mirror of
https://github.com/pgvector/pgvector.git
synced 2026-07-22 12:07:34 +08:00
Compare commits
1 Commits
ci-alpine
...
parallel-i
| Author | SHA1 | Date | |
|---|---|---|---|
|
f6058beeda |
246
.github/workflows/build.yml
vendored
246
.github/workflows/build.yml
vendored
@@ -1,120 +1,120 @@
|
||||
name: build
|
||||
on: [push, pull_request]
|
||||
jobs:
|
||||
# ubuntu:
|
||||
# runs-on: ${{ matrix.os }}
|
||||
# if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
||||
# strategy:
|
||||
# fail-fast: false
|
||||
# matrix:
|
||||
# include:
|
||||
# - postgres: 19
|
||||
# os: ubuntu-24.04
|
||||
# - postgres: 18
|
||||
# os: ubuntu-24.04
|
||||
# - postgres: 17
|
||||
# os: ubuntu-24.04
|
||||
# - postgres: 16
|
||||
# os: ubuntu-24.04-arm
|
||||
# - postgres: 15
|
||||
# os: ubuntu-22.04
|
||||
# - postgres: 14
|
||||
# os: ubuntu-22.04-arm
|
||||
# - postgres: 13
|
||||
# os: ubuntu-22.04
|
||||
# steps:
|
||||
# - uses: actions/checkout@v5
|
||||
# - uses: ankane/setup-postgres@v1
|
||||
# with:
|
||||
# postgres-version: ${{ matrix.postgres }}
|
||||
# dev-files: true
|
||||
# - run: make
|
||||
# env:
|
||||
# PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare ${{ matrix.postgres >= 18 && '-Wno-missing-field-initializers' || '' }}
|
||||
# - run: |
|
||||
# export PG_CONFIG=`which pg_config`
|
||||
# sudo --preserve-env=PG_CONFIG make install
|
||||
# - run: make installcheck
|
||||
# - if: ${{ failure() }}
|
||||
# run: cat regression.diffs
|
||||
# - run: |
|
||||
# sudo apt-get update
|
||||
# sudo apt-get install libipc-run-perl
|
||||
# - run: make prove_installcheck
|
||||
# mac:
|
||||
# runs-on: ${{ matrix.os }}
|
||||
# if: ${{ !startsWith(github.ref_name, 'windows') }}
|
||||
# strategy:
|
||||
# fail-fast: false
|
||||
# matrix:
|
||||
# include:
|
||||
# - postgres: 18
|
||||
# os: macos-26
|
||||
# - postgres: 14
|
||||
# os: macos-15-intel
|
||||
# steps:
|
||||
# - uses: actions/checkout@v5
|
||||
# - uses: ankane/setup-postgres@v1
|
||||
# with:
|
||||
# postgres-version: ${{ matrix.postgres }}
|
||||
# - run: make
|
||||
# env:
|
||||
# PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-unknown-warning-option ${{ matrix.postgres >= 18 && '-Wno-missing-field-initializers' || '' }}
|
||||
# - run: make install
|
||||
# - run: make installcheck
|
||||
# - if: ${{ failure() }}
|
||||
# run: cat regression.diffs
|
||||
# # Homebrew Postgres does not enable TAP tests, so need to download
|
||||
# - run: |
|
||||
# brew install cpanm
|
||||
# cpanm --notest IPC::Run
|
||||
# wget -q https://github.com/postgres/postgres/archive/refs/tags/$TAG.tar.gz
|
||||
# tar xf $TAG.tar.gz
|
||||
# mv postgres-$TAG postgres
|
||||
# env:
|
||||
# TAG: ${{ matrix.postgres == 18 && 'REL_18_0' || 'REL_14_19' }}
|
||||
# - 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
|
||||
# env:
|
||||
# LLVM_VERSION: ${{ matrix.os == 'macos-26' && 20 || 18 }}
|
||||
# PG_CFLAGS: -DUSE_ASSERT_CHECKING
|
||||
# windows:
|
||||
# runs-on: ${{ matrix.os }}
|
||||
# if: ${{ !startsWith(github.ref_name, 'mac') }}
|
||||
# strategy:
|
||||
# fail-fast: false
|
||||
# matrix:
|
||||
# include:
|
||||
# - postgres: 17
|
||||
# os: windows-2025
|
||||
# - postgres: 14
|
||||
# os: windows-2022
|
||||
# steps:
|
||||
# - uses: actions/checkout@v5
|
||||
# - uses: ankane/setup-postgres@v1
|
||||
# with:
|
||||
# postgres-version: ${{ matrix.postgres }}
|
||||
# - run: |
|
||||
# call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvars64.bat" && ^
|
||||
# nmake /NOLOGO /F Makefile.win && ^
|
||||
# nmake /NOLOGO /F Makefile.win install && ^
|
||||
# nmake /NOLOGO /F Makefile.win installcheck ${{ matrix.postgres != 17 && 'PG_REGRESS=$(PGROOT)\bin\pg_regress' || '' }} && ^
|
||||
# nmake /NOLOGO /F Makefile.win clean && ^
|
||||
# nmake /NOLOGO /F Makefile.win uninstall
|
||||
# shell: cmd
|
||||
# - if: ${{ failure() }}
|
||||
# run: cat regression.diffs
|
||||
ubuntu:
|
||||
runs-on: ${{ matrix.os }}
|
||||
if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
include:
|
||||
- postgres: 19
|
||||
os: ubuntu-24.04
|
||||
- postgres: 18
|
||||
os: ubuntu-24.04
|
||||
- postgres: 17
|
||||
os: ubuntu-24.04
|
||||
- postgres: 16
|
||||
os: ubuntu-24.04-arm
|
||||
- postgres: 15
|
||||
os: ubuntu-22.04
|
||||
- postgres: 14
|
||||
os: ubuntu-22.04-arm
|
||||
- postgres: 13
|
||||
os: ubuntu-22.04
|
||||
steps:
|
||||
- uses: actions/checkout@v4
|
||||
- uses: ankane/setup-postgres@v1
|
||||
with:
|
||||
postgres-version: ${{ matrix.postgres }}
|
||||
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
|
||||
- run: |
|
||||
export PG_CONFIG=`which pg_config`
|
||||
sudo --preserve-env=PG_CONFIG make install
|
||||
- run: make installcheck
|
||||
- if: ${{ failure() }}
|
||||
run: cat regression.diffs
|
||||
- run: |
|
||||
sudo apt-get update
|
||||
sudo apt-get install libipc-run-perl
|
||||
- run: make prove_installcheck
|
||||
mac:
|
||||
runs-on: ${{ matrix.os }}
|
||||
if: ${{ !startsWith(github.ref_name, 'windows') }}
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
include:
|
||||
- postgres: 17
|
||||
os: macos-26
|
||||
- postgres: 14
|
||||
os: macos-15-intel
|
||||
steps:
|
||||
- uses: actions/checkout@v4
|
||||
- uses: ankane/setup-postgres@v1
|
||||
with:
|
||||
postgres-version: ${{ matrix.postgres }}
|
||||
- run: make
|
||||
env:
|
||||
PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-unknown-warning-option
|
||||
- run: make install
|
||||
- run: make installcheck
|
||||
- if: ${{ failure() }}
|
||||
run: cat regression.diffs
|
||||
# Homebrew Postgres does not enable TAP tests, so need to download
|
||||
- run: |
|
||||
brew install cpanm
|
