mirror of
https://github.com/pgvector/pgvector.git
synced 2026-07-22 20:15:46 +08:00
Compare commits
44 Commits
parallel-i
...
ci-alpine
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
9250ddce87 | ||
|
|
bb431d531a | ||
|
|
2821648051 | ||
|
|
e583329be6 | ||
|
|
9d4fbeb7df | ||
|
|
eab825ae21 | ||
|
|
0d2ef735e5 | ||
|
|
1697c9f6d0 | ||
|
|
77814eb0ee | ||
|
|
1a381ffdcd | ||
|
|
250d453f8c | ||
|
|
498c23aa0c | ||
|
|
f09ef49dbf | ||
|
|
544686feb1 | ||
|
|
1a8ebb1129 | ||
|
|
aca89aec73 | ||
|
|
c9875f5150 | ||
|
|
df2c3d301f | ||
|
|
d557f1c9b1 | ||
|
|
3d6dd20469 | ||
|
|
c711da411c | ||
|
|
3975e212cc | ||
|
|
2b741ef068 | ||
|
|
f85a7d3fd0 | ||
|
|
5c25f81789 | ||
|
|
d72ee71f23 | ||
|
|
b7c0085afe | ||
|
|
2b742576b4 | ||
|
|
a126c02184 | ||
|
|
744362305b | ||
|
|
69e78e36a3 | ||
|
|
8dd37d4fb2 | ||
|
|
a24125839a | ||
|
|
2ee3113417 | ||
|
|
3ebb9a506e | ||
|
|
727d4836d2 | ||
|
|
d823c44591 | ||
|
|
8f4aa0751c | ||
|
|
1ff30720c7 | ||
|
|
4cef9213ec | ||
|
|
90add68d6f | ||
|
|
bc05dbf312 | ||
|
|
02fefc0e3f | ||
|
|
e7899b1250 |
246
.github/workflows/build.yml
vendored
246
.github/workflows/build.yml
vendored
@@ -1,120 +1,120 @@
|
|||||||
name: build
|
name: build
|
||||||
on: [push, pull_request]
|
on: [push, pull_request]
|
||||||
jobs:
|
jobs:
|
||||||
ubuntu:
|
# ubuntu:
|
||||||
runs-on: ${{ matrix.os }}
|
# runs-on: ${{ matrix.os }}
|
||||||
if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
# if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
||||||
strategy:
|
# strategy:
|
||||||
fail-fast: false
|
# fail-fast: false
|
||||||
matrix:
|
# matrix:
|
||||||
include:
|
# include:
|
||||||
- postgres: 19
|
# - postgres: 19
|
||||||
os: ubuntu-24.04
|
# os: ubuntu-24.04
|
||||||
- postgres: 18
|
# - postgres: 18
|
||||||
os: ubuntu-24.04
|
# os: ubuntu-24.04
|
||||||
- postgres: 17
|
# - postgres: 17
|
||||||
os: ubuntu-24.04
|
# os: ubuntu-24.04
|
||||||
- postgres: 16
|
# - postgres: 16
|
||||||
os: ubuntu-24.04-arm
|
# os: ubuntu-24.04-arm
|
||||||
- postgres: 15
|
# - postgres: 15
|
||||||
os: ubuntu-22.04
|
# os: ubuntu-22.04
|
||||||
- postgres: 14
|
# - postgres: 14
|
||||||
os: ubuntu-22.04-arm
|
# os: ubuntu-22.04-arm
|
||||||
- postgres: 13
|
# - postgres: 13
|
||||||
os: ubuntu-22.04
|
# os: ubuntu-22.04
|
||||||
steps:
|
# steps:
|
||||||
- uses: actions/checkout@v4
|
# - uses: actions/checkout@v5
|
||||||
- uses: ankane/setup-postgres@v1
|
# - uses: ankane/setup-postgres@v1
|
||||||
with:
|
# with:
|
||||||
postgres-version: ${{ matrix.postgres }}
|
# postgres-version: ${{ matrix.postgres }}
|
||||||
dev-files: true
|
# dev-files: true
|
||||||
- run: make
|
# - run: make
|
||||||
env:
|
# 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 ${{ matrix.postgres >= 18 && '-Wno-missing-field-initializers' || '' }}
|
||||||
- run: |
|
# - run: |
|
||||||
export PG_CONFIG=`which pg_config`
|
# export PG_CONFIG=`which pg_config`
|
||||||
sudo --preserve-env=PG_CONFIG make install
|
# sudo --preserve-env=PG_CONFIG make install
|
||||||
- run: make installcheck
|
# - run: make installcheck
|
||||||
- if: ${{ failure() }}
|
# - if: ${{ failure() }}
|
||||||
run: cat regression.diffs
|
# run: cat regression.diffs
|
||||||
- run: |
|
# - run: |
|
||||||
sudo apt-get update
|
# sudo apt-get update
|
||||||
sudo apt-get install libipc-run-perl
|
# sudo apt-get install libipc-run-perl
|
||||||
- run: make prove_installcheck
|
# - run: make prove_installcheck
|
||||||
mac:
|
# mac:
|
||||||
runs-on: ${{ matrix.os }}
|
# runs-on: ${{ matrix.os }}
|
||||||
if: ${{ !startsWith(github.ref_name, 'windows') }}
|
# if: ${{ !startsWith(github.ref_name, 'windows') }}
|
||||||
strategy:
|
# strategy:
|
||||||
fail-fast: false
|
# fail-fast: false
|
||||||
matrix:
|
# matrix:
|
||||||
include:
|
# include:
|
||||||
- postgres: 17
|
# - postgres: 18
|
||||||
os: macos-26
|
# os: macos-26
|
||||||
- postgres: 14
|
# - postgres: 14
|
||||||
os: macos-15-intel
|
# os: macos-15-intel
|
||||||
steps:
|
# steps:
|
||||||
- uses: actions/checkout@v4
|
# - uses: actions/checkout@v5
|
||||||
- uses: ankane/setup-postgres@v1
|
# - uses: ankane/setup-postgres@v1
|
||||||
with:
|
# with:
|
||||||
postgres-version: ${{ matrix.postgres }}
|
# postgres-version: ${{ matrix.postgres }}
|
||||||
- run: make
|
# - run: make
|
||||||
env:
|
# 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 -Wno-unknown-warning-option ${{ matrix.postgres >= 18 && '-Wno-missing-field-initializers' || '' }}
|
||||||
- run: make install
|
# - run: make install
|
||||||
- run: make installcheck
|
# - run: make installcheck
|
||||||
- if: ${{ failure() }}
|
# - if: ${{ failure() }}
|
||||||
run: cat regression.diffs
|
# run: cat regression.diffs
|
||||||
# Homebrew Postgres does not enable TAP tests, so need to download
|
# # Homebrew Postgres does not enable TAP tests, so need to download
|
||||||
- run: |
|
# - run: |
|
||||||
brew install cpanm
|
# brew install cpanm
|
||||||
cpanm --notest IPC::Run
|
# cpanm --notest IPC::Run
|
||||||
wget -q https://github.com/postgres/postgres/archive/refs/tags/$TAG.tar.gz
|
# wget -q https://github.com/postgres/postgres/archive/refs/tags/$TAG.tar.gz
|
||||||
tar xf $TAG.tar.gz
|
# tar xf $TAG.tar.gz
|
||||||
mv postgres-$TAG postgres
|
# mv postgres-$TAG postgres
|
||||||
env:
|
# env:
|
||||||
TAG: ${{ matrix.postgres == 17 && 'REL_17_6' || 'REL_14_19' }}
|
# TAG: ${{ matrix.postgres == 18 && 'REL_18_0' || 'REL_14_19' }}
|
||||||
- run: make prove_installcheck PROVE_FLAGS="-I ./postgres/src/test/perl -I ./test/perl"
|
# - run: make prove_installcheck PROVE_FLAGS="-I ./postgres/src/test/perl -I ./test/perl"
|
||||||
env:
|
# env:
|
||||||
PERL5LIB: /Users/runner/perl5/lib/perl5
|
# PERL5LIB: /Users/runner/perl5/lib/perl5
|
||||||
- run: make clean && $(brew --prefix llvm@$LLVM_VERSION)/bin/scan-build --status-bugs make
|
# - run: make clean && $(brew --prefix llvm@$LLVM_VERSION)/bin/scan-build --status-bugs make
|
||||||
env:
|
# env:
|
||||||
LLVM_VERSION: ${{ matrix.os == 'macos-26' && 20 || 18 }}
|
# LLVM_VERSION: ${{ matrix.os == 'macos-26' && 20 || 18 }}
|
||||||
PG_CFLAGS: -DUSE_ASSERT_CHECKING
|
# PG_CFLAGS: -DUSE_ASSERT_CHECKING
|
||||||
windows:
|
# windows:
|
||||||
runs-on: ${{ matrix.os }}
|
# runs-on: ${{ matrix.os }}
|
||||||
if: ${{ !startsWith(github.ref_name, 'mac') }}
|
# if: ${{ !startsWith(github.ref_name, 'mac') }}
|
||||||
strategy:
|
# strategy:
|
||||||
fail-fast: false
|
# fail-fast: false
|
||||||
matrix:
|
# matrix:
|
||||||
include:
|
# include:
|
||||||
- postgres: 17
|
# - postgres: 17
|
||||||
os: windows-2025
|
# os: windows-2025
|
||||||
- postgres: 14
|
# - postgres: 14
|
||||||
os: windows-2022
|
# os: windows-2022
|
||||||
steps:
|
# steps:
|
||||||
- uses: actions/checkout@v4
|
# - uses: actions/checkout@v5
|
||||||
- uses: ankane/setup-postgres@v1
|
# - uses: ankane/setup-postgres@v1
|
||||||
with:
|
# with:
|
||||||
postgres-version: ${{ matrix.postgres }}
|
# postgres-version: ${{ matrix.postgres }}
|
||||||
- run: |
|
# - run: |
|
||||||
call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvars64.bat" && ^
|
# call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvars64.bat" && ^
|
||||||
nmake /NOLOGO /F Makefile.win && ^
|
# nmake /NOLOGO /F Makefile.win && ^
|
||||||
nmake /NOLOGO /F Makefile.win install && ^
|
# 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 installcheck ${{ matrix.postgres != 17 && 'PG_REGRESS=$(PGROOT)\bin\pg_regress' || '' }} && ^
|
||||||
nmake /NOLOGO /F Makefile.win clean && ^
|
# nmake /NOLOGO /F Makefile.win clean && ^
|
||||||
nmake /NOLOGO /F Makefile.win uninstall
|
# nmake /NOLOGO /F Makefile.win uninstall
|
||||||
shell: cmd
|
# shell: cmd
|
||||||
- if: ${{ failure() }}
|
# - if: ${{ failure() }}
|
||||||
run: cat regression.diffs
|
# run: cat regression.diffs
|
||||||
i386:
|
i386:
|
||||||
if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
||||||
runs-on: ubuntu-latest
|
runs-on: ubuntu-latest
|
||||||
container:
|
container:
|
||||||
image: debian:12
|
image: alpine:3.23
|
||||||
options: --platform linux/386
|
options: --platform linux/386
|
||||||
steps:
|
steps:
|
||||||
- run: apt-get update && apt-get install -y build-essential git libipc-run-perl postgresql-15 postgresql-server-dev-15 sudo
|
- run: apk add --update build-base git openrc perl perl-ipc-run perl-test-harness-utils postgresql18 postgresql18-dev
|
||||||
- run: service postgresql start
|
- run: mkdir /run/openrc && touch /run/openrc/softlevel && rc-update add postgresql && rc-service postgresql start
|
||||||
- run: |
|
- run: |
|
||||||
git clone https://github.com/${{ github.repository }}.git pgvector
|
git clone https://github.com/${{ github.repository }}.git pgvector
|
||||||
cd pgvector
|
cd pgvector
|
||||||
@@ -123,21 +123,21 @@ jobs:
|
|||||||
make
|
make
|
||||||
make install
|
make install
|
||||||
chown -R postgres .
