_overflow_error(void)
 {
@@ -108,31 +160,15 @@ float_overflow_error(void)
 			(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
 			 errmsg("value out of range: overflow")));
 }
-#endif
 
-/*
- * Print vector - useful for debugging
- */
-void
-PrintVector(char *msg, Vector * vector)
+static pg_noinline void
+float_underflow_error(void)
 {
-	StringInfoData buf;
-	int			dim = vector->dim;
-	int			i;
-
-	initStringInfo(&buf);
-
-	appendStringInfoChar(&buf, '[');
-	for (i = 0; i < dim; i++)
-	{
-		if (i > 0)
-			appendStringInfoString(&buf, ",");
-		appendStringInfoString(&buf, float8out_internal(vector->x[i]));
-	}
-	appendStringInfoChar(&buf, ']');
-
-	elog(INFO, "%s = %s", msg, buf.data);
+	ereport(ERROR,
+			(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
+			 errmsg("value out of range: underflow")));
 }
+#endif
 
 /*
  * Convert textual representation to internal representation
@@ -143,17 +179,20 @@ vector_in(PG_FUNCTION_ARGS)
 {
 	char	   *str = PG_GETARG_CSTRING(0);
 	int32		typmod = PG_GETARG_INT32(2);
-	int			i;
 	float		x[VECTOR_MAX_DIM];
 	int			dim = 0;
 	char	   *pt;
 	char	   *stringEnd;
 	Vector	   *result;
+	char	   *lit = pstrdup(str);
+
+	while (vector_isspace(*str))
+		str++;
 
 	if (*str != '[')
 		ereport(ERROR,
 				(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-				 errmsg("malformed vector literal: \"%s\"", str),
+				 errmsg("malformed vector literal: \"%s\"", lit),
 				 errdetail("Vector contents must start with \"[\".")));
 
 	str++;
@@ -167,6 +206,15 @@ vector_in(PG_FUNCTION_ARGS)
 					(errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
 					 errmsg("vector cannot have more than %d dimensions", VECTOR_MAX_DIM)));
 
+		while (vector_isspace(*pt))
+			pt++;
+
+		/* Check for empty string like float4in */
+		if (*pt == '\0')
+			ereport(ERROR,
+					(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+					 errmsg("invalid input syntax for type vector: \"%s\"", lit)));
+
 		/* Use strtof like float4in to avoid a double-rounding problem */
 		x[dim] = strtof(pt, &stringEnd);
 		CheckElement(x[dim]);
@@ -175,37 +223,57 @@ vector_in(PG_FUNCTION_ARGS)
 		if (stringEnd == pt)
 			ereport(ERROR,
 					(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-					 errmsg("invalid input syntax for type vector: \"%s\"", pt)));
+					 errmsg("invalid input syntax for type vector: \"%s\"", lit)));
+
+		while (vector_isspace(*stringEnd))
+			stringEnd++;
 
 		if (*stringEnd != '\0' && *stringEnd != ']')
 			ereport(ERROR,
 					(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-					 errmsg("invalid input syntax for type vector: \"%s\"", pt)));
+					 errmsg("invalid input syntax for type vector: \"%s\"", lit)));
 
 		pt = strtok(NULL, ",");
 	}
 
-	if (*stringEnd != ']')
+	if (stringEnd == NULL || *stringEnd != ']')
 		ereport(ERROR,
 				(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-				 errmsg("malformed vector literal"),
+				 errmsg("malformed vector literal: \"%s\"", lit),
 				 errdetail("Unexpected end of input.")));
 
-	if (stringEnd[1] != '\0')
+	stringEnd++;
+
+	/* Only whitespace is allowed after the closing brace */
+	while (vector_isspace(*stringEnd))
+		stringEnd++;
+
+	if (*stringEnd != '\0')
 		ereport(ERROR,
 				(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-				 errmsg("malformed vector literal"),
+				 errmsg("malformed vector literal: \"%s\"", lit),
 				 errdetail("Junk after closing right brace.")));
 
+	/* Ensure no consecutive delimiters since strtok skips */
+	for (pt = lit + 1; *pt != '\0'; pt++)
+	{
+		if (pt[-1] == ',' && *pt == ',')
+			ereport(ERROR,
+					(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+					 errmsg("malformed vector literal: \"%s\"", lit)));
+	}
+
 	if (dim < 1)
 		ereport(ERROR,
 				(errcode(ERRCODE_DATA_EXCEPTION),
 				 errmsg("vector must have at least 1 dimension")));
 
+	pfree(lit);
+
 	CheckExpectedDim(typmod, dim);
 
 	result = InitVector(dim);
-	for (i = 0; i < dim; i++)
+	for (int i = 0; i < dim; i++)
 		result->x[i] = x[i];
 
 	PG_RETURN_POINTER(result);
@@ -222,7 +290,6 @@ vector_out(PG_FUNCTION_ARGS)
 	int			dim = vector->dim;
 	char	   *buf;
 	char	   *ptr;
-	int			i;
 	int			n;
 
 #if PG_VERSION_NUM < 120000
@@ -249,7 +316,7 @@ vector_out(PG_FUNCTION_ARGS)
 
 	*ptr = '[';
 	ptr++;
-	for (i = 0; i < dim; i++)
+	for (int i = 0; i < dim; i++)
 	{
 		if (i > 0)
 		{
@@ -272,6 +339,18 @@ vector_out(PG_FUNCTION_ARGS)
 	PG_RETURN_CSTRING(buf);
 }
 
+/*
+ * Print vector - useful for debugging
+ */
+void
+PrintVector(char *msg, Vector * vector)
+{
+	char	   *out = DatumGetPointer(DirectFunctionCall1(vector_out, PointerGetDatum(vector)));
+
+	elog(INFO, "%s = %s", msg, out);
+	pfree(out);
+}
+
 /*
  * Convert type modifier
  */
@@ -315,7 +394,6 @@ vector_recv(PG_FUNCTION_ARGS)
 	Vector	   *result;
 	int16		dim;
 	int16		unused;
-	int			i;
 
 	dim = pq_getmsgint(buf, sizeof(int16));
 	unused = pq_getmsgint(buf, sizeof(int16));
@@ -329,8 +407,11 @@ vector_recv(PG_FUNCTION_ARGS)
 				 errmsg("expected unused to be 0, not %d", unused)));
 
 	result = InitVector(dim);
-	for (i = 0; i < dim; i++)
+	for (int i = 0; i < dim; i++)
+	{
 		result->x[i] = pq_getmsgfloat4(buf);
+		CheckElement(result->x[i]);
+	}
 
 	PG_RETURN_POINTER(result);
 }
@@ -344,12 +425,11 @@ vector_send(PG_FUNCTION_ARGS)
 {
 	Vector	   *vec = PG_GETARG_VECTOR_P(0);
 	StringInfoData buf;
-	int			i;
 
 	pq_begintypsend(&buf);
 	pq_sendint(&buf, vec->dim, sizeof(int16));
 	pq_sendint(&buf, vec->unused, sizeof(int16));
-	for (i = 0; i < vec->dim; i++)
+	for (int i = 0; i < vec->dim; i++)
 		pq_sendfloat4(&buf, vec->x[i]);
 
 	PG_RETURN_BYTEA_P(pq_endtypsend(&buf));
@@ -357,17 +437,18 @@ vector_send(PG_FUNCTION_ARGS)
 
 /*
  * Convert vector to vector
+ * This is needed to check the type modifier
  */
 PGDLLEXPORT PG_FUNCTION_INFO_V1(vector);
 Datum
 vector(PG_FUNCTION_ARGS)
 {
-	Vector	   *arg = PG_GETARG_VECTOR_P(0);
+	Vector	   *vec = PG_GETARG_VECTOR_P(0);
 	int32		typmod = PG_GETARG_INT32(1);
 
-	CheckExpectedDim(typmod, arg->dim);
+	CheckExpectedDim(typmod, vec->dim);
 
-	PG_RETURN_POINTER(arg);
+	PG_RETURN_POINTER(vec);
 }
 
 /*
@@ -379,13 +460,11 @@ array_to_vector(PG_FUNCTION_ARGS)
 {
 	ArrayType  *array = PG_GETARG_ARRAYTYPE_P(0);
 	int32		typmod = PG_GETARG_INT32(1);
-	int			i;
 	Vector	   *result;
 	int16		typlen;
 	bool		typbyval;
 	char		typalign;
 	Datum	   *elemsp;
-	bool	   *nullsp;
 	int			nelemsp;
 
 	if (ARR_NDIM(array) > 1)
@@ -393,37 +472,55 @@ array_to_vector(PG_FUNCTION_ARGS)
 				(errcode(ERRCODE_DATA_EXCEPTION),
 				 errmsg("array must be 1-D")));
 
-	get_typlenbyvalalign(ARR_ELEMTYPE(array), &typlen, &typbyval, &typalign);
-	deconstruct_array(array, ARR_ELEMTYPE(array), typlen, typbyval, typalign, &elemsp, &nullsp, &nelemsp);
+	if (ARR_HASNULL(array) && array_contains_nulls(array))
+		ereport(ERROR,
+				(errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
+				 errmsg("array must not contain nulls")));
 
-	if (typmod == -1)
-		CheckDim(nelemsp);
-	else
-		CheckExpectedDim(typmod, nelemsp);
+	get_typlenbyvalalign(ARR_ELEMTYPE(array), &typlen, &typbyval, &typalign);
+	deconstruct_array(array, ARR_ELEMTYPE(array), typlen, typbyval, typalign, &elemsp, NULL, &nelemsp);
+
+	CheckDim(nelemsp);
+	CheckExpectedDim(typmod, nelemsp);
 
 	result = InitVector(nelemsp);
-	for (i = 0; i < nelemsp; i++)
+
+	if (ARR_ELEMTYPE(array) == INT4OID)
 	{
-		if (nullsp[i])
-			ereport(ERROR,
-					(errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
-					 errmsg("array must not containing NULLs")));
-
-		if (ARR_ELEMTYPE(array) == INT4OID)
+		for (int i = 0; i < nelemsp; i++)
 			result->x[i] = DatumGetInt32(elemsp[i]);
-		else if (ARR_ELEMTYPE(array) == FLOAT8OID)
-			result->x[i] = DatumGetFloat8(elemsp[i]);
-		else if (ARR_ELEMTYPE(array) == FLOAT4OID)
-			result->x[i] = DatumGetFloat4(elemsp[i]);
-		else if (ARR_ELEMTYPE(array) == NUMERICOID)
-			result->x[i] = DatumGetFloat4(DirectFunctionCall1(numeric_float4, NumericGetDatum(elemsp[i])));
-		else
-			ereport(ERROR,
-					(errcode(ERRCODE_DATA_EXCEPTION),
-					 errmsg("unsupported array type")));
-
-		CheckElement(result->x[i]);
 	}
+	else if (ARR_ELEMTYPE(array) == FLOAT8OID)
+	{
+		for (int i = 0; i < nelemsp; i++)
+			result->x[i] = DatumGetFloat8(elemsp[i]);
+	}
+	else if (ARR_ELEMTYPE(array) == FLOAT4OID)
+	{
+		for (int i = 0; i < nelemsp; i++)
+			result->x[i] = DatumGetFloat4(elemsp[i]);
+	}
+	else if (ARR_ELEMTYPE(array) == NUMERICOID)
+	{
+		for (int i = 0; i < nelemsp; i++)
+			result->x[i] = DatumGetFloat4(DirectFunctionCall1(numeric_float4, elemsp[i]));
+	}
+	else
+	{
+		ereport(ERROR,
+				(errcode(ERRCODE_DATA_EXCEPTION),
+				 errmsg("unsupported array type")));
+	}
+
+	/*
+	 * Free allocation from deconstruct_array. Do not free individual elements
+	 * when pass-by-reference since they point to original array.
+	 */
+	pfree(elemsp);
+
+	/* Check elements */
+	for (int i = 0; i < result->dim; i++)
+		CheckElement(result->x[i]);
 