||||
cpanm --notest IPC::Run
|
||||
wget -q https://github.com/postgres/postgres/archive/refs/tags/$TAG.tar.gz
|
||||
tar xf $TAG.tar.gz
|
||||
mv postgres-$TAG postgres
|
||||
env:
|
||||
TAG: ${{ matrix.postgres == 17 && 'REL_17_6' || 'REL_14_19' }}
|
||||
- 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
|
||||
env:
|
||||
LLVM_VERSION: ${{ matrix.os == 'macos-26' && 20 || 18 }}
|
||||
PG_CFLAGS: -DUSE_ASSERT_CHECKING
|
||||
windows:
|
||||
runs-on: ${{ matrix.os }}
|
||||
if: ${{ !startsWith(github.ref_name, 'mac') }}
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
include:
|
||||
- postgres: 17
|
||||
os: windows-2025
|
||||
- postgres: 14
|
||||
os: windows-2022
|
||||
steps:
|
||||
- uses: actions/checkout@v4
|
||||
- uses: ankane/setup-postgres@v1
|
||||
with:
|
||||
postgres-version: ${{ matrix.postgres }}
|
||||
- run: |
|
||||
call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvars64.bat" && ^
|
||||
nmake /NOLOGO /F Makefile.win && ^
|
||||
nmake /NOLOGO /F Makefile.win install && ^
|
||||
nmake /NOLOGO /F Makefile.win installcheck ${{ matrix.postgres != 17 && 'PG_REGRESS=$(PGROOT)\bin\pg_regress' || '' }} && ^
|
||||
nmake /NOLOGO /F Makefile.win clean && ^
|
||||
nmake /NOLOGO /F Makefile.win uninstall
|
||||
shell: cmd
|
||||
- if: ${{ failure() }}
|
||||
run: cat regression.diffs
|
||||
i386:
|
||||
if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
||||
runs-on: ubuntu-latest
|
||||
container:
|
||||
image: alpine:3.23
|
||||
image: debian:12
|
||||
options: --platform linux/386
|
||||
steps:
|
||||
- run: apk add --update build-base git openrc perl perl-ipc-run perl-test-harness-utils postgresql18 postgresql18-dev
|
||||
- run: mkdir /run/openrc && touch /run/openrc/softlevel && rc-update add postgresql && rc-service postgresql start
|
||||
- run: apt-get update && apt-get install -y build-essential git libipc-run-perl postgresql-15 postgresql-server-dev-15 sudo
|
||||
- run: service postgresql start
|
||||
- run: |
|
||||
git clone https://github.com/${{ github.repository }}.git pgvector
|
||||
cd pgvector
|
||||
@@ -123,21 +123,21 @@ jobs:
|
||||
make
|
||||
make install
|
||||
chown -R postgres .
|
||||
su postgres -c "make installcheck"
|
||||
su postgres -c "make prove_installcheck"
|
||||
sudo -u postgres make installcheck
|
||||
sudo -u postgres make prove_installcheck
|
||||
env:
|
||||
PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare -Wno-missing-field-initializers
|
||||
PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
|
||||
- if: ${{ failure() }}
|
||||
run: cat pgvector/regression.diffs
|
||||
# valgrind:
|
||||
# if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
||||
# runs-on: ubuntu-latest
|
||||
# steps:
|
||||
# - uses: actions/checkout@v5
|
||||
# - uses: ankane/setup-postgres-valgrind@v1
|
||||
# with:
|
||||
# postgres-version: 18
|
||||
# check-ub: yes
|
||||
# - run: make OPTFLAGS=""
|
||||
# - run: sudo --preserve-env=PG_CONFIG make install
|
||||
# - run: make installcheck
|
||||
valgrind:
|
||||
if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- uses: actions/checkout@v4
|
||||
- uses: ankane/setup-postgres-valgrind@v1
|
||||
with:
|
||||
postgres-version: 17
|
||||
check-ub: yes
|
||||
- run: make OPTFLAGS=""
|
||||
- run: sudo --preserve-env=PG_CONFIG make install
|
||||
- run: make installcheck
|
||||
|
||||
@@ -1,8 +1,3 @@
|
||||
## 0.8.2 (unreleased)
|
||||
|
||||
- Improved `install` target on Windows
|
||||
- Fixed `Index Searches` in `EXPLAIN` output for Postgres 18
|
||||
|
||||
## 0.8.1 (2025-09-04)
|
||||
|
||||
- Added support for Postgres 18 rc1
|
||||
|
||||
2
LICENSE
2
LICENSE
@@ -1,4 +1,4 @@
|
||||
Portions Copyright (c) 1996-2026, PostgreSQL Global Development Group
|
||||
Portions Copyright (c) 1996-2025, PostgreSQL Global Development Group
|
||||
|
||||
Portions Copyright (c) 1994, The Regents of the University of California
|
||||
|
||||
|
||||
5
Makefile
5
Makefile
@@ -27,11 +27,6 @@ ifneq ($(filter ppc64%, $(shell uname -m)), )
|
||||
OPTFLAGS =
|
||||
endif
|
||||
|
||||
# RISC-V64 doesn't support -march=native
|
||||
ifeq ($(shell uname -m), riscv64)
|
||||
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
|
||||
|
||||
@@ -57,7 +57,7 @@ install: all
|
||||
copy $(SHLIB) "$(PKGLIBDIR)"
|
||||
copy $(EXTENSION).control "$(SHAREDIR)\extension"
|
||||
copy sql\$(EXTENSION)--*.sql "$(SHAREDIR)\extension"
|
||||
if not exist "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)" mkdir "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
|
||||
mkdir "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
|
||||
for %f in ($(HEADERS)) do copy %f "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
|
||||
|
||||
installcheck:
|
||||
|
||||
59
README.md
59
README.md
@@ -29,14 +29,14 @@ make install # may need sudo
|
||||
|
||||
See the [installation notes](#installation-notes---linux-and-mac) if you run into issues
|
||||
|
||||
You can also install it with [Docker](#docker), [Homebrew](#homebrew), [PGXN](#pgxn), [APT](#apt), [Yum](#yum), [pkg](#pkg), [APK](#apk), 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).
|
||||
You can also install it with [Docker](#docker), [Homebrew](#homebrew), [PGXN](#pgxn), [APT](#apt), [Yum](#yum), [pkg](#pkg), 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).
|
||||
|
||||
### 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:
|
||||
|
||||
```cmd
|
||||
set "PGROOT=C:\Program Files\PostgreSQL\18"
|
||||
set "PGROOT=C:\Program Files\PostgreSQL\17"
|
||||
cd %TEMP%
|
||||
git clone --branch v0.8.1 https://github.com/pgvector/pgvector.git
|
||||
cd pgvector
|
||||
@@ -695,10 +695,10 @@ CREATE INDEX CONCURRENTLY ...