|
chown -R postgres .
|
||||||
sudo -u postgres make installcheck
|
su postgres -c "make installcheck"
|
||||||
sudo -u postgres make prove_installcheck
|
su postgres -c "make prove_installcheck"
|
||||||
env:
|
env:
|
||||||
PG_CFLAGS: -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 -Wno-missing-field-initializers
|
||||||
- if: ${{ failure() }}
|
- if: ${{ failure() }}
|
||||||
run: cat pgvector/regression.diffs
|
run: cat pgvector/regression.diffs
|
||||||
valgrind:
|
# valgrind:
|
||||||
if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
# if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
||||||
runs-on: ubuntu-latest
|
# runs-on: ubuntu-latest
|
||||||
steps:
|
# steps:
|
||||||
- uses: actions/checkout@v4
|
# - uses: actions/checkout@v5
|
||||||
- uses: ankane/setup-postgres-valgrind@v1
|
# - uses: ankane/setup-postgres-valgrind@v1
|
||||||
with:
|
# with:
|
||||||
postgres-version: 17
|
# postgres-version: 18
|
||||||
check-ub: yes
|
# check-ub: yes
|
||||||
- run: make OPTFLAGS=""
|
# - run: make OPTFLAGS=""
|
||||||
- run: sudo --preserve-env=PG_CONFIG make install
|
# - run: sudo --preserve-env=PG_CONFIG make install
|
||||||
- run: make installcheck
|
# - run: make installcheck
|
||||||
|
|||||||
@@ -1,3 +1,8 @@
|
|||||||
|
## 0.8.2 (unreleased)
|
||||||
|
|
||||||
|
- Improved `install` target on Windows
|
||||||
|
- Fixed `Index Searches` in `EXPLAIN` output for Postgres 18
|
||||||
|
|
||||||
## 0.8.1 (2025-09-04)
|
## 0.8.1 (2025-09-04)
|
||||||
|
|
||||||
- Added support for Postgres 18 rc1
|
- Added support for Postgres 18 rc1
|
||||||
|
|||||||
2
LICENSE
2
LICENSE
@@ -1,4 +1,4 @@
|
|||||||
Portions Copyright (c) 1996-2025, PostgreSQL Global Development Group
|
Portions Copyright (c) 1996-2026, PostgreSQL Global Development Group
|
||||||
|
|
||||||
Portions Copyright (c) 1994, The Regents of the University of California
|
Portions Copyright (c) 1994, The Regents of the University of California
|
||||||
|
|
||||||
|
|||||||
5
Makefile
5
Makefile
@@ -27,6 +27,11 @@ ifneq ($(filter ppc64%, $(shell uname -m)), )
|
|||||||
OPTFLAGS =
|
OPTFLAGS =
|
||||||
endif
|
endif
|
||||||
|
|
||||||
|
# RISC-V64 doesn't support -march=native
|
||||||
|
ifeq ($(shell uname -m), riscv64)
|
||||||
|
OPTFLAGS =
|
||||||
|
endif
|
||||||
|
|
||||||
# For auto-vectorization:
|
# For auto-vectorization:
|
||||||
# - GCC (needs -ftree-vectorize OR -O3) - https://gcc.gnu.org/projects/tree-ssa/vectorization.html
|
# - 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
|
# - Clang (could use pragma instead) - https://llvm.org/docs/Vectorizers.html
|
||||||
|
|||||||
@@ -57,7 +57,7 @@ install: all
|
|||||||
copy $(SHLIB) "$(PKGLIBDIR)"
|
copy $(SHLIB) "$(PKGLIBDIR)"
|
||||||
copy $(EXTENSION).control "$(SHAREDIR)\extension"
|
copy $(EXTENSION).control "$(SHAREDIR)\extension"
|
||||||
copy sql\$(EXTENSION)--*.sql "$(SHAREDIR)\extension"
|
copy sql\$(EXTENSION)--*.sql "$(SHAREDIR)\extension"
|
||||||
mkdir "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
|
if not exist "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)" mkdir "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
|
||||||
for %f in ($(HEADERS)) do copy %f "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
|
for %f in ($(HEADERS)) do copy %f "$(INCLUDEDIR_SERVER)\extension\$(EXTENSION)"
|
||||||
|
|
||||||
installcheck:
|
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
|
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), 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), [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).
|
||||||
|
|
||||||
### Windows
|
### 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 `x64 Native Tools Command Prompt for VS [version]` as administrator. Then use `nmake` to build:
|
||||||
|
|
||||||
```cmd
|
```cmd
|
||||||
set "PGROOT=C:\Program Files\PostgreSQL\17"
|
set "PGROOT=C:\Program Files\PostgreSQL\18"
|
||||||
cd %TEMP%
|
cd %TEMP%
|
||||||
git clone --branch v0.8.1 https://github.com/pgvector/pgvector.git
|
git clone --branch v0.8.1 https://github.com/pgvector/pgvector.git
|
||||||
cd pgvector
|
cd pgvector
|
||||||
@@ -695,10 +695,10 @@ CREATE INDEX CONCURRENTLY ...
|
|||||||
|
|
||||||
### Querying
|
### Querying
|
||||||
|
|
||||||
Use `EXPLAIN ANALYZE` to debug performance.
|
Use `EXPLAIN (ANALYZE, BUFFERS)` to debug performance.