 	PG_RETURN_POINTER(result);
 }
@@ -436,17 +533,18 @@ Datum
 vector_to_float4(PG_FUNCTION_ARGS)
 {
 	Vector	   *vec = PG_GETARG_VECTOR_P(0);
-	Datum	   *d;
+	Datum	   *datums;
 	ArrayType  *result;
-	int			i;
 
-	d = (Datum *) palloc(sizeof(Datum) * vec->dim);
+	datums = (Datum *) palloc(sizeof(Datum) * vec->dim);
 
-	for (i = 0; i < vec->dim; i++)
-		d[i] = Float4GetDatum(vec->x[i]);
+	for (int i = 0; i < vec->dim; i++)
+		datums[i] = Float4GetDatum(vec->x[i]);
 
 	/* Use TYPALIGN_INT for float4 */
-	result = construct_array(d, vec->dim, FLOAT4OID, sizeof(float4), true, TYPALIGN_INT);
+	result = construct_array(datums, vec->dim, FLOAT4OID, sizeof(float4), true, TYPALIGN_INT);
+
+	pfree(datums);
 
 	PG_RETURN_POINTER(result);
 }
@@ -462,18 +560,19 @@ l2_distance(PG_FUNCTION_ARGS)
 	Vector	   *b = PG_GETARG_VECTOR_P(1);
 	float	   *ax = a->x;
 	float	   *bx = b->x;
-	double		distance = 0.0;
-	double		diff;
+	float		distance = 0.0;
+	float		diff;
 
 	CheckDims(a, b);
 
+	/* Auto-vectorized */
 	for (int i = 0; i < a->dim; i++)
 	{
 		diff = ax[i] - bx[i];
 		distance += diff * diff;
 	}
 
-	PG_RETURN_FLOAT8(sqrt(distance));
+	PG_RETURN_FLOAT8(sqrt((double) distance));
 }
 
 /*
@@ -488,18 +587,19 @@ vector_l2_squared_distance(PG_FUNCTION_ARGS)
 	Vector	   *b = PG_GETARG_VECTOR_P(1);
 	float	   *ax = a->x;
 	float	   *bx = b->x;
-	double		distance = 0.0;
-	double		diff;
+	float		distance = 0.0;
+	float		diff;
 
 	CheckDims(a, b);
 
+	/* Auto-vectorized */
 	for (int i = 0; i < a->dim; i++)
 	{
 		diff = ax[i] - bx[i];
 		distance += diff * diff;
 	}
 
-	PG_RETURN_FLOAT8(distance);
+	PG_RETURN_FLOAT8((double) distance);
 }
 
 /*
@@ -513,14 +613,15 @@ inner_product(PG_FUNCTION_ARGS)
 	Vector	   *b = PG_GETARG_VECTOR_P(1);
 	float	   *ax = a->x;
 	float	   *bx = b->x;
-	double		distance = 0.0;
+	float		distance = 0.0;
 
 	CheckDims(a, b);
 
+	/* Auto-vectorized */
 	for (int i = 0; i < a->dim; i++)
 		distance += ax[i] * bx[i];
 
-	PG_RETURN_FLOAT8(distance);
+	PG_RETURN_FLOAT8((double) distance);
 }
 
 /*
@@ -534,14 +635,15 @@ vector_negative_inner_product(PG_FUNCTION_ARGS)
 	Vector	   *b = PG_GETARG_VECTOR_P(1);
 	float	   *ax = a->x;
 	float	   *bx = b->x;
-	double		distance = 0.0;
+	float		distance = 0.0;
 
 	CheckDims(a, b);
 
+	/* Auto-vectorized */
 	for (int i = 0; i < a->dim; i++)
 		distance += ax[i] * bx[i];
 
-	PG_RETURN_FLOAT8(distance * -1);
+	PG_RETURN_FLOAT8((double) distance * -1);
 }
 
 /*
@@ -555,12 +657,14 @@ cosine_distance(PG_FUNCTION_ARGS)
 	Vector	   *b = PG_GETARG_VECTOR_P(1);
 	float	   *ax = a->x;
 	float	   *bx = b->x;
-	double		distance = 0.0;
-	double		norma = 0.0;
-	double		normb = 0.0;
+	float		distance = 0.0;
+	float		norma = 0.0;
+	float		normb = 0.0;
+	double		similarity;
 
 	CheckDims(a, b);
 
+	/* Auto-vectorized */
 	for (int i = 0; i < a->dim; i++)
 	{
 		distance += ax[i] * bx[i];
@@ -568,7 +672,22 @@ cosine_distance(PG_FUNCTION_ARGS)
 		normb += bx[i] * bx[i];
 	}
 
-	PG_RETURN_FLOAT8(1 - (distance / (sqrt(norma) * sqrt(normb))));
+	/* Use sqrt(a * b) over sqrt(a) * sqrt(b) */
+	similarity = (double) distance / sqrt((double) norma * (double) 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 = 1.0;
+	else if (similarity < -1)
+		similarity = -1.0;
+
+	PG_RETURN_FLOAT8(1.0 - similarity);
 }
 
 /*
@@ -582,12 +701,18 @@ vector_spherical_distance(PG_FUNCTION_ARGS)
 {
 	Vector	   *a = PG_GETARG_VECTOR_P(0);
 	Vector	   *b = PG_GETARG_VECTOR_P(1);
-	double		distance = 0.0;
+	float	   *ax = a->x;
+	float	   *bx = b->x;
+	float		dp = 0.0;
+	double		distance;
 
 	CheckDims(a, b);
 
+	/* Auto-vectorized */
 	for (int i = 0; i < a->dim; i++)
-		distance += a->x[i] * b->x[i];
+		dp += ax[i] * bx[i];
+
+	distance = (double) dp;
 
 	/* Prevent NaN with acos with loss of precision */
 	if (distance > 1)
@@ -598,6 +723,28 @@ vector_spherical_distance(PG_FUNCTION_ARGS)
 	PG_RETURN_FLOAT8(acos(distance) / M_PI);
 }
 
+/*
+ * Get the L1 distance between vectors
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(l1_distance);
+Datum
+l1_distance(PG_FUNCTION_ARGS)
+{
+	Vector	   *a = PG_GETARG_VECTOR_P(0);
+	Vector	   *b = PG_GETARG_VECTOR_P(1);
+	float	   *ax = a->x;
+	float	   *bx = b->x;
+	float		distance = 0.0;
+
+	CheckDims(a, b);
+
+	/* Auto-vectorized */
+	for (int i = 0; i < a->dim; i++)
+		distance += fabsf(ax[i] - bx[i]);
+
+	PG_RETURN_FLOAT8((double) distance);
+}
+
 /*
  * Get the dimensions of a vector
  */
@@ -621,8 +768,9 @@ vector_norm(PG_FUNCTION_ARGS)
 	float	   *ax = a->x;
 	double		norm = 0.0;
 
+	/* Auto-vectorized */
 	for (int i = 0; i < a->dim; i++)
-		norm += ax[i] * ax[i];
+		norm += (double) ax[i] * (double) ax[i];
 
 	PG_RETURN_FLOAT8(sqrt(norm));
 }
@@ -645,9 +793,18 @@ vector_add(PG_FUNCTION_ARGS)
 
 	result = InitVector(a->dim);
 	rx = result->x;
+
+	/* Auto-vectorized */
 	for (int i = 0, imax = a->dim; i < imax; i++)
 		rx[i] = ax[i] + bx[i];
 
+	/* Check for overflow */
+	for (int i = 0, imax = a->dim; i < imax; i++)
+	{
+		if (isinf(rx[i]))
+			float_overflow_error();
+	}
+
 	PG_RETURN_POINTER(result);
 }
 
@@ -669,9 +826,54 @@ vector_sub(PG_FUNCTION_ARGS)
 
 	result = InitVector(a->dim);
 	rx = result->x;
+
+	/* Auto-vectorized */
 	for (int i = 0, imax = a->dim; i < imax; i++)
 		rx[i] = ax[i] - bx[i];
 
+	/* Check for overflow */
+	for (int i = 0, imax = a->dim; i < imax; i++)
+	{
+		if (isinf(rx[i]))
+			float_overflow_error();
+	}
+
+	PG_RETURN_POINTER(result);
+}
+
+/*
+ * Multiply vectors
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(vector_mul);
+Datum
+vector_mul(PG_FUNCTION_ARGS)
+{
+	Vector	   *a = PG_GETARG_VECTOR_P(0);
+	Vector	   *b = PG_GETARG_VECTOR_P(1);
+	float	   *ax = a->x;
+	float	   *bx = b->x;
+	Vector	   *result;
+	float	   *rx;
+
+	CheckDims(a, b);
+
+	result = InitVector(a->dim);
+	rx = result->x;
+
+	/* Auto-vectorized */
+	for (int i = 0, imax = a->dim; i < imax; i++)
+		rx[i] = ax[i] * bx[i];
+
+	/* Check for overflow and underflow */
+	for (int i = 0, imax = a->dim; i < imax; i++)
+	{
+		if (isinf(rx[i]))
+			float_overflow_error();
+
+		if (rx[i] == 0 && !(ax[i] == 0 || bx[i] == 0))
+			float_underflow_error();
+	}
+
 	PG_RETURN_POINTER(result);
 }
 
@@ -681,11 +883,9 @@ vector_sub(PG_FUNCTION_ARGS)
 int
 vector_cmp_internal(Vector * a, Vector * b)
 {
-	int			i;
-
 	CheckDims(a, b);
 
