Removed duplicated macros [skip ci]

.dockerignore

@@ -1,15 +1,8 @@
 /.git/
 /dist/
-/log/
 /results/
 /tmp_check/
 /sql/vector--?.?.?.sql
 regression.*
 *.o
 *.so
-*.bc
-*.dll
-*.dylib
-*.obj
-*.lib
-*.exp

.github/workflows/build.yml

@@ -8,18 +8,18 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - postgres: 18
+          # - postgres: 18
-            os: ubuntu-24.04
+          #   os: ubuntu-24.04
           - postgres: 17
             os: ubuntu-24.04
           - postgres: 16
-            os: ubuntu-24.04-arm
+            os: ubuntu-22.04
           - postgres: 15
             os: ubuntu-22.04
           - postgres: 14
-            os: ubuntu-22.04-arm
+            os: ubuntu-20.04
           - postgres: 13
-            os: ubuntu-22.04
+            os: ubuntu-20.04
     steps:
       - uses: actions/checkout@v4
       - uses: ankane/setup-postgres@v1
@@ -28,7 +28,7 @@ jobs:
           dev-files: true
       - run: make
         env:
-          PG_CFLAGS: ${{ matrix.postgres == 18 && '-Wno-missing-field-initializers' || '' }} -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
+          PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
       - run: |
           export PG_CONFIG=`which pg_config`
           sudo --preserve-env=PG_CONFIG make install
@@ -46,10 +46,10 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - postgres: 17
+          - postgres: 16
-            os: macos-15
+            os: macos-14
           - postgres: 14
-            os: macos-13
+            os: macos-12
     steps:
       - uses: actions/checkout@v4
       - uses: ankane/setup-postgres@v1
@@ -57,7 +57,7 @@ jobs:
           postgres-version: ${{ matrix.postgres }}
       - run: make
         env:
-          PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-unknown-warning-option
+          PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter
       - run: make install
       - run: make installcheck
       - if: ${{ failure() }}
@@ -70,13 +70,12 @@ jobs:
           tar xf $TAG.tar.gz
           mv postgres-$TAG postgres
         env:
-          TAG: ${{ matrix.postgres == 17 && 'REL_17_2' || 'REL_14_15' }}
+          TAG: ${{ matrix.postgres == 16 && 'REL_16_2' || 'REL_14_11' }}
       - run: make prove_installcheck PROVE_FLAGS="-I ./postgres/src/test/perl -I ./test/perl"
         env:
           PERL5LIB: /Users/runner/perl5/lib/perl5
-      - run: make clean && $(brew --prefix llvm@$LLVM_VERSION)/bin/scan-build --status-bugs make
+      - run: make clean && $(brew --prefix llvm@15)/bin/scan-build --status-bugs make
         env:
-          LLVM_VERSION: ${{ matrix.os == 'macos-15' && 18 || 15 }}
           PG_CFLAGS: -DUSE_ASSERT_CHECKING
   windows:
     runs-on: windows-latest
@@ -126,7 +125,7 @@ jobs:
       - uses: actions/checkout@v4
       - uses: ankane/setup-postgres-valgrind@v1
         with:
-          postgres-version: 17
+          postgres-version: 16
           check-ub: yes
       - run: make OPTFLAGS=""
       - run: sudo --preserve-env=PG_CONFIG make install

CHANGELOG.md

@@ -1,14 +1,10 @@
-## 0.8.1 (unreleased)
+## 0.8.0 (unreleased)
-
-- Improved performance of HNSW index scans for Postgres 17
-
-## 0.8.0 (2024-10-30)
 
 - Added support for iterative index scans
 - Added casts for arrays to `sparsevec`
-- Improved cost estimation for better index selection when filtering
+- Improved cost estimation
-- Improved performance of HNSW index scans
 - Improved performance of HNSW inserts and on-disk index builds
+- Reduced memory usage for HNSW index scans
 - Dropped support for Postgres 12
 
 ## 0.7.4 (2024-08-05)

LICENSE

@@ -1,4 +1,4 @@
-Portions Copyright (c) 1996-2025, PostgreSQL Global Development Group
+Portions Copyright (c) 1996-2024, 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": "0.8.0",
+	"version": "0.7.4",
 	"maintainer": [
 		"Andrew Kane <andrew@ankane.org>"
 	],
@@ -12,7 +12,7 @@
 	"prereqs": {
 		"runtime": {
 			"requires": {
-				"PostgreSQL": "13.0.0"
+				"PostgreSQL": "12.0.0"
 			}
 		}
 	},
@@ -20,7 +20,7 @@
 		"vector": {
 			"file": "sql/vector.sql",
 			"docfile": "README.md",
-			"version": "0.8.0",
+			"version": "0.7.4",
 			"abstract": "Open-source vector similarity search for Postgres"
 		}
 	},

Makefile

@@ -1,5 +1,5 @@
 EXTENSION = vector
-EXTVERSION = 0.8.0
+EXTVERSION = 0.7.4
 
 MODULE_big = vector
 DATA = $(wildcard sql/*--*--*.sql)

Makefile.win

@@ -1,5 +1,5 @@
 EXTENSION = vector
-EXTVERSION = 0.8.0
+EXTVERSION = 0.7.4
 
 DATA_built = sql\$(EXTENSION)--$(EXTVERSION).sql
 OBJS = src\bitutils.obj src\bitvec.obj src\halfutils.obj src\halfvec.obj src\hnsw.obj src\hnswbuild.obj src\hnswinsert.obj src\hnswscan.obj src\hnswutils.obj src\hnswvacuum.obj src\ivfbuild.obj src\ivfflat.obj src\ivfinsert.obj src\ivfkmeans.obj src\ivfscan.obj src\ivfutils.obj src\ivfvacuum.obj src\sparsevec.obj src\vector.obj

README.md

@@ -17,11 +17,11 @@ Plus [ACID](https://en.wikipedia.org/wiki/ACID) compliance, point-in-time recove
 
 ### Linux and Mac
 
-Compile and install the extension (supports Postgres 13+)
+Compile and install the extension (supports Postgres 12+)
 
 ```sh
 cd /tmp
-git clone --branch v0.8.0 https://github.com/pgvector/pgvector.git
+git clone --branch v0.7.4 https://github.com/pgvector/pgvector.git
 cd pgvector
 make
 make install # may need sudo
@@ -33,17 +33,27 @@ You can also install it with [Docker](#docker), [Homebrew](#homebrew), [PGXN](#p
 
 ### Windows
 
-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 `x64 Native Tools Command Prompt for VS [version]` as administrator. Then use `nmake` to build:
+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
-set "PGROOT=C:\Program Files\PostgreSQL\17"
+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\16"
 cd %TEMP%
-git clone --branch v0.8.0 https://github.com/pgvector/pgvector.git
+git clone --branch v0.7.4 https://github.com/pgvector/pgvector.git
 cd pgvector
 nmake /F Makefile.win
 nmake /F Makefile.win install
 ```
 
+Note: Postgres 17 is not supported yet due to an upstream issue
+
 See the [installation notes](#installation-notes---windows) if you run into issues
 
 You can also install it with [Docker](#docker) or [conda-forge](#conda-forge).
@@ -74,7 +84,7 @@ Get the nearest neighbors by L2 distance
 SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
 ```
 
-Also supports inner product (`<#>`), cosine distance (`<=>`), and L1 distance (`<+>`)
+Also supports inner product (`<#>`), cosine distance (`<=>`), and L1 distance (`<+>`, added in 0.7.0)
 
 Note: `<#>` returns the negative inner product since Postgres only supports `ASC` order index scans on operators
 
@@ -138,9 +148,9 @@ Supported distance functions are:
 - `<->` - L2 distance
 - `<#>` - (negative) inner product
 - `<=>` - cosine distance
-- `<+>` - L1 distance
+- `<+>` - L1 distance (added in 0.7.0)
-- `<~>` - Hamming distance (binary vectors)
+- `<~>` - Hamming distance (binary vectors, added in 0.7.0)
-- `<%>` - Jaccard distance (binary vectors)
+- `<%>` - Jaccard distance (binary vectors, added in 0.7.0)
 
 Get the nearest neighbors to a row
 
@@ -227,19 +237,19 @@ Cosine distance
 CREATE INDEX ON items USING hnsw (embedding vector_cosine_ops);
 ```
 
-L1 distance
+L1 distance - added in 0.7.0
 
 ```sql
 CREATE INDEX ON items USING hnsw (embedding vector_l1_ops);
 ```
 
-Hamming distance
+Hamming distance - added in 0.7.0
 
 ```sql
 CREATE INDEX ON items USING hnsw (embedding bit_hamming_ops);
 ```
 
-Jaccard distance
+Jaccard distance - added in 0.7.0
 
 ```sql
 CREATE INDEX ON items USING hnsw (embedding bit_jaccard_ops);
@@ -248,9 +258,9 @@ CREATE INDEX ON items USING hnsw (embedding bit_jaccard_ops);
 Supported types are:
 
 - `vector` - up to 2,000 dimensions
-- `halfvec` - up to 4,000 dimensions
+- `halfvec` - up to 4,000 dimensions (added in 0.7.0)
-- `bit` - up to 64,000 dimensions
+- `bit` - up to 64,000 dimensions (added in 0.7.0)
-- `sparsevec` - up to 1,000 non-zero elements
+- `sparsevec` - up to 1,000 non-zero elements (added in 0.7.0)
 
 ### Index Options
 
@@ -304,19 +314,17 @@ Note: Do not set `maintenance_work_mem` so high that it exhausts the memory on t
 
 Like other index types, it’s faster to create an index after loading your initial data
 
-You can also speed up index creation by increasing the number of parallel workers (2 by default)
+Starting with 0.6.0, you can also speed up index creation by increasing the number of parallel workers (2 by default)
 
 ```sql
 SET max_parallel_maintenance_workers = 7; -- plus leader
 ```
 
-For a large number of workers, you may need to increase `max_parallel_workers` (8 by default)
+For a large number of workers, you may also need to increase `max_parallel_workers` (8 by default)
-
-The [index options](#index-options) also have a significant impact on build time (use the defaults unless seeing low recall)
 
 ### Indexing Progress
 
-Check [indexing progress](https://www.postgresql.org/docs/current/progress-reporting.html#CREATE-INDEX-PROGRESS-REPORTING)
+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;
@@ -359,7 +367,7 @@ Cosine distance
 CREATE INDEX ON items USING ivfflat (embedding vector_cosine_ops) WITH (lists = 100);
 ```
 
-Hamming distance
+Hamming distance - added in 0.7.0
 
 ```sql
 CREATE INDEX ON items USING ivfflat (embedding bit_hamming_ops) WITH (lists = 100);
@@ -368,8 +376,8 @@ CREATE INDEX ON items USING ivfflat (embedding bit_hamming_ops) WITH (lists = 10
 Supported types are:
 
 - `vector` - up to 2,000 dimensions
-- `halfvec` - up to 4,000 dimensions
+- `halfvec` - up to 4,000 dimensions (added in 0.7.0)
-- `bit` - up to 64,000 dimensions
+- `bit` - up to 64,000 dimensions (added in 0.7.0)
 
 ### Query Options
 
@@ -402,7 +410,7 @@ For a large number of workers, you may also need to increase `max_parallel_worke
 
 ### Indexing Progress
 
-Check [indexing progress](https://www.postgresql.org/docs/current/progress-reporting.html#CREATE-INDEX-PROGRESS-REPORTING)
+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 * tuples_done / nullif(tuples_total, 0), 1) AS "%" FROM pg_stat_progress_create_index;
@@ -419,115 +427,92 @@ Note: `%` is only populated during the `loading tuples` phase
 
 ## Filtering
 
-There are a few ways to index nearest neighbor queries with a `WHERE` clause.
+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;
 ```
 
-A good place to start is creating an index on the filter column. This can provide fast, exact nearest neighbor search in many cases. Postgres has a number of [index types](https://www.postgresql.org/docs/current/indexes-types.html) for this: B-tree (default), hash, GiST, SP-GiST, GIN, and BRIN.
+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 (category_id);
 ```
 
-For multiple columns, consider a [multicolumn index](https://www.postgresql.org/docs/current/indexes-multicolumn.html).
+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 (location_id, category_id);
-```
-
-Exact indexes work well for conditions that match a low percentage of rows. Otherwise, [approximate indexes](#indexing) can work better.
-
-```sql
-CREATE INDEX ON items USING hnsw (embedding vector_l2_ops);
-```
-
-With approximate indexes, filtering is applied *after* the index is scanned. If a condition matches 10% of rows, with HNSW and the default `hnsw.ef_search` of 40, only 4 rows will match on average. For more rows, increase `hnsw.ef_search`.
-
-```sql
-SET hnsw.ef_search = 200;
-```
-
-Starting with 0.8.0, you can enable [iterative index scans](#iterative-index-scans), which will automatically scan more of the index when needed.
-
-```sql
-SET hnsw.iterative_scan = strict_order;
-```
-
-If filtering by only a few distinct values, consider [partial indexing](https://www.postgresql.org/docs/current/indexes-partial.html).
 