|
||||||
|
|
||||||
```sql
|
```sql
|
||||||
EXPLAIN ANALYZE SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
|
EXPLAIN (ANALYZE, BUFFERS) SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
|
||||||
```
|
```
|
||||||
|
|
||||||
#### Exact Search
|
#### Exact Search
|
||||||
@@ -771,9 +771,12 @@ Use pgvector from any language with a Postgres client. You can even generate and
|
|||||||
|
|
||||||
Language | Libraries / Examples
|
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-c](https://github.com/pgvector/pgvector-c)
|
||||||
C++ | [pgvector-cpp](https://github.com/pgvector/pgvector-cpp)
|
C++ | [pgvector-cpp](https://github.com/pgvector/pgvector-cpp)
|
||||||
C#, F#, Visual Basic | [pgvector-dotnet](https://github.com/pgvector/pgvector-dotnet)
|
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)
|
Crystal | [pgvector-crystal](https://github.com/pgvector/pgvector-crystal)
|
||||||
D | [pgvector-d](https://github.com/pgvector/pgvector-d)
|
D | [pgvector-d](https://github.com/pgvector/pgvector-d)
|
||||||
Dart | [pgvector-dart](https://github.com/pgvector/pgvector-dart)
|
Dart | [pgvector-dart](https://github.com/pgvector/pgvector-dart)
|
||||||
@@ -790,14 +793,18 @@ Lisp | [pgvector-lisp](https://github.com/pgvector/pgvector-lisp)
|
|||||||
Lua | [pgvector-lua](https://github.com/pgvector/pgvector-lua)
|
Lua | [pgvector-lua](https://github.com/pgvector/pgvector-lua)
|
||||||
Nim | [pgvector-nim](https://github.com/pgvector/pgvector-nim)
|
Nim | [pgvector-nim](https://github.com/pgvector/pgvector-nim)
|
||||||
OCaml | [pgvector-ocaml](https://github.com/pgvector/pgvector-ocaml)
|
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)
|
Perl | [pgvector-perl](https://github.com/pgvector/pgvector-perl)
|
||||||
PHP | [pgvector-php](https://github.com/pgvector/pgvector-php)
|
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)
|
Python | [pgvector-python](https://github.com/pgvector/pgvector-python)
|
||||||
R | [pgvector-r](https://github.com/pgvector/pgvector-r)
|
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)
|
Raku | [pgvector-raku](https://github.com/pgvector/pgvector-raku)
|
||||||
Ruby | [pgvector-ruby](https://github.com/pgvector/pgvector-ruby), [Neighbor](https://github.com/ankane/neighbor)
|
Ruby | [pgvector-ruby](https://github.com/pgvector/pgvector-ruby), [Neighbor](https://github.com/ankane/neighbor)
|
||||||
Rust | [pgvector-rust](https://github.com/pgvector/pgvector-rust)
|
Rust | [pgvector-rust](https://github.com/pgvector/pgvector-rust)
|
||||||
Swift | [pgvector-swift](https://github.com/pgvector/pgvector-swift)
|
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)
|
Zig | [pgvector-zig](https://github.com/pgvector/pgvector-zig)
|
||||||
|
|
||||||
## Frequently Asked Questions
|
## Frequently Asked Questions
|
||||||
@@ -812,7 +819,7 @@ 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?
|
#### What if I want to index vectors with more than 2,000 dimensions?
|
||||||
|
|
||||||
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).
|
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).
|
||||||
|
|
||||||
#### Can I store vectors with different dimensions in the same column?
|
#### Can I store vectors with different dimensions in the same column?
|
||||||
|
|
||||||
@@ -1064,7 +1071,7 @@ 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:
|
If your machine has multiple Postgres installations, specify the path to [pg_config](https://www.postgresql.org/docs/current/app-pgconfig.html) with:
|
||||||
|
|
||||||
```sh
|
```sh
|
||||||
export PG_CONFIG=/Library/PostgreSQL/17/bin/pg_config
|
export PG_CONFIG=/Library/PostgreSQL/18/bin/pg_config
|
||||||
```
|
```
|
||||||
|
|
||||||
Then re-run the installation instructions (run `make clean` before `make` if needed). If `sudo` is needed for `make install`, use:
|
Then re-run the installation instructions (run `make clean` before `make` if needed). If `sudo` is needed for `make install`, use:
|
||||||
@@ -1075,11 +1082,11 @@ sudo --preserve-env=PG_CONFIG make install
|
|||||||
|
|
||||||
A few common paths on Mac are:
|
A few common paths on Mac are:
|
||||||
|
|
||||||
- EDB installer - `/Library/PostgreSQL/17/bin/pg_config`
|
- EDB installer - `/Library/PostgreSQL/18/bin/pg_config`
|
||||||
- Homebrew (arm64) - `/opt/homebrew/opt/postgresql@17/bin/pg_config`
|
- Homebrew (arm64) - `/opt/homebrew/opt/postgresql@18/bin/pg_config`
|
||||||
- Homebrew (x86-64) - `/usr/local/opt/postgresql@17/bin/pg_config`
|
- Homebrew (x86-64) - `/usr/local/opt/postgresql@18/bin/pg_config`
|
||||||
|
|
||||||
Note: Replace `17` with your Postgres server version
|
Note: Replace `18` with your Postgres server version
|
||||||
|
|
||||||
### Missing Header
|
### Missing Header
|
||||||
|
|
||||||
@@ -1088,10 +1095,10 @@ If compilation fails with `fatal error: postgres.h: No such file or directory`,
|
|||||||
For Ubuntu and Debian, use:
|
For Ubuntu and Debian, use:
|
||||||
|
|
||||||
```sh
|
```sh
|
||||||
sudo apt install postgresql-server-dev-17
|
sudo apt install postgresql-server-dev-18
|
||||||
```
|
```
|
||||||
|
|
||||||
Note: Replace `17` with your Postgres server version
|
Note: Replace `18` with your Postgres server version
|
||||||
|
|
||||||
### Missing SDK
|
### Missing SDK
|
||||||
|
|
||||||
@@ -1138,13 +1145,15 @@ If installation fails with `Access is denied`, re-run the installation instructi
|
|||||||
Get the [Docker image](https://hub.docker.com/r/pgvector/pgvector) with:
|
Get the [Docker image](https://hub.docker.com/r/pgvector/pgvector) with:
|
||||||
|
|
||||||
```sh
|
```sh
|
||||||
docker pull pgvector/pgvector:pg17-trixie
|
docker pull pgvector/pgvector:pg18-trixie
|
||||||
```
|
```
|
||||||
|
|
||||||
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).
|
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).
|
||||||
|
|
||||||
Supported tags are:
|
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-trixie`, `0.8.1-pg17-trixie`
|
||||||
- `pg17-bookworm`, `0.8.1-pg17-bookworm`, `pg17`, `0.8.1-pg17`
|
- `pg17-bookworm`, `0.8.1-pg17-bookworm`, `pg17`, `0.8.1-pg17`
|
||||||
- `pg16-trixie`, `0.8.1-pg16-trixie`
|
- `pg16-trixie`, `0.8.1-pg16-trixie`
|
||||||
@@ -1161,7 +1170,7 @@ You can also build the image manually:
|
|||||||
```sh
|
```sh
|
||||||
git clone --branch v0.8.1 https://github.com/pgvector/pgvector.git
|
git clone --branch v0.8.1 https://github.com/pgvector/pgvector.git
|
||||||
cd pgvector
|
cd pgvector
|
||||||
docker build --pull --build-arg PG_MAJOR=17 -t myuser/pgvector .
|
docker build --pull --build-arg PG_MAJOR=18 -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.
|
If you increase `maintenance_work_mem`, make sure `--shm-size` is at least that size to avoid an error with parallel HNSW index builds.
|
||||||
@@ -1178,7 +1187,7 @@ With Homebrew Postgres, you can use:
|
|||||||
brew install pgvector
|
brew install pgvector
|
||||||
```
|
```
|
||||||
|
|
||||||
Note: This only adds it to the `postgresql@17` and `postgresql@14` formulas
|
Note: This only adds it to the `postgresql@18` and `postgresql@17` formulas
|
||||||
|
|
||||||
### PGXN
|
### PGXN
|
||||||
|
|
||||||
@@ -1193,22 +1202,22 @@ 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:
|
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
|
```sh
|
||||||
sudo apt install postgresql-17-pgvector
|
sudo apt install postgresql-18-pgvector
|
||||||
```
|
```
|
||||||
|
|
||||||
Note: Replace `17` with your Postgres server version
|
Note: Replace `18` with your Postgres server version
|
||||||
|
|
||||||
### Yum
|
### 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:
|
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
|
```sh
|
||||||
sudo yum install pgvector_17
|