-	for (i = 0; i < a->dim; i++)
+	for (int i = 0; i < a->dim; i++)
 	{
 		if (a->x[i] < b->x[i])
 			return -1;
@@ -817,12 +1017,12 @@ vector_accum(PG_FUNCTION_ARGS)
 	n = statevalues[0] + 1.0;
 
 	statedatums = CreateStateDatums(dim);
-	statedatums[0] = Float8GetDatumFast(n);
+	statedatums[0] = Float8GetDatum(n);
 
 	if (newarr)
 	{
 		for (int i = 0; i < dim; i++)
-			statedatums[i + 1] = Float8GetDatumFast(x[i]);
+			statedatums[i + 1] = Float8GetDatum((double) x[i]);
 	}
 	else
 	{
@@ -830,10 +1030,11 @@ vector_accum(PG_FUNCTION_ARGS)
 		{
 			double		v = statevalues[i + 1] + x[i];
 
+			/* Check for overflow */
 			if (isinf(v))
 				float_overflow_error();
 
-			statedatums[i + 1] = Float8GetDatumFast(v);
+			statedatums[i + 1] = Float8GetDatum(v);
 		}
 	}
 
@@ -878,7 +1079,7 @@ vector_combine(PG_FUNCTION_ARGS)
 		dim = STATE_DIMS(statearray2);
 		statedatums = CreateStateDatums(dim);
 		for (int i = 1; i <= dim; i++)
-			statedatums[i] = Float8GetDatumFast(statevalues2[i]);
+			statedatums[i] = Float8GetDatum(statevalues2[i]);
 	}
 	else if (n2 == 0.0)
 	{
@@ -886,7 +1087,7 @@ vector_combine(PG_FUNCTION_ARGS)
 		dim = STATE_DIMS(statearray1);
 		statedatums = CreateStateDatums(dim);
 		for (int i = 1; i <= dim; i++)
-			statedatums[i] = Float8GetDatumFast(statevalues1[i]);
+			statedatums[i] = Float8GetDatum(statevalues1[i]);
 	}
 	else
 	{
@@ -898,14 +1099,15 @@ vector_combine(PG_FUNCTION_ARGS)
 		{
 			double		v = statevalues1[i] + statevalues2[i];
 
+			/* Check for overflow */
 			if (isinf(v))
 				float_overflow_error();
 
-			statedatums[i] = Float8GetDatumFast(v);
+			statedatums[i] = Float8GetDatum(v);
 		}
 	}
 
-	statedatums[0] = Float8GetDatumFast(n);
+	statedatums[0] = Float8GetDatum(n);
 
 	result = construct_array(statedatums, dim + 1,
 							 FLOAT8OID,
@@ -928,7 +1130,6 @@ vector_avg(PG_FUNCTION_ARGS)
 	float8		n;
 	uint16		dim;
 	Vector	   *result;
-	float		v;
 
 	/* Check array before using */
 	statevalues = CheckStateArray(statearray, "vector_avg");
@@ -940,12 +1141,12 @@ vector_avg(PG_FUNCTION_ARGS)
 
 	/* Create vector */
 	dim = STATE_DIMS(statearray);
+	CheckDim(dim);
 	result = InitVector(dim);
 	for (int i = 0; i < dim; i++)
 	{
-		v = statevalues[i + 1] / n;
-		CheckElement(v);
-		result->x[i] = v;
+		result->x[i] = statevalues[i + 1] / n;
+		CheckElement(result->x[i]);
 	}
 
 	PG_RETURN_POINTER(result);

src/vector.h

@@ -1,8 +1,6 @@
 #ifndef VECTOR_H
 #define VECTOR_H
 
-#include "postgres.h"
-
 #define VECTOR_MAX_DIM 16000
 
 #define VECTOR_SIZE(_dim)		(offsetof(Vector, x) + sizeof(float)*(_dim))
@@ -18,24 +16,8 @@ typedef struct Vector
 	float		x[FLEXIBLE_ARRAY_MEMBER];
 }			Vector;
 
+Vector	   *InitVector(int dim);
 void		PrintVector(char *msg, Vector * vector);
 int			vector_cmp_internal(Vector * a, Vector * b);
 
-/*
- * Allocate and initialize a new vector
- */
-static inline Vector *
-InitVector(int dim)
-{
-	Vector	   *result;
-	int			size;
-
-	size = VECTOR_SIZE(dim);
-	result = (Vector *) palloc0(size);
-	SET_VARSIZE(result, size);
-	result->dim = dim;
-
-	return result;
-}
-
 #endif

test/expected/cast.out

@@ -22,8 +22,14 @@ SELECT ARRAY[1,2,3]::float8[]::vector;
  [1,2,3]
 (1 row)
 
+SELECT ARRAY[1,2,3]::numeric[]::vector;
+  array  
+---------
+ [1,2,3]
+(1 row)
+
 SELECT '{NULL}'::real[]::vector;
-ERROR:  array must not containing NULLs
+ERROR:  array must not contain nulls
 SELECT '{NaN}'::real[]::vector;
 ERROR:  NaN not allowed in vector
 SELECT '{Infinity}'::real[]::vector;
@@ -32,6 +38,8 @@ SELECT '{-Infinity}'::real[]::vector;
 ERROR:  infinite value not allowed in vector
 SELECT '{}'::real[]::vector;
 ERROR:  vector must have at least 1 dimension
+SELECT '{{1}}'::real[]::vector;
+ERROR:  array must be 1-D
 SELECT '[1,2,3]'::vector::real[];
  float4  
 ---------
@@ -40,6 +48,8 @@ SELECT '[1,2,3]'::vector::real[];
 
 SELECT array_agg(n)::vector FROM generate_series(1, 16001) n;
 ERROR:  vector cannot have more than 16000 dimensions
+SELECT array_to_vector(array_agg(n), 16001, false) FROM generate_series(1, 16001) n;
+ERROR:  vector cannot have more than 16000 dimensions
 -- ensure no error
 SELECT ARRAY[1,2,3] = ARRAY[1,2,3];
  ?column? 

test/expected/functions.out

@@ -4,12 +4,26 @@ SELECT '[1,2,3]'::vector + '[4,5,6]';
  [5,7,9]
 (1 row)
 
+SELECT '[3e38]'::vector + '[3e38]';
+ERROR:  value out of range: overflow
 SELECT '[1,2,3]'::vector - '[4,5,6]';
   ?column?  
 ------------
  [-3,-3,-3]
 (1 row)
 
+SELECT '[-3e38]'::vector - '[3e38]';
+ERROR:  value out of range: overflow
+SELECT '[1,2,3]'::vector * '[4,5,6]';
+ ?column?  
+-----------
+ [4,10,18]
+(1 row)
+
+SELECT '[1e37]'::vector * '[1e37]';
+ERROR:  value out of range: overflow
+SELECT '[1e-37]'::vector * '[1e-37]';
+ERROR:  value out of range: underflow
 SELECT vector_dims('[1,2,3]');
  vector_dims 
 -------------
@@ -22,14 +36,44 @@ SELECT round(vector_norm('[1,1]')::numeric, 5);
  1.41421
 (1 row)
 
-SELECT round(l2_distance('[1,2]', '[0,0]')::numeric, 5);
-  round  
----------
- 2.23607
+SELECT vector_norm('[3,4]');
+ vector_norm 
+-------------
+           5
+(1 row)
+
+SELECT vector_norm('[0,1]');
+ vector_norm 
+-------------
+           1
+(1 row)
+
+SELECT vector_norm('[3e37,4e37]')::real;
+ vector_norm 
+-------------
+       5e+37
+(1 row)
+
+SELECT l2_distance('[0,0]', '[3,4]');
+ l2_distance 
+-------------
+           5
+(1 row)
+
+SELECT l2_distance('[0,0]', '[0,1]');
+ l2_distance 
+-------------
+           1
 (1 row)
 
 SELECT l2_distance('[1,2]', '[3]');
 ERROR:  different vector dimensions 2 and 1
+SELECT l2_distance('[3e38]', '[-3e38]');
+ l2_distance 
+-------------
+    Infinity
+(1 row)
+
 SELECT inner_product('[1,2]', '[3,4]');
  inner_product 
 ---------------
@@ -38,10 +82,16 @@ SELECT inner_product('[1,2]', '[3,4]');
 
 SELECT inner_product('[1,2]', '[3]');
 ERROR:  different vector dimensions 2 and 1
-SELECT round(cosine_distance('[1,2]', '[2,4]')::numeric, 5);
-  round  
----------
- 0.00000
+SELECT inner_product('[3e38]', '[3e38]');
+ inner_product 
+---------------
+      Infinity
+(1 row)
+
+SELECT cosine_distance('[1,2]', '[2,4]');
+ cosine_distance 
+-----------------
+               0
 (1 row)
 
 SELECT cosine_distance('[1,2]', '[0,0]');
@@ -50,8 +100,64 @@ SELECT cosine_distance('[1,2]', '[0,0]');
              NaN
 (1 row)
 
+SELECT cosine_distance('[1,1]', '[1,1]');
+ cosine_distance 
+-----------------
+               0
+(1 row)
+
+SELECT cosine_distance('[1,0]', '[0,2]');
+ cosine_distance 
+-----------------
+               1
+(1 row)
+
+SELECT cosine_distance('[1,1]', '[-1,-1]');
+ cosine_distance 
+-----------------
+               2
+(1 row)
+
 SELECT cosine_distance('[1,2]', '[3]');
 ERROR:  different vector dimensions 2 and 1
+SELECT cosine_distance('[1,1]', '[1.1,1.1]');
+ cosine_distance 
+-----------------
+               0
+(1 row)
+
+SELECT cosine_distance('[1,1]', '[-1.1,-1.1]');
+ cosine_distance 
+-----------------
+               2
+(1 row)
+
+SELECT cosine_distance('[3e38]', '[3e38]');
+ cosine_distance 
+-----------------
+             NaN
+(1 row)
+
+SELECT l1_distance('[0,0]', '[3,4]');
+ l1_distance 
+-------------
+           7
+(1 row)
+
+SELECT l1_distance('[0,0]', '[0,1]');
+ l1_distance 
+-------------
+           1
+(1 row)
+
+SELECT l1_distance('[1,2]', '[3]');
+ERROR:  different vector dimensions 2 and 1
+SELECT l1_distance('[3e38]', '[-3e38]');
+ l1_distance 
+-------------
+    Infinity
+(1 row)
+
 SELECT avg(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]']) v;
     avg    
 -----------
@@ -72,3 +178,33 @@ SELECT avg(v) FROM unnest(ARRAY[]::vector[]) v;
 