 ```sql
 CREATE INDEX ON items USING hnsw (embedding vector_l2_ops) WHERE (category_id = 123);
 ```
 
-If filtering by many different values, consider [partitioning](https://www.postgresql.org/docs/current/ddl-partitioning.html).
+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);
 ```
 
-## Iterative Index Scans
+## Iterative Search [unreleased]
 
-With approximate indexes, queries with filtering can return less results since filtering is applied *after* the index is scanned. Starting with 0.8.0, you can enable iterative index scans, which will automatically scan more of the index until enough results are found (or it reaches `hnsw.max_scan_tuples` or `ivfflat.max_probes`).
+*Added in 0.8.0*
 
-Iterative scans can use strict or relaxed ordering.
+With approximate indexes, you can end up with less results than expected due to filtering conditions in the query.
 
-Strict ensures results are in the exact order by distance
+Starting with 0.8.0, you can enable iterative search. If too few results from the initial index scan match the query filters, it will resume scanning until enough results are found. This can significantly improve recall (at the cost of speed).
 
-```sql
+```tsql
-SET hnsw.iterative_scan = strict_order;
+SET hnsw.streaming = on;
+-- or
+SET ivfflat.streaming = on;
 ```
 
-Relaxed allows results to be slightly out of order by distance, but provides better recall
+However, there are some important caveats.
+
+### Iterative Caveats
+
+With iterative search, it’s possible for rows to be slightly out of order by distance. For strict ordering, use:
 
 ```sql
-SET hnsw.iterative_scan = relaxed_order;
+WITH approx_order AS MATERIALIZED (
-# or
+    SELECT *, embedding <-> '[1,2,3]' AS distance FROM items WHERE ... ORDER BY distance LIMIT 5
-SET ivfflat.iterative_scan = relaxed_order;
+) SELECT * FROM approx_order ORDER BY distance;
 ```
 
-With relaxed ordering, you can use a [materialized CTE](https://www.postgresql.org/docs/current/queries-with.html#QUERIES-WITH-CTE-MATERIALIZATION) to get strict ordering
+For distance filters, use a CTE and place the filter outside it.
 
 ```sql
-WITH relaxed_results AS MATERIALIZED (
+WITH approx_order AS MATERIALIZED (
-    SELECT id, embedding <-> '[1,2,3]' AS distance FROM items WHERE category_id = 123 ORDER BY distance LIMIT 5
+    SELECT *, embedding <-> '[1,2,3]' AS distance FROM items WHERE ... ORDER BY distance LIMIT 5
-) SELECT * FROM relaxed_results ORDER BY distance + 0;
+) SELECT * FROM approx_order WHERE distance < 0.1 ORDER BY distance;
 ```
 
-Note: `+ 0` is needed for Postgres 17+
+### Iterative Options
 
-For queries that filter by distance, use a materialized CTE and place the distance filter outside of it for best performance (due to the [current behavior](https://www.postgresql.org/message-id/flat/CAOdR5yGUoMQ6j7M5hNUXrySzaqZVGf_Ne%2B8fwZMRKTFxU1nbJg%40mail.gmail.com) of the Postgres executor)
+Since scanning a large portion of the index is expensive, there are options to control when the scan ends.
-
-```sql
-WITH nearest_results AS MATERIALIZED (
-    SELECT id, embedding <-> '[1,2,3]' AS distance FROM items ORDER BY distance LIMIT 5
-) SELECT * FROM nearest_results WHERE distance < 5 ORDER BY distance;
-```
-
-Note: Place any other filters inside the CTE
-
-### Iterative Scan Options
-
-Since scanning a large portion of an approximate index is expensive, there are options to control when a scan ends.
 
 #### HNSW
 
-Specify the max number of tuples to visit (20,000 by default)
+Specify the max number of additional tuples visited
 
 ```sql
-SET hnsw.max_scan_tuples = 20000;
+SET hnsw.ef_stream = 10000;
 ```
 
-Note: This is approximate and does not affect the initial scan
+The scan will also end if reaches `work_mem`. You can see when this happens by enabling debug messages.
-
-Specify the max amount of memory to use, as a multiple of `work_mem` (1 by default)
 
 ```sql
-SET hnsw.scan_mem_multiplier = 2;
+SET client_min_messages = debug1;
 ```
 
-Note: Try increasing this if increasing `hnsw.max_scan_tuples` does not improve recall
+```text
+DEBUG:  hnsw index scan exceeded work_mem after 10000 tuples
+HINT:  Increase work_mem to scan more tuples.
+```
+
+If the server has enough memory, you can adjust this with:
+
+```sql
+SET work_mem = '8MB';
+```
 
 #### IVFFlat
 
@@ -537,10 +522,10 @@ Specify the max number of probes
 SET ivfflat.max_probes = 100;
 ```
 
-Note: If this is lower than `ivfflat.probes`, `ivfflat.probes` will be used
-
 ## Half-Precision Vectors
 
+*Added in 0.7.0*
+
 Use the `halfvec` type to store half-precision vectors
 
 ```sql
@@ -549,6 +534,8 @@ CREATE TABLE items (id bigserial PRIMARY KEY, embedding halfvec(3));
 
 ## Half-Precision Indexing
 
+*Added in 0.7.0*
+
 Index vectors at half precision for smaller indexes
 
 ```sql
@@ -570,16 +557,24 @@ CREATE TABLE items (id bigserial PRIMARY KEY, embedding bit(3));
 INSERT INTO items (embedding) VALUES ('000'), ('111');
 ```
 
-Get the nearest neighbors by Hamming distance
+Get the nearest neighbors by Hamming distance (added in 0.7.0)
 
 ```sql
 SELECT * FROM items ORDER BY embedding <~> '101' LIMIT 5;
 ```
 
+Or (before 0.7.0)
+
+```sql
+SELECT * FROM items ORDER BY bit_count(embedding # '101') LIMIT 5;
+```
+
 Also supports Jaccard distance (`<%>`)
 
 ## Binary Quantization
 
+*Added in 0.7.0*
+
 Use expression indexing for binary quantization
 
 ```sql
@@ -602,6 +597,8 @@ SELECT * FROM (
 
 ## Sparse Vectors
 
+*Added in 0.7.0*
+
 Use the `sparsevec` type to store sparse vectors
 
 ```sql
@@ -635,6 +632,8 @@ You can use [Reciprocal Rank Fusion](https://github.com/pgvector/pgvector-python
 
 ## Indexing Subvectors
 
+*Added in 0.7.0*
+
 Use expression indexing to index subvectors
 
 ```sql
@@ -748,6 +747,8 @@ SELECT query, calls, ROUND((total_plan_time + total_exec_time) / calls) AS avg_t
     FROM pg_stat_statements ORDER BY total_plan_time + total_exec_time DESC LIMIT 20;
 ```
 
+Note: Replace `total_plan_time + total_exec_time` with `total_time` for Postgres < 13
+
 Monitor recall by comparing results from approximate search with exact search.
 
 ```sql
@@ -775,17 +776,13 @@ C | [pgvector-c](https://github.com/pgvector/pgvector-c)
 C++ | [pgvector-cpp](https://github.com/pgvector/pgvector-cpp)
 C#, F#, Visual Basic | [pgvector-dotnet](https://github.com/pgvector/pgvector-dotnet)
 Crystal | [pgvector-crystal](https://github.com/pgvector/pgvector-crystal)
-D | [pgvector-d](https://github.com/pgvector/pgvector-d)
 Dart | [pgvector-dart](https://github.com/pgvector/pgvector-dart)
 Elixir | [pgvector-elixir](https://github.com/pgvector/pgvector-elixir)
-Erlang | [pgvector-erlang](https://github.com/pgvector/pgvector-erlang)
-Fortran | [pgvector-fortran](https://github.com/pgvector/pgvector-fortran)
-Gleam | [pgvector-gleam](https://github.com/pgvector/pgvector-gleam)
 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.jl](https://github.com/pgvector/Pgvector.jl)
+Julia | [pgvector-julia](https://github.com/pgvector/pgvector-julia)
 Lisp | [pgvector-lisp](https://github.com/pgvector/pgvector-lisp)
 Lua | [pgvector-lua](https://github.com/pgvector/pgvector-lua)
 Nim | [pgvector-nim](https://github.com/pgvector/pgvector-nim)
@@ -794,7 +791,6 @@ 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)
-Raku | [pgvector-raku](https://github.com/pgvector/pgvector-raku)
 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)
@@ -816,7 +812,7 @@ You can use [half-precision indexing](#half-precision-indexing) to index up to 4
 
 #### Can I store vectors with different dimensions in the same column?
 
-You can use `vector` as the type (instead of `vector(n)`).
+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));
@@ -916,7 +912,7 @@ ALTER TABLE items ALTER COLUMN embedding SET STORAGE PLAIN;
 
 #### Why are there less results for a query after adding an HNSW index?
 
-Results are limited by the size of the dynamic candidate list (`hnsw.ef_search`), which is 40 by default. There may be even less results due to dead tuples or filtering conditions in the query. Enabling [iterative index scans](#iterative-index-scans) can help address this.
+Results are limited by the size of the dynamic candidate list (`hnsw.ef_search`). There may be even less results due to dead tuples or filtering conditions in the query. We recommend setting `hnsw.ef_search` to at least twice the `LIMIT` of the query. If you need more than 500 results, use an IVFFlat index instead.
 
 Also, note that `NULL` vectors are not indexed (as well as zero vectors for cosine distance).
 
@@ -928,7 +924,7 @@ The index was likely created with too little data for the number of lists. Drop
 DROP INDEX index_name;
 ```
 
-Results can also be limited by the number of probes (`ivfflat.probes`). Enabling [iterative index scans](#iterative-index-scans) can address this.
+Results can also be limited by the number of probes (`ivfflat.probes`).
 
 Also, note that `NULL` vectors are not indexed (as well as zero vectors for cosine distance).
 
@@ -1095,13 +1091,7 @@ Note: Replace `17` with your Postgres server version
 
 ### Missing SDK
 
-If compilation fails and the output includes `warning: no such sysroot directory` on Mac, your Postgres installation points to a path that no longer exists.
+If compilation fails and the output includes `warning: no such sysroot directory` on Mac, reinstall Xcode Command Line Tools.
-
-```sh
-pg_config --cppflags
-```
-
-Reinstall Postgres to fix this.
 
 ### Portability
 
@@ -1119,14 +1109,6 @@ make OPTFLAGS=""
 
 If compilation fails with `Cannot open include file: 'postgres.h': No such file or directory`, make sure `PGROOT` is correct.
 
-### Mismatched Architecture
-
-If compilation fails with `error C2196: case value '4' already used`, make sure you’re using the `x64 Native Tools Command Prompt`. Then run `nmake /F Makefile.win clean` and re-run the installation instructions.
-
-### Missing Symbol
-
-If linking fails with `unresolved external symbol float_to_shortest_decimal_bufn` with Postgres 17.0-17.2, upgrade to Postgres 17.3+.
-
 ### Permissions
 
 If installation fails with `Access is denied`, re-run the installation instructions as an administrator.
@@ -1146,17 +1128,11 @@ This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (r
 You can also build the image manually:
 
 ```sh
-git clone --branch v0.8.0 https://github.com/pgvector/pgvector.git
+git clone --branch v0.7.4 https://github.com/pgvector/pgvector.git
 cd pgvector
 docker build --pull --build-arg PG_MAJOR=17 -t myuser/pgvector .
 ```
 
-If you increase `maintenance_work_mem`, make sure `--shm-size` is at least that size to avoid an error with parallel HNSW index builds.
-
-```sh
-docker run --shm-size=1g ...
-```
-
 ### Homebrew
 
 With Homebrew Postgres, you can use:
@@ -1202,7 +1178,7 @@ Note: Replace `17` with your Postgres server version
 Install the FreeBSD package with:
 
 ```sh
-pkg install postgresql16-pgvector
+pkg install postgresql15-pgvector
 ```
 
 or the port with:
@@ -1244,6 +1220,36 @@ You can check the version in the current database with:
 SELECT extversion FROM pg_extension WHERE extname = 'vector';
 ```
 