|
||||
|
||||
### Querying
|
||||
|
||||
Use `EXPLAIN (ANALYZE, BUFFERS)` to debug performance.
|
||||
Use `EXPLAIN ANALYZE` to debug performance.
|
||||
|
||||
```sql
|
||||
EXPLAIN (ANALYZE, BUFFERS) SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
|
||||
EXPLAIN ANALYZE SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
|
||||
```
|
||||
|
||||
#### Exact Search
|
||||
@@ -771,12 +771,9 @@ Use pgvector from any language with a Postgres client. You can even generate and
|
||||
|
||||
Language | Libraries / Examples
|
||||
--- | ---
|
||||
Ada | [pgvector-ada](https://github.com/pgvector/pgvector-ada)
|
||||
Algol | [pgvector-algol](https://github.com/pgvector/pgvector-algol)
|
||||
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)
|
||||
COBOL | [pgvector-cobol](https://github.com/pgvector/pgvector-cobol)
|
||||
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)
|
||||
@@ -793,18 +790,14 @@ 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)
|
||||
OCaml | [pgvector-ocaml](https://github.com/pgvector/pgvector-ocaml)
|
||||
Pascal | [pgvector-pascal](https://github.com/pgvector/pgvector-pascal)
|
||||
Perl | [pgvector-perl](https://github.com/pgvector/pgvector-perl)
|
||||
PHP | [pgvector-php](https://github.com/pgvector/pgvector-php)
|
||||
Prolog | [pgvector-prolog](https://github.com/pgvector/pgvector-prolog)
|
||||
Python | [pgvector-python](https://github.com/pgvector/pgvector-python)
|
||||
R | [pgvector-r](https://github.com/pgvector/pgvector-r)
|
||||
Racket | [pgvector-racket](https://github.com/pgvector/pgvector-racket)
|
||||
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)
|
||||
Tcl | [pgvector-tcl](https://github.com/pgvector/pgvector-tcl)
|
||||
Zig | [pgvector-zig](https://github.com/pgvector/pgvector-zig)
|
||||
|
||||
## Frequently Asked Questions
|
||||
@@ -819,7 +812,7 @@ 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?
|
||||
|
||||
You can use [half-precision vectors](#half-precision-vectors) or [half-precision indexing](#half-precision-indexing) to index up to 4,000 dimensions or [binary quantization](#binary-quantization) to index up to 64,000 dimensions. Other options are [indexing subvectors](#indexing-subvectors) (for models that support it) or [dimensionality reduction](https://en.wikipedia.org/wiki/Dimensionality_reduction).
|
||||
You can use [half-precision indexing](#half-precision-indexing) to index up to 4,000 dimensions or [binary quantization](#binary-quantization) to index up to 64,000 dimensions. Another option is [dimensionality reduction](https://en.wikipedia.org/wiki/Dimensionality_reduction).
|
||||
|
||||
#### Can I store vectors with different dimensions in the same column?
|
||||
|
||||
@@ -1071,7 +1064,7 @@ l2_normalize(sparsevec) → sparsevec | Normalize with Euclidean norm | 0.7.0
|
||||
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=/Library/PostgreSQL/18/bin/pg_config
|
||||
export PG_CONFIG=/Library/PostgreSQL/17/bin/pg_config
|
||||
```
|
||||
|
||||
Then re-run the installation instructions (run `make clean` before `make` if needed). If `sudo` is needed for `make install`, use:
|
||||
@@ -1082,11 +1075,11 @@ sudo --preserve-env=PG_CONFIG make install
|
||||
|
||||
A few common paths on Mac are:
|
||||
|
||||
- EDB installer - `/Library/PostgreSQL/18/bin/pg_config`
|
||||
- Homebrew (arm64) - `/opt/homebrew/opt/postgresql@18/bin/pg_config`
|
||||
- Homebrew (x86-64) - `/usr/local/opt/postgresql@18/bin/pg_config`
|
||||
- EDB installer - `/Library/PostgreSQL/17/bin/pg_config`
|
||||
- Homebrew (arm64) - `/opt/homebrew/opt/postgresql@17/bin/pg_config`
|
||||
- Homebrew (x86-64) - `/usr/local/opt/postgresql@17/bin/pg_config`
|
||||
|
||||
Note: Replace `18` with your Postgres server version
|
||||
Note: Replace `17` with your Postgres server version
|
||||
|
||||
### Missing Header
|
||||
|
||||
@@ -1095,10 +1088,10 @@ If compilation fails with `fatal error: postgres.h: No such file or directory`,
|
||||
For Ubuntu and Debian, use:
|
||||
|
||||
```sh
|
||||
sudo apt install postgresql-server-dev-18
|
||||
sudo apt install postgresql-server-dev-17
|
||||
```
|
||||
|
||||
Note: Replace `18` with your Postgres server version
|
||||
Note: Replace `17` with your Postgres server version
|
||||
|
||||
### Missing SDK
|
||||
|
||||
@@ -1145,15 +1138,13 @@ If installation fails with `Access is denied`, re-run the installation instructi
|
||||
Get the [Docker image](https://hub.docker.com/r/pgvector/pgvector) with:
|
||||
|
||||
```sh
|
||||
docker pull pgvector/pgvector:pg18-trixie
|
||||
docker pull pgvector/pgvector:pg17-trixie
|
||||
```
|
||||
|
||||
This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (replace `18` with your Postgres server version, and run it the same way).
|
||||
This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (replace `17` with your Postgres server version, and run it the same way).
|
||||
|
||||
Supported tags are:
|
||||
|
||||
- `pg18-trixie`, `0.8.1-pg18-trixie`
|
||||
- `pg18-bookworm`, `0.8.1-pg18-bookworm`, `pg18`, `0.8.1-pg18`
|
||||
- `pg17-trixie`, `0.8.1-pg17-trixie`
|
||||
- `pg17-bookworm`, `0.8.1-pg17-bookworm`, `pg17`, `0.8.1-pg17`
|
||||
- `pg16-trixie`, `0.8.1-pg16-trixie`
|
||||
@@ -1170,7 +1161,7 @@ You can also build the image manually:
|
||||
```sh
|
||||
git clone --branch v0.8.1 https://github.com/pgvector/pgvector.git
|
||||
cd pgvector
|
||||
docker build --pull --build-arg PG_MAJOR=18 -t myuser/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.