 SELECT avg(v) FROM unnest(ARRAY['[1,2]'::vector, '[3]']) v;
 ERROR:  expected 2 dimensions, not 1
+SELECT avg(v) FROM unnest(ARRAY['[3e38]'::vector, '[3e38]']) v;
+   avg   
+---------
+ [3e+38]
+(1 row)
+
+SELECT vector_avg(array_agg(n)) FROM generate_series(1, 16002) n;
+ERROR:  vector cannot have more than 16000 dimensions
+SELECT sum(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]']) v;
+   sum    
+----------
+ [4,7,10]
+(1 row)
+
+SELECT sum(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]', NULL]) v;
+   sum    
+----------
+ [4,7,10]
+(1 row)
+
+SELECT sum(v) FROM unnest(ARRAY[]::vector[]) v;
+ sum 
+-----
+ 
+(1 row)
+
+SELECT sum(v) FROM unnest(ARRAY['[1,2]'::vector, '[3]']) v;
+ERROR:  different vector dimensions 2 and 1
+SELECT sum(v) FROM unnest(ARRAY['[3e38]'::vector, '[3e38]']) v;
+ERROR:  value out of range: overflow

test/expected/hnsw_cosine.out

@@ -0,0 +1,26 @@
+SET enable_seqscan = off;
+CREATE TABLE t (val vector(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING hnsw (val vector_cosine_ops);
+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::vector)) t2;
+ count 
+-------
+     3
+(1 row)
+
+DROP TABLE t;

test/expected/hnsw_ip.out

@@ -0,0 +1,21 @@
+SET enable_seqscan = off;
+CREATE TABLE t (val vector(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING hnsw (val vector_ip_ops);
+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::vector)) t2;
+ count 
+-------
+     4
+(1 row)
+
+DROP TABLE t;

test/expected/hnsw_l2.out

@@ -0,0 +1,36 @@
+SET enable_seqscan = off;
+CREATE TABLE t (val vector(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING hnsw (val vector_l2_ops);
+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 * FROM t ORDER BY val <-> (SELECT NULL::vector);
+   val   
+---------
+ [0,0,0]
+ [1,1,1]
+ [1,2,3]
+ [1,2,4]
+(4 rows)
+
+SELECT COUNT(*) FROM t;
+ count 
+-------
+     5
+(1 row)
+
+TRUNCATE t;
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+ val 
+-----
+(0 rows)
+
+DROP TABLE t;

test/expected/hnsw_options.out

@@ -0,0 +1,26 @@
+CREATE TABLE t (val vector(3));
+CREATE INDEX ON t USING hnsw (val vector_l2_ops) WITH (m = 1);
+ERROR:  value 1 out of bounds for option "m"
+DETAIL:  Valid values are between "2" and "100".
+CREATE INDEX ON t USING hnsw (val vector_l2_ops) WITH (m = 101);
+ERROR:  value 101 out of bounds for option "m"
+DETAIL:  Valid values are between "2" and "100".
+CREATE INDEX ON t USING hnsw (val vector_l2_ops) WITH (ef_construction = 3);
+ERROR:  value 3 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 (ef_construction = 1001);
+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
+SHOW hnsw.ef_search;
+ hnsw.ef_search 
+----------------
+ 40
+(1 row)
+
+SET hnsw.ef_search = 0;
+ERROR:  0 is outside the valid range for parameter "hnsw.ef_search" (1 .. 1000)
+SET hnsw.ef_search = 1001;
+ERROR:  1001 is outside the valid range for parameter "hnsw.ef_search" (1 .. 1000)
+DROP TABLE t;

test/expected/hnsw_unlogged.out

@@ -0,0 +1,13 @@
+SET enable_seqscan = off;
+CREATE UNLOGGED TABLE t (val vector(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING hnsw (val vector_l2_ops);
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+   val   
+---------
+ [1,2,3]
+ [1,1,1]
+ [0,0,0]
+(3 rows)
+
+DROP TABLE t;

test/expected/input.out

@@ -4,10 +4,22 @@ SELECT '[1,2,3]'::vector;
  [1,2,3]
 (1 row)
 
-SELECT '[-1,2,3]'::vector;
-  vector  
-----------
- [-1,2,3]
+SELECT '[-1,-2,-3]'::vector;
+   vector   
+------------
+ [-1,-2,-3]
+(1 row)
+
+SELECT '[1.,2.,3.]'::vector;
+ vector  
+---------
+ [1,2,3]
+(1 row)
+
+SELECT ' [ 1,  2 ,    3  ] '::vector;
+ vector  
+---------
+ [1,2,3]
 (1 row)
 
 SELECT '[1.23456]'::vector;
@@ -17,7 +29,7 @@ SELECT '[1.23456]'::vector;
 (1 row)
 
 SELECT '[hello,1]'::vector;
-ERROR:  invalid input syntax for type vector: "hello"
+ERROR:  invalid input syntax for type vector: "[hello,1]"
 LINE 1: SELECT '[hello,1]'::vector;
                ^
 SELECT '[NaN,1]'::vector;
@@ -32,13 +44,35 @@ SELECT '[-Infinity,1]'::vector;
 ERROR:  infinite value not allowed in vector
 LINE 1: SELECT '[-Infinity,1]'::vector;
                ^
+SELECT '[1.5e38,-1.5e38]'::vector;
+       vector       
+--------------------
+ [1.5e+38,-1.5e+38]
+(1 row)
+
+SELECT '[1.5e+38,-1.5e+38]'::vector;
+       vector       
+--------------------
+ [1.5e+38,-1.5e+38]
+(1 row)
+
+SELECT '[1.5e-38,-1.5e-38]'::vector;
+       vector       
+--------------------
+ [1.5e-38,-1.5e-38]
+(1 row)
+
+SELECT '[4e38,1]'::vector;
+ERROR:  infinite value not allowed in vector
+LINE 1: SELECT '[4e38,1]'::vector;
+               ^
 SELECT '[1,2,3'::vector;
-ERROR:  malformed vector literal
+ERROR:  malformed vector literal: "[1,2,3"
 LINE 1: SELECT '[1,2,3'::vector;
                ^
 DETAIL:  Unexpected end of input.
 SELECT '[1,2,3]9'::vector;
-ERROR:  malformed vector literal
+ERROR:  malformed vector literal: "[1,2,3]9"
 LINE 1: SELECT '[1,2,3]9'::vector;
                ^
 DETAIL:  Junk after closing right brace.
@@ -47,14 +81,41 @@ ERROR:  malformed vector literal: "1,2,3"
 LINE 1: SELECT '1,2,3'::vector;
                ^
 DETAIL:  Vector contents must start with "[".
+SELECT ''::vector;
+ERROR:  malformed vector literal: ""
+LINE 1: SELECT ''::vector;
+               ^
+DETAIL:  Vector contents must start with "[".
+SELECT '['::vector;
+ERROR:  malformed vector literal: "["
+LINE 1: SELECT '['::vector;
+               ^
+DETAIL:  Unexpected end of input.
+SELECT '[,'::vector;
+ERROR:  malformed vector literal: "[,"
+LINE 1: SELECT '[,'::vector;
+               ^
+DETAIL:  Unexpected end of input.
 SELECT '[]'::vector;
 ERROR:  vector must have at least 1 dimension
 LINE 1: SELECT '[]'::vector;
                ^
 SELECT '[1,]'::vector;
-ERROR:  invalid input syntax for type vector: "]"
+ERROR:  invalid input syntax for type vector: "[1,]"
 LINE 1: SELECT '[1,]'::vector;
                ^
+SELECT '[1a]'::vector;
+ERROR:  invalid input syntax for type vector: "[1a]"
+LINE 1: SELECT '[1a]'::vector;
+               ^
+SELECT '[1,,3]'::vector;
+ERROR:  malformed vector literal: "[1,,3]"
+LINE 1: SELECT '[1,,3]'::vector;
+               ^
+SELECT '[1, ,3]'::vector;
+ERROR:  invalid input syntax for type vector: "[1, ,3]"
+LINE 1: SELECT '[1, ,3]'::vector;
+               ^
 SELECT '[1,2,3]'::vector(2);
 ERROR:  expected 2 dimensions, not 3
 SELECT unnest('{"[1,2,3]", "[4,5,6]"}'::vector[]);

test/expected/ivfflat_cosine.out

@@ -11,9 +11,16 @@ SELECT * FROM t ORDER BY val <=> '[3,3,3]';
  [1,2,4]
 (3 rows)
 
-SELECT * FROM t ORDER BY val <=> (SELECT NULL::vector);
- val 
------
-(0 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::vector)) t2;
+ count 
+-------
+     3
+(1 row)
 
 DROP TABLE t;

test/expected/ivfflat_ip.out

@@ -12,9 +12,10 @@ SELECT * FROM t ORDER BY val <#> '[3,3,3]';
  [0,0,0]
 (4 rows)
 
-SELECT * FROM t ORDER BY val <#> (SELECT NULL::vector);
- val 
------
-(0 rows)
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <#> (SELECT NULL::vector)) t2;
+ count 
+-------
+     4
+(1 row)
 
 DROP TABLE t;

test/expected/ivfflat_l2.out

@@ -1,7 +1,7 @@
 SET enable_seqscan = off;
 CREATE TABLE t (val vector(3));
 INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
-CREATE INDEX ON t USING ivfflat (val) WITH (lists = 1);
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops) WITH (lists = 1);
 INSERT INTO t (val) VALUES ('[1,2,4]');
 SELECT * FROM t ORDER BY val <-> '[3,3,3]';
    val   
@@ -13,9 +13,13 @@ SELECT * FROM t ORDER BY val <-> '[3,3,3]';
 (4 rows)
 
 SELECT * FROM t ORDER BY val <-> (SELECT NULL::vector);
- val 
------
-(0 rows)
+   val   
+---------
+ [0,0,0]
+ [1,1,1]
+ [1,2,3]
+ [1,2,4]
+(4 rows)
 
 SELECT COUNT(*) FROM t;
  count 
@@ -23,4 +27,13 @@ SELECT COUNT(*) FROM t;
      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;

test/expected/ivfflat_options.out

@@ -1,9 +1,8 @@
-SET enable_seqscan = off;
 CREATE TABLE t (val vector(3));
-CREATE INDEX ON t USING ivfflat (val) WITH (lists = 0);
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops) WITH (lists = 0);
 ERROR:  value 0 out of bounds for option "lists"
 DETAIL:  Valid values are between "1" and "32768".
-CREATE INDEX ON t USING ivfflat (val) WITH (lists = 32769);
+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".
 SHOW ivfflat.probes;

test/expected/ivfflat_unlogged.out

@@ -1,7 +1,7 @@
 SET enable_seqscan = off;
 CREATE UNLOGGED TABLE t (val vector(3));
 INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
-CREATE INDEX ON t USING ivfflat (val) WITH (lists = 1);
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops) WITH (lists = 1);
 SELECT * FROM t ORDER BY val <-> '[3,3,3]';
    val   
 ---------

test/sql/cast.sql

@@ -2,13 +2,16 @@ SELECT ARRAY[1,2,3]::vector;
 SELECT ARRAY[1.0,2.0,3.0]::vector;
 SELECT ARRAY[1,2,3]::float4[]::vector;
 SELECT ARRAY[1,2,3]::float8[]::vector;
+SELECT ARRAY[1,2,3]::numeric[]::vector;
 SELECT '{NULL}'::real[]::vector;
 SELECT '{NaN}'::real[]::vector;
 SELECT '{Infinity}'::real[]::vector;
 SELECT '{-Infinity}'::real[]::vector;
 SELECT '{}'::real[]::vector;
+SELECT '{{1}}'::real[]::vector;
 SELECT '[1,2,3]'::vector::real[];
 SELECT array_agg(n)::vector FROM generate_series(1, 16001) n;
+SELECT array_to_vector(array_agg(n), 16001, false) FROM generate_series(1, 16001) n;
 