+## Upgrade Notes
+
+### 0.6.0
+
+#### Postgres 12
+
+If upgrading with Postgres 12, 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;`.
+
+#### Docker
+
+The Docker image is now published in the `pgvector` org, and there are tags for each supported version of Postgres (rather than a `latest` tag).
+
+```sh
+docker pull pgvector/pgvector:pg16
+# or
+docker pull pgvector/pgvector:0.6.0-pg16
+```
+
+Also, if you’ve increased `maintenance_work_mem`, make sure `--shm-size` is at least that size to avoid an error with parallel HNSW index builds.
+
+```sh
+docker run --shm-size=1g ...
+```
+
 ## Thanks
 
 Thanks to:

src/hnsw.c

@@ -18,17 +18,18 @@
 #define MarkGUCPrefixReserved(x) EmitWarningsOnPlaceholders(x)
 #endif
 
-static const struct config_enum_entry hnsw_iterative_scan_options[] = {
+static const struct config_enum_entry hnsw_iterative_search_options[] = {
-	{"off", HNSW_ITERATIVE_SCAN_OFF, false},
+	{"off", HNSW_ITERATIVE_SEARCH_OFF, false},
-	{"relaxed_order", HNSW_ITERATIVE_SCAN_RELAXED, false},
+	{"strict", HNSW_ITERATIVE_SEARCH_STRICT, false},
-	{"strict_order", HNSW_ITERATIVE_SCAN_STRICT, false},
+	{"relaxed", HNSW_ITERATIVE_SEARCH_RELAXED, false},
+	/* TODO Change to strict before merging */
+	{"on", HNSW_ITERATIVE_SEARCH_RELAXED, false},
 	{NULL, 0, false}
 };
 
 int			hnsw_ef_search;
-int			hnsw_iterative_scan;
+int			hnsw_max_iterative_tuples;
-int			hnsw_max_scan_tuples;
+int			hnsw_iterative_search;
-double		hnsw_scan_mem_multiplier;
 int			hnsw_lock_tranche_id;
 static relopt_kind hnsw_relopt_kind;
 
@@ -79,19 +80,16 @@ HnswInit(void)
 							"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);
 
-	DefineCustomEnumVariable("hnsw.iterative_scan", "Sets the mode for iterative scans",
+	/* TODO Change name */
-							 NULL, &hnsw_iterative_scan,
+	DefineCustomEnumVariable("hnsw.streaming", "Iterative search mode",
-							 HNSW_ITERATIVE_SCAN_OFF, hnsw_iterative_scan_options, PGC_USERSET, 0, NULL, NULL, NULL);
+							 NULL, &hnsw_iterative_search,
+							 HNSW_ITERATIVE_SEARCH_OFF, hnsw_iterative_search_options, PGC_USERSET, 0, NULL, NULL, NULL);
 
-	/* This is approximate and does not affect the initial scan */
+	/* TODO Change name */
-	DefineCustomIntVariable("hnsw.max_scan_tuples", "Sets the max number of tuples to visit for iterative scans",
+	/* TODO Ensure ivfflat.max_probes uses same value for "all" */
-							NULL, &hnsw_max_scan_tuples,
+	DefineCustomIntVariable("hnsw.ef_stream", "Sets the max number of additional candidates to visit for streaming search",
-							20000, 1, INT_MAX, PGC_USERSET, 0, NULL, NULL, NULL);
+							"-1 means all", &hnsw_max_iterative_tuples,
-
+							HNSW_DEFAULT_EF_STREAM, HNSW_MIN_EF_STREAM, HNSW_MAX_EF_STREAM, PGC_USERSET, 0, NULL, NULL, NULL);
-	/* Same range as hash_mem_multiplier */
-	DefineCustomRealVariable("hnsw.scan_mem_multiplier", "Sets the multiple of work_mem to use for iterative scans",
-							 NULL, &hnsw_scan_mem_multiplier,
-							 1, 1, 1000, PGC_USERSET, 0, NULL, NULL, NULL);
 
 	MarkGUCPrefixReserved("hnsw");
 }
@@ -137,10 +135,6 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
 		*indexSelectivity = 0;
 		*indexCorrelation = 0;
 		*indexPages = 0;
-#if PG_VERSION_NUM >= 180000
-		/* See "On disable_cost" thread on pgsql-hackers */
-		path->path.disabled_nodes = 2;
-#endif
 		return;
 	}
 
@@ -259,11 +253,6 @@ hnswhandler(PG_FUNCTION_ARGS)
 	amroutine->amoptsprocnum = 0;
 	amroutine->amcanorder = false;
 	amroutine->amcanorderbyop = true;
-#if PG_VERSION_NUM >= 180000
-	amroutine->amcanhash = false;
-	amroutine->amconsistentequality = false;
-	amroutine->amconsistentordering = false;
-#endif
 	amroutine->amcanbackward = false;	/* can change direction mid-scan */
 	amroutine->amcanunique = false;
 	amroutine->amcanmulticol = false;
@@ -296,9 +285,6 @@ hnswhandler(PG_FUNCTION_ARGS)
 	amroutine->amvacuumcleanup = hnswvacuumcleanup;
 	amroutine->amcanreturn = NULL;
 	amroutine->amcostestimate = hnswcostestimate;
-#if PG_VERSION_NUM >= 180000
-	amroutine->amgettreeheight = NULL;
-#endif
 	amroutine->amoptions = hnswoptions;
 	amroutine->amproperty = NULL;	/* TODO AMPROP_DISTANCE_ORDERABLE */
 	amroutine->ambuildphasename = hnswbuildphasename;
@@ -319,10 +305,5 @@ hnswhandler(PG_FUNCTION_ARGS)
 	amroutine->aminitparallelscan = NULL;
 	amroutine->amparallelrescan = NULL;
 
-#if PG_VERSION_NUM >= 180000
-	amroutine->amtranslatestrategy = NULL;
-	amroutine->amtranslatecmptype = NULL;
-#endif
-
 	PG_RETURN_POINTER(amroutine);
 }

src/hnsw.h

@@ -42,6 +42,9 @@
 #define HNSW_DEFAULT_EF_SEARCH	40
 #define HNSW_MIN_EF_SEARCH		1
 #define HNSW_MAX_EF_SEARCH		1000
+#define HNSW_DEFAULT_EF_STREAM	-1
+#define HNSW_MIN_EF_STREAM		-1
+#define HNSW_MAX_EF_STREAM		INT_MAX
 
 /* Tuple types */
 #define HNSW_ELEMENT_TUPLE_TYPE  1
@@ -109,24 +112,23 @@
 
 /* Variables */
 extern int	hnsw_ef_search;
-extern int	hnsw_iterative_scan;
+extern int	hnsw_max_iterative_tuples;
-extern int	hnsw_max_scan_tuples;
+extern int	hnsw_iterative_search;
-extern double hnsw_scan_mem_multiplier;
 extern int	hnsw_lock_tranche_id;
 
-typedef enum HnswIterativeScanMode
+typedef enum HnswIterativeSearchType
 {
-	HNSW_ITERATIVE_SCAN_OFF,
+	HNSW_ITERATIVE_SEARCH_OFF,
-	HNSW_ITERATIVE_SCAN_RELAXED,
+	HNSW_ITERATIVE_SEARCH_STRICT,
-	HNSW_ITERATIVE_SCAN_STRICT
+	HNSW_ITERATIVE_SEARCH_RELAXED
-}			HnswIterativeScanMode;
+}			HnswIterativeSearchType;
 
 typedef struct HnswElementData HnswElementData;
 typedef struct HnswNeighborArray HnswNeighborArray;
 
 #define HnswPtrDeclare(type, relptrtype, ptrtype) \
 	relptr_declare(type, relptrtype); \
-	typedef union { type *ptr; relptrtype relptr; } ptrtype
+	typedef union { type *ptr; relptrtype relptr; } ptrtype;
 
 /* Pointers that can be absolute or relative */
 /* Use char for DatumPtr so works with Pointer */
@@ -362,13 +364,6 @@ typedef union
 	ItemPointerData indextid;
 }			HnswUnvisited;
 
-typedef struct HnswReadStreamData
-{
-	HnswUnvisited *unvisited;
-	int			unvisitedLength;
-	int			visited;
-}			HnswReadStreamData;
-
 typedef struct HnswScanOpaqueData
 {
 	const		HnswTypeInfo *typeInfo;
@@ -380,7 +375,6 @@ typedef struct HnswScanOpaqueData
 	int			m;
 	int64		tuples;
 	double		previousDistance;
-	Size		maxMemory;
 	MemoryContext tmpCtx;
 
 	/* Support functions */
@@ -424,13 +418,13 @@ bool		HnswCheckNorm(HnswSupport * support, Datum value);
 Buffer		HnswNewBuffer(Relation index, ForkNumber forkNum);
 void		HnswInitPage(Buffer buf, Page page);
 void		HnswInit(void);
-List	   *HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation index, HnswSupport * support, int m, bool inserting, HnswElement skipElement, visited_hash * v, pairingheap **discarded, bool initVisited, int64 *tuples, bool maintenance);
+List	   *HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation index, HnswSupport * support, int m, bool inserting, HnswElement skipElement, visited_hash * v, pairingheap **discarded, bool initVisited, int64 *tuples);
 HnswElement HnswGetEntryPoint(Relation index);
 void		HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint);
 void	   *HnswAlloc(HnswAllocator * allocator, Size size);
 HnswElement HnswInitElement(char *base, ItemPointer tid, int m, double ml, int maxLevel, HnswAllocator * alloc);
 HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno);
-void		HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint, Relation index, HnswSupport * support, int m, int efConstruction, bool existing, bool maintenance);
+void		HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint, Relation index, HnswSupport * support, int m, int efConstruction, bool existing);
 HnswSearchCandidate *HnswEntryCandidate(char *base, HnswElement em, HnswQuery * q, Relation rel, HnswSupport * support, bool loadVec);
 void		HnswUpdateMetaPage(Relation index, int updateEntry, HnswElement entryPoint, BlockNumber insertPage, ForkNumber forkNum, bool building);
 void		HnswSetNeighborTuple(char *base, HnswNeighborTuple ntup, HnswElement e, int m);

src/hnswbuild.c

@@ -457,7 +457,7 @@ InsertTupleInMemory(HnswBuildState * buildstate, HnswElement element)
 	}
 
 	/* Find neighbors for element */
-	HnswFindElementNeighbors(base, element, entryPoint, NULL, support, m, efConstruction, false, true);
+	HnswFindElementNeighbors(base, element, entryPoint, NULL, support, m, efConstruction, false);
 
 	/* Update graph in memory */
 	UpdateGraphInMemory(support, element, m, efConstruction, entryPoint, buildstate);

src/hnswinsert.c

@@ -725,7 +725,7 @@ HnswInsertTupleOnDisk(Relation index, HnswSupport * support, Datum value, ItemPo
 	}
 
 	/* Find neighbors for element */
-	HnswFindElementNeighbors(base, element, entryPoint, index, support, m, efConstruction, false, building);
+	HnswFindElementNeighbors(base, element, entryPoint, index, support, m, efConstruction, false);
 
 	/* Update graph on disk */
 	UpdateGraphOnDisk(index, support, element, m, efConstruction, entryPoint, building);

src/hnswscan.c

@@ -37,11 +37,11 @@ GetScanItems(IndexScanDesc scan, Datum value)
 
 	for (int lc = entryPoint->level; lc >= 1; lc--)
 	{
-		w = HnswSearchLayer(base, q, ep, 1, lc, index, support, m, false, NULL, NULL, NULL, true, NULL, false);
+		w = HnswSearchLayer(base, q, ep, 1, lc, index, support, m, false, NULL, NULL, NULL, true, NULL);
 		ep = w;
 	}
 
-	return HnswSearchLayer(base, q, ep, hnsw_ef_search, 0, index, support, m, false, NULL, &so->v, hnsw_iterative_scan != HNSW_ITERATIVE_SCAN_OFF ? &so->discarded : NULL, true, &so->tuples, false);
+	return HnswSearchLayer(base, q, ep, hnsw_ef_search, 0, index, support, m, false, NULL, &so->v, hnsw_iterative_search != HNSW_ITERATIVE_SEARCH_OFF ? &so->discarded : NULL, true, &so->tuples);
 }
 
 /*
@@ -72,7 +72,7 @@ ResumeScanItems(IndexScanDesc scan)
 		ep = lappend(ep, sc);
 	}
 
-	return HnswSearchLayer(base, &so->q, ep, batch_size, 0, index, &so->support, so->m, false, NULL, &so->v, &so->discarded, false, &so->tuples, false);
+	return HnswSearchLayer(base, &so->q, ep, batch_size, 0, index, &so->support, so->m, false, NULL, &so->v, &so->discarded, false, &so->tuples);
 }
 