|
||||
@@ -1187,7 +1178,7 @@ With Homebrew Postgres, you can use:
|
||||
brew install pgvector
|
||||
```
|
||||
|
||||
Note: This only adds it to the `postgresql@18` and `postgresql@17` formulas
|
||||
Note: This only adds it to the `postgresql@17` and `postgresql@14` formulas
|
||||
|
||||
### PGXN
|
||||
|
||||
@@ -1202,22 +1193,22 @@ pgxn install vector
|
||||
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-18-pgvector
|
||||
sudo apt install postgresql-17-pgvector
|
||||
```
|
||||
|
||||
Note: Replace `18` with your Postgres server version
|
||||
Note: Replace `17` 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_18
|
||||
sudo yum install pgvector_17
|
||||
# or
|
||||
sudo dnf install pgvector_18
|
||||
sudo dnf install pgvector_17
|
||||
```
|
||||
|
||||
Note: Replace `18` with your Postgres server version
|
||||
Note: Replace `17` with your Postgres server version
|
||||
|
||||
### pkg
|
||||
|
||||
@@ -1234,14 +1225,6 @@ cd /usr/ports/databases/pgvector
|
||||
make install
|
||||
```
|
||||
|
||||
### APK
|
||||
|
||||
Install the Alpine package with:
|
||||
|
||||
```sh
|
||||
apk add postgresql-pgvector
|
||||
```
|
||||
|
||||
### conda-forge
|
||||
|
||||
With Conda Postgres, install from [conda-forge](https://anaconda.org/conda-forge/pgvector) with:
|
||||
|
||||
@@ -2,7 +2,6 @@
|
||||
|
||||
#include "bitutils.h"
|
||||
#include "bitvec.h"
|
||||
#include "fmgr.h"
|
||||
#include "utils/varbit.h"
|
||||
#include "vector.h"
|
||||
|
||||
|
||||
@@ -1,7 +1,5 @@
|
||||
#include "postgres.h"
|
||||
|
||||
#include <math.h>
|
||||
|
||||
#include "halfutils.h"
|
||||
#include "halfvec.h"
|
||||
|
||||
|
||||
@@ -13,20 +13,12 @@
|
||||
#include "port.h" /* for strtof() */
|
||||
#include "sparsevec.h"
|
||||
#include "utils/array.h"
|
||||
#include "utils/builtins.h"
|
||||
#include "utils/float.h"
|
||||
#include "utils/fmgrprotos.h"
|
||||
#include "utils/lsyscache.h"
|
||||
#include "utils/varbit.h"
|
||||
#include "utils/numeric.h"
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
#if PG_VERSION_NUM >= 170000
|
||||
#include "parser/scansup.h"
|
||||
#endif
|
||||
|
||||
#define STATE_DIMS(x) (ARR_DIMS(x)[0] - 1)
|
||||
#define CreateStateDatums(dim) palloc(sizeof(Datum) * (dim + 1))
|
||||
|
||||
@@ -137,9 +129,9 @@ InitHalfVector(int dim)
|
||||
return result;
|
||||
}
|
||||
|
||||
#if PG_VERSION_NUM >= 170000
|
||||
#define halfvec_isspace(ch) scanner_isspace(ch)
|
||||
#else
|
||||
/*
|
||||
* Check for whitespace, since array_isspace() is static
|
||||
*/
|
||||
static inline bool
|
||||
halfvec_isspace(char ch)
|
||||
{
|
||||
@@ -152,7 +144,6 @@ halfvec_isspace(char ch)
|
||||
return true;
|
||||
return false;
|
||||
}
|
||||
#endif
|
||||
|
||||
/*
|
||||
* Check state array
|
||||
|
||||
@@ -1,24 +1,18 @@
|
||||
#include "postgres.h"
|
||||
|
||||
#include <float.h>
|
||||
#include <limits.h>
|
||||
#include <math.h>
|
||||
|
||||
#include "access/amapi.h"
|
||||
#include "access/genam.h"
|
||||
#include "access/reloptions.h"
|
||||
#include "commands/progress.h"
|
||||
#include "commands/vacuum.h"
|
||||
#include "fmgr.h"
|
||||
#include "hnsw.h"
|
||||
#include "miscadmin.h"
|
||||
#include "nodes/pg_list.h"
|
||||
#include "utils/float.h"
|
||||
#include "utils/guc.h"
|
||||
#include "utils/relcache.h"
|
||||
#include "utils/selfuncs.h"
|
||||
#include "utils/spccache.h"
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM < 150000
|
||||
#define MarkGUCPrefixReserved(x) EmitWarningsOnPlaceholders(x)
|
||||
|
||||
@@ -3,8 +3,6 @@
|
||||
|
||||
#include "postgres.h"
|
||||
|
||||
#include <math.h>
|
||||
|
||||
#include "access/genam.h"
|
||||
#include "access/parallel.h"
|
||||
#include "lib/pairingheap.h"
|
||||
@@ -14,10 +12,6 @@
|
||||
#include "utils/sampling.h"
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 190000
|
||||
typedef Pointer Item;
|
||||
#endif
|
||||
|
||||
#define HNSW_MAX_DIM 2000
|
||||
#define HNSW_MAX_NNZ 1000
|
||||
|
||||
@@ -430,7 +424,7 @@ 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);
|
||||
HnswSearchCandidate *HnswEntryCandidate(char *base, HnswElement entryPoint, HnswQuery * q, Relation index, HnswSupport * support, bool loadVec);
|
||||
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);
|
||||
void HnswAddHeapTid(HnswElement element, ItemPointer heaptid);
|
||||
|
||||
@@ -36,12 +36,11 @@
|
||||
*/
|
||||
#include "postgres.h"
|
||||
|
||||
#include "access/genam.h"
|
||||
#include <math.h>
|
||||
|
||||
#include "access/parallel.h"
|
||||
#include "access/relscan.h"
|
||||
#include "access/table.h"
|
||||
#include "access/tableam.h"
|
||||
#include "access/tupdesc.h"
|
||||
#include "access/xact.h"
|
||||
#include "access/xloginsert.h"
|
||||
#include "catalog/index.h"
|
||||
@@ -49,18 +48,11 @@
|
||||
#include "commands/progress.h"
|
||||
#include "hnsw.h"
|
||||
#include "miscadmin.h"
|
||||
#include "nodes/execnodes.h"
|
||||
#include "optimizer/optimizer.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "tcop/tcopprot.h"
|
||||
#include "utils/datum.h"
|
||||
#include "utils/memutils.h"
|
||||
#include "utils/rel.h"
|
||||
#include "utils/snapmgr.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
#if PG_VERSION_NUM >= 140000
|
||||
#include "utils/backend_progress.h"
|
||||
@@ -406,7 +398,7 @@ UpdateNeighborsInMemory(char *base, HnswSupport * support, HnswElement e, int m)
|
||||
* Update graph in memory
|
||||
*/
|
||||
static void
|
||||
UpdateGraphInMemory(HnswSupport * support, HnswElement element, int m, HnswElement entryPoint, HnswBuildState * buildstate)
|
||||
UpdateGraphInMemory(HnswSupport * support, HnswElement element, int m, int efConstruction, HnswElement entryPoint, HnswBuildState * buildstate)