 -- ensure no error
 SELECT ARRAY[1,2,3] = ARRAY[1,2,3];

test/sql/functions.sql

@@ -1,20 +1,51 @@
 SELECT '[1,2,3]'::vector + '[4,5,6]';
+SELECT '[3e38]'::vector + '[3e38]';
 SELECT '[1,2,3]'::vector - '[4,5,6]';
+SELECT '[-3e38]'::vector - '[3e38]';
+SELECT '[1,2,3]'::vector * '[4,5,6]';
+SELECT '[1e37]'::vector * '[1e37]';
+SELECT '[1e-37]'::vector * '[1e-37]';
 
 SELECT vector_dims('[1,2,3]');
-SELECT round(vector_norm('[1,1]')::numeric, 5);
 
-SELECT round(l2_distance('[1,2]', '[0,0]')::numeric, 5);
+SELECT round(vector_norm('[1,1]')::numeric, 5);
+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 inner_product('[1,2]', '[3,4]');
 SELECT inner_product('[1,2]', '[3]');
+SELECT inner_product('[3e38]', '[3e38]');
 
-SELECT round(cosine_distance('[1,2]', '[2,4]')::numeric, 5);
+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 l1_distance('[0,0]', '[3,4]');
+SELECT l1_distance('[0,0]', '[0,1]');
+SELECT l1_distance('[1,2]', '[3]');
+SELECT l1_distance('[3e38]', '[-3e38]');
 
 SELECT avg(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]']) v;
 SELECT avg(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]', NULL]) v;
 SELECT avg(v) FROM unnest(ARRAY[]::vector[]) v;
 SELECT avg(v) FROM unnest(ARRAY['[1,2]'::vector, '[3]']) v;
+SELECT avg(v) FROM unnest(ARRAY['[3e38]'::vector, '[3e38]']) v;
+SELECT vector_avg(array_agg(n)) FROM generate_series(1, 16002) n;
+
+SELECT sum(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]']) v;
+SELECT sum(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]', NULL]) v;
+SELECT sum(v) FROM unnest(ARRAY[]::vector[]) v;
+SELECT sum(v) FROM unnest(ARRAY['[1,2]'::vector, '[3]']) v;
+SELECT sum(v) FROM unnest(ARRAY['[3e38]'::vector, '[3e38]']) v;

test/sql/hnsw_cosine.sql

@@ -0,0 +1,13 @@
+SET enable_seqscan = off;
+
+CREATE TABLE t (val vector(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING hnsw (val vector_cosine_ops);
+
+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::vector)) t2;
+
+DROP TABLE t;

test/sql/hnsw_ip.sql

@@ -0,0 +1,12 @@
+SET enable_seqscan = off;
+
+CREATE TABLE t (val vector(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING hnsw (val vector_ip_ops);
+
+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::vector)) t2;
+
+DROP TABLE t;

test/sql/hnsw_l2.sql

@@ -0,0 +1,16 @@
+SET enable_seqscan = off;
+
+CREATE TABLE t (val vector(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING hnsw (val vector_l2_ops);
+
+INSERT INTO t (val) VALUES ('[1,2,4]');
+
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+SELECT * FROM t ORDER BY val <-> (SELECT NULL::vector);
+SELECT COUNT(*) FROM t;
+
+TRUNCATE t;
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+
+DROP TABLE t;

test/sql/hnsw_options.sql

@@ -0,0 +1,13 @@
+CREATE TABLE t (val vector(3));
+CREATE INDEX ON t USING hnsw (val vector_l2_ops) WITH (m = 1);
+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);
+
+SHOW hnsw.ef_search;
+
+SET hnsw.ef_search = 0;
+SET hnsw.ef_search = 1001;
+
+DROP TABLE t;

test/sql/hnsw_unlogged.sql

@@ -0,0 +1,9 @@
+SET enable_seqscan = off;
+
+CREATE UNLOGGED TABLE t (val vector(3));
+INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
+CREATE INDEX ON t USING hnsw (val vector_l2_ops);
+
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+
+DROP TABLE t;

test/sql/input.sql

@@ -1,15 +1,27 @@
 SELECT '[1,2,3]'::vector;
-SELECT '[-1,2,3]'::vector;
+SELECT '[-1,-2,-3]'::vector;
+SELECT '[1.,2.,3.]'::vector;
+SELECT ' [ 1,  2 ,    3  ] '::vector;
 SELECT '[1.23456]'::vector;
 SELECT '[hello,1]'::vector;
 SELECT '[NaN,1]'::vector;
 SELECT '[Infinity,1]'::vector;
 SELECT '[-Infinity,1]'::vector;
+SELECT '[1.5e38,-1.5e38]'::vector;
+SELECT '[1.5e+38,-1.5e+38]'::vector;
+SELECT '[1.5e-38,-1.5e-38]'::vector;
+SELECT '[4e38,1]'::vector;
 SELECT '[1,2,3'::vector;
 SELECT '[1,2,3]9'::vector;
 SELECT '1,2,3'::vector;
+SELECT ''::vector;
+SELECT '['::vector;
+SELECT '[,'::vector;
 SELECT '[]'::vector;
 SELECT '[1,]'::vector;
+SELECT '[1a]'::vector;
+SELECT '[1,,3]'::vector;
+SELECT '[1, ,3]'::vector;
 SELECT '[1,2,3]'::vector(2);
 
 SELECT unnest('{"[1,2,3]", "[4,5,6]"}'::vector[]);

test/sql/ivfflat_cosine.sql

@@ -7,6 +7,7 @@ CREATE INDEX ON t USING ivfflat (val vector_cosine_ops) WITH (lists = 1);
 INSERT INTO t (val) VALUES ('[1,2,4]');
 
 SELECT * FROM t ORDER BY val <=> '[3,3,3]';
-SELECT * FROM t ORDER BY val <=> (SELECT NULL::vector);
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> '[0,0,0]') t2;
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> (SELECT NULL::vector)) t2;
 
 DROP TABLE t;

test/sql/ivfflat_ip.sql

@@ -7,6 +7,6 @@ CREATE INDEX ON t USING ivfflat (val vector_ip_ops) WITH (lists = 1);
 INSERT INTO t (val) VALUES ('[1,2,4]');
 
 SELECT * FROM t ORDER BY val <#> '[3,3,3]';
-SELECT * FROM t ORDER BY val <#> (SELECT NULL::vector);
+SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <#> (SELECT NULL::vector)) t2;
 
 DROP TABLE t;

test/sql/ivfflat_l2.sql

@@ -2,7 +2,7 @@ SET enable_seqscan = off;
 
 CREATE TABLE t (val vector(3));
 INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
-CREATE INDEX ON t USING ivfflat (val) WITH (lists = 1);
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops) WITH (lists = 1);
 
 INSERT INTO t (val) VALUES ('[1,2,4]');
 
@@ -10,4 +10,7 @@ SELECT * FROM t ORDER BY val <-> '[3,3,3]';
 SELECT * FROM t ORDER BY val <-> (SELECT NULL::vector);
 SELECT COUNT(*) FROM t;
 
+TRUNCATE t;
+SELECT * FROM t ORDER BY val <-> '[3,3,3]';
+
 DROP TABLE t;

test/sql/ivfflat_options.sql

@@ -1,8 +1,6 @@
-SET enable_seqscan = off;
-
 CREATE TABLE t (val vector(3));
-CREATE INDEX ON t USING ivfflat (val) WITH (lists = 0);
-CREATE INDEX ON t USING ivfflat (val) WITH (lists = 32769);
+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);
 
 SHOW ivfflat.probes;
 

test/sql/ivfflat_unlogged.sql

@@ -2,7 +2,7 @@ SET enable_seqscan = off;
 
 CREATE UNLOGGED TABLE t (val vector(3));
 INSERT INTO t (val) VALUES ('[0,0,0]'), ('[1,2,3]'), ('[1,1,1]'), (NULL);
-CREATE INDEX ON t USING ivfflat (val) WITH (lists = 1);
+CREATE INDEX ON t USING ivfflat (val vector_l2_ops) WITH (lists = 1);
 
 SELECT * FROM t ORDER BY val <-> '[3,3,3]';
 

test/t/001_ivfflat_wal.pl

@@ -5,7 +5,7 @@ use strict;
 use warnings;
 use PostgresNode;
 use TestLib;
-use Test::More tests => 31;
+use Test::More;
 
 my $dim = 32;
 
@@ -19,14 +19,13 @@ sub test_index_replay
 
 	# Wait for replica to catch up
 	my $applname = $node_replica->name;
-
-	my $server_version_num = $node_primary->safe_psql("postgres", "SHOW server_version_num");
 	my $caughtup_query = "SELECT pg_current_wal_lsn() <= replay_lsn FROM pg_stat_replication WHERE application_name = '$applname';";
 	$node_primary->poll_query_until('postgres', $caughtup_query)
 	  or die "Timed out while waiting for replica 1 to catch up";
 
 	my @r = ();
-	for (1 .. $dim) {
+	for (1 .. $dim)
+	{
 		push(@r, rand());
 	}
 	my $sql = join(",", @r);
@@ -52,11 +51,13 @@ my $array_sql = join(",", ('random()') x $dim);
 # Initialize primary node
 $node_primary = get_new_node('primary');
 $node_primary->init(allows_streaming => 1);
-if ($dim > 32) {
+if ($dim > 32)
+{
 	# TODO use wal_keep_segments for Postgres < 13
 	$node_primary->append_conf('postgresql.conf', qq(wal_keep_size = 1GB));
 }
-if ($dim > 1500) {
+if ($dim > 1500)
+{
 	$node_primary->append_conf('postgresql.conf', qq(maintenance_work_mem = 128MB));
 }
 $node_primary->start;
@@ -67,8 +68,7 @@ $node_primary->backup($backup_name);
 
 # Create streaming replica linking to primary
 $node_replica = get_new_node('replica');
-$node_replica->init_from_backup($node_primary, $backup_name,
-	has_streaming => 1);
+$node_replica->init_from_backup($node_primary, $backup_name, has_streaming => 1);
 $node_replica->start;
 