 /*
@@ -102,17 +102,6 @@ GetScanValue(IndexScanDesc scan)
 	return value;
 }
 
-#if defined(HNSW_MEMORY)
-/*
- * Show memory usage
- */
-static void
-ShowMemoryUsage(HnswScanOpaque so)
-{
-	elog(INFO, "memory: %zu KB, tuples: " INT64_FORMAT, MemoryContextMemAllocated(so->tmpCtx, false) / 1024, so->tuples);
-}
-#endif
-
 /*
  * Prepare for an index scan
  */
@@ -121,29 +110,21 @@ hnswbeginscan(Relation index, int nkeys, int norderbys)
 {
 	IndexScanDesc scan;
 	HnswScanOpaque so;
-	double		maxMemory;
 
 	scan = RelationGetIndexScan(index, nkeys, norderbys);
 
 	so = (HnswScanOpaque) palloc(sizeof(HnswScanOpaqueData));
 	so->typeInfo = HnswGetTypeInfo(index);
+	so->first = true;
+	so->v.tids = NULL;
+	so->discarded = NULL;
+	so->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
+									   "Hnsw scan temporary context",
+									   ALLOCSET_DEFAULT_SIZES);
 
 	/* Set support functions */
 	HnswInitSupport(&so->support, index);
 
-	/*
-	 * Use a lower max allocation size than default to allow scanning more
-	 * tuples for iterative search before exceeding work_mem
-	 */
-	so->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
-									   "Hnsw scan temporary context",
-									   0, 8 * 1024, 256 * 1024);
-
-	/* Calculate max memory */
-	/* Add 256 extra bytes to fill last block when close */
-	maxMemory = (double) work_mem * hnsw_scan_mem_multiplier * 1024.0 + 256;
-	so->maxMemory = Min(maxMemory, (double) SIZE_MAX);
-
 	scan->opaque = so;
 
 	return scan;
@@ -157,10 +138,13 @@ hnswrescan(IndexScanDesc scan, ScanKey keys, int nkeys, ScanKey orderbys, int no
 {
 	HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
 
+	if (so->v.tids != NULL)
+		tidhash_reset(so->v.tids);
+
+	if (so->discarded != NULL)
+		pairingheap_reset(so->discarded);
+
 	so->first = true;
-	/* v and discarded are allocated in tmpCtx */
-	so->v.tids = NULL;
-	so->discarded = NULL;
 	so->tuples = 0;
 	so->previousDistance = -get_float8_infinity();
 	MemoryContextReset(so->tmpCtx);
@@ -220,7 +204,7 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 		so->first = false;
 
 #if defined(HNSW_MEMORY)
-		ShowMemoryUsage(so);
+		elog(INFO, "memory: %zu KB", MemoryContextMemAllocated(so->tmpCtx, false) / 1024);
 #endif
 	}
 
@@ -233,15 +217,15 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 
 		if (list_length(so->w) == 0)
 		{
-			if (hnsw_iterative_scan == HNSW_ITERATIVE_SCAN_OFF)
+			if (hnsw_iterative_search == HNSW_ITERATIVE_SEARCH_OFF)
 				break;
 
 			/* Empty index */
 			if (so->discarded == NULL)
 				break;
 
-			/* Reached max number of tuples or memory limit */
+			/* Reached max number of additional tuples */
-			if (so->tuples >= hnsw_max_scan_tuples || MemoryContextMemAllocated(so->tmpCtx, false) > so->maxMemory)
+			if (hnsw_max_iterative_tuples != -1 && so->tuples >= hnsw_ef_search + hnsw_max_iterative_tuples)
 			{
 				if (pairingheap_is_empty(so->discarded))
 					break;
@@ -249,6 +233,21 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 				/* Return remaining tuples */
 				so->w = lappend(so->w, HnswGetSearchCandidate(w_node, pairingheap_remove_first(so->discarded)));
 			}
+			/* Prevent scans from consuming too much memory */
+			else if (MemoryContextMemAllocated(so->tmpCtx, false) > (Size) work_mem * 1024L)
+			{
+				if (pairingheap_is_empty(so->discarded))
+				{
+					ereport(DEBUG1,
+							(errmsg("hnsw index scan exceeded work_mem after " INT64_FORMAT " tuples", so->tuples),
+							 errhint("Increase work_mem to scan more tuples.")));
+
+					break;
+				}
+
+				/* Return remaining tuples */
+				so->w = lappend(so->w, HnswGetSearchCandidate(w_node, pairingheap_remove_first(so->discarded)));
+			}
 			else
 			{
 				/*
@@ -267,7 +266,7 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 				UnlockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
 
 #if defined(HNSW_MEMORY)
-				ShowMemoryUsage(so);
+				elog(INFO, "memory: %zu KB", MemoryContextMemAllocated(so->tmpCtx, false) / 1024);
 #endif
 			}
 
@@ -284,7 +283,7 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 			so->w = list_delete_last(so->w);
 
 			/* Mark memory as free for next iteration */
-			if (hnsw_iterative_scan != HNSW_ITERATIVE_SCAN_OFF)
+			if (hnsw_iterative_search != HNSW_ITERATIVE_SEARCH_OFF)
 			{
 				pfree(element);
 				pfree(sc);
@@ -295,7 +294,7 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
 
 		heaptid = &element->heaptids[--element->heaptidsLength];
 
-		if (hnsw_iterative_scan == HNSW_ITERATIVE_SCAN_STRICT)
+		if (hnsw_iterative_search == HNSW_ITERATIVE_SEARCH_STRICT)
 		{
 			if (sc->distance < so->previousDistance)
 				continue;

src/hnswutils.c

@@ -15,10 +15,6 @@
 #include "utils/memdebug.h"
 #include "utils/rel.h"
 
-#if PG_VERSION_NUM >= 170000
-#include "storage/read_stream.h"
-#endif
-
 #if PG_VERSION_NUM < 170000
 static inline uint64
 murmurhash64(uint64 data)
@@ -529,12 +525,14 @@ HnswGetDistance(Datum a, Datum b, HnswSupport * support)
  * Load an element and optionally get its distance from q
  */
 static void
-HnswLoadElementImpl(Buffer buf, OffsetNumber offno, double *distance, HnswQuery * q, Relation index, HnswSupport * support, bool loadVec, double *maxDistance, HnswElement * element)
+HnswLoadElementImpl(BlockNumber blkno, OffsetNumber offno, double *distance, HnswQuery * q, Relation index, HnswSupport * support, bool loadVec, double *maxDistance, HnswElement * element)
 {
+	Buffer		buf;
 	Page		page;
 	HnswElementTuple etup;
 
 	/* Read vector */
+	buf = ReadBuffer(index, blkno);
 	LockBuffer(buf, BUFFER_LOCK_SHARE);
 	page = BufferGetPage(buf);
 
@@ -555,7 +553,7 @@ HnswLoadElementImpl(Buffer buf, OffsetNumber offno, double *distance, HnswQuery
 	if (distance == NULL || maxDistance == NULL || *distance < *maxDistance)
 	{
 		if (*element == NULL)
-			*element = HnswInitElementFromBlock(BufferGetBlockNumber(buf), offno);
+			*element = HnswInitElementFromBlock(blkno, offno);
 
 		HnswLoadElementFromTuple(*element, etup, true, loadVec);
 	}
@@ -569,9 +567,7 @@ HnswLoadElementImpl(Buffer buf, OffsetNumber offno, double *distance, HnswQuery
 void
 HnswLoadElement(HnswElement element, double *distance, HnswQuery * q, Relation index, HnswSupport * support, bool loadVec, double *maxDistance)
 {
-	Buffer		buf = ReadBuffer(index, element->blkno);
+	HnswLoadElementImpl(element->blkno, element->offno, distance, q, index, support, loadVec, maxDistance, &element);
-
-	HnswLoadElementImpl(buf, element->offno, distance, q, index, support, loadVec, maxDistance, &element);
 }
 
 /*
@@ -585,34 +581,21 @@ GetElementDistance(char *base, HnswElement element, HnswQuery * q, HnswSupport *
 	return HnswGetDistance(q->value, value, support);
 }
 
-/*
- * Allocate a search candidate
- */
-static HnswSearchCandidate *
-HnswInitSearchCandidate(char *base, HnswElement element, double distance)
-{
-	HnswSearchCandidate *sc = palloc(sizeof(HnswSearchCandidate));
-
-	HnswPtrStore(base, sc->element, element);
-	sc->distance = distance;
-	return sc;
-}
-
 /*
  * Create a candidate for the entry point
  */
 HnswSearchCandidate *
 HnswEntryCandidate(char *base, HnswElement entryPoint, HnswQuery * q, Relation index, HnswSupport * support, bool loadVec)
 {
+	HnswSearchCandidate *sc = palloc(sizeof(HnswSearchCandidate));
 	bool		inMemory = index == NULL;
-	double		distance;
 
+	HnswPtrStore(base, sc->element, entryPoint);
 	if (inMemory)
-		distance = GetElementDistance(base, entryPoint, q, support);
+		sc->distance = GetElementDistance(base, entryPoint, q, support);
 	else
-		HnswLoadElement(entryPoint, &distance, q, index, support, loadVec, NULL);
+		HnswLoadElement(entryPoint, &sc->distance, q, index, support, loadVec, NULL);
-
+	return sc;
-	return HnswInitSearchCandidate(base, entryPoint, distance);
 }
 
 /*
@@ -811,31 +794,11 @@ HnswLoadUnvisitedFromDisk(HnswElement element, HnswUnvisited * unvisited, int *u
 	}
 }
 
-#if PG_VERSION_NUM >= 170000
-/*
- * Get next block number for read stream
- */
-static BlockNumber
-HnswReadStreamNextBlock(ReadStream *stream, void *callback_private_data, void *per_buffer_data)
-{
-	HnswReadStreamData *streamData = callback_private_data;
-	OffsetNumber *offno = per_buffer_data;
-	HnswUnvisited *uv;
-
-	if (streamData->visited == streamData->unvisitedLength)
-		return InvalidBlockNumber;
-
-	uv = &streamData->unvisited[streamData->visited++];
-	*offno = ItemPointerGetOffsetNumber(&uv->indextid);
-	return ItemPointerGetBlockNumber(&uv->indextid);
-}
-#endif
-
 /*
  * Algorithm 2 from paper
  */
 List *
-HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation index, HnswSupport * support, int m, bool inserting, HnswElement skipElement, visited_hash * v, pairingheap **discarded, bool initVisited, int64 *tuples, bool maintenance)
+HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation index, HnswSupport * support, int m, bool inserting, HnswElement skipElement, visited_hash * v, pairingheap **discarded, bool initVisited, int64 *tuples)
 {
 	List	   *w = NIL;
 	pairingheap *C = pairingheap_allocate(CompareNearestCandidates, NULL);
@@ -850,18 +813,6 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
 	int			unvisitedLength;
 	bool		inMemory = index == NULL;
 
-#if PG_VERSION_NUM >= 170000
-	HnswReadStreamData streamData;
-	ReadStream *stream = NULL;
-
-	if (!inMemory)
-	{
-		int			flags = maintenance ? READ_STREAM_MAINTENANCE : READ_STREAM_DEFAULT;
-
-		stream = read_stream_begin_relation(flags, NULL, index, MAIN_FORKNUM, HnswReadStreamNextBlock, &streamData, sizeof(OffsetNumber));
-	}
-#endif
-
 	if (v == NULL)
 	{
 		v = &vh;
@@ -893,7 +844,6 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
 		{
 			AddToVisited(base, v, sc->element, inMemory, &found);
 
-			/* OK to count elements instead of tuples */
 			if (tuples != NULL)
 				(*tuples)++;
 		}
@@ -924,19 +874,8 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
 		if (inMemory)
 			HnswLoadUnvisitedFromMemory(base, cElement, unvisited, &unvisitedLength, v, lc, localNeighborhood, neighborhoodSize);
 		else
-		{
 			HnswLoadUnvisitedFromDisk(cElement, unvisited, &unvisitedLength, v, index, m, lm, lc);
 
-#if PG_VERSION_NUM >= 170000
-			read_stream_reset(stream);
-
-			streamData.unvisited = unvisited;
-			streamData.unvisitedLength = unvisitedLength;
-			streamData.visited = 0;
-#endif
-		}
-
-		/* OK to count elements instead of tuples */
 		if (tuples != NULL)
 			(*tuples) += unvisitedLength;
 