sudo yum install pgvector_18
|
||||||
# or
|
# or
|
||||||
sudo dnf install pgvector_17
|
sudo dnf install pgvector_18
|
||||||
```
|
```
|
||||||
|
|
||||||
Note: Replace `17` with your Postgres server version
|
Note: Replace `18` with your Postgres server version
|
||||||
|
|
||||||
### pkg
|
### pkg
|
||||||
|
|
||||||
@@ -1225,6 +1234,14 @@ cd /usr/ports/databases/pgvector
|
|||||||
make install
|
make install
|
||||||
```
|
```
|
||||||
|
|
||||||
|
### APK
|
||||||
|
|
||||||
|
Install the Alpine package with:
|
||||||
|
|
||||||
|
```sh
|
||||||
|
apk add postgresql-pgvector
|
||||||
|
```
|
||||||
|
|
||||||
### conda-forge
|
### conda-forge
|
||||||
|
|
||||||
With Conda Postgres, install from [conda-forge](https://anaconda.org/conda-forge/pgvector) with:
|
With Conda Postgres, install from [conda-forge](https://anaconda.org/conda-forge/pgvector) with:
|
||||||
|
|||||||
@@ -2,6 +2,7 @@
|
|||||||
|
|
||||||
#include "bitutils.h"
|
#include "bitutils.h"
|
||||||
#include "bitvec.h"
|
#include "bitvec.h"
|
||||||
|
#include "fmgr.h"
|
||||||
#include "utils/varbit.h"
|
#include "utils/varbit.h"
|
||||||
#include "vector.h"
|
#include "vector.h"
|
||||||
|
|
||||||
|
|||||||
@@ -1,5 +1,7 @@
|
|||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
|
#include <math.h>
|
||||||
|
|
||||||
#include "halfutils.h"
|
#include "halfutils.h"
|
||||||
#include "halfvec.h"
|
#include "halfvec.h"
|
||||||
|
|
||||||
|
|||||||
@@ -13,12 +13,20 @@
|
|||||||
#include "port.h" /* for strtof() */
|
#include "port.h" /* for strtof() */
|
||||||
#include "sparsevec.h"
|
#include "sparsevec.h"
|
||||||
#include "utils/array.h"
|
#include "utils/array.h"
|
||||||
#include "utils/builtins.h"
|
|
||||||
#include "utils/float.h"
|
#include "utils/float.h"
|
||||||
|
#include "utils/fmgrprotos.h"
|
||||||
#include "utils/lsyscache.h"
|
#include "utils/lsyscache.h"
|
||||||
#include "utils/numeric.h"
|
#include "utils/varbit.h"
|
||||||
#include "vector.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 STATE_DIMS(x) (ARR_DIMS(x)[0] - 1)
|
||||||
#define CreateStateDatums(dim) palloc(sizeof(Datum) * (dim + 1))
|
#define CreateStateDatums(dim) palloc(sizeof(Datum) * (dim + 1))
|
||||||
|
|
||||||
@@ -129,9 +137,9 @@ InitHalfVector(int dim)
|
|||||||
return result;
|
return result;
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
#if PG_VERSION_NUM >= 170000
|
||||||
* Check for whitespace, since array_isspace() is static
|
#define halfvec_isspace(ch) scanner_isspace(ch)
|
||||||
*/
|
#else
|
||||||
static inline bool
|
static inline bool
|
||||||
halfvec_isspace(char ch)
|
halfvec_isspace(char ch)
|
||||||
{
|
{
|
||||||
@@ -144,6 +152,7 @@ halfvec_isspace(char ch)
|
|||||||
return true;
|
return true;
|
||||||
return false;
|
return false;
|
||||||
}
|
}
|
||||||
|
#endif
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Check state array
|
* Check state array
|
||||||
|
|||||||
@@ -1,18 +1,24 @@
|
|||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
#include <float.h>
|
#include <float.h>
|
||||||
|
#include <limits.h>
|
||||||
#include <math.h>
|
#include <math.h>
|
||||||
|
|
||||||
#include "access/amapi.h"
|
#include "access/amapi.h"
|
||||||
|
#include "access/genam.h"
|
||||||
#include "access/reloptions.h"
|
#include "access/reloptions.h"
|
||||||
#include "commands/progress.h"
|
#include "commands/progress.h"
|
||||||
#include "commands/vacuum.h"
|
#include "commands/vacuum.h"
|
||||||
|
#include "fmgr.h"
|
||||||
#include "hnsw.h"
|
#include "hnsw.h"
|
||||||
#include "miscadmin.h"
|
#include "miscadmin.h"
|
||||||
|
#include "nodes/pg_list.h"
|
||||||
#include "utils/float.h"
|
#include "utils/float.h"
|
||||||
#include "utils/guc.h"
|
#include "utils/guc.h"
|
||||||
|
#include "utils/relcache.h"
|
||||||
#include "utils/selfuncs.h"
|
#include "utils/selfuncs.h"
|
||||||
#include "utils/spccache.h"
|
#include "utils/spccache.h"
|
||||||
|
#include "vector.h"
|
||||||
|
|
||||||
#if PG_VERSION_NUM < 150000
|
#if PG_VERSION_NUM < 150000
|
||||||
#define MarkGUCPrefixReserved(x) EmitWarningsOnPlaceholders(x)
|
#define MarkGUCPrefixReserved(x) EmitWarningsOnPlaceholders(x)
|
||||||
|
|||||||
@@ -3,6 +3,8 @@
|
|||||||
|
|
||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
|
#include <math.h>
|
||||||
|
|
||||||
#include "access/genam.h"
|
#include "access/genam.h"
|
||||||
#include "access/parallel.h"
|
#include "access/parallel.h"
|
||||||
#include "lib/pairingheap.h"
|
#include "lib/pairingheap.h"
|
||||||
@@ -12,6 +14,10 @@
|
|||||||
#include "utils/sampling.h"
|
#include "utils/sampling.h"
|
||||||
#include "vector.h"
|
#include "vector.h"
|
||||||
|
|
||||||
|
#if PG_VERSION_NUM >= 190000
|
||||||
|
typedef Pointer Item;
|
||||||
|
#endif
|
||||||
|
|
||||||
#define HNSW_MAX_DIM 2000
|
#define HNSW_MAX_DIM 2000
|
||||||
#define HNSW_MAX_NNZ 1000
|
#define HNSW_MAX_NNZ 1000
|
||||||
|
|
||||||
@@ -424,7 +430,7 @@ void *HnswAlloc(HnswAllocator * allocator, Size size);
|
|||||||
HnswElement HnswInitElement(char *base, ItemPointer tid, int m, double ml, int maxLevel, HnswAllocator * alloc);
|
HnswElement HnswInitElement(char *base, ItemPointer tid, int m, double ml, int maxLevel, HnswAllocator * alloc);
|
||||||
HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno);
|
HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno);
|
||||||
void HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint, Relation index, HnswSupport * support, int m, int efConstruction, bool existing);
|
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);
|
HnswSearchCandidate *HnswEntryCandidate(char *base, HnswElement entryPoint, HnswQuery * q, Relation index, HnswSupport * support, bool loadVec);
|
||||||
void HnswUpdateMetaPage(Relation index, int updateEntry, HnswElement entryPoint, BlockNumber insertPage, ForkNumber forkNum, bool building);
|
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 HnswSetNeighborTuple(char *base, HnswNeighborTuple ntup, HnswElement e, int m);
|
||||||
void HnswAddHeapTid(HnswElement element, ItemPointer heaptid);
|
void HnswAddHeapTid(HnswElement element, ItemPointer heaptid);
|
||||||
|
|||||||
@@ -36,11 +36,12 @@
|
|||||||
*/
|
*/
|
||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
#include <math.h>
|
#include "access/genam.h"
|
||||||
|
|
||||||
#include "access/parallel.h"
|
#include "access/parallel.h"
|
||||||
|
#include "access/relscan.h"
|
||||||
#include "access/table.h"
|
#include "access/table.h"
|
||||||
#include "access/tableam.h"
|
#include "access/tableam.h"
|
||||||
|
#include "access/tupdesc.h"
|
||||||
#include "access/xact.h"
|
#include "access/xact.h"
|
||||||
#include "access/xloginsert.h"
|
#include "access/xloginsert.h"
|
||||||
#include "catalog/index.h"
|
#include "catalog/index.h"
|
||||||
@@ -48,11 +49,18 @@
|
|||||||
#include "commands/progress.h"
|
#include "commands/progress.h"
|
||||||
#include "hnsw.h"
|
#include "hnsw.h"
|
||||||
#include "miscadmin.h"
|
#include "miscadmin.h"
|
||||||
|
#include "nodes/execnodes.h"
|
||||||
#include "optimizer/optimizer.h"
|
#include "optimizer/optimizer.h"
|
||||||
#include "storage/bufmgr.h"
|
#include "storage/bufmgr.h"
|
||||||
#include "tcop/tcopprot.h"
|
#include "tcop/tcopprot.h"
|
||||||
#include "utils/datum.h"
|
#include "utils/datum.h"
|
||||||
#include "utils/memutils.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
|
#if PG_VERSION_NUM >= 140000
|
||||||
#include "utils/backend_progress.h"
|
#include "utils/backend_progress.h"
|
||||||
@@ -398,7 +406,7 @@ UpdateNeighborsInMemory(char *base, HnswSupport * support, HnswElement e, int m)
|
|||||||
* Update graph in memory
|
* Update graph in memory
|
||||||
*/
|
*/
|
||||||
static void
|
static void
|
||||||