|
||||
{
|
||||
HnswGraph *graph = buildstate->graph;
|
||||
char *base = buildstate->hnswarea;
|
||||
@@ -468,7 +460,7 @@ InsertTupleInMemory(HnswBuildState * buildstate, HnswElement element)
|
||||
HnswFindElementNeighbors(base, element, entryPoint, NULL, support, m, efConstruction, false);
|
||||
|
||||
/* Update graph in memory */
|
||||
UpdateGraphInMemory(support, element, m, entryPoint, buildstate);
|
||||
UpdateGraphInMemory(support, element, m, efConstruction, entryPoint, buildstate);
|
||||
|
||||
/* Release entry lock */
|
||||
LWLockRelease(entryLock);
|
||||
@@ -553,7 +545,7 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heaptid, Hn
|
||||
|
||||
/* Copy the datum */
|
||||
memcpy(valuePtr, DatumGetPointer(value), valueSize);
|
||||
HnswPtrStore(base, element->value, (char *) valuePtr);
|
||||
HnswPtrStore(base, element->value, valuePtr);
|
||||
|
||||
/* Create a lock for the element */
|
||||
LWLockInitialize(&element->lock, hnsw_lock_tranche_id);
|
||||
@@ -1062,7 +1054,7 @@ ComputeParallelWorkers(Relation heap, Relation index)
|
||||
* Build graph
|
||||
*/
|
||||
static void
|
||||
BuildGraph(HnswBuildState * buildstate)
|
||||
BuildGraph(HnswBuildState * buildstate, ForkNumber forkNum)
|
||||
{
|
||||
int parallel_workers = 0;
|
||||
|
||||
@@ -1110,7 +1102,7 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
|
||||
|
||||
InitBuildState(buildstate, heap, index, indexInfo, forkNum);
|
||||
|
||||
BuildGraph(buildstate);
|
||||
BuildGraph(buildstate, forkNum);
|
||||
|
||||
if (RelationNeedsWAL(index) || forkNum == INIT_FORKNUM)
|
||||
log_newpage_range(index, forkNum, 0, RelationGetNumberOfBlocksInFork(index, forkNum), true);
|
||||
|
||||
@@ -1,18 +1,13 @@
|
||||
#include "postgres.h"
|
||||
|
||||
#include "access/genam.h"
|
||||
#include <math.h>
|
||||
|
||||
#include "access/generic_xlog.h"
|
||||
#include "hnsw.h"
|
||||
#include "nodes/execnodes.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "storage/lmgr.h"
|
||||
#include "utils/datum.h"
|
||||
#include "utils/memutils.h"
|
||||
#include "utils/rel.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
/*
|
||||
* Get the insert page
|
||||
@@ -665,7 +660,7 @@ FindDuplicateOnDisk(Relation index, HnswElement element, bool building)
|
||||
* Update graph on disk
|
||||
*/
|
||||
static void
|
||||
UpdateGraphOnDisk(Relation index, HnswSupport * support, HnswElement element, int m, HnswElement entryPoint, bool building)
|
||||
UpdateGraphOnDisk(Relation index, HnswSupport * support, HnswElement element, int m, int efConstruction, HnswElement entryPoint, bool building)
|
||||
{
|
||||
BlockNumber newInsertPage = InvalidBlockNumber;
|
||||
|
||||
@@ -713,7 +708,7 @@ HnswInsertTupleOnDisk(Relation index, HnswSupport * support, Datum value, ItemPo
|
||||
|
||||
/* Create an element */
|
||||
element = HnswInitElement(base, heaptid, m, HnswGetMl(m), HnswGetMaxLevel(m), NULL);
|
||||
HnswPtrStore(base, element->value, (char *) DatumGetPointer(value));
|
||||
HnswPtrStore(base, element->value, DatumGetPointer(value));
|
||||
|
||||
/* Prevent concurrent inserts when likely updating entry point */
|
||||
if (entryPoint == NULL || element->level > entryPoint->level)
|
||||
@@ -733,7 +728,7 @@ HnswInsertTupleOnDisk(Relation index, HnswSupport * support, Datum value, ItemPo
|
||||
HnswFindElementNeighbors(base, element, entryPoint, index, support, m, efConstruction, false);
|
||||
|
||||
/* Update graph on disk */
|
||||
UpdateGraphOnDisk(index, support, element, m, entryPoint, building);
|
||||
UpdateGraphOnDisk(index, support, element, m, efConstruction, entryPoint, building);
|
||||
|
||||
/* Release lock */
|
||||
UnlockPage(index, HNSW_UPDATE_LOCK, lockmode);
|
||||
|
||||
@@ -1,21 +1,12 @@
|
||||
#include "postgres.h"
|
||||
|
||||
#include "access/genam.h"
|
||||
#include "access/relscan.h"
|
||||
#include "hnsw.h"
|
||||
#include "lib/pairingheap.h"
|
||||
#include "miscadmin.h"
|
||||
#include "nodes/pg_list.h"
|
||||
#include "pgstat.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "storage/lmgr.h"
|
||||
#include "utils/float.h"
|
||||
#include "utils/memutils.h"
|
||||
#include "utils/relcache.h"
|
||||
#include "utils/snapmgr.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
/*
|
||||
* Algorithm 5 from paper
|
||||
@@ -202,10 +193,6 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
|
||||
|
||||
/* Count index scan for stats */
|
||||
pgstat_count_index_scan(scan->indexRelation);
|
||||
#if PG_VERSION_NUM >= 180000
|
||||
if (scan->instrument)
|
||||
scan->instrument->nsearches++;
|
||||
#endif
|
||||
|
||||
/* Safety check */
|
||||
if (scan->orderByData == NULL)
|
||||
|
||||
@@ -2,24 +2,18 @@
|
||||
|
||||
#include <math.h>
|
||||
|
||||
#include "access/genam.h"
|
||||
#include "access/generic_xlog.h"
|
||||
#include "catalog/pg_type.h"
|
||||
#include "catalog/pg_type_d.h"
|
||||
#include "common/hashfn.h"
|
||||
#include "fmgr.h"
|
||||
#include "hnsw.h"
|
||||
#include "lib/pairingheap.h"
|
||||
#include "nodes/pg_list.h"
|
||||
#include "port/atomics.h"
|
||||
#include "sparsevec.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "utils/datum.h"
|
||||
#include "utils/memdebug.h"
|
||||
#include "utils/rel.h"
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
#if PG_VERSION_NUM < 170000
|
||||
static inline uint64
|
||||
@@ -262,7 +256,7 @@ HnswInitElement(char *base, ItemPointer heaptid, int m, double ml, int maxLevel,
|
||||
|
||||
HnswInitNeighbors(base, element, m, allocator);
|
||||
|
||||
HnswPtrStore(base, element->value, (char *) NULL);
|
||||
HnswPtrStore(base, element->value, (Pointer) NULL);
|
||||
|
||||
return element;
|
||||
}
|
||||