 # Create ivfflat index on primary
@@ -77,7 +77,7 @@ $node_primary->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector($dim));
 $node_primary->safe_psql("postgres",
 	"INSERT INTO tst SELECT i % 10, ARRAY[$array_sql] FROM generate_series(1, 100000) i;"
 );
-$node_primary->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v);");
+$node_primary->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops);");
 
 # Test that queries give same result
 test_index_replay('initial');
@@ -95,3 +95,5 @@ for my $i (1 .. 10)
 	);
 	test_index_replay("insert $i");
 }
+
+done_testing();

test/t/002_ivfflat_vacuum.pl

@@ -2,12 +2,13 @@ use strict;
 use warnings;
 use PostgresNode;
 use TestLib;
-use Test::More tests => 1;
+use Test::More;
 
 my $dim = 3;
 
 my @r = ();
-for (1 .. $dim) {
+for (1 .. $dim)
+{
 	my $v = int(rand(1000)) + 1;
 	push(@r, "i % $v");
 }
@@ -24,7 +25,7 @@ $node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector($dim));");
 $node->safe_psql("postgres",
 	"INSERT INTO tst SELECT i % 10, ARRAY[$array_sql] FROM generate_series(1, 100000) i;"
 );
-$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v);");
+$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops);");
 
 # Get size
 my $size = $node->safe_psql("postgres", "SELECT pg_total_relation_size('tst_v_idx');");
@@ -39,3 +40,5 @@ $node->safe_psql("postgres",
 # Check size
 my $new_size = $node->safe_psql("postgres", "SELECT pg_total_relation_size('tst_v_idx');");
 is($size, $new_size, "size does not change");
+
+done_testing();

test/t/003_ivfflat_build_recall.pl

@@ -0,0 +1,128 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+my $node;
+my @queries = ();
+my @expected;
+my $limit = 20;
+
+sub test_recall
+{
+	my ($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 %actual_set = map { $_ => 1 } @actual_ids;
+
+		my @expected_ids = split("\n", $expected[$i]);
+
+		foreach (@expected_ids)
+		{
+			if (exists($actual_set{$_}))
+			{
+				$correct++;
+			}
+			$total++;
+		}
+	}
+
+	cmp_ok($correct / $total, ">=", $min, $operator);
+}
+
+# Initialize node
+$node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector(3));");
+$node->safe_psql("postgres",
+	"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 100000) i;"
+);
+
+# Generate queries
+for (1 .. 20)
+{
+	my $r1 = rand();
+	my $r2 = rand();
+	my $r3 = rand();
+	push(@queries, "[$r1,$r2,$r3]");
+}
+
+# Check each index type
+my @operators = ("<->", "<#>", "<=>");
+my @opclasses = ("vector_l2_ops", "vector_ip_ops", "vector_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", "SELECT i FROM tst ORDER BY v $operator '$_' LIMIT $limit;");
+		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, 0.71, $operator);
+		test_recall(10, 0.95, $operator);
+	}
+	# Account for equal distances
+	test_recall(100, 0.9925, $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, 0.71, $operator);
+		test_recall(10, 0.95, $operator);
+	}
+	# Account for equal distances
+	test_recall(100, 0.9925, $operator);
+
+	$node->safe_psql("postgres", "DROP INDEX idx;");
+}
+
+done_testing();

test/t/003_recall.pl

@@ -1,88 +0,0 @@
-use strict;
-use warnings;
-use PostgresNode;
-use TestLib;
-use Test::More tests => 9;
-
-my $node;
-my @queries = ();
-my @expected;
-my $limit = 20;
-
-sub test_recall
-{
-	my ($probes, $min, $operator) = @_;
-	my $correct = 0;
-	my $total = 0;
-
-	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 %actual_set = map { $_ => 1 } @actual_ids;
-
-		my @expected_ids = split("\n", $expected[$i]);
-
-		foreach (@expected_ids) {
-			if (exists($actual_set{$_})) {
-				$correct++;
-			}
-			$total++;
-		}
-	}
-
-	cmp_ok($correct / $total, ">=", $min, $operator);
-}
-
-# Initialize node
-$node = get_new_node('node');
-$node->init;
-$node->start;
-
-# Create table
-$node->safe_psql("postgres", "CREATE EXTENSION vector;");
-$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector(3));");
-$node->safe_psql("postgres",
-	"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 100000) i;"
-);
-
-# Generate queries
-for (1..20) {
-	my $r1 = rand();
-	my $r2 = rand();
-	my $r3 = rand();
-	push(@queries, "[$r1,$r2,$r3]");
-}
-
-# Check each index type
-my @operators = ("<->", "<#>", "<=>");
-
-foreach (@operators) {
-	my $operator = $_;
-
-	# Get exact results
-	@expected = ();
-	foreach (@queries) {
-		my $res = $node->safe_psql("postgres", "SELECT i FROM tst ORDER BY v $operator '$_' LIMIT $limit;");
-		push(@expected, $res);
-	}
-
-	# Add index
-	my $opclass;
-	if ($operator == "<->") {
-		$opclass = "vector_l2_ops";
-	} elsif ($operator == "<#>") {
-		$opclass = "vector_ip_ops";
-	} else {
-		$opclass = "vector_cosine_ops";
-	}
-	$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v $opclass);");
-
-	# Test approximate results
-	test_recall(1, 0.75, $operator);
-	test_recall(10, 0.95, $operator);
-	test_recall(100, 1.0, $operator);
-}

test/t/004_ivfflat_centers.pl

@@ -2,7 +2,7 @@ use strict;
 use warnings;
 use PostgresNode;
 use TestLib;
-use Test::More tests => 3;
+use Test::More;
 
 # Initialize node
 my $node = get_new_node('node');
@@ -20,7 +20,7 @@ sub test_centers
 {
 	my ($lists, $min) = @_;
 
-	my ($ret, $stdout, $stderr) = $node->psql("postgres", "CREATE INDEX ON tst USING ivfflat (v) WITH (lists = $lists);");
+	my ($ret, $stdout, $stderr) = $node->psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops) WITH (lists = $lists);");
 	is($ret, 0, $stderr);
 }
 
@@ -34,3 +34,5 @@ $node->safe_psql("postgres",
 
 # Test no error for duplicate centers
 test_centers(10);
+
+done_testing();

test/t/005_ivfflat_query_recall.pl

@@ -2,7 +2,7 @@ use strict;
 use warnings;
 use PostgresNode;
 use TestLib;
-use Test::More tests => 60;
+use Test::More;
 
 # Initialize node
 my $node = get_new_node('node');
@@ -18,24 +18,21 @@ $node->safe_psql("postgres",
 
 # Check each index type
 my @operators = ("<->", "<#>", "<=>");
-foreach (@operators) {
-	my $operator = $_;
+my @opclasses = ("vector_l2_ops", "vector_ip_ops", "vector_cosine_ops");
+
+for my $i (0 .. $#operators)
+{
+	my $operator = $operators[$i];
+	my $opclass = $opclasses[$i];
 
 	# Add index
-	my $opclass;
-	if ($operator == "<->") {
-		$opclass = "vector_l2_ops";
-	} elsif ($operator == "<#>") {
-		$opclass = "vector_ip_ops";
-	} else {
-		$opclass = "vector_cosine_ops";
-	}
 	$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v $opclass);");
 
 	# Test 100% recall
-	for (1..20) {
-		my $i = int(rand() * 100000);
-		my $query = $node->safe_psql("postgres", "SELECT v FROM tst WHERE i = $i;");
+	for (1 .. 20)
+	{
+		my $id = int(rand() * 100000);
+		my $query = $node->safe_psql("postgres", "SELECT v FROM tst WHERE i = $id;");
 		my $res = $node->safe_psql("postgres", qq(
 			SET enable_seqscan = off;
 			SELECT v FROM tst ORDER BY v <-> '$query' LIMIT 1;
@@ -43,3 +40,5 @@ foreach (@operators) {
 		is($res, $query);
 	}
 }
+
+done_testing();

test/t/006_ivfflat_lists.pl

@@ -2,7 +2,7 @@ use strict;
 use warnings;
 use PostgresNode;
 use TestLib;
-use Test::More tests => 3;
+use Test::More;
 
 # Initialize node
 my $node = get_new_node('node');
@@ -16,8 +16,8 @@ $node->safe_psql("postgres",
 	"INSERT INTO tst SELECT ARRAY[random(), random(), random()] FROM generate_series(1, 100000) i;"
 );
 
-$node->safe_psql("postgres", "CREATE INDEX lists50 ON tst USING ivfflat (v) WITH (lists = 50);");
-$node->safe_psql("postgres", "CREATE INDEX lists100 ON tst USING ivfflat (v) WITH (lists = 100);");
+$node->safe_psql("postgres", "CREATE INDEX lists50 ON tst USING ivfflat (v vector_l2_ops) WITH (lists = 50);");
+$node->safe_psql("postgres", "CREATE INDEX lists100 ON tst USING ivfflat (v vector_l2_ops) WITH (lists = 100);");
 
 # Test prefers more lists
 my $res = $node->safe_psql("postgres", "EXPLAIN SELECT v FROM tst ORDER BY v <-> '[0.5,0.5,0.5]' LIMIT 10;");
@@ -26,6 +26,8 @@ unlike($res, qr/lists50/);
 
 # Test errors with too much memory
 my ($ret, $stdout, $stderr) = $node->psql("postgres",
-	"CREATE INDEX lists10000 ON tst USING ivfflat (v) WITH (lists = 10000);"
+	"CREATE INDEX lists10000 ON tst USING ivfflat (v vector_l2_ops) WITH (lists = 10000);"
 );
 like($stderr, qr/memory required is/);
+
+done_testing();

test/t/007_ivfflat_inserts.pl

@@ -2,7 +2,7 @@ use strict;
 use warnings;
 use PostgresNode;
 use TestLib;
-use Test::More tests => 5;
+use Test::More;
 
 my $dim = 768;
 
@@ -19,7 +19,7 @@ $node->safe_psql("postgres", "CREATE TABLE tst (v vector($dim));");
 $node->safe_psql("postgres",
 	"INSERT INTO tst SELECT ARRAY[$array_sql] FROM generate_series(1, 10000) i;"
 );
-$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v);");
+$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops);");
 
 $node->pgbench(
 	"--no-vacuum --client=5 --transactions=100",
@@ -28,14 +28,23 @@ $node->pgbench(
 	[qr{^$}],
 	"concurrent INSERTs",
 	{
-		"007_inserts" => "INSERT INTO tst SELECT ARRAY[$array_sql] FROM generate_series(1, 10) i;"
+		"007_ivfflat_inserts" => "INSERT INTO tst SELECT ARRAY[$array_sql] FROM generate_series(1, 10) i;"
 	}
 );
 