UpdateGraphInMemory(HnswSupport * support, HnswElement element, int m, int efConstruction, HnswElement entryPoint, HnswBuildState * buildstate)
|
UpdateGraphInMemory(HnswSupport * support, HnswElement element, int m, HnswElement entryPoint, HnswBuildState * buildstate)
|
||||||
{
|
{
|
||||||
HnswGraph *graph = buildstate->graph;
|
HnswGraph *graph = buildstate->graph;
|
||||||
char *base = buildstate->hnswarea;
|
char *base = buildstate->hnswarea;
|
||||||
@@ -460,7 +468,7 @@ InsertTupleInMemory(HnswBuildState * buildstate, HnswElement element)
|
|||||||
HnswFindElementNeighbors(base, element, entryPoint, NULL, support, m, efConstruction, false);
|
HnswFindElementNeighbors(base, element, entryPoint, NULL, support, m, efConstruction, false);
|
||||||
|
|
||||||
/* Update graph in memory */
|
/* Update graph in memory */
|
||||||
UpdateGraphInMemory(support, element, m, efConstruction, entryPoint, buildstate);
|
UpdateGraphInMemory(support, element, m, entryPoint, buildstate);
|
||||||
|
|
||||||
/* Release entry lock */
|
/* Release entry lock */
|
||||||
LWLockRelease(entryLock);
|
LWLockRelease(entryLock);
|
||||||
@@ -545,7 +553,7 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heaptid, Hn
|
|||||||
|
|
||||||
/* Copy the datum */
|
/* Copy the datum */
|
||||||
memcpy(valuePtr, DatumGetPointer(value), valueSize);
|
memcpy(valuePtr, DatumGetPointer(value), valueSize);
|
||||||
HnswPtrStore(base, element->value, valuePtr);
|
HnswPtrStore(base, element->value, (char *) valuePtr);
|
||||||
|
|
||||||
/* Create a lock for the element */
|
/* Create a lock for the element */
|
||||||
LWLockInitialize(&element->lock, hnsw_lock_tranche_id);
|
LWLockInitialize(&element->lock, hnsw_lock_tranche_id);
|
||||||
@@ -1054,7 +1062,7 @@ ComputeParallelWorkers(Relation heap, Relation index)
|
|||||||
* Build graph
|
* Build graph
|
||||||
*/
|
*/
|
||||||
static void
|
static void
|
||||||
BuildGraph(HnswBuildState * buildstate, ForkNumber forkNum)
|
BuildGraph(HnswBuildState * buildstate)
|
||||||
{
|
{
|
||||||
int parallel_workers = 0;
|
int parallel_workers = 0;
|
||||||
|
|
||||||
@@ -1102,7 +1110,7 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
|
|||||||
|
|
||||||
InitBuildState(buildstate, heap, index, indexInfo, forkNum);
|
InitBuildState(buildstate, heap, index, indexInfo, forkNum);
|
||||||
|
|
||||||
BuildGraph(buildstate, forkNum);
|
BuildGraph(buildstate);
|
||||||
|
|
||||||
if (RelationNeedsWAL(index) || forkNum == INIT_FORKNUM)
|
if (RelationNeedsWAL(index) || forkNum == INIT_FORKNUM)
|
||||||
log_newpage_range(index, forkNum, 0, RelationGetNumberOfBlocksInFork(index, forkNum), true);
|
log_newpage_range(index, forkNum, 0, RelationGetNumberOfBlocksInFork(index, forkNum), true);
|
||||||
|
|||||||
@@ -1,13 +1,18 @@
|
|||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
#include <math.h>
|
#include "access/genam.h"
|
||||||
|
|
||||||
#include "access/generic_xlog.h"
|
#include "access/generic_xlog.h"
|
||||||
#include "hnsw.h"
|
#include "hnsw.h"
|
||||||
|
#include "nodes/execnodes.h"
|
||||||
#include "storage/bufmgr.h"
|
#include "storage/bufmgr.h"
|
||||||
#include "storage/lmgr.h"
|
#include "storage/lmgr.h"
|
||||||
#include "utils/datum.h"
|
#include "utils/datum.h"
|
||||||
#include "utils/memutils.h"
|
#include "utils/memutils.h"
|
||||||
|
#include "utils/rel.h"
|
||||||
|
|
||||||
|
#if PG_VERSION_NUM >= 160000
|
||||||
|
#include "varatt.h"
|
||||||
|
#endif
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Get the insert page
|
* Get the insert page
|
||||||
@@ -660,7 +665,7 @@ FindDuplicateOnDisk(Relation index, HnswElement element, bool building)
|
|||||||
* Update graph on disk
|
* Update graph on disk
|
||||||
*/
|
*/
|
||||||
static void
|
static void
|
||||||
UpdateGraphOnDisk(Relation index, HnswSupport * support, HnswElement element, int m, int efConstruction, HnswElement entryPoint, bool building)
|
UpdateGraphOnDisk(Relation index, HnswSupport * support, HnswElement element, int m, HnswElement entryPoint, bool building)
|
||||||
{
|
{
|
||||||
BlockNumber newInsertPage = InvalidBlockNumber;
|
BlockNumber newInsertPage = InvalidBlockNumber;
|
||||||
|
|
||||||
@@ -708,7 +713,7 @@ HnswInsertTupleOnDisk(Relation index, HnswSupport * support, Datum value, ItemPo
|
|||||||
|
|
||||||
/* Create an element */
|
/* Create an element */
|
||||||
element = HnswInitElement(base, heaptid, m, HnswGetMl(m), HnswGetMaxLevel(m), NULL);
|
element = HnswInitElement(base, heaptid, m, HnswGetMl(m), HnswGetMaxLevel(m), NULL);
|
||||||
HnswPtrStore(base, element->value, DatumGetPointer(value));
|
HnswPtrStore(base, element->value, (char *) DatumGetPointer(value));
|
||||||
|
|
||||||
/* Prevent concurrent inserts when likely updating entry point */
|
/* Prevent concurrent inserts when likely updating entry point */
|
||||||
if (entryPoint == NULL || element->level > entryPoint->level)
|
if (entryPoint == NULL || element->level > entryPoint->level)
|
||||||
@@ -728,7 +733,7 @@ HnswInsertTupleOnDisk(Relation index, HnswSupport * support, Datum value, ItemPo
|
|||||||
HnswFindElementNeighbors(base, element, entryPoint, index, support, m, efConstruction, false);
|
HnswFindElementNeighbors(base, element, entryPoint, index, support, m, efConstruction, false);
|
||||||
|
|
||||||
/* Update graph on disk */
|
/* Update graph on disk */
|
||||||
UpdateGraphOnDisk(index, support, element, m, efConstruction, entryPoint, building);
|
UpdateGraphOnDisk(index, support, element, m, entryPoint, building);
|
||||||
|
|
||||||
/* Release lock */
|
/* Release lock */
|
||||||
UnlockPage(index, HNSW_UPDATE_LOCK, lockmode);
|
UnlockPage(index, HNSW_UPDATE_LOCK, lockmode);
|
||||||
|
|||||||
@@ -1,12 +1,21 @@
|
|||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
|
#include "access/genam.h"
|
||||||
#include "access/relscan.h"
|
#include "access/relscan.h"
|
||||||
#include "hnsw.h"
|
#include "hnsw.h"
|
||||||
|
#include "lib/pairingheap.h"
|
||||||
|
#include "miscadmin.h"
|
||||||
|
#include "nodes/pg_list.h"
|
||||||
#include "pgstat.h"
|
#include "pgstat.h"
|
||||||
#include "storage/bufmgr.h"
|
|
||||||
#include "storage/lmgr.h"
|
#include "storage/lmgr.h"
|
||||||
#include "utils/float.h"
|
#include "utils/float.h"
|
||||||
#include "utils/memutils.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
|
* Algorithm 5 from paper
|
||||||
@@ -193,6 +202,10 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
|
|||||||
|
|
||||||
/* Count index scan for stats */
|
/* Count index scan for stats */
|
||||||
pgstat_count_index_scan(scan->indexRelation);
|
pgstat_count_index_scan(scan->indexRelation);
|
||||||
|
#if PG_VERSION_NUM >= 180000
|
||||||
|
if (scan->instrument)
|
||||||
|
scan->instrument->nsearches++;
|
||||||
|
#endif
|
||||||
|
|
||||||
/* Safety check */
|
/* Safety check */
|
||||||
if (scan->orderByData == NULL)
|
if (scan->orderByData == NULL)
|
||||||
|
|||||||
@@ -2,18 +2,24 @@
|
|||||||
|
|
||||||
#include <math.h>
|
#include <math.h>
|
||||||
|
|
||||||
|
#include "access/genam.h"
|
||||||
#include "access/generic_xlog.h"
|
#include "access/generic_xlog.h"
|
||||||
#include "catalog/pg_type.h"
|
|
||||||
#include "catalog/pg_type_d.h"
|
|
||||||
#include "common/hashfn.h"
|
#include "common/hashfn.h"
|
||||||
#include "fmgr.h"
|
#include "fmgr.h"
|
||||||
#include "hnsw.h"
|
#include "hnsw.h"
|
||||||
#include "lib/pairingheap.h"
|
#include "lib/pairingheap.h"
|
||||||
|
#include "nodes/pg_list.h"
|
||||||
|
#include "port/atomics.h"
|
||||||
#include "sparsevec.h"
|
#include "sparsevec.h"
|
||||||
#include "storage/bufmgr.h"
|
#include "storage/bufmgr.h"
|
||||||
#include "utils/datum.h"
|
#include "utils/datum.h"
|
||||||
#include "utils/memdebug.h"