@@ -956,24 +895,13 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
 			}
 			else
 			{
-				Buffer		buf;
-				OffsetNumber offno;
-
-#if PG_VERSION_NUM >= 170000
-				void	   *offnoPtr;
-
-				buf = read_stream_next_buffer(stream, &offnoPtr);
-				offno = *((OffsetNumber *) offnoPtr);
-#else
 				ItemPointer indextid = &unvisited[i].indextid;
-
+				BlockNumber blkno = ItemPointerGetBlockNumber(indextid);
-				buf = ReadBuffer(index, ItemPointerGetBlockNumber(indextid));
+				OffsetNumber offno = ItemPointerGetOffsetNumber(indextid);
-				offno = ItemPointerGetOffsetNumber(indextid);
-#endif
 
 				/* Avoid any allocations if not adding */
 				eElement = NULL;
-				HnswLoadElementImpl(buf, offno, &eDistance, q, index, support, inserting, alwaysAdd || discarded != NULL ? NULL : &f->distance, &eElement);
+				HnswLoadElementImpl(blkno, offno, &eDistance, q, index, support, inserting, alwaysAdd || discarded != NULL ? NULL : &f->distance, &eElement);
 
 				if (eElement == NULL)
 					continue;
@@ -984,7 +912,9 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
 				if (discarded != NULL)
 				{
 					/* Create a new candidate */
-					e = HnswInitSearchCandidate(base, eElement, eDistance);
+					e = palloc(sizeof(HnswSearchCandidate));
+					HnswPtrStore(base, e->element, eElement);
+					e->distance = eDistance;
 					pairingheap_add(*discarded, &e->w_node);
 				}
 
@@ -996,7 +926,9 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
 				continue;
 
 			/* Create a new candidate */
-			e = HnswInitSearchCandidate(base, eElement, eDistance);
+			e = palloc(sizeof(HnswSearchCandidate));
+			HnswPtrStore(base, e->element, eElement);
+			e->distance = eDistance;
 			pairingheap_add(C, &e->c_node);
 			pairingheap_add(W, &e->w_node);
 
@@ -1029,11 +961,6 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
 		w = lappend(w, sc);
 	}
 
-#if PG_VERSION_NUM >= 170000
-	if (!inMemory)
-		read_stream_end(stream);
-#endif
-
 	return w;
 }
 
@@ -1329,7 +1256,7 @@ PrecomputeHash(char *base, HnswElement element)
  * Algorithm 1 from paper
  */
 void
-HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint, Relation index, HnswSupport * support, int m, int efConstruction, bool existing, bool maintenance)
+HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint, Relation index, HnswSupport * support, int m, int efConstruction, bool existing)
 {
 	List	   *ep;
 	List	   *w;
@@ -1356,7 +1283,7 @@ HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint
 	/* 1st phase: greedy search to insert level */
 	for (int lc = entryLevel; lc >= level + 1; lc--)
 	{
-		w = HnswSearchLayer(base, &q, ep, 1, lc, index, support, m, true, skipElement, NULL, NULL, true, NULL, maintenance);
+		w = HnswSearchLayer(base, &q, ep, 1, lc, index, support, m, true, skipElement, NULL, NULL, true, NULL);
 		ep = w;
 	}
 
@@ -1375,7 +1302,7 @@ HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint
 		List	   *lw = NIL;
 		ListCell   *lc2;
 
-		w = HnswSearchLayer(base, &q, ep, efConstruction, lc, index, support, m, true, skipElement, NULL, NULL, true, NULL, maintenance);
+		w = HnswSearchLayer(base, &q, ep, efConstruction, lc, index, support, m, true, skipElement, NULL, NULL, true, NULL);
 
 		/* Convert search candidates to candidates */
 		foreach(lc2, w)
@@ -1455,7 +1382,7 @@ hnsw_halfvec_support(PG_FUNCTION_ARGS)
 	};
 
 	PG_RETURN_POINTER(&typeInfo);
-}
+};
 
 FUNCTION_PREFIX PG_FUNCTION_INFO_V1(hnsw_bit_support);
 Datum
@@ -1468,7 +1395,7 @@ hnsw_bit_support(PG_FUNCTION_ARGS)
 	};
 
 	PG_RETURN_POINTER(&typeInfo);
-}
+};
 
 FUNCTION_PREFIX PG_FUNCTION_INFO_V1(hnsw_sparsevec_support);
 Datum
@@ -1481,4 +1408,4 @@ hnsw_sparsevec_support(PG_FUNCTION_ARGS)
 	};
 
 	PG_RETURN_POINTER(&typeInfo);
-}
+};

src/hnswvacuum.c

@@ -9,10 +9,6 @@
 #include "storage/lmgr.h"
 #include "utils/memutils.h"
 
-#if PG_VERSION_NUM >= 180000
-#define vacuum_delay_point() vacuum_delay_point(false)
-#endif
-
 /*
  * Check if deleted list contains an index TID
  */
@@ -208,7 +204,7 @@ RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswEleme
 	element->heaptidsLength = 0;
 
 	/* Find neighbors for element, skipping itself */
-	HnswFindElementNeighbors(base, element, entryPoint, index, support, m, efConstruction, true, true);
+	HnswFindElementNeighbors(base, element, entryPoint, index, support, m, efConstruction, true);
 
 	/* Zero memory for each element */
 	MemSet(ntup, 0, HNSW_TUPLE_ALLOC_SIZE);

src/ivfbuild.c

@@ -228,11 +228,11 @@ BuildCallback(Relation index, ItemPointer tid, Datum *values,
 static inline void
 GetNextTuple(Tuplesortstate *sortstate, TupleDesc tupdesc, TupleTableSlot *slot, IndexTuple *itup, int *list)
 {
+	Datum		value;
+	bool		isnull;
+
 	if (tuplesort_gettupleslot(sortstate, true, false, slot, NULL))
 	{
-		Datum		value;
-		bool		isnull;
-
 		*list = DatumGetInt32(slot_getattr(slot, 1, &isnull));
 		value = slot_getattr(slot, 3, &isnull);
 
@@ -254,8 +254,8 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
 	IndexTuple	itup = NULL;	/* silence compiler warning */
 	int64		inserted = 0;
 
-	TupleTableSlot *slot = MakeSingleTupleTableSlot(buildstate->sortdesc, &TTSOpsMinimalTuple);
+	TupleTableSlot *slot = MakeSingleTupleTableSlot(buildstate->tupdesc, &TTSOpsMinimalTuple);
-	TupleDesc	tupdesc = buildstate->tupdesc;
+	TupleDesc	tupdesc = RelationGetDescr(index);
 
 	pgstat_progress_update_param(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_LOAD);
 
@@ -319,7 +319,6 @@ InitBuildState(IvfflatBuildState * buildstate, Relation heap, Relation index, In
 	buildstate->index = index;
 	buildstate->indexInfo = indexInfo;
 	buildstate->typeInfo = IvfflatGetTypeInfo(index);
-	buildstate->tupdesc = RelationGetDescr(index);
 
 	buildstate->lists = IvfflatGetLists(index);
 	buildstate->dimensions = TupleDescAttr(index->rd_att, 0)->atttypmod;
@@ -357,12 +356,12 @@ InitBuildState(IvfflatBuildState * buildstate, Relation heap, Relation index, In
 				 errmsg("dimensions must be greater than one for this opclass")));
 
 	/* Create tuple description for sorting */
-	buildstate->sortdesc = CreateTemplateTupleDesc(3);
+	buildstate->tupdesc = CreateTemplateTupleDesc(3);
-	TupleDescInitEntry(buildstate->sortdesc, (AttrNumber) 1, "list", INT4OID, -1, 0);
+	TupleDescInitEntry(buildstate->tupdesc, (AttrNumber) 1, "list", INT4OID, -1, 0);
-	TupleDescInitEntry(buildstate->sortdesc, (AttrNumber) 2, "tid", TIDOID, -1, 0);
+	TupleDescInitEntry(buildstate->tupdesc, (AttrNumber) 2, "tid", TIDOID, -1, 0);
-	TupleDescInitEntry(buildstate->sortdesc, (AttrNumber) 3, "vector", TupleDescAttr(buildstate->tupdesc, 0)->atttypid, -1, 0);
+	TupleDescInitEntry(buildstate->tupdesc, (AttrNumber) 3, "vector", RelationGetDescr(index)->attrs[0].atttypid, -1, 0);
 
-	buildstate->slot = MakeSingleTupleTableSlot(buildstate->sortdesc, &TTSOpsVirtual);
+	buildstate->slot = MakeSingleTupleTableSlot(buildstate->tupdesc, &TTSOpsVirtual);
 
 	buildstate->centers = VectorArrayInit(buildstate->lists, buildstate->dimensions, buildstate->typeInfo->itemSize(buildstate->dimensions));
 	buildstate->listInfo = palloc(sizeof(ListInfo) * buildstate->lists);
@@ -634,7 +633,7 @@ IvfflatParallelScanAndSort(IvfflatSpool * ivfspool, IvfflatShared * ivfshared, S
 	InitBuildState(&buildstate, ivfspool->heap, ivfspool->index, indexInfo);
 	memcpy(buildstate.centers->items, ivfcenters, buildstate.centers->itemsize * buildstate.centers->maxlen);
 	buildstate.centers->length = buildstate.centers->maxlen;
-	ivfspool->sortstate = InitBuildSortState(buildstate.sortdesc, sortmem, coordinate);
+	ivfspool->sortstate = InitBuildSortState(buildstate.tupdesc, sortmem, coordinate);
 	buildstate.sortstate = ivfspool->sortstate;
 	scan = table_beginscan_parallel(ivfspool->heap,
 									ParallelTableScanFromIvfflatShared(ivfshared));
@@ -951,7 +950,7 @@ AssignTuples(IvfflatBuildState * buildstate)
 	}
 
 	/* Begin serial/leader tuplesort */
-	buildstate->sortstate = InitBuildSortState(buildstate->sortdesc, maintenance_work_mem, coordinate);
+	buildstate->sortstate = InitBuildSortState(buildstate->tupdesc, maintenance_work_mem, coordinate);
 
 	/* Add tuples to sort */
 	if (buildstate->heap != NULL)
@@ -1023,10 +1022,6 @@ ivfflatbuild(Relation heap, Relation index, IndexInfo *indexInfo)
 	IndexBuildResult *result;
 	IvfflatBuildState buildstate;
 
-#ifdef IVFFLAT_BENCH
-	SeedRandom(42);
-#endif
-
 	BuildIndex(heap, index, indexInfo, &buildstate, MAIN_FORKNUM);
 
 	result = (IndexBuildResult *) palloc(sizeof(IndexBuildResult));

src/ivfflat.c

@@ -17,16 +17,8 @@
 #endif
 
 int			ivfflat_probes;
-int			ivfflat_iterative_scan;
-int			ivfflat_max_probes;
 static relopt_kind ivfflat_relopt_kind;
 
-static const struct config_enum_entry ivfflat_iterative_scan_options[] = {
-	{"off", IVFFLAT_ITERATIVE_SCAN_OFF, false},
-	{"relaxed_order", IVFFLAT_ITERATIVE_SCAN_RELAXED, false},
-	{NULL, 0, false}
-};
-
 /*
  * Initialize index options and variables
  */
@@ -41,15 +33,6 @@ IvfflatInit(void)
 							"Valid range is 1..lists.", &ivfflat_probes,
 							IVFFLAT_DEFAULT_PROBES, IVFFLAT_MIN_LISTS, IVFFLAT_MAX_LISTS, PGC_USERSET, 0, NULL, NULL, NULL);
 
-	DefineCustomEnumVariable("ivfflat.iterative_scan", "Sets the mode for iterative scans",
-							 NULL, &ivfflat_iterative_scan,
-							 IVFFLAT_ITERATIVE_SCAN_OFF, ivfflat_iterative_scan_options, PGC_USERSET, 0, NULL, NULL, NULL);
-
-	/* If this is less than probes, probes is used */
-	DefineCustomIntVariable("ivfflat.max_probes", "Sets the max number of probes for iterative scans",
-							NULL, &ivfflat_max_probes,
-							IVFFLAT_MAX_LISTS, IVFFLAT_MIN_LISTS, IVFFLAT_MAX_LISTS, PGC_USERSET, 0, NULL, NULL, NULL);
-
 	MarkGUCPrefixReserved("ivfflat");
 }
 
@@ -99,10 +82,6 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
 		*indexSelectivity = 0;
 		*indexCorrelation = 0;
 		*indexPages = 0;
-#if PG_VERSION_NUM >= 180000
-		/* See "On disable_cost" thread on pgsql-hackers */
-		path->path.disabled_nodes = 2;
-#endif
 		return;
 	}
 