+sub idx_scan
+{
+	# Stats do not update instantaneously
+	# https://www.postgresql.org/docs/current/monitoring-stats.html#MONITORING-STATS-VIEWS
+	sleep(1);
+	$node->safe_psql("postgres", "SELECT idx_scan FROM pg_stat_user_indexes WHERE indexrelid = 'tst_v_idx'::regclass;");
+}
+
 my $expected = 10000 + 5 * 100 * 10;
 
 my $count = $node->safe_psql("postgres", "SELECT COUNT(*) FROM tst;");
 is($count, $expected);
+is(idx_scan(), 0);
 
 $count = $node->safe_psql("postgres", qq(
 	SET enable_seqscan = off;
@@ -43,3 +52,6 @@ $count = $node->safe_psql("postgres", qq(
 	SELECT COUNT(*) FROM (SELECT v FROM tst ORDER BY v <-> (SELECT v FROM tst LIMIT 1)) t;
 ));
 is($count, $expected);
+is(idx_scan(), 1);
+
+done_testing();

test/t/008_aggregates.pl

@@ -0,0 +1,49 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+# Initialize node
+my $node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (r1 real, r2 real, r3 real, v vector(3));");
+$node->safe_psql("postgres", qq(
+	INSERT INTO tst SELECT r1, r2, r3, ARRAY[r1, r2, r3] FROM (
+		SELECT random() + 1.01 AS r1, random() + 2.01 AS r2, random() + 3.01 AS r3 FROM generate_series(1, 1000000) t
+	) i;
+));
+
+sub test_aggregate
+{
+	my ($agg) = @_;
+
+	# Test value
+	my $res = $node->safe_psql("postgres", "SELECT $agg(v) FROM tst;");
+	like($res, qr/\[1\.5/);
+	like($res, qr/,2\.5/);
+	like($res, qr/,3\.5/);
+
+	# Test matches real for avg
+	# Cannot test sum since sum(real) varies between calls
+	if ($agg eq 'avg')
+	{
+		my $r1 = $node->safe_psql("postgres", "SELECT $agg(r1)::float4 FROM tst;");
+		my $r2 = $node->safe_psql("postgres", "SELECT $agg(r2)::float4 FROM tst;");
+		my $r3 = $node->safe_psql("postgres", "SELECT $agg(r3)::float4 FROM tst;");
+		is($res, "[$r1,$r2,$r3]");
+	}
+
+	# Test explain
+	my $explain = $node->safe_psql("postgres", "EXPLAIN SELECT $agg(v) FROM tst;");
+	like($explain, qr/Partial Aggregate/);
+}
+
+test_aggregate('avg');
+test_aggregate('sum');
+
+done_testing();

test/t/008_avg.pl

@@ -1,35 +0,0 @@
-use strict;
-use warnings;
-use PostgresNode;
-use TestLib;
-use Test::More tests => 5;
-
-# Initialize node
-my $node = get_new_node('node');
-$node->init;
-$node->start;
-
-# Create table
-$node->safe_psql("postgres", "CREATE EXTENSION vector;");
-$node->safe_psql("postgres", "CREATE TABLE tst (r1 real, r2 real, r3 real, v vector(3));");
-$node->safe_psql("postgres", qq(
-	INSERT INTO tst SELECT r1, r2, r3, ARRAY[r1, r2, r3] FROM (
-		SELECT random() + 1.01 AS r1, random() + 2.01 AS r2, random() + 3.01 AS r3 FROM generate_series(1, 1000000) t
-	) i;
-));
-
-# Test avg
-my $avg = $node->safe_psql("postgres", "SELECT AVG(v) FROM tst;");
-like($avg, qr/\[1\.5/);
-like($avg, qr/,2\.5/);
-like($avg, qr/,3\.5/);
-
-# Test matches real
-my $r1 = $node->safe_psql("postgres", "SELECT AVG(r1)::float4 FROM tst;");
-my $r2 = $node->safe_psql("postgres", "SELECT AVG(r2)::float4 FROM tst;");
-my $r3 = $node->safe_psql("postgres", "SELECT AVG(r3)::float4 FROM tst;");
-is($avg, "[$r1,$r2,$r3]");
-
-# Test explain
-my $explain = $node->safe_psql("postgres", "EXPLAIN SELECT AVG(v) FROM tst;");
-like($explain, qr/Partial Aggregate/);

test/t/009_storage.pl

@@ -2,7 +2,7 @@ use strict;
 use warnings;
 use PostgresNode;
 use TestLib;
-use Test::More tests => 1;
+use Test::More;
 
 my $dim = 1024;
 
@@ -30,3 +30,5 @@ my ($ret, $stdout, $stderr) = $node->psql("postgres",
 	"INSERT INTO tst SELECT array_agg(n), array_agg(n), array_agg(n) FROM generate_series(1, $dim) n"
 );
 like($stderr, qr/row is too big/);
+
+done_testing();

test/t/010_hnsw_wal.pl

@@ -0,0 +1,99 @@
+# Based on postgres/contrib/bloom/t/001_wal.pl
+
+# Test generic xlog record work for hnsw index replication.
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+my $dim = 32;
+
+my $node_primary;
+my $node_replica;
+
+# Run few queries on both primary and replica and check their results match.
+sub test_index_replay
+{
+	my ($test_name) = @_;
+
+	# Wait for replica to catch up
+	my $applname = $node_replica->name;
+	my $caughtup_query = "SELECT pg_current_wal_lsn() <= replay_lsn FROM pg_stat_replication WHERE application_name = '$applname';";
+	$node_primary->poll_query_until('postgres', $caughtup_query)
+	  or die "Timed out while waiting for replica 1 to catch up";
+
+	my @r = ();
+	for (1 .. $dim)
+	{
+		push(@r, rand());
+	}
+	my $sql = join(",", @r);
+
+	my $queries = qq(
+		SET enable_seqscan = off;
+		SELECT * FROM tst ORDER BY v <-> '[$sql]' LIMIT 10;
+	);
+
+	# Run test queries and compare their result
+	my $primary_result = $node_primary->safe_psql("postgres", $queries);
+	my $replica_result = $node_replica->safe_psql("postgres", $queries);
+
+	is($primary_result, $replica_result, "$test_name: query result matches");
+	return;
+}
+
+# Use ARRAY[random(), random(), random(), ...] over
+# SELECT array_agg(random()) FROM generate_series(1, $dim)
+# to generate different values for each row
+my $array_sql = join(",", ('random()') x $dim);
+
+# Initialize primary node
+$node_primary = get_new_node('primary');
+$node_primary->init(allows_streaming => 1);
+if ($dim > 32)
+{
+	# TODO use wal_keep_segments for Postgres < 13
+	$node_primary->append_conf('postgresql.conf', qq(wal_keep_size = 1GB));
+}
+if ($dim > 1500)
+{
+	$node_primary->append_conf('postgresql.conf', qq(maintenance_work_mem = 128MB));
+}
+$node_primary->start;
+my $backup_name = 'my_backup';
+
+# Take backup
+$node_primary->backup($backup_name);
+
+# Create streaming replica linking to primary
+$node_replica = get_new_node('replica');
+$node_replica->init_from_backup($node_primary, $backup_name, has_streaming => 1);
+$node_replica->start;
+
+# Create hnsw index on primary
+$node_primary->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node_primary->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector($dim));");
+$node_primary->safe_psql("postgres",
+	"INSERT INTO tst SELECT i % 10, ARRAY[$array_sql] FROM generate_series(1, 1000) i;"
+);
+$node_primary->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v vector_l2_ops);");
+
+# Test that queries give same result
+test_index_replay('initial');
+
+# Run 10 cycles of table modification. Run test queries after each modification.
+for my $i (1 .. 10)
+{
+	$node_primary->safe_psql("postgres", "DELETE FROM tst WHERE i = $i;");
+	test_index_replay("delete $i");
+	$node_primary->safe_psql("postgres", "VACUUM tst;");
+	test_index_replay("vacuum $i");
+	my ($start, $end) = (1001 + ($i - 1) * 100, 1000 + $i * 100);
+	$node_primary->safe_psql("postgres",
+		"INSERT INTO tst SELECT i % 10, ARRAY[$array_sql] FROM generate_series($start, $end) i;"
+	);
+	test_index_replay("insert $i");
+}
+
+done_testing();

test/t/011_hnsw_vacuum.pl

@@ -0,0 +1,54 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+my $dim = 3;
+
+my @r = ();
+for (1 .. $dim)
+{
+	my $v = int(rand(1000)) + 1;
+	push(@r, "i % $v");
+}
+my $array_sql = join(", ", @r);
+
+# Initialize node
+my $node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table and index
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector($dim));");
+$node->safe_psql("postgres",
+	"INSERT INTO tst SELECT i, ARRAY[$array_sql] FROM generate_series(1, 10000) i;"
+);
+$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v vector_l2_ops);");
+
+# Get size
+my $size = $node->safe_psql("postgres", "SELECT pg_total_relation_size('tst_v_idx');");
+
+# Delete all, vacuum, and insert same data
+$node->safe_psql("postgres", "DELETE FROM tst;");
+$node->safe_psql("postgres", "VACUUM tst;");
+$node->safe_psql("postgres",
+	"INSERT INTO tst SELECT i, ARRAY[$array_sql] FROM generate_series(1, 10000) i;"
+);
+
+# Check size
+# May increase some due to different levels
+my $new_size = $node->safe_psql("postgres", "SELECT pg_total_relation_size('tst_v_idx');");
+cmp_ok($new_size, "<=", $size * 1.02, "size does not increase too much");
+
+# Delete all but one
+$node->safe_psql("postgres", "DELETE FROM tst WHERE i != 123;");
+$node->safe_psql("postgres", "VACUUM tst;");
+my $res = $node->safe_psql("postgres", qq(
+	SET enable_seqscan = off;
+	SELECT i FROM tst ORDER BY v <-> '[0,0,0]' LIMIT 10;
+));
+is($res, 123);
+
+done_testing();