@@ -288,7 +282,7 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
|
||||
element->blkno = blkno;
|
||||
element->offno = offno;
|
||||
HnswPtrStore(base, element->neighbors, (HnswNeighborArrayPtr *) NULL);
|
||||
HnswPtrStore(base, element->value, (char *) NULL);
|
||||
HnswPtrStore(base, element->value, (Pointer) NULL);
|
||||
return element;
|
||||
}
|
||||
|
||||
@@ -514,7 +508,7 @@ HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHe
|
||||
char *base = NULL;
|
||||
Datum value = datumCopy(PointerGetDatum(&etup->data), false, -1);
|
||||
|
||||
HnswPtrStore(base, element->value, (char *) DatumGetPointer(value));
|
||||
HnswPtrStore(base, element->value, DatumGetPointer(value));
|
||||
}
|
||||
}
|
||||
|
||||
@@ -928,7 +922,7 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
|
||||
continue;
|
||||
}
|
||||
|
||||
if (!(eDistance < f->distance || alwaysAdd))
|
||||
if (eElement == NULL || !(eDistance < f->distance || alwaysAdd))
|
||||
{
|
||||
if (discarded != NULL)
|
||||
{
|
||||
|
||||
@@ -1,18 +1,13 @@
|
||||
#include "postgres.h"
|
||||
|
||||
#include "access/genam.h"
|
||||
#include <math.h>
|
||||
|
||||
#include "access/generic_xlog.h"
|
||||
#include "commands/vacuum.h"
|
||||
#include "hnsw.h"
|
||||
#include "nodes/pg_list.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "storage/lmgr.h"
|
||||
#include "utils/memutils.h"
|
||||
#include "utils/rel.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
#if PG_VERSION_NUM >= 180000
|
||||
#define vacuum_delay_point() vacuum_delay_point(false)
|
||||
@@ -179,12 +174,7 @@ NeedsUpdated(HnswVacuumState * vacuumstate, HnswElement element)
|
||||
/* Also update if layer 0 is not full */
|
||||
/* This could indicate too many candidates being deleted during insert */
|
||||
if (!needsUpdated)
|
||||
{
|
||||
/* Keep clang-tidy happy */
|
||||
Assert(ntup->count > 0);
|
||||
|
||||
needsUpdated = !ItemPointerIsValid(&ntup->indextids[ntup->count - 1]);
|
||||
}
|
||||
|
||||
UnlockReleaseBuffer(buf);
|
||||
|
||||
@@ -534,9 +524,8 @@ MarkDeleted(HnswVacuumState * vacuumstate)
|
||||
ntup = (HnswNeighborTuple) PageGetItem(npage, PageGetItemId(npage, neighborOffno));
|
||||
|
||||
/* Overwrite element */
|
||||
/* Use memset instead of MemSet to keep clang-tidy happy */
|
||||
etup->deleted = 1;
|
||||
memset(&etup->data, 0, VARSIZE_ANY(&etup->data));
|
||||
MemSet(&etup->data, 0, VARSIZE_ANY(&etup->data));
|
||||
|
||||
/* Overwrite neighbors */
|
||||
for (int i = 0; i < ntup->count; i++)
|
||||
|
||||
@@ -2,36 +2,23 @@
|
||||
|
||||
#include <float.h>
|
||||
|
||||
#include "access/genam.h"
|
||||
#include "access/generic_xlog.h"
|
||||
#include "access/itup.h"
|
||||
#include "access/relscan.h"
|
||||
#include "access/table.h"
|
||||
#include "access/tableam.h"
|
||||
#include "access/tupdesc.h"
|
||||
#include "access/parallel.h"
|
||||
#include "access/xact.h"
|
||||
#include "access/xloginsert.h"
|
||||
#include "bitvec.h"
|
||||
#include "catalog/index.h"
|
||||
#include "catalog/pg_operator_d.h"
|
||||
#include "catalog/pg_type_d.h"
|
||||
#include "commands/progress.h"
|
||||
#include "fmgr.h"
|
||||
#include "halfvec.h"
|
||||
#include "ivfflat.h"
|
||||
#include "miscadmin.h"
|
||||
#include "nodes/execnodes.h"
|
||||
#include "optimizer/optimizer.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "tcop/tcopprot.h"
|
||||
#include "utils/memutils.h"
|
||||
#include "utils/rel.h"
|
||||
#include "utils/sampling.h"
|
||||
#include "utils/snapmgr.h"
|
||||
#include "utils/tuplesort.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 140000
|
||||
#include "utils/backend_progress.h"
|
||||
@@ -151,7 +138,7 @@ SampleRows(IvfflatBuildState * buildstate)
|
||||
* Add tuple to sort
|
||||
*/
|
||||
static void
|
||||
AddTupleToSort(ItemPointer tid, Datum *values, IvfflatBuildState * buildstate)
|
||||
AddTupleToSort(Relation index, ItemPointer tid, Datum *values, IvfflatBuildState * buildstate)
|
||||
{
|
||||
double distance;
|
||||
double minDistance = DBL_MAX;
|
||||
@@ -228,7 +215,7 @@ BuildCallback(Relation index, ItemPointer tid, Datum *values,
|
||||
oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
|
||||
|
||||
/* Add tuple to sort */
|
||||
AddTupleToSort(tid, values, buildstate);
|
||||
AddTupleToSort(index, tid, values, buildstate);
|
||||
|
||||
/* Reset memory context */
|
||||
MemoryContextSwitchTo(oldCtx);
|
||||
@@ -483,8 +470,8 @@ CreateMetaPage(Relation index, int dimensions, int lists, ForkNumber forkNum)
|
||||
* Create list pages
|
||||
*/
|
||||
static void
|
||||
CreateListPages(Relation index, VectorArray centers, int lists,
|
||||
ForkNumber forkNum, ListInfo * *listInfo)
|
||||
CreateListPages(Relation index, VectorArray centers, int dimensions,
|
||||
int lists, ForkNumber forkNum, ListInfo * *listInfo)
|
||||
{
|
||||
Buffer buf;
|
||||
Page page;
|
||||
@@ -1017,7 +1004,7 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
|
||||
|
||||
/* Create pages */
|
||||
CreateMetaPage(index, buildstate->dimensions, buildstate->lists, forkNum);
|
||||
CreateListPages(index, buildstate->centers, buildstate->lists, forkNum, &buildstate->listInfo);
|
||||
CreateListPages(index, buildstate->centers, buildstate->dimensions, buildstate->lists, forkNum, &buildstate->listInfo);
|
||||
CreateEntryPages(buildstate, forkNum);
|
||||
|
||||
/* Write WAL for initialization fork since GenericXLog functions do not */
|
||||
|
||||
@@ -3,19 +3,14 @@
|
||||
#include <float.h>
|
||||
|
||||
#include "access/amapi.h"
|
||||
#include "access/genam.h"
|
||||
#include "access/reloptions.h"
|
||||
#include "commands/progress.h"
|
||||
#include "commands/vacuum.h"
|
||||
#include "fmgr.h"
|
||||
#include "ivfflat.h"
|
||||