@@ -186,11 +165,6 @@ ivfflathandler(PG_FUNCTION_ARGS)
 	amroutine->amoptsprocnum = 0;
 	amroutine->amcanorder = false;
 	amroutine->amcanorderbyop = true;
-#if PG_VERSION_NUM >= 180000
-	amroutine->amcanhash = false;
-	amroutine->amconsistentequality = false;
-	amroutine->amconsistentordering = false;
-#endif
 	amroutine->amcanbackward = false;	/* can change direction mid-scan */
 	amroutine->amcanunique = false;
 	amroutine->amcanmulticol = false;
@@ -223,9 +197,6 @@ ivfflathandler(PG_FUNCTION_ARGS)
 	amroutine->amvacuumcleanup = ivfflatvacuumcleanup;
 	amroutine->amcanreturn = NULL;	/* tuple not included in heapsort */
 	amroutine->amcostestimate = ivfflatcostestimate;
-#if PG_VERSION_NUM >= 180000
-	amroutine->amgettreeheight = NULL;
-#endif
 	amroutine->amoptions = ivfflatoptions;
 	amroutine->amproperty = NULL;	/* TODO AMPROP_DISTANCE_ORDERABLE */
 	amroutine->ambuildphasename = ivfflatbuildphasename;
@@ -246,10 +217,5 @@ ivfflathandler(PG_FUNCTION_ARGS)
 	amroutine->aminitparallelscan = NULL;
 	amroutine->amparallelrescan = NULL;
 
-#if PG_VERSION_NUM >= 180000
-	amroutine->amtranslatestrategy = NULL;
-	amroutine->amtranslatecmptype = NULL;
-#endif
-
 	PG_RETURN_POINTER(amroutine);
 }

src/ivfflat.h

@@ -73,23 +73,13 @@
 #if PG_VERSION_NUM >= 150000
 #define RandomDouble() pg_prng_double(&pg_global_prng_state)
 #define RandomInt() pg_prng_uint32(&pg_global_prng_state)
-#define SeedRandom(seed) pg_prng_seed(&pg_global_prng_state, seed)
 #else
 #define RandomDouble() (((double) random()) / MAX_RANDOM_VALUE)
 #define RandomInt() random()
-#define SeedRandom(seed) srandom(seed)
 #endif
 
 /* Variables */
 extern int	ivfflat_probes;
-extern int	ivfflat_iterative_scan;
-extern int	ivfflat_max_probes;
-
-typedef enum IvfflatIterativeScanMode
-{
-	IVFFLAT_ITERATIVE_SCAN_OFF,
-	IVFFLAT_ITERATIVE_SCAN_RELAXED
-}			IvfflatIterativeScanMode;
 
 typedef struct VectorArrayData
 {
@@ -175,7 +165,6 @@ typedef struct IvfflatBuildState
 	Relation	index;
 	IndexInfo  *indexInfo;
 	const		IvfflatTypeInfo *typeInfo;
-	TupleDesc	tupdesc;
 
 	/* Settings */
 	int			dimensions;
@@ -209,7 +198,7 @@ typedef struct IvfflatBuildState
 
 	/* Sorting */
 	Tuplesortstate *sortstate;
-	TupleDesc	sortdesc;
+	TupleDesc	tupdesc;
 	TupleTableSlot *slot;
 
 	/* Memory */
@@ -258,11 +247,8 @@ typedef struct IvfflatScanOpaqueData
 {
 	const		IvfflatTypeInfo *typeInfo;
 	int			probes;
-	int			maxProbes;
 	int			dimensions;
 	bool		first;
-	Datum		value;
-	MemoryContext tmpCtx;
 
 	/* Sorting */
 	Tuplesortstate *sortstate;
@@ -279,9 +265,7 @@ typedef struct IvfflatScanOpaqueData
 
 	/* Lists */
 	pairingheap *listQueue;
-	BlockNumber *listPages;
+	IvfflatScanList lists[FLEXIBLE_ARRAY_MEMBER];	/* must come last */
-	int			listIndex;
-	IvfflatScanList *lists;
 }			IvfflatScanOpaqueData;
 
 typedef IvfflatScanOpaqueData * IvfflatScanOpaque;

src/ivfinsert.c

@@ -98,7 +98,7 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
 	IvfflatGetMetaPageInfo(index, NULL, NULL);
 
 	/* Find the insert page - sets the page and list info */
-	FindInsertPage(index, &value, &insertPage, &listInfo);
+	FindInsertPage(index, values, &insertPage, &listInfo);
 	Assert(BlockNumberIsValid(insertPage));
 	originalInsertPage = insertPage;
 

src/ivfscan.c

@@ -10,7 +10,10 @@
 #include "miscadmin.h"
 #include "pgstat.h"
 #include "storage/bufmgr.h"
+
+#ifdef IVFFLAT_MEMORY
 #include "utils/memutils.h"
+#endif
 
 #define GetScanList(ptr) pairingheap_container(IvfflatScanList, ph_node, ptr)
 #define GetScanListConst(ptr) pairingheap_const_container(IvfflatScanList, ph_node, ptr)
@@ -62,7 +65,7 @@ GetScanLists(IndexScanDesc scan, Datum value)
 			/* Use procinfo from the index instead of scan key for performance */
 			distance = DatumGetFloat8(so->distfunc(so->procinfo, so->collation, PointerGetDatum(&list->center), value));
 
-			if (listCount < so->maxProbes)
+			if (listCount < so->probes)
 			{
 				IvfflatScanList *scanlist;
 
@@ -75,7 +78,7 @@ GetScanLists(IndexScanDesc scan, Datum value)
 				pairingheap_add(so->listQueue, &scanlist->ph_node);
 
 				/* Calculate max distance */
-				if (listCount == so->maxProbes)
+				if (listCount == so->probes)
 					maxDistance = GetScanList(pairingheap_first(so->listQueue))->distance;
 			}
 			else if (distance < maxDistance)
@@ -99,11 +102,6 @@ GetScanLists(IndexScanDesc scan, Datum value)
 
 		UnlockReleaseBuffer(cbuf);
 	}
-
-	for (int i = listCount - 1; i >= 0; i--)
-		so->listPages[i] = GetScanList(pairingheap_remove_first(so->listQueue))->startPage;
-
-	Assert(pairingheap_is_empty(so->listQueue));
 }
 
 /*
@@ -114,15 +112,13 @@ GetScanItems(IndexScanDesc scan, Datum value)
 {
 	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
 	TupleDesc	tupdesc = RelationGetDescr(scan->indexRelation);
+	double		tuples = 0;
 	TupleTableSlot *slot = so->vslot;
-	int			batchProbes = 0;
-
-	tuplesort_reset(so->sortstate);
 
 	/* Search closest probes lists */
-	while (so->listIndex < so->maxProbes && (++batchProbes) <= so->probes)
+	while (!pairingheap_is_empty(so->listQueue))
 	{
-		BlockNumber searchPage = so->listPages[so->listIndex++];
+		BlockNumber searchPage = GetScanList(pairingheap_remove_first(so->listQueue))->startPage;
 
 		/* Search all entry pages for list */
 		while (BlockNumberIsValid(searchPage))
@@ -160,6 +156,8 @@ GetScanItems(IndexScanDesc scan, Datum value)
 				ExecStoreVirtualTuple(slot);
 
 				tuplesort_puttupleslot(so->sortstate, slot);
+
+				tuples++;
 			}
 
 			searchPage = IvfflatPageGetOpaque(page)->nextblkno;
@@ -168,11 +166,13 @@ GetScanItems(IndexScanDesc scan, Datum value)
 		}
 	}
 
-	tuplesort_performsort(so->sortstate);
+	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.")));
 
-#if defined(IVFFLAT_MEMORY)
+	tuplesort_performsort(so->sortstate);
-	elog(INFO, "memory: %zu MB", MemoryContextMemAllocated(CurrentMemoryContext, true) / (1024 * 1024));
-#endif
 }
 
 /*
@@ -209,13 +209,7 @@ GetScanValue(IndexScanDesc scan)
 
 		/* Normalize if needed */
 		if (so->normprocinfo != NULL)
-		{
-			MemoryContext oldCtx = MemoryContextSwitchTo(so->tmpCtx);
-
 			value = IvfflatNormValue(so->typeInfo, so->collation, value);
-
-			MemoryContextSwitchTo(oldCtx);
-		}
 	}
 
 	return value;
@@ -246,30 +240,19 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 	int			lists;
 	int			dimensions;
 	int			probes = ivfflat_probes;
-	int			maxProbes;
-	MemoryContext oldCtx;
 
 	scan = RelationGetIndexScan(index, nkeys, norderbys);
 
 	/* Get lists and dimensions from metapage */
 	IvfflatGetMetaPageInfo(index, &lists, &dimensions);
 
-	if (ivfflat_iterative_scan != IVFFLAT_ITERATIVE_SCAN_OFF)
-		maxProbes = Max(ivfflat_max_probes, probes);
-	else
-		maxProbes = probes;
-
 	if (probes > lists)
 		probes = lists;
 
-	if (maxProbes > lists)
+	so = (IvfflatScanOpaque) palloc(offsetof(IvfflatScanOpaqueData, lists) + probes * sizeof(IvfflatScanList));
-		maxProbes = lists;
-
-	so = (IvfflatScanOpaque) palloc(sizeof(IvfflatScanOpaqueData));
 	so->typeInfo = IvfflatGetTypeInfo(index);
 	so->first = true;
 	so->probes = probes;
-	so->maxProbes = maxProbes;
 	so->dimensions = dimensions;
 
 	/* Set support functions */
@@ -277,12 +260,6 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 	so->normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC);
 	so->collation = index->rd_indcollation[0];
 
-	so->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
-									   "Ivfflat scan temporary context",
-									   ALLOCSET_DEFAULT_SIZES);
-
-	oldCtx = MemoryContextSwitchTo(so->tmpCtx);
-
 	/* Create tuple description for sorting */
 	so->tupdesc = CreateTemplateTupleDesc(2);
 	TupleDescInitEntry(so->tupdesc, (AttrNumber) 1, "distance", FLOAT8OID, -1, 0);
@@ -303,11 +280,6 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
 	so->bas = GetAccessStrategy(BAS_BULKREAD);
 
 	so->listQueue = pairingheap_allocate(CompareLists, scan);
-	so->listPages = palloc(maxProbes * sizeof(BlockNumber));
-	so->listIndex = 0;
-	so->lists = palloc(maxProbes * sizeof(IvfflatScanList));
-
-	MemoryContextSwitchTo(oldCtx);
 
 	scan->opaque = so;
 
@@ -322,9 +294,11 @@ ivfflatrescan(IndexScanDesc scan, ScanKey keys, int nkeys, ScanKey orderbys, int
 {
 	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
 
+	if (!so->first)
+		tuplesort_reset(so->sortstate);
+
 	so->first = true;
 	pairingheap_reset(so->listQueue);
-	so->listIndex = 0;
 
 	if (keys && scan->numberOfKeys > 0)
 		memmove(scan->keyData, keys, scan->numberOfKeys * sizeof(ScanKeyData));
@@ -340,8 +314,6 @@ bool
 ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
 {
 	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
-	ItemPointer heaptid;
-	bool		isnull;
 
 	/*
 	 * Index can be used to scan backward, but Postgres doesn't support
@@ -369,23 +341,28 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
 		IvfflatBench("GetScanLists", GetScanLists(scan, value));
 		IvfflatBench("GetScanItems", GetScanItems(scan, value));
 		so->first = false;
-		so->value = value;
+
+#if defined(IVFFLAT_MEMORY)
+		elog(INFO, "memory: %zu MB", MemoryContextMemAllocated(CurrentMemoryContext, true) / (1024 * 1024));
+#endif
+
+		/* Clean up if we allocated a new value */
+		if (value != scan->orderByData->sk_argument)
+			pfree(DatumGetPointer(value));
 	}
 
-	while (!tuplesort_gettupleslot(so->sortstate, true, false, so->mslot, NULL))
+	if (tuplesort_gettupleslot(so->sortstate, true, false, so->mslot, NULL))
 	{
-		if (so->listIndex == so->maxProbes)
+		bool		isnull;
-			return false;
+		ItemPointer heaptid = (ItemPointer) DatumGetPointer(slot_getattr(so->mslot, 2, &isnull));
 
-		IvfflatBench("GetScanItems", GetScanItems(scan, so->value));
+		scan->xs_heaptid = *heaptid;
+		scan->xs_recheck = false;
+		scan->xs_recheckorderby = false;
+		return true;
 	}
 
-	heaptid = (ItemPointer) DatumGetPointer(slot_getattr(so->mslot, 2, &isnull));
+	return false;
-
-	scan->xs_heaptid = *heaptid;
-	scan->xs_recheck = false;
-	scan->xs_recheckorderby = false;
-	return true;
 }
 