test/t/012_hnsw_build_recall.pl

@@ -0,0 +1,114 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+my $node;
+my @queries = ();
+my @expected;
+my $limit = 20;
+
+sub test_recall
+{
+	my ($min, $operator) = @_;
+	my $correct = 0;
+	my $total = 0;
+
+	my $explain = $node->safe_psql("postgres", qq(
+		SET enable_seqscan = off;
+		EXPLAIN ANALYZE SELECT i FROM tst ORDER BY v $operator '$queries[0]' LIMIT $limit;
+	));
+	like($explain, qr/Index Scan/);
+
+	for my $i (0 .. $#queries)
+	{
+		my $actual = $node->safe_psql("postgres", qq(
+			SET enable_seqscan = off;
+			SELECT i FROM tst ORDER BY v $operator '$queries[$i]' LIMIT $limit;
+		));
+		my @actual_ids = split("\n", $actual);
+		my %actual_set = map { $_ => 1 } @actual_ids;
+
+		my @expected_ids = split("\n", $expected[$i]);
+
+		foreach (@expected_ids)
+		{
+			if (exists($actual_set{$_}))
+			{
+				$correct++;
+			}
+			$total++;
+		}
+	}
+
+	cmp_ok($correct / $total, ">=", $min, $operator);
+}
+
+# Initialize node
+$node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector(3));");
+$node->safe_psql("postgres",
+	"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 10000) i;"
+);
+
+# Generate queries
+for (1 .. 20)
+{
+	my $r1 = rand();
+	my $r2 = rand();
+	my $r3 = rand();
+	push(@queries, "[$r1,$r2,$r3]");
+}
+
+# Check each index type
+my @operators = ("<->", "<#>", "<=>");
+my @opclasses = ("vector_l2_ops", "vector_ip_ops", "vector_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", "SELECT i FROM tst ORDER BY v $operator '$_' LIMIT $limit;");
+		push(@expected, $res);
+	}
+
+	# Build index serially
+	$node->safe_psql("postgres", qq(
+		SET max_parallel_maintenance_workers = 0;
+		CREATE INDEX idx ON tst USING hnsw (v $opclass);
+	));
+
+	# Test approximate results
+	my $min = $operator eq "<#>" ? 0.80 : 0.99;
+	test_recall($min, $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;
+		SET hnsw.enable_parallel_build = on;
+		CREATE INDEX idx ON tst USING hnsw (v $opclass);
+	));
+	is($ret, 0, $stderr);
+	like($stderr, qr/using \d+ parallel workers/);
+
+	# Test approximate results
+	test_recall($min, $operator);
+
+	$node->safe_psql("postgres", "DROP INDEX idx;");
+}
+
+done_testing();

test/t/013_hnsw_insert_recall.pl

@@ -0,0 +1,108 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+my $node;
+my @queries = ();
+my @expected;
+my $limit = 20;
+
+sub test_recall
+{
+	my ($min, $operator) = @_;
+	my $correct = 0;
+	my $total = 0;
+
+	my $explain = $node->safe_psql("postgres", qq(
+		SET enable_seqscan = off;
+		EXPLAIN ANALYZE SELECT i FROM tst ORDER BY v $operator '$queries[0]' LIMIT $limit;
+	));
+	like($explain, qr/Index Scan/);
+
+	for my $i (0 .. $#queries)
+	{
+		my $actual = $node->safe_psql("postgres", qq(
+			SET enable_seqscan = off;
+			SELECT i FROM tst ORDER BY v $operator '$queries[$i]' LIMIT $limit;
+		));
+		my @actual_ids = split("\n", $actual);
+		my %actual_set = map { $_ => 1 } @actual_ids;
+
+		my @expected_ids = split("\n", $expected[$i]);
+
+		foreach (@expected_ids)
+		{
+			if (exists($actual_set{$_}))
+			{
+				$correct++;
+			}
+			$total++;
+		}
+	}
+
+	cmp_ok($correct / $total, ">=", $min, $operator);
+}
+
+# Initialize node
+$node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i serial, v vector(3));");
+
+# Generate queries
+for (1 .. 20)
+{
+	my $r1 = rand();
+	my $r2 = rand();
+	my $r3 = rand();
+	push(@queries, "[$r1,$r2,$r3]");
+}
+
+# Check each index type
+my @operators = ("<->", "<#>", "<=>");
+my @opclasses = ("vector_l2_ops", "vector_ip_ops", "vector_cosine_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",
+		{
+			"013_hnsw_insert_recall_$opclass" => "INSERT INTO tst (v) VALUES (ARRAY[random(), random(), random()]);"
+		}
+	);
+
+	# 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);
+	}
+
+	my $min = $operator eq "<#>" ? 0.80 : 0.99;
+	test_recall($min, $operator);
+
+	$node->safe_psql("postgres", "DROP INDEX idx;");
+	$node->safe_psql("postgres", "TRUNCATE tst;");
+}
+
+done_testing();

test/t/014_hnsw_inserts.pl

@@ -0,0 +1,74 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+# Ensures elements and neighbors on both same and different pages
+my $dim = 1900;
+
+my $array_sql = join(",", ('random()') x $dim);
+
+# Initialize node
+my $node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table and index
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (v vector($dim));");
+$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v vector_l2_ops);");
+
+sub idx_scan
+{
+	# Stats do not update instantaneously
+	# https://www.postgresql.org/docs/current/monitoring-stats.html#MONITORING-STATS-VIEWS
+	sleep(1);
+	$node->safe_psql("postgres", "SELECT idx_scan FROM pg_stat_user_indexes WHERE indexrelid = 'tst_v_idx'::regclass;");
+}
+
+for my $i (1 .. 20)
+{
+	$node->pgbench(
+		"--no-vacuum --client=10 --transactions=1",
+		0,
+		[qr{actually processed}],
+		[qr{^$}],
+		"concurrent INSERTs",
+		{
+			"014_hnsw_inserts_$i" => "INSERT INTO tst VALUES (ARRAY[$array_sql]);"
+		}
+	);
+
+	my $count = $node->safe_psql("postgres", qq(
+		SET enable_seqscan = off;
+		SELECT COUNT(*) FROM (SELECT v FROM tst ORDER BY v <-> (SELECT v FROM tst LIMIT 1)) t;
+	));
+	is($count, 10);
+
+	$node->safe_psql("postgres", "TRUNCATE tst;");
+}
+
+$node->pgbench(
+	"--no-vacuum --client=20 --transactions=5",
+	0,
+	[qr{actually processed}],
+	[qr{^$}],
+	"concurrent INSERTs",
+	{
+		"014_hnsw_inserts" => "INSERT INTO tst SELECT ARRAY[$array_sql] FROM generate_series(1, 10) i;"
+	}
+);
+
+my $count = $node->safe_psql("postgres", qq(
+	SET enable_seqscan = off;
+	SET hnsw.ef_search = 1000;
+	SELECT COUNT(*) FROM (SELECT v FROM tst ORDER BY v <-> (SELECT v FROM tst LIMIT 1)) t;
+));
+# Elements may lose all incoming connections with the HNSW algorithm
+# Vacuuming can fix this if one of the elements neighbors is deleted
+cmp_ok($count, ">=", 997);
+
+is(idx_scan(), 21);
+
+done_testing();

test/t/015_hnsw_duplicates.pl

@@ -0,0 +1,58 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+# Initialize node
+my $node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (v vector(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();
+$node->safe_psql("postgres", "CREATE INDEX idx ON tst USING hnsw (v vector_l2_ops);");
+test_duplicates();
+
+# Reset
+$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",
+	{
+		"015_hnsw_duplicates" => "INSERT INTO tst VALUES ('[1,1,1]');"
+	}
+);
+
+done_testing();

test/t/016_hnsw_vacuum_recall.pl

@@ -0,0 +1,97 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+my $node;
+my @queries = ();
+my @expected;
+my $limit = 20;
+
+sub test_recall
+{
+	my ($min, $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 = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector(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
+$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v vector_l2_ops) WITH (m = 4, ef_construction = 8);");
+
+# Delete data
+$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(0.20, $limit, "before vacuum");
+test_recall(0.95, 100, "before vacuum");
+
+# TODO Test concurrent inserts with vacuum
+$node->safe_psql("postgres", "VACUUM tst;");
+
+test_recall(0.95, $limit, "after vacuum");
+
+done_testing();

test/t/017_ivfflat_insert_recall.pl

@@ -0,0 +1,117 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+my $node;
+my @queries = ();
+my @expected;
+my $limit = 20;
+
+sub test_recall
+{
+	my ($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 %actual_set = map { $_ => 1 } @actual_ids;
+
+		my @expected_ids = split("\n", $expected[$i]);
+
+		foreach (@expected_ids)
+		{
+			if (exists($actual_set{$_}))
+			{
+				$correct++;
+			}
+			$total++;
+		}
+	}
+
+	cmp_ok($correct / $total, ">=", $min, $operator);
+}
+
+# Initialize node
+$node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i serial, v vector(3));");
+
+# Generate queries
+for (1 .. 20)
+{
+	my $r1 = rand();
+	my $r2 = rand();
+	my $r3 = rand();
+	push(@queries, "[$r1,$r2,$r3]");
+}
+
+# Check each index type
+my @operators = ("<->", "<#>", "<=>");
+my @opclasses = ("vector_l2_ops", "vector_ip_ops", "vector_cosine_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 ivfflat (v $opclass);");
+
+	# Use concurrent inserts
+	$node->pgbench(
+		"--no-vacuum --client=10 --transactions=1000",
+		0,
+		[qr{actually processed}],
+		[qr{^$}],
+		"concurrent INSERTs",
+		{
+			"017_ivfflat_insert_recall_$opclass" => "INSERT INTO tst (v) SELECT ARRAY[random(), random(), random()] FROM generate_series(1, 10) i;"
+		}
+	);
+
+	# 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, $res);
+	}
+
+	# Test approximate results
+	if ($operator ne "<#>")
+	{
+		# TODO Fix test (uniform random vectors all have similar inner product)
+		test_recall(1, 0.71, $operator);
+		test_recall(10, 0.95, $operator);
+	}
+	# Account for equal distances
+	test_recall(100, 0.9925, $operator);
+
+	$node->safe_psql("postgres", "DROP INDEX idx;");
+	$node->safe_psql("postgres", "TRUNCATE tst;");
+}
+
+done_testing();

test/t/018_ivfflat_deletes.pl

@@ -0,0 +1,43 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+my $dim = 3;
+
+my $array_sql = join(",", ('random()') x $dim);
+
+# Initialize node
+my $node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table and index
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i serial, v vector($dim));");
+$node->safe_psql("postgres",
+	"INSERT INTO tst (v) SELECT ARRAY[$array_sql] FROM generate_series(1, 10000) i;"
+);
+$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops);");
+
+# Delete data
+$node->safe_psql("postgres", "DELETE FROM tst WHERE i % 100 != 0;");
+
+my $exp = $node->safe_psql("postgres", qq(
+	SET enable_indexscan = off;
+	SELECT i FROM tst ORDER BY v <-> '[0,0,0]';
+));
+
+# Run twice to make sure correct tuples marked as dead
+for (1 .. 2)
+{
+	my $res = $node->safe_psql("postgres", qq(
+		SET enable_seqscan = off;
+		SET ivfflat.probes = 100;
+		SELECT i FROM tst ORDER BY v <-> '[0,0,0]';
+	));
+	is($res, $exp);
+}
+
+done_testing();

vector.control

@@ -1,4 +1,4 @@
-comment = 'vector data type and ivfflat access method'
-default_version = '0.4.0'
+comment = 'vector data type and ivfflat and hnsw access methods'
+default_version = '0.5.1'
 module_pathname = '$libdir/vector'
 relocatable = true