#include "nodes/pg_list.h"
|
||||
#include "utils/float.h"
|
||||
#include "utils/guc.h"
|
||||
#include "utils/relcache.h"
|
||||
#include "utils/selfuncs.h"
|
||||
#include "utils/spccache.h"
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM < 150000
|
||||
#define MarkGUCPrefixReserved(x) EmitWarningsOnPlaceholders(x)
|
||||
@@ -148,6 +143,10 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
|
||||
*indexSelectivity = costs.indexSelectivity;
|
||||
*indexCorrelation = costs.indexCorrelation;
|
||||
*indexPages = costs.numIndexPages;
|
||||
|
||||
elog(INFO, "ivfflatcostestimate = %f", costs.indexTotalCost);
|
||||
/* Cost estimates for parallel workers applied outside of amcostestimate */
|
||||
elog(INFO, "parallel_workers = %d, parallel aware = %d", path->path.parallel_workers, path->path.parallel_aware);
|
||||
}
|
||||
|
||||
/*
|
||||
@@ -175,6 +174,25 @@ ivfflatvalidate(Oid opclassoid)
|
||||
return true;
|
||||
}
|
||||
|
||||
static Size
|
||||
ivfflatestimateparallelscan()
|
||||
{
|
||||
elog(INFO, "ivfflatestimateparallelscan");
|
||||
return 0;
|
||||
}
|
||||
|
||||
static void
|
||||
ivfflatinitparallelscan(void *target)
|
||||
{
|
||||
elog(INFO, "ivfflatinitparallelscan");
|
||||
}
|
||||
|
||||
static void
|
||||
ivfflatparallelrescan(IndexScanDesc scan)
|
||||
{
|
||||
elog(INFO, "ivfflatparallelrescan");
|
||||
}
|
||||
|
||||
/*
|
||||
* Define index handler
|
||||
*
|
||||
@@ -205,7 +223,7 @@ ivfflathandler(PG_FUNCTION_ARGS)
|
||||
amroutine->amstorage = false;
|
||||
amroutine->amclusterable = false;
|
||||
amroutine->ampredlocks = false;
|
||||
amroutine->amcanparallel = false;
|
||||
amroutine->amcanparallel = true;
|
||||
#if PG_VERSION_NUM >= 170000
|
||||
amroutine->amcanbuildparallel = true;
|
||||
#endif
|
||||
@@ -247,9 +265,9 @@ ivfflathandler(PG_FUNCTION_ARGS)
|
||||
amroutine->amrestrpos = NULL;
|
||||
|
||||
/* Interface functions to support parallel index scans */
|
||||
amroutine->amestimateparallelscan = NULL;
|
||||
amroutine->aminitparallelscan = NULL;
|
||||
amroutine->amparallelrescan = NULL;
|
||||
amroutine->amestimateparallelscan = ivfflatestimateparallelscan;
|
||||
amroutine->aminitparallelscan = ivfflatinitparallelscan;
|
||||
amroutine->amparallelrescan = ivfflatparallelrescan;
|
||||
|
||||
#if PG_VERSION_NUM >= 180000
|
||||
amroutine->amtranslatestrategy = NULL;
|
||||
|
||||
@@ -13,10 +13,6 @@
|
||||
#include "utils/tuplesort.h"
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
#if PG_VERSION_NUM >= 150000
|
||||
#include "common/pg_prng.h"
|
||||
#endif
|
||||
@@ -25,10 +21,6 @@
|
||||
#include "portability/instr_time.h"
|
||||
#endif
|
||||
|
||||
#if PG_VERSION_NUM >= 190000
|
||||
typedef Pointer Item;
|
||||
#endif
|
||||
|
||||
#define IVFFLAT_MAX_DIM 2000
|
||||
|
||||
/* Support functions */
|
||||
|
||||
@@ -2,16 +2,11 @@
|
||||
|
||||
#include <float.h>
|
||||
|
||||
#include "access/genam.h"
|
||||
#include "access/generic_xlog.h"
|
||||
#include "access/itup.h"
|
||||
#include "fmgr.h"
|
||||
#include "ivfflat.h"
|
||||
#include "nodes/execnodes.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "storage/lmgr.h"
|
||||
#include "utils/memutils.h"
|
||||
#include "utils/rel.h"
|
||||
|
||||
/*
|
||||
* Find the list that minimizes the distance function
|
||||
@@ -70,7 +65,7 @@ FindInsertPage(Relation index, Datum *values, BlockNumber *insertPage, ListInfo
|
||||
* Insert a tuple into the index
|
||||
*/
|
||||
static void
|
||||
InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid)
|
||||
InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
|
||||
{
|
||||
const IvfflatTypeInfo *typeInfo = IvfflatGetTypeInfo(index);
|
||||
IndexTuple itup;
|
||||
@@ -209,7 +204,7 @@ ivfflatinsert(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid,
|
||||
oldCtx = MemoryContextSwitchTo(insertCtx);
|
||||
|
||||
/* Insert tuple */
|
||||
InsertTuple(index, values, isnull, heap_tid);
|
||||
InsertTuple(index, values, isnull, heap_tid, heap);
|
||||
|
||||
/* Delete memory context */
|
||||
MemoryContextSwitchTo(oldCtx);
|
||||
|
||||
@@ -1,19 +1,17 @@
|
||||
#include "postgres.h"
|
||||
|
||||
#include <float.h>
|
||||
#include <limits.h>
|
||||
#include <math.h>
|
||||
|
||||
#include "access/genam.h"
|
||||
#include "fmgr.h"
|
||||
#include "bitvec.h"
|
||||
#include "halfutils.h"
|
||||
#include "halfvec.h"
|
||||
#include "ivfflat.h"
|
||||
#include "miscadmin.h"
|
||||
#include "utils/builtins.h"
|
||||
#include "utils/datum.h"
|
||||
#include "utils/memutils.h"
|
||||
#include "utils/relcache.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
#include "vector.h"
|
||||
|
||||
/*
|
||||
* Initialize with kmeans++
|
||||
|
||||
@@ -2,26 +2,15 @@
|
||||
|
||||
#include <float.h>
|
||||
|
||||
#include "access/genam.h"
|
||||
#include "access/itup.h"
|
||||
#include "access/relscan.h"
|
||||
#include "access/tupdesc.h"
|
||||
#include "catalog/pg_operator_d.h"
|
||||
#include "catalog/pg_type_d.h"
|
||||
#include "fmgr.h"
|
||||
#include "lib/pairingheap.h"
|
||||
#include "ivfflat.h"
|
||||
#include "miscadmin.h"
|
||||
#include "pgstat.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "utils/memutils.h"
|
||||
#include "utils/rel.h"
|
||||
#include "utils/snapmgr.h"
|
||||
#include "utils/tuplesort.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
#define GetScanList(ptr) pairingheap_container(IvfflatScanList, ph_node, ptr)
|
||||
#define GetScanListConst(ptr) pairingheap_const_container(IvfflatScanList, ph_node, ptr)
|
||||
@@ -130,6 +119,9 @@ GetScanItems(IndexScanDesc scan, Datum value)
|
||||
|
||||
tuplesort_reset(so->sortstate);
|
||||
|
||||
if (scan->parallel_scan != NULL)
|
||||
elog(INFO, "parallel scan");