|
#include "utils/memdebug.h"
|
||||||
#include "utils/rel.h"
|
#include "utils/rel.h"
|
||||||
|
#include "vector.h"
|
||||||
|
|
||||||
|
#if PG_VERSION_NUM >= 160000
|
||||||
|
#include "varatt.h"
|
||||||
|
#endif
|
||||||
|
|
||||||
#if PG_VERSION_NUM < 170000
|
#if PG_VERSION_NUM < 170000
|
||||||
static inline uint64
|
static inline uint64
|
||||||
@@ -256,7 +262,7 @@ HnswInitElement(char *base, ItemPointer heaptid, int m, double ml, int maxLevel,
|
|||||||
|
|
||||||
HnswInitNeighbors(base, element, m, allocator);
|
HnswInitNeighbors(base, element, m, allocator);
|
||||||
|
|
||||||
HnswPtrStore(base, element->value, (Pointer) NULL);
|
HnswPtrStore(base, element->value, (char *) NULL);
|
||||||
|
|
||||||
return element;
|
return element;
|
||||||
}
|
}
|
||||||
@@ -282,7 +288,7 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
|
|||||||
element->blkno = blkno;
|
element->blkno = blkno;
|
||||||
element->offno = offno;
|
element->offno = offno;
|
||||||
HnswPtrStore(base, element->neighbors, (HnswNeighborArrayPtr *) NULL);
|
HnswPtrStore(base, element->neighbors, (HnswNeighborArrayPtr *) NULL);
|
||||||
HnswPtrStore(base, element->value, (Pointer) NULL);
|
HnswPtrStore(base, element->value, (char *) NULL);
|
||||||
return element;
|
return element;
|
||||||
}
|
}
|
||||||
|
|
||||||
@@ -508,7 +514,7 @@ HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHe
|
|||||||
char *base = NULL;
|
char *base = NULL;
|
||||||
Datum value = datumCopy(PointerGetDatum(&etup->data), false, -1);
|
Datum value = datumCopy(PointerGetDatum(&etup->data), false, -1);
|
||||||
|
|
||||||
HnswPtrStore(base, element->value, DatumGetPointer(value));
|
HnswPtrStore(base, element->value, (char *) DatumGetPointer(value));
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
@@ -922,7 +928,7 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
|
|||||||
continue;
|
continue;
|
||||||
}
|
}
|
||||||
|
|
||||||
if (eElement == NULL || !(eDistance < f->distance || alwaysAdd))
|
if (!(eDistance < f->distance || alwaysAdd))
|
||||||
{
|
{
|
||||||
if (discarded != NULL)
|
if (discarded != NULL)
|
||||||
{
|
{
|
||||||
|
|||||||
@@ -1,13 +1,18 @@
|
|||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
#include <math.h>
|
#include "access/genam.h"
|
||||||
|
|
||||||
#include "access/generic_xlog.h"
|
#include "access/generic_xlog.h"
|
||||||
#include "commands/vacuum.h"
|
#include "commands/vacuum.h"
|
||||||
#include "hnsw.h"
|
#include "hnsw.h"
|
||||||
|
#include "nodes/pg_list.h"
|
||||||
#include "storage/bufmgr.h"
|
#include "storage/bufmgr.h"
|
||||||
#include "storage/lmgr.h"
|
#include "storage/lmgr.h"
|
||||||
#include "utils/memutils.h"
|
#include "utils/memutils.h"
|
||||||
|
#include "utils/rel.h"
|
||||||
|
|
||||||
|
#if PG_VERSION_NUM >= 160000
|
||||||
|
#include "varatt.h"
|
||||||
|
#endif
|
||||||
|
|
||||||
#if PG_VERSION_NUM >= 180000
|
#if PG_VERSION_NUM >= 180000
|
||||||
#define vacuum_delay_point() vacuum_delay_point(false)
|
#define vacuum_delay_point() vacuum_delay_point(false)
|
||||||
@@ -174,7 +179,12 @@ NeedsUpdated(HnswVacuumState * vacuumstate, HnswElement element)
|
|||||||
/* Also update if layer 0 is not full */
|
/* Also update if layer 0 is not full */
|
||||||
/* This could indicate too many candidates being deleted during insert */
|
/* This could indicate too many candidates being deleted during insert */
|
||||||
if (!needsUpdated)
|
if (!needsUpdated)
|
||||||
|
{
|
||||||
|
/* Keep clang-tidy happy */
|
||||||
|
Assert(ntup->count > 0);
|
||||||
|
|
||||||
needsUpdated = !ItemPointerIsValid(&ntup->indextids[ntup->count - 1]);
|
needsUpdated = !ItemPointerIsValid(&ntup->indextids[ntup->count - 1]);
|
||||||
|
}
|
||||||
|
|
||||||
UnlockReleaseBuffer(buf);
|
UnlockReleaseBuffer(buf);
|
||||||
|
|
||||||
@@ -524,8 +534,9 @@ MarkDeleted(HnswVacuumState * vacuumstate)
|
|||||||
ntup = (HnswNeighborTuple) PageGetItem(npage, PageGetItemId(npage, neighborOffno));
|
ntup = (HnswNeighborTuple) PageGetItem(npage, PageGetItemId(npage, neighborOffno));
|
||||||
|
|
||||||
/* Overwrite element */
|
/* Overwrite element */
|
||||||
|
/* Use memset instead of MemSet to keep clang-tidy happy */
|
||||||
etup->deleted = 1;
|
etup->deleted = 1;
|
||||||
MemSet(&etup->data, 0, VARSIZE_ANY(&etup->data));
|
memset(&etup->data, 0, VARSIZE_ANY(&etup->data));
|
||||||
|
|
||||||
/* Overwrite neighbors */
|
/* Overwrite neighbors */
|
||||||
for (int i = 0; i < ntup->count; i++)
|
for (int i = 0; i < ntup->count; i++)
|
||||||
|
|||||||
@@ -2,23 +2,36 @@
|
|||||||
|
|
||||||
#include <float.h>
|
#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/table.h"
|
||||||
#include "access/tableam.h"
|
#include "access/tableam.h"
|
||||||
|
#include "access/tupdesc.h"
|
||||||
#include "access/parallel.h"
|
#include "access/parallel.h"
|
||||||
#include "access/xact.h"
|
#include "access/xact.h"
|
||||||
#include "bitvec.h"
|
#include "access/xloginsert.h"
|
||||||
#include "catalog/index.h"
|
#include "catalog/index.h"
|
||||||
#include "catalog/pg_operator_d.h"
|
#include "catalog/pg_operator_d.h"
|
||||||
#include "catalog/pg_type_d.h"
|
#include "catalog/pg_type_d.h"
|
||||||
#include "commands/progress.h"
|
#include "commands/progress.h"
|
||||||
#include "halfvec.h"
|
#include "fmgr.h"
|
||||||
#include "ivfflat.h"
|
#include "ivfflat.h"
|
||||||
#include "miscadmin.h"
|
#include "miscadmin.h"
|
||||||
|
#include "nodes/execnodes.h"
|
||||||
#include "optimizer/optimizer.h"
|
#include "optimizer/optimizer.h"
|
||||||
#include "storage/bufmgr.h"
|
#include "storage/bufmgr.h"
|
||||||
#include "tcop/tcopprot.h"
|
#include "tcop/tcopprot.h"
|
||||||
#include "utils/memutils.h"
|
#include "utils/memutils.h"
|
||||||
#include "vector.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
|
||||||
|
|
||||||
#if PG_VERSION_NUM >= 140000
|
#if PG_VERSION_NUM >= 140000
|
||||||
#include "utils/backend_progress.h"
|
#include "utils/backend_progress.h"
|
||||||
@@ -138,7 +151,7 @@ SampleRows(IvfflatBuildState * buildstate)
|
|||||||
* Add tuple to sort
|
* Add tuple to sort
|
||||||
*/
|
*/
|
||||||
static void
|
static void
|
||||||
AddTupleToSort(Relation index, ItemPointer tid, Datum *values, IvfflatBuildState * buildstate)
|
AddTupleToSort(ItemPointer tid, Datum *values, IvfflatBuildState * buildstate)
|
||||||
{
|
{
|
||||||
double distance;
|
double distance;
|
||||||
double minDistance = DBL_MAX;
|
double minDistance = DBL_MAX;
|
||||||
@@ -215,7 +228,7 @@ BuildCallback(Relation index, ItemPointer tid, Datum *values,
|
|||||||
oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
|
oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
|
||||||
|
|
||||||
/* Add tuple to sort */
|
/* Add tuple to sort */
|
||||||
AddTupleToSort(index, tid, values, buildstate);
|
AddTupleToSort(tid, values, buildstate);
|
||||||
|
|
||||||
/* Reset memory context */
|
/* Reset memory context */
|
||||||
MemoryContextSwitchTo(oldCtx);
|
MemoryContextSwitchTo(oldCtx);
|
||||||
@@ -470,8 +483,8 @@ CreateMetaPage(Relation index, int dimensions, int lists, ForkNumber forkNum)
|
|||||||
* Create list pages
|
* Create list pages
|
||||||
*/
|
*/
|
||||||
static void
|
static void
|
||||||
CreateListPages(Relation index, VectorArray centers, int dimensions,
|
CreateListPages(Relation index, VectorArray centers, int lists,
|
||||||
int lists, ForkNumber forkNum, ListInfo * *listInfo)
|
ForkNumber forkNum, ListInfo * *listInfo)
|
||||||
{
|
{
|
||||||
Buffer buf;
|
Buffer buf;
|
||||||
Page page;
|
Page page;
|
||||||
@@ -1004,7 +1017,7 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
|
|||||||
|
|
||||||
/* Create pages */
|
/* Create pages */
|
||||||
CreateMetaPage(index, buildstate->dimensions, buildstate->lists, forkNum);