 /*
@@ -396,10 +373,12 @@ ivfflatendscan(IndexScanDesc scan)
 {
 	IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
 
-	/* Free any temporary files */
+	pairingheap_free(so->listQueue);
 	tuplesort_end(so->sortstate);
+	FreeAccessStrategy(so->bas);
+	FreeTupleDesc(so->tupdesc);
 
-	MemoryContextDelete(so->tmpCtx);
+	/* TODO Free vslot and mslot without freeing TupleDesc */
 
 	pfree(so);
 	scan->opaque = NULL;

src/ivfutils.c

@@ -259,8 +259,8 @@ VectorUpdateCenter(Pointer v, int dimensions, float *x)
 	SET_VARSIZE(vec, VECTOR_SIZE(dimensions));
 	vec->dim = dimensions;
 
-	for (int i = 0; i < dimensions; i++)
+	for (int k = 0; k < dimensions; k++)
-		vec->x[i] = x[i];
+		vec->x[k] = x[k];
 }
 
 static void
@@ -271,8 +271,8 @@ HalfvecUpdateCenter(Pointer v, int dimensions, float *x)
 	SET_VARSIZE(vec, HALFVEC_SIZE(dimensions));
 	vec->dim = dimensions;
 
-	for (int i = 0; i < dimensions; i++)
+	for (int k = 0; k < dimensions; k++)
-		vec->x[i] = Float4ToHalfUnchecked(x[i]);
+		vec->x[k] = Float4ToHalfUnchecked(x[k]);
 }
 
 static void
@@ -284,33 +284,29 @@ BitUpdateCenter(Pointer v, int dimensions, float *x)
 	SET_VARSIZE(vec, VARBITTOTALLEN(dimensions));
 	VARBITLEN(vec) = dimensions;
 
-	for (uint32 i = 0; i < VARBITBYTES(vec); i++)
+	for (uint32 k = 0; k < VARBITBYTES(vec); k++)
-		nx[i] = 0;
+		nx[k] = 0;
 
-	for (int i = 0; i < dimensions; i++)
+	for (int k = 0; k < dimensions; k++)
-		nx[i / 8] |= (x[i] > 0.5 ? 1 : 0) << (7 - (i % 8));
+		nx[k / 8] |= (x[k] > 0.5 ? 1 : 0) << (7 - (k % 8));
 }
 
 static void
 VectorSumCenter(Pointer v, float *x)
 {
 	Vector	   *vec = (Vector *) v;
-	int			dim = vec->dim;
 
-	/* Auto-vectorized */
+	for (int k = 0; k < vec->dim; k++)
-	for (int i = 0; i < dim; i++)
+		x[k] += vec->x[k];
-		x[i] += vec->x[i];
 }
 
 static void
 HalfvecSumCenter(Pointer v, float *x)
 {
 	HalfVector *vec = (HalfVector *) v;
-	int			dim = vec->dim;
 
-	/* Auto-vectorized on aarch64 */
+	for (int k = 0; k < vec->dim; k++)
-	for (int i = 0; i < dim; i++)
+		x[k] += HalfToFloat4(vec->x[k]);
-		x[i] += HalfToFloat4(vec->x[i]);
 }
 
 static void
@@ -318,8 +314,8 @@ BitSumCenter(Pointer v, float *x)
 {
 	VarBit	   *vec = (VarBit *) v;
 
-	for (int i = 0; i < VARBITLEN(vec); i++)
+	for (int k = 0; k < VARBITLEN(vec); k++)
-		x[i] += (float) (((VARBITS(vec)[i / 8]) >> (7 - (i % 8))) & 0x01);
+		x[k] += (float) (((VARBITS(vec)[k / 8]) >> (7 - (k % 8))) & 0x01);
 }
 
 /*
@@ -359,7 +355,7 @@ ivfflat_halfvec_support(PG_FUNCTION_ARGS)
 	};
 
 	PG_RETURN_POINTER(&typeInfo);
-}
+};
 
 FUNCTION_PREFIX PG_FUNCTION_INFO_V1(ivfflat_bit_support);
 Datum
@@ -374,4 +370,4 @@ ivfflat_bit_support(PG_FUNCTION_ARGS)
 	};
 
 	PG_RETURN_POINTER(&typeInfo);
-}
+};

src/ivfvacuum.c

@@ -5,10 +5,6 @@
 #include "ivfflat.h"
 #include "storage/bufmgr.h"
 
-#if PG_VERSION_NUM >= 180000
-#define vacuum_delay_point() vacuum_delay_point(false)
-#endif
-
 /*
  * Bulk delete tuples from the index
  */
@@ -30,7 +26,7 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
 		Page		cpage;
 		OffsetNumber coffno;
 		OffsetNumber cmaxoffno;
-		BlockNumber listPages[MaxOffsetNumber];
+		BlockNumber startPages[MaxOffsetNumber];
 		ListInfo	listInfo;
 
 		cbuf = ReadBuffer(index, blkno);
@@ -44,7 +40,7 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
 		{
 			IvfflatList list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, coffno));
 
-			listPages[coffno - FirstOffsetNumber] = list->startPage;
+			startPages[coffno - FirstOffsetNumber] = list->startPage;
 		}
 
 		listInfo.blkno = blkno;
@@ -54,7 +50,7 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
 
 		for (coffno = FirstOffsetNumber; coffno <= cmaxoffno; coffno = OffsetNumberNext(coffno))
 		{
-			BlockNumber searchPage = listPages[coffno - FirstOffsetNumber];
+			BlockNumber searchPage = startPages[coffno - FirstOffsetNumber];
 			BlockNumber insertPage = InvalidBlockNumber;
 
 			/* Iterate over entry pages */

src/sparsevec.c

@@ -4,7 +4,6 @@
 #include <math.h>
 
 #include "catalog/pg_type.h"
-#include "common/shortest_dec.h"
 #include "common/string.h"
 #include "fmgr.h"
 #include "halfutils.h"
@@ -13,10 +12,17 @@
 #include "sparsevec.h"
 #include "utils/array.h"
 #include "utils/builtins.h"
-#include "utils/float.h"
 #include "utils/lsyscache.h"
 #include "vector.h"
 
+#if PG_VERSION_NUM >= 120000
+#include "common/shortest_dec.h"
+#include "utils/float.h"
+#else
+#include <float.h>
+#include "utils/builtins.h"
+#endif
+
 typedef struct SparseInputElement
 {
 	int32		index;

src/vector.c

@@ -916,19 +916,15 @@ vector_concat(PG_FUNCTION_ARGS)
 	Vector	   *b = PG_GETARG_VECTOR_P(1);
 	Vector	   *result;
 	int			dim = a->dim + b->dim;
-	int			dim_a = a->dim;
-	int			dim_b = b->dim;
 
 	CheckDim(dim);
 	result = InitVector(dim);
 
-	/* Auto-vectorized */
+	for (int i = 0; i < a->dim; i++)
-	for (int i = 0; i < dim_a; i++)
 		result->x[i] = a->x[i];
 
-	/* Auto-vectorized */
+	for (int i = 0; i < b->dim; i++)
-	for (int i = 0; i < dim_b; i++)
+		result->x[i + a->dim] = b->x[i];
-		result->x[i + dim_a] = b->x[i];
 
 	PG_RETURN_POINTER(result);
 }

test/expected/hnsw_vector.out

@@ -99,38 +99,6 @@ SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <+> (SELECT NULL::vector)) t2
      4
 (1 row)
 
-DROP TABLE t;
--- iterative
-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);
-SET hnsw.iterative_scan = strict_order;
-SET hnsw.ef_search = 1;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-   val   
----------
- [1,2,3]
- [1,1,1]
- [0,0,0]
-(3 rows)
-
-SET hnsw.iterative_scan = relaxed_order;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-   val   
----------
- [1,2,3]
- [1,1,1]
- [0,0,0]
-(3 rows)
-
-TRUNCATE t;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
- val 
------
-(0 rows)
-
-RESET hnsw.iterative_scan;
-RESET hnsw.ef_search;
 DROP TABLE t;
 -- unlogged
 CREATE UNLOGGED TABLE t (val vector(3));
@@ -171,31 +139,4 @@ 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)
-SHOW hnsw.iterative_scan;
- hnsw.iterative_scan 
----------------------
- off
-(1 row)
-
-SET hnsw.iterative_scan = on;
-ERROR:  invalid value for parameter "hnsw.iterative_scan": "on"
-HINT:  Available values: off, relaxed_order, strict_order.
-SHOW hnsw.max_scan_tuples;
- hnsw.max_scan_tuples 
-----------------------
- 20000
-(1 row)
-
-SET hnsw.max_scan_tuples = 0;
-ERROR:  0 is outside the valid range for parameter "hnsw.max_scan_tuples" (1 .. 2147483647)
-SHOW hnsw.scan_mem_multiplier;
- hnsw.scan_mem_multiplier 
---------------------------
- 1
-(1 row)
-
-SET hnsw.scan_mem_multiplier = 0;
-ERROR:  0 is outside the valid range for parameter "hnsw.scan_mem_multiplier" (1 .. 1000)
-SET hnsw.scan_mem_multiplier = 1001;
-ERROR:  1001 is outside the valid range for parameter "hnsw.scan_mem_multiplier" (1 .. 1000)
 DROP TABLE t;

test/expected/ivfflat_vector.out

@@ -81,46 +81,6 @@ SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> (SELECT NULL::vector)) t2
      3
 (1 row)
 
-DROP TABLE t;
--- iterative
-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 vector_l2_ops) WITH (lists = 3);
-SET ivfflat.iterative_scan = relaxed_order;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-   val   
----------
- [1,2,3]
- [1,1,1]
- [0,0,0]
-(3 rows)
-
-SET ivfflat.max_probes = 1;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-   val   
----------
- [1,2,3]
-(1 row)
-
-SET ivfflat.max_probes = 2;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-   val   
----------
- [1,2,3]
- [1,1,1]
-(2 rows)
-
-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)
-
-RESET ivfflat.iterative_scan;
-RESET ivfflat.max_probes;
 DROP TABLE t;
 -- unlogged
 CREATE UNLOGGED TABLE t (val vector(3));
@@ -149,27 +109,4 @@ SHOW ivfflat.probes;
  1
 (1 row)
 
-SET ivfflat.probes = 0;
-ERROR:  0 is outside the valid range for parameter "ivfflat.probes" (1 .. 32768)
-SET ivfflat.probes = 32769;
-ERROR:  32769 is outside the valid range for parameter "ivfflat.probes" (1 .. 32768)
-SHOW ivfflat.iterative_scan;
- ivfflat.iterative_scan 
-------------------------
- off
-(1 row)
-
-SET ivfflat.iterative_scan = on;
-ERROR:  invalid value for parameter "ivfflat.iterative_scan": "on"
-HINT:  Available values: off, relaxed_order.
-SHOW ivfflat.max_probes;
- ivfflat.max_probes 
---------------------
- 32768
-(1 row)
-
-SET ivfflat.max_probes = 0;
-ERROR:  0 is outside the valid range for parameter "ivfflat.max_probes" (1 .. 32768)
-SET ivfflat.max_probes = 32769;
-ERROR:  32769 is outside the valid range for parameter "ivfflat.max_probes" (1 .. 32768)
 DROP TABLE t;

test/sql/hnsw_vector.sql

@@ -57,26 +57,6 @@ SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <+> (SELECT NULL::vector)) t2
 
 DROP TABLE t;
 
--- iterative
-
-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);
-
-SET hnsw.iterative_scan = strict_order;
-SET hnsw.ef_search = 1;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-
-SET hnsw.iterative_scan = relaxed_order;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-
-TRUNCATE t;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-
-RESET hnsw.iterative_scan;
-RESET hnsw.ef_search;
-DROP TABLE t;
-
 -- unlogged
 
 CREATE UNLOGGED TABLE t (val vector(3));
@@ -101,17 +81,4 @@ SHOW hnsw.ef_search;
 SET hnsw.ef_search = 0;
 SET hnsw.ef_search = 1001;
 
-SHOW hnsw.iterative_scan;
-
-SET hnsw.iterative_scan = on;
-
-SHOW hnsw.max_scan_tuples;
-
-SET hnsw.max_scan_tuples = 0;
-
-SHOW hnsw.scan_mem_multiplier;
-
-SET hnsw.scan_mem_multiplier = 0;
-SET hnsw.scan_mem_multiplier = 1001;
-
 DROP TABLE t;

test/sql/ivfflat_vector.sql

@@ -44,28 +44,6 @@ SELECT COUNT(*) FROM (SELECT * FROM t ORDER BY val <=> (SELECT NULL::vector)) t2
 
 DROP TABLE t;
 