|
||||
|
||||
/* Search closest probes lists */
|
||||
while (so->listIndex < so->maxProbes && (++batchProbes) <= so->probes)
|
||||
{
|
||||
@@ -366,10 +358,6 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
|
||||
|
||||
/* Count index scan for stats */
|
||||
pgstat_count_index_scan(scan->indexRelation);
|
||||
#if PG_VERSION_NUM >= 180000
|
||||
if (scan->instrument)
|
||||
scan->instrument->nsearches++;
|
||||
#endif
|
||||
|
||||
/* Safety check */
|
||||
if (scan->orderByData == NULL)
|
||||
|
||||
@@ -1,19 +1,13 @@
|
||||
#include "postgres.h"
|
||||
|
||||
#include "access/genam.h"
|
||||
#include "access/generic_xlog.h"
|
||||
#include "bitvec.h"
|
||||
#include "catalog/pg_type.h"
|
||||
#include "fmgr.h"
|
||||
#include "halfutils.h"
|
||||
#include "halfvec.h"
|
||||
#include "ivfflat.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "utils/relcache.h"
|
||||
#include "utils/varbit.h"
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
/*
|
||||
* Allocate a vector array
|
||||
|
||||
@@ -1,12 +1,9 @@
|
||||
#include "postgres.h"
|
||||
|
||||
#include "access/genam.h"
|
||||
#include "access/generic_xlog.h"
|
||||
#include "access/itup.h"
|
||||
#include "commands/vacuum.h"
|
||||
#include "ivfflat.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "utils/relcache.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 180000
|
||||
#define vacuum_delay_point() vacuum_delay_point(false)
|
||||
|
||||
@@ -5,27 +5,18 @@
|
||||
|
||||
#include "catalog/pg_type.h"
|
||||
#include "common/shortest_dec.h"
|
||||
#include "common/string.h"
|
||||
#include "fmgr.h"
|
||||
#include "halfutils.h"
|
||||
#include "halfvec.h"
|
||||
#include "lib/stringinfo.h"
|
||||
#include "libpq/pqformat.h"
|
||||
#include "sparsevec.h"
|
||||
#include "utils/array.h"
|
||||
#include "utils/builtins.h"
|
||||
#include "utils/float.h"
|
||||
#include "utils/fmgrprotos.h"
|
||||
#include "utils/lsyscache.h"
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
#if PG_VERSION_NUM >= 170000
|
||||
#include "parser/scansup.h"
|
||||
#endif
|
||||
|
||||
typedef struct SparseInputElement
|
||||
{
|
||||
int32 index;
|
||||
@@ -159,9 +150,9 @@ InitSparseVector(int dim, int nnz)
|
||||
return result;
|
||||
}
|
||||
|
||||
#if PG_VERSION_NUM >= 170000
|
||||
#define sparsevec_isspace(ch) scanner_isspace(ch)
|
||||
#else
|
||||
/*
|
||||
* Check for whitespace, since array_isspace() is static
|
||||
*/
|
||||
static inline bool
|
||||
sparsevec_isspace(char ch)
|
||||
{
|
||||
@@ -174,7 +165,6 @@ sparsevec_isspace(char ch)
|
||||
return true;
|
||||
return false;
|
||||
}
|
||||
#endif
|
||||
|
||||
/*
|
||||
* Compare indices
|
||||
|
||||
15
src/vector.c
15
src/vector.c
@@ -16,20 +16,16 @@
|
||||
#include "port.h" /* for strtof() */
|
||||
#include "sparsevec.h"
|
||||
#include "utils/array.h"
|
||||
#include "utils/builtins.h"
|
||||
#include "utils/float.h"
|
||||
#include "utils/fmgrprotos.h"
|
||||
#include "utils/lsyscache.h"
|
||||
#include "utils/varbit.h"
|
||||
#include "utils/numeric.h"
|
||||
#include "vector.h"
|
||||
|
||||
#if PG_VERSION_NUM >= 160000
|
||||
#include "varatt.h"
|
||||
#endif
|
||||
|
||||
#if PG_VERSION_NUM >= 170000
|
||||
#include "parser/scansup.h"
|
||||
#endif
|
||||
|
||||
#define STATE_DIMS(x) (ARR_DIMS(x)[0] - 1)
|
||||
#define CreateStateDatums(dim) palloc(sizeof(Datum) * (dim + 1))
|
||||
|
||||
@@ -133,9 +129,9 @@ InitVector(int dim)
|
||||
return result;
|
||||
}
|
||||
|
||||
#if PG_VERSION_NUM >= 170000
|
||||
#define vector_isspace(ch) scanner_isspace(ch)
|
||||
#else
|
||||
/*
|
||||
* Check for whitespace, since array_isspace() is static
|
||||
*/
|
||||
static inline bool
|
||||
vector_isspace(char ch)
|
||||
{
|
||||
@@ -148,7 +144,6 @@ vector_isspace(char ch)
|
||||
return true;
|
||||
return false;
|
||||
}
|
||||
#endif
|
||||
|
||||
/*
|
||||
* Check state array
|
||||
|
||||
16
test/sql/ivfflat_parallel.sql
Normal file
16
test/sql/ivfflat_parallel.sql
Normal file
@@ -0,0 +1,16 @@
|
||||
-- SET force_parallel_mode = on;
|
||||
SET parallel_setup_cost = 10;
|
||||
SET parallel_tuple_cost = 0.000001;
|
||||
SET min_parallel_table_scan_size = 1;
|
||||
SET min_parallel_index_scan_size = 1;
|
||||
|
||||
CREATE TABLE t (id integer, val vector(3));
|
||||
ALTER TABLE t ALTER COLUMN val SET STORAGE PLAIN;
|
||||
INSERT INTO t (id, val) SELECT n, ARRAY[random(), random(), random()] FROM generate_series(1,1000000) n;
|
||||
CREATE INDEX ON t USING ivfflat (val) WITH (lists = 10);
|
||||
SET ivfflat.probes = 4;
|
||||
|
||||
EXPLAIN SELECT * FROM t ORDER BY val <-> '[0.5,0.5,0.5]' LIMIT 5;
|
||||
SELECT * FROM t ORDER BY val <-> '[0.5,0.5,0.5]' LIMIT 5;
|
||||
|
||||
DROP TABLE t;
|
||||
@@ -16,19 +16,29 @@ $node->safe_psql("postgres",
|
||||
"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 100000) i;"
|
||||
);
|
||||
|
||||
# Add index
|
||||
$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops);");
|
||||
# Check each index type
|
||||
my @operators = ("<->", "<#>", "<=>");
|
||||
my @opclasses = ("vector_l2_ops", "vector_ip_ops", "vector_cosine_ops");
|
||||
|
||||
# Test 100% recall
|
||||
for (1 .. 20)
|
||||
for my $i (0 .. $#operators)
|
||||
{
|
||||
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;
|
||||
));
|
||||
is($res, $query);
|
||||
my $operator = $operators[$i];
|
||||
my $opclass = $opclasses[$i];
|
||||
|
||||
# Add index
|
||||
$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v $opclass);");
|
||||
|
||||
# Test 100% recall
|
||||
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;
|
||||
));
|
||||
is($res, $query);
|
||||
}
|
||||
}
|
||||
|
||||
done_testing();
|
||||
|
||||
Reference in New Issue
Block a user