|
CreateMetaPage(index, buildstate->dimensions, buildstate->lists, forkNum);
|
||||||
CreateListPages(index, buildstate->centers, buildstate->dimensions, buildstate->lists, forkNum, &buildstate->listInfo);
|
CreateListPages(index, buildstate->centers, buildstate->lists, forkNum, &buildstate->listInfo);
|
||||||
CreateEntryPages(buildstate, forkNum);
|
CreateEntryPages(buildstate, forkNum);
|
||||||
|
|
||||||
/* Write WAL for initialization fork since GenericXLog functions do not */
|
/* Write WAL for initialization fork since GenericXLog functions do not */
|
||||||
|
|||||||
@@ -3,14 +3,19 @@
|
|||||||
#include <float.h>
|
#include <float.h>
|
||||||
|
|
||||||
#include "access/amapi.h"
|
#include "access/amapi.h"
|
||||||
|
#include "access/genam.h"
|
||||||
#include "access/reloptions.h"
|
#include "access/reloptions.h"
|
||||||
#include "commands/progress.h"
|
#include "commands/progress.h"
|
||||||
#include "commands/vacuum.h"
|
#include "commands/vacuum.h"
|
||||||
|
#include "fmgr.h"
|
||||||
#include "ivfflat.h"
|
#include "ivfflat.h"
|
||||||
|
#include "nodes/pg_list.h"
|
||||||
#include "utils/float.h"
|
#include "utils/float.h"
|
||||||
#include "utils/guc.h"
|
#include "utils/guc.h"
|
||||||
|
#include "utils/relcache.h"
|
||||||
#include "utils/selfuncs.h"
|
#include "utils/selfuncs.h"
|
||||||
#include "utils/spccache.h"
|
#include "utils/spccache.h"
|
||||||
|
#include "vector.h"
|
||||||
|
|
||||||
#if PG_VERSION_NUM < 150000
|
#if PG_VERSION_NUM < 150000
|
||||||
#define MarkGUCPrefixReserved(x) EmitWarningsOnPlaceholders(x)
|
#define MarkGUCPrefixReserved(x) EmitWarningsOnPlaceholders(x)
|
||||||
@@ -143,10 +148,6 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
|
|||||||
*indexSelectivity = costs.indexSelectivity;
|
*indexSelectivity = costs.indexSelectivity;
|
||||||
*indexCorrelation = costs.indexCorrelation;
|
*indexCorrelation = costs.indexCorrelation;
|
||||||
*indexPages = costs.numIndexPages;
|
*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);
|
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
@@ -174,25 +175,6 @@ ivfflatvalidate(Oid opclassoid)
|
|||||||
return true;
|
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
|
* Define index handler
|
||||||
*
|
*
|
||||||
@@ -223,7 +205,7 @@ ivfflathandler(PG_FUNCTION_ARGS)
|
|||||||
amroutine->amstorage = false;
|
amroutine->amstorage = false;
|
||||||
amroutine->amclusterable = false;
|
amroutine->amclusterable = false;
|
||||||
amroutine->ampredlocks = false;
|
amroutine->ampredlocks = false;
|
||||||
amroutine->amcanparallel = true;
|
amroutine->amcanparallel = false;
|
||||||
#if PG_VERSION_NUM >= 170000
|
#if PG_VERSION_NUM >= 170000
|
||||||
amroutine->amcanbuildparallel = true;
|
amroutine->amcanbuildparallel = true;
|
||||||
#endif
|
#endif
|
||||||
@@ -265,9 +247,9 @@ ivfflathandler(PG_FUNCTION_ARGS)
|
|||||||
amroutine->amrestrpos = NULL;
|
amroutine->amrestrpos = NULL;
|
||||||
|
|
||||||
/* Interface functions to support parallel index scans */
|
/* Interface functions to support parallel index scans */
|
||||||
amroutine->amestimateparallelscan = ivfflatestimateparallelscan;
|
amroutine->amestimateparallelscan = NULL;
|
||||||
amroutine->aminitparallelscan = ivfflatinitparallelscan;
|
amroutine->aminitparallelscan = NULL;
|
||||||
amroutine->amparallelrescan = ivfflatparallelrescan;
|
amroutine->amparallelrescan = NULL;
|
||||||
|
|
||||||
#if PG_VERSION_NUM >= 180000
|
#if PG_VERSION_NUM >= 180000
|
||||||
amroutine->amtranslatestrategy = NULL;
|
amroutine->amtranslatestrategy = NULL;
|
||||||
|
|||||||
@@ -13,6 +13,10 @@
|
|||||||
#include "utils/tuplesort.h"
|
#include "utils/tuplesort.h"
|
||||||
#include "vector.h"
|
#include "vector.h"
|
||||||
|
|
||||||
|
#if PG_VERSION_NUM >= 160000
|
||||||
|
#include "varatt.h"
|
||||||
|
#endif
|
||||||
|
|
||||||
#if PG_VERSION_NUM >= 150000
|
#if PG_VERSION_NUM >= 150000
|
||||||
#include "common/pg_prng.h"
|
#include "common/pg_prng.h"
|
||||||
#endif
|
#endif
|
||||||
@@ -21,6 +25,10 @@
|
|||||||
#include "portability/instr_time.h"
|
#include "portability/instr_time.h"
|
||||||
#endif
|
#endif
|
||||||
|
|
||||||
|
#if PG_VERSION_NUM >= 190000
|
||||||
|
typedef Pointer Item;
|
||||||
|
#endif
|
||||||
|
|
||||||
#define IVFFLAT_MAX_DIM 2000
|
#define IVFFLAT_MAX_DIM 2000
|
||||||
|
|
||||||
/* Support functions */
|
/* Support functions */
|
||||||
|
|||||||
@@ -2,11 +2,16 @@
|
|||||||
|
|
||||||
#include <float.h>
|
#include <float.h>
|
||||||
|
|
||||||
|
#include "access/genam.h"
|
||||||
#include "access/generic_xlog.h"
|
#include "access/generic_xlog.h"
|
||||||
|
#include "access/itup.h"
|
||||||
|
#include "fmgr.h"
|
||||||
#include "ivfflat.h"
|
#include "ivfflat.h"
|
||||||
|
#include "nodes/execnodes.h"
|
||||||
#include "storage/bufmgr.h"
|
#include "storage/bufmgr.h"
|
||||||
#include "storage/lmgr.h"
|
#include "storage/lmgr.h"
|
||||||
#include "utils/memutils.h"
|
#include "utils/memutils.h"
|
||||||
|
#include "utils/rel.h"
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Find the list that minimizes the distance function
|
* Find the list that minimizes the distance function
|
||||||
@@ -65,7 +70,7 @@ FindInsertPage(Relation index, Datum *values, BlockNumber *insertPage, ListInfo
|
|||||||
* Insert a tuple into the index
|
* Insert a tuple into the index
|
||||||
*/
|
*/
|
||||||
static void
|
static void
|
||||||
InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
|
InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid)
|
||||||
{
|
{
|
||||||
const IvfflatTypeInfo *typeInfo = IvfflatGetTypeInfo(index);
|
const IvfflatTypeInfo *typeInfo = IvfflatGetTypeInfo(index);
|
||||||
IndexTuple itup;
|
IndexTuple itup;
|
||||||
@@ -204,7 +209,7 @@ ivfflatinsert(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid,
|
|||||||
oldCtx = MemoryContextSwitchTo(insertCtx);
|
oldCtx = MemoryContextSwitchTo(insertCtx);
|
||||||
|
|
||||||
/* Insert tuple */
|
/* Insert tuple */
|
||||||
InsertTuple(index, values, isnull, heap_tid, heap);
|
InsertTuple(index, values, isnull, heap_tid);
|
||||||
|
|
||||||
/* Delete memory context */
|
/* Delete memory context */
|
||||||
MemoryContextSwitchTo(oldCtx);
|
MemoryContextSwitchTo(oldCtx);
|
||||||
|
|||||||
@@ -1,17 +1,19 @@
|
|||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
#include <float.h>
|
#include <float.h>
|
||||||
|
#include <limits.h>
|
||||||
#include <math.h>
|
#include <math.h>
|
||||||
|
|
||||||
#include "bitvec.h"
|
#include "access/genam.h"
|
||||||
#include "halfutils.h"
|
#include "fmgr.h"
|
||||||
#include "halfvec.h"
|
|
||||||
#include "ivfflat.h"
|
#include "ivfflat.h"
|
||||||
#include "miscadmin.h"
|
#include "miscadmin.h"
|
||||||
#include "utils/builtins.h"
|
|
||||||
#include "utils/datum.h"
|
|
||||||
#include "utils/memutils.h"
|
#include "utils/memutils.h"
|
||||||
#include "vector.h"
|
#include "utils/relcache.h"
|
||||||
|
|
||||||
|
#if PG_VERSION_NUM >= 160000
|
||||||
|
#include "varatt.h"
|
||||||
|
#endif
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Initialize with kmeans++
|
* Initialize with kmeans++
|
||||||
|
|||||||
@@ -2,15 +2,26 @@
|
|||||||
|
|
||||||
#include <float.h>
|
#include <float.h>
|
||||||
|
|
||||||
|
#include "access/genam.h"
|
||||||
|
#include "access/itup.h"
|
||||||
#include "access/relscan.h"
|
#include "access/relscan.h"
|
||||||
|
#include "access/tupdesc.h"
|
||||||
#include "catalog/pg_operator_d.h"
|
#include "catalog/pg_operator_d.h"
|
||||||
#include "catalog/pg_type_d.h"
|
#include "catalog/pg_type_d.h"
|
||||||
|
#include "fmgr.h"
|
||||||
#include "lib/pairingheap.h"
|
#include "lib/pairingheap.h"
|
||||||
#include "ivfflat.h"
|
#include "ivfflat.h"