--- iterative
-
-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 vector_l2_ops) WITH (lists = 3);
-
-SET ivfflat.iterative_scan = relaxed_order;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-
-SET ivfflat.max_probes = 1;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-
-SET ivfflat.max_probes = 2;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-
-TRUNCATE t;
-SELECT * FROM t ORDER BY val <-> '[3,3,3]';
-
-RESET ivfflat.iterative_scan;
-RESET ivfflat.max_probes;
-DROP TABLE t;
-
 -- unlogged
 
 CREATE UNLOGGED TABLE t (val vector(3));
@@ -84,16 +62,4 @@ CREATE INDEX ON t USING ivfflat (val vector_l2_ops) WITH (lists = 32769);
 
 SHOW ivfflat.probes;
 
-SET ivfflat.probes = 0;
-SET ivfflat.probes = 32769;
-
-SHOW ivfflat.iterative_scan;
-
-SET ivfflat.iterative_scan = on;
-
-SHOW ivfflat.max_probes;
-
-SET ivfflat.max_probes = 0;
-SET ivfflat.max_probes = 32769;
-
 DROP TABLE t;

test/t/002_ivfflat_vacuum.pl

@@ -6,7 +6,13 @@ use Test::More;
 
 my $dim = 3;
 
-my $array_sql = join(",", ('random()') x $dim);
+my @r = ();
+for (1 .. $dim)
+{
+	my $v = int(rand(1000)) + 1;
+	push(@r, "i % $v");
+}
+my $array_sql = join(", ", @r);
 
 # Initialize node
 my $node = PostgreSQL::Test::Cluster->new('node');
@@ -17,20 +23,19 @@ $node->start;
 $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, 100000) i;"
+	"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 vector_l2_ops);");
 
 # Get size
 my $size = $node->safe_psql("postgres", "SELECT pg_total_relation_size('tst_v_idx');");
 
-# Store values
-$node->safe_psql("postgres", "CREATE TABLE tmp AS SELECT * FROM tst;");
-
 # 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 * FROM tmp;");
+$node->safe_psql("postgres",
+	"INSERT INTO tst SELECT i % 10, ARRAY[$array_sql] FROM generate_series(1, 100000) i;"
+);
 
 # Check size
 my $new_size = $node->safe_psql("postgres", "SELECT pg_total_relation_size('tst_v_idx');");

test/t/041_hnsw_streaming.pl

@@ -26,26 +26,25 @@ $node->safe_psql("postgres", qq(
 
 my $count = $node->safe_psql("postgres", qq(
 	SET enable_seqscan = off;
-	SET hnsw.iterative_scan = relaxed_order;
+	SET hnsw.streaming = on;
-	SET hnsw.max_scan_tuples = 100000;
+	SET work_mem = '8MB';
-	SET hnsw.scan_mem_multiplier = 2;
 	SELECT COUNT(*) FROM (SELECT v FROM tst WHERE i % 10000 = 0 ORDER BY v <-> (SELECT v FROM tst LIMIT 1) LIMIT 11) t;
 ));
 is($count, 10);
 
 foreach ((30000, 50000, 70000))
 {
-	my $max_tuples = $_;
+	my $ef_stream = $_;
-	my $expected = $max_tuples / 10000;
+	my $expected = $ef_stream / 10000;
 	my $sum = 0;
 
 	for my $i (1 .. 20)
 	{
 		$count = $node->safe_psql("postgres", qq(
 			SET enable_seqscan = off;
-			SET hnsw.iterative_scan = relaxed_order;
+			SET hnsw.streaming = on;
-			SET hnsw.max_scan_tuples = $max_tuples;
+			SET hnsw.ef_stream = $ef_stream;
-			SET hnsw.scan_mem_multiplier = 2;
+			SET work_mem = '8MB';
 			SELECT COUNT(*) FROM (SELECT v FROM tst WHERE i % 10000 = 0 ORDER BY v <-> (SELECT v FROM tst WHERE i = $i) LIMIT 11) t;
 		));
 		$sum += $count;
@@ -56,4 +55,13 @@ foreach ((30000, 50000, 70000))
 	cmp_ok($avg, '<', $expected + 2);
 }
 
+my ($ret, $stdout, $stderr) = $node->psql("postgres", qq(
+	SET enable_seqscan = off;
+	SET hnsw.streaming = on;
+	SET client_min_messages = debug1;
+	SET work_mem = '2MB';
+	SELECT COUNT(*) FROM (SELECT v FROM tst WHERE i % 10000 = 0 ORDER BY v <-> (SELECT v FROM tst LIMIT 1) LIMIT 11) t;
+));
+like($stderr, qr/hnsw index scan exceeded work_mem after \d+ tuples/);
+
 done_testing();

test/t/041_ivfflat_iterative_scan.pl

@@ -1,54 +0,0 @@
-use strict;
-use warnings FATAL => 'all';
-use PostgreSQL::Test::Cluster;
-use PostgreSQL::Test::Utils;
-use Test::More;
-
-my $dim = 3;
-my $array_sql = join(",", ('random()') x $dim);
-
-# Initialize node
-my $node = PostgreSQL::Test::Cluster->new('node');
-$node->init;
-$node->start;
-
-# Create table
-$node->safe_psql("postgres", "CREATE EXTENSION vector;");
-$node->safe_psql("postgres", "CREATE TABLE tst (i int4 PRIMARY KEY, v vector($dim));");
-$node->safe_psql("postgres",
-	"INSERT INTO tst SELECT i, ARRAY[$array_sql] FROM generate_series(1, 100000) i;"
-);
-$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops);");
-
-my $count = $node->safe_psql("postgres", qq(
-	SET enable_seqscan = off;
-	SET ivfflat.probes = 10;
-	SET ivfflat.iterative_scan = relaxed_order;
-	SELECT COUNT(*) FROM (SELECT v FROM tst WHERE i % 10000 = 0 ORDER BY v <-> (SELECT v FROM tst LIMIT 1) LIMIT 11) t;
-));
-is($count, 10);
-
-foreach ((30, 50, 70))
-{
-	my $max_probes = $_;
-	my $expected = $max_probes / 10;
-	my $sum = 0;
-
-	for my $i (1 .. 20)
-	{
-		$count = $node->safe_psql("postgres", qq(
-			SET enable_seqscan = off;
-			SET ivfflat.probes = 10;
-			SET ivfflat.iterative_scan = relaxed_order;
-			SET ivfflat.max_probes = $max_probes;
-			SELECT COUNT(*) FROM (SELECT v FROM tst WHERE i % 10000 = 0 ORDER BY v <-> (SELECT v FROM tst WHERE i = $i) LIMIT 11) t;
-		));
-		$sum += $count;
-	}
-
-	my $avg = $sum / 20;
-	cmp_ok($avg, '>', $expected - 2);
-	cmp_ok($avg, '<', $expected + 2);
-}
-
-done_testing();

test/t/042_hnsw_streaming_recall.pl

@@ -10,18 +10,18 @@ my @expected;
 my $limit = 20;
 my $dim = 3;
 my $array_sql = join(",", ('random()') x $dim);
-my @cs = (50, 500);
+my @cs = (100, 1000);
 
 sub test_recall
 {
-	my ($c, $ef_search, $min, $operator, $mode) = @_;
+	my ($c, $ef_search, $min, $operator) = @_;
 	my $correct = 0;
 	my $total = 0;
 
 	my $explain = $node->safe_psql("postgres", qq(
 		SET enable_seqscan = off;
 		SET hnsw.ef_search = $ef_search;
-		SET hnsw.iterative_scan = $mode;
+		SET hnsw.streaming = on;
 		EXPLAIN ANALYZE SELECT i FROM tst WHERE i % $c = 0 ORDER BY v $operator '$queries[0]' LIMIT $limit;
 	));
 	like($explain, qr/Index Scan using idx on tst/);
@@ -31,7 +31,7 @@ sub test_recall
 		my $actual = $node->safe_psql("postgres", qq(
 			SET enable_seqscan = off;
 			SET hnsw.ef_search = $ef_search;
-			SET hnsw.iterative_scan = $mode;
+			SET hnsw.streaming = on;
 			SELECT i FROM tst WHERE i % $c = 0 ORDER BY v $operator '$queries[$i]' LIMIT $limit;
 		));
 		my @actual_ids = split("\n", $actual);
@@ -50,7 +50,7 @@ sub test_recall
 		$total += $limit;
 	}
 
-	cmp_ok($correct / $total, ">=", $min, "$operator $mode $c");
+	cmp_ok($correct / $total, ">=", $min, $operator);
 }
 
 # Initialize node
@@ -62,7 +62,7 @@ $node->start;
 $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, 50000) i;"
+	"INSERT INTO tst SELECT i, ARRAY[$array_sql] FROM generate_series(1, 100000) i;"
 );
 
 # Generate queries
@@ -108,8 +108,21 @@ for my $i (0 .. $#operators)
 			push(@expected, $res);
 		}
 
-		test_recall($c, 40, 0.99, $operator, "strict_order");
+		if ($c == 100)
-		test_recall($c, 40, 0.99, $operator, "relaxed_order");
+		{
+			test_recall($c, 40, 0.99, $operator);
+		}
+		else
+		{
+			if ($operator eq "<->")
+			{
+				test_recall($c, 40, 0.99, $operator);
+			}
+			else
+			{
+				test_recall($c, 40, 0.99, $operator);
+			}
+		}
 	}
 
 	$node->safe_psql("postgres", "DROP INDEX idx;");

test/t/042_ivfflat_iterative_scan_recall.pl

@@ -1,125 +0,0 @@
-use strict;
-use warnings FATAL => 'all';
-use PostgreSQL::Test::Cluster;
-use PostgreSQL::Test::Utils;
-use Test::More;
-
-my $node;
-my @queries = ();
-my @expected;
-my $limit = 20;
-my @cs = (100, 1000);
-
-sub test_recall
-{
-	my ($c, $probes, $min, $operator) = @_;
-	my $correct = 0;
-	my $total = 0;
-
-	my $explain = $node->safe_psql("postgres", qq(
-		SET enable_seqscan = off;
-		SET ivfflat.probes = $probes;
-		SET ivfflat.iterative_scan = relaxed_order;
-		EXPLAIN ANALYZE SELECT i FROM tst WHERE i % $c = 0 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;
-			SET ivfflat.iterative_scan = relaxed_order;
-			SELECT i FROM tst WHERE i % $c = 0 ORDER BY v $operator '$queries[$i]' LIMIT $limit;
-		));
-		my @actual_ids = split("\n", $actual);
-
-		my @expected_ids = split("\n", $expected[$i]);
-		my %expected_set = map { $_ => 1 } @expected_ids;
-
-		foreach (@actual_ids)
-		{
-			if (exists($expected_set{$_}))
-			{
-				$correct++;
-			}
-		}
-
-		$total += $limit;
-	}
-
-	cmp_ok($correct / $total, ">=", $min, "$operator $c");
-}
-
-# Initialize node
-$node = PostgreSQL::Test::Cluster->new('node');
-$node->init;
-$node->start;
-
-# Create table
-$node->safe_psql("postgres", "CREATE EXTENSION vector;");
-$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v 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_cosine_ops");
-
-for my $i (0 .. $#operators)
-{
-	my $operator = $operators[$i];
-	my $opclass = $opclasses[$i];
-
-	$node->safe_psql("postgres", "CREATE INDEX idx ON tst USING ivfflat (v $opclass);");
-
-	foreach (@cs)
-	{
-		my $c = $_;
-
-		# Get exact results
-		@expected = ();
-		foreach (@queries)
-		{
-			my $res = $node->safe_psql("postgres", qq(
-				SET enable_indexscan = off;
-				WITH top AS (
-					SELECT v $operator '$_' AS distance FROM tst WHERE i % $c = 0 ORDER BY distance LIMIT $limit
-				)
-				SELECT i FROM tst WHERE (v $operator '$_') <= (SELECT MAX(distance) FROM top)
-			));
-			push(@expected, $res);
-		}
-
-		if ($c == 100)
-		{
-			test_recall($c, 1, 0.57, $operator);
-			test_recall($c, 10, 0.98, $operator);
-		}
-		else
-		{
-			if ($operator eq "<->")
-			{
-				test_recall($c, 1, 0.80, $operator);
-			}
-			else
-			{
-				test_recall($c, 1, 0.88, $operator);
-			}
-		}
-	}
-
-	$node->safe_psql("postgres", "DROP INDEX idx;");
-}
-
-done_testing();

vector.control

@@ -1,4 +1,4 @@
 comment = 'vector data type and ivfflat and hnsw access methods'
-default_version = '0.8.0'
+default_version = '0.7.4'
 module_pathname = '$libdir/vector'
 relocatable = true