|
||||||
#include "miscadmin.h"
|
#include "miscadmin.h"
|
||||||
#include "pgstat.h"
|
#include "pgstat.h"
|
||||||
#include "storage/bufmgr.h"
|
#include "storage/bufmgr.h"
|
||||||
#include "utils/memutils.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 GetScanList(ptr) pairingheap_container(IvfflatScanList, ph_node, ptr)
|
||||||
#define GetScanListConst(ptr) pairingheap_const_container(IvfflatScanList, ph_node, ptr)
|
#define GetScanListConst(ptr) pairingheap_const_container(IvfflatScanList, ph_node, ptr)
|
||||||
@@ -119,9 +130,6 @@ GetScanItems(IndexScanDesc scan, Datum value)
|
|||||||
|
|
||||||
tuplesort_reset(so->sortstate);
|
tuplesort_reset(so->sortstate);
|
||||||
|
|
||||||
if (scan->parallel_scan != NULL)
|
|
||||||
elog(INFO, "parallel scan");
|
|
||||||
|
|
||||||
/* Search closest probes lists */
|
/* Search closest probes lists */
|
||||||
while (so->listIndex < so->maxProbes && (++batchProbes) <= so->probes)
|
while (so->listIndex < so->maxProbes && (++batchProbes) <= so->probes)
|
||||||
{
|
{
|
||||||
@@ -358,6 +366,10 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
|
|||||||
|
|
||||||
/* Count index scan for stats */
|
/* Count index scan for stats */
|
||||||
pgstat_count_index_scan(scan->indexRelation);
|
pgstat_count_index_scan(scan->indexRelation);
|
||||||
|
#if PG_VERSION_NUM >= 180000
|
||||||
|
if (scan->instrument)
|
||||||
|
scan->instrument->nsearches++;
|
||||||
|
#endif
|
||||||
|
|
||||||
/* Safety check */
|
/* Safety check */
|
||||||
if (scan->orderByData == NULL)
|
if (scan->orderByData == NULL)
|
||||||
|
|||||||
@@ -1,13 +1,19 @@
|
|||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
|
#include "access/genam.h"
|
||||||
#include "access/generic_xlog.h"
|
#include "access/generic_xlog.h"
|
||||||
#include "bitvec.h"
|
|
||||||
#include "catalog/pg_type.h"
|
|
||||||
#include "fmgr.h"
|
#include "fmgr.h"
|
||||||
#include "halfutils.h"
|
#include "halfutils.h"
|
||||||
#include "halfvec.h"
|
#include "halfvec.h"
|
||||||
#include "ivfflat.h"
|
#include "ivfflat.h"
|
||||||
#include "storage/bufmgr.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
|
* Allocate a vector array
|
||||||
|
|||||||
@@ -1,9 +1,12 @@
|
|||||||
#include "postgres.h"
|
#include "postgres.h"
|
||||||
|
|
||||||
|
#include "access/genam.h"
|
||||||
#include "access/generic_xlog.h"
|
#include "access/generic_xlog.h"
|
||||||
|
#include "access/itup.h"
|
||||||
#include "commands/vacuum.h"
|
#include "commands/vacuum.h"
|
||||||
#include "ivfflat.h"
|
#include "ivfflat.h"
|
||||||
#include "storage/bufmgr.h"
|
#include "storage/bufmgr.h"
|
||||||
|
#include "utils/relcache.h"
|
||||||
|
|
||||||
#if PG_VERSION_NUM >= 180000
|
#if PG_VERSION_NUM >= 180000
|
||||||
#define vacuum_delay_point() vacuum_delay_point(false)
|
#define vacuum_delay_point() vacuum_delay_point(false)
|
||||||
|
|||||||
@@ -5,18 +5,27 @@
|
|||||||
|
|
||||||
#include "catalog/pg_type.h"
|
#include "catalog/pg_type.h"
|
||||||
#include "common/shortest_dec.h"
|
#include "common/shortest_dec.h"
|
||||||
#include "common/string.h"
|
|
||||||
#include "fmgr.h"
|
#include "fmgr.h"
|
||||||
#include "halfutils.h"
|
#include "halfutils.h"
|
||||||
#include "halfvec.h"
|
#include "halfvec.h"
|
||||||
|
#include "lib/stringinfo.h"
|
||||||
#include "libpq/pqformat.h"
|
#include "libpq/pqformat.h"
|
||||||
#include "sparsevec.h"
|
#include "sparsevec.h"
|
||||||
#include "utils/array.h"
|
#include "utils/array.h"
|
||||||
#include "utils/builtins.h"
|
#include "utils/builtins.h"
|
||||||
#include "utils/float.h"
|
#include "utils/float.h"
|
||||||
|
#include "utils/fmgrprotos.h"
|
||||||
#include "utils/lsyscache.h"
|
#include "utils/lsyscache.h"
|
||||||
#include "vector.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
|
typedef struct SparseInputElement
|
||||||
{
|
{
|
||||||
int32 index;
|
int32 index;
|
||||||
@@ -150,9 +159,9 @@ InitSparseVector(int dim, int nnz)
|
|||||||
return result;
|
return result;
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
#if PG_VERSION_NUM >= 170000
|
||||||
* Check for whitespace, since array_isspace() is static
|
#define sparsevec_isspace(ch) scanner_isspace(ch)
|
||||||
*/
|
#else
|
||||||
static inline bool
|
static inline bool
|
||||||
sparsevec_isspace(char ch)
|
sparsevec_isspace(char ch)
|
||||||
{
|
{
|
||||||
@@ -165,6 +174,7 @@ sparsevec_isspace(char ch)
|
|||||||
return true;
|
return true;
|
||||||
return false;
|
return false;
|
||||||
}
|
}
|
||||||
|
#endif
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Compare indices
|
* Compare indices
|
||||||
|
|||||||
15
src/vector.c
15
src/vector.c
@@ -16,16 +16,20 @@
|
|||||||
#include "port.h" /* for strtof() */
|
#include "port.h" /* for strtof() */
|
||||||
#include "sparsevec.h"
|
#include "sparsevec.h"
|
||||||
#include "utils/array.h"
|
#include "utils/array.h"
|
||||||
#include "utils/builtins.h"
|
|
||||||
#include "utils/float.h"
|
#include "utils/float.h"
|
||||||
|
#include "utils/fmgrprotos.h"
|
||||||
#include "utils/lsyscache.h"
|
#include "utils/lsyscache.h"
|
||||||
#include "utils/numeric.h"
|
#include "utils/varbit.h"
|
||||||
#include "vector.h"
|
#include "vector.h"
|
||||||
|
|
||||||
#if PG_VERSION_NUM >= 160000
|
#if PG_VERSION_NUM >= 160000
|
||||||
#include "varatt.h"
|
#include "varatt.h"
|
||||||
#endif
|
#endif
|
||||||
|
|
||||||
|
#if PG_VERSION_NUM >= 170000
|
||||||
|
#include "parser/scansup.h"
|
||||||
|
#endif
|
||||||
|
|
||||||
#define STATE_DIMS(x) (ARR_DIMS(x)[0] - 1)
|
#define STATE_DIMS(x) (ARR_DIMS(x)[0] - 1)
|
||||||
#define CreateStateDatums(dim) palloc(sizeof(Datum) * (dim + 1))
|
#define CreateStateDatums(dim) palloc(sizeof(Datum) * (dim + 1))
|
||||||
|
|
||||||
@@ -129,9 +133,9 @@ InitVector(int dim)
|
|||||||
return result;
|
return result;
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
#if PG_VERSION_NUM >= 170000
|
||||||
* Check for whitespace, since array_isspace() is static
|
#define vector_isspace(ch) scanner_isspace(ch)
|
||||||
*/
|
#else
|
||||||
static inline bool
|
static inline bool
|
||||||
vector_isspace(char ch)
|
vector_isspace(char ch)
|
||||||
{
|
{
|
||||||
@@ -144,6 +148,7 @@ vector_isspace(char ch)
|
|||||||
return true;
|
return true;
|
||||||
return false;
|
return false;
|
||||||
}
|
}
|
||||||
|
#endif
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Check state array
|
* Check state array
|
||||||
|
|||||||
@@ -1,16 +0,0 @@
|
|||||||
-- 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;
|
|
||||||
@@ -16,29 +16,19 @@ $node->safe_psql("postgres",
|
|||||||
"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 100000) i;"
|
"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 100000) i;"
|
||||||
);
|
);
|
||||||
|
|
||||||
# Check each index type
|
# Add index
|
||||||
my @operators = ("<->", "<#>", "<=>");
|
$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops);");
|
||||||
my @opclasses = ("vector_l2_ops", "vector_ip_ops", "vector_cosine_ops");
|
|
||||||
|
|
||||||
for my $i (0 .. $#operators)
|
# Test 100% recall
|
||||||
|
for (1 .. 20)
|
||||||
{
|
{
|
||||||
my $operator = $operators[$i];
|
my $id = int(rand() * 100000);
|
||||||
my $opclass = $opclasses[$i];
|
my $query = $node->safe_psql("postgres", "SELECT v FROM tst WHERE i = $id;");
|
||||||
|
my $res = $node->safe_psql("postgres", qq(
|
||||||
# Add index
|
SET enable_seqscan = off;
|
||||||
$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v $opclass);");
|
SELECT v FROM tst ORDER BY v <-> '$query' LIMIT 1;
|
||||||
|
));
|
||||||
# Test 100% recall
|
is($res, $query);
|
||||||
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();
|
done_testing();
|
||||||
|
|||||||
Reference in New Issue
Block a user