Compare commits

...

44 Commits

Author SHA1 Message Date
Andrew Kane
007defb6dd Updated [skip ci] 2025-12-12 14:34:20 -08:00
Andrew Kane
627d3fcc27 Merge branch 'master' into includes 2025-12-12 14:33:42 -08:00
Andrew Kane
d557f1c9b1 Added link to pgvector-tcl [skip ci] 2025-12-11 23:22:56 -08:00
Andrew Kane
3d6dd20469 Removed unneeded check 2025-12-11 16:28:45 -08:00
Andrew Kane
c711da411c Improved includes for indexes 2025-12-11 15:35:37 -08:00
Andrew Kane
3975e212cc Improved includes for ivfkmeans [skip ci] 2025-12-11 13:26:36 -08:00
Andrew Kane
8983103d58 Merge branch 'master' into includes 2025-12-11 00:37:44 -08:00
Andrew Kane
2b741ef068 Improved includes for types 2025-12-11 00:35:18 -08:00
Andrew Kane
6aba9f4422 Added more headers 2025-12-10 23:49:14 -08:00
Andrew Kane
364e4149a6 Added more headers [skip ci] 2025-12-10 23:45:25 -08:00
Andrew Kane
08412c5704 Added more headers 2025-12-10 23:38:46 -08:00
Andrew Kane
88f9c495ba Added more headers [skip ci] 2025-12-10 23:37:12 -08:00
Andrew Kane
fc8c736858 Added more headers [skip ci] 2025-12-10 23:32:21 -08:00
Andrew Kane
a233d68365 Added more headers [skip ci] 2025-12-10 23:19:52 -08:00
Andrew Kane
831ba3f3de Added more headers [skip ci] 2025-12-10 23:13:40 -08:00
Andrew Kane
0c5e1e89d3 Added more headers [skip ci] 2025-12-10 23:05:28 -08:00
Andrew Kane
eda94a46d8 Added more headers 2025-12-10 22:55:54 -08:00
Andrew Kane
8ca5e37c83 Fixed CI 2025-12-10 22:30:32 -08:00
Andrew Kane
18f8562ce9 Improved includes 2025-12-10 22:25:38 -08:00
Andrew Kane
f85a7d3fd0 Keep clang-tidy happy [skip ci] 2025-12-10 20:03:26 -08:00
Andrew Kane
5c25f81789 Fixed inconsistent parameter names [skip ci] 2025-12-10 18:49:00 -08:00
Andrew Kane
d72ee71f23 Fixed compilation errors with Postgres 19 2025-12-10 14:59:29 -08:00
Andrew Kane
b7c0085afe Added instructions for APK [skip ci] 2025-12-01 22:52:23 -08:00
Andrew Kane
2b742576b4 Added links [skip ci] 2025-11-24 12:41:30 -08:00
Andrew Kane
a126c02184 Added link to pgvector-ada [skip ci] 2025-10-30 22:29:36 -07:00
Andrew Kane
744362305b Updated changelog [skip ci] 2025-10-30 11:48:44 -07:00
Andrew Kane
69e78e36a3 Added link to pgvetor-pascal [skip ci] 2025-10-30 11:40:29 -07:00
Andrew Kane
8dd37d4fb2 Improved install target for Windows - #917 [skip ci] 2025-10-30 11:03:59 -07:00
Andrew Kane
a24125839a Updated readme [skip ci] 2025-10-28 10:25:34 -07:00
Andrew Kane
2ee3113417 Updated readme [skip ci] 2025-10-22 13:10:58 -07:00
Andrew Kane
3ebb9a506e Added varatt headers for Postgres 16+ 2025-10-22 11:56:05 -07:00
Andrew Kane
727d4836d2 Fixed Index Searches in EXPLAIN output for Postgres 18 2025-10-22 11:47:11 -07:00
Andrew Kane
d823c44591 Removed unused parameter [skip ci] 2025-09-27 15:58:18 -07:00
Andrew Kane
8f4aa0751c Removed unused parameters [skip ci] 2025-09-27 15:51:32 -07:00
Andrew Kane
1ff30720c7 Updated readme [skip ci] 2025-09-26 13:02:37 -07:00
Andrew Kane
4cef9213ec Added Docker images for Postgres 18 [skip ci] 2025-09-25 20:00:10 -07:00
Andrew Kane
90add68d6f Updated style to be consistent with Mac [skip ci] 2025-09-25 15:55:30 -07:00
Andrew Kane
bc05dbf312 Updated CI for Postgres 18 [skip ci] 2025-09-25 15:54:37 -07:00
Andrew Kane
02fefc0e3f Updated readme for Postgres 18 [skip ci] 2025-09-25 15:46:28 -07:00
Andrew Kane
e7899b1250 Updated checkout action [skip ci] 2025-09-22 11:55:01 -07:00
Andrew Kane
9e29dff78c Updated CI [skip ci] 2025-09-19 19:16:08 -07:00
Andrew Kane
13c0dbf530 Test with macos-15-intel on CI [skip ci] 2025-09-19 18:34:51 -07:00
Andrew Kane
c205a12107 Updated CI [skip ci] 2025-09-05 17:03:37 -07:00
Andrew Kane
db3755a58f Added windows-2025 to CI [skip ci] 2025-09-05 16:41:22 -07:00
24 changed files with 385 additions and 134 deletions

View File

@@ -23,24 +23,24 @@ jobs:
- postgres: 13
os: ubuntu-22.04
steps:
- uses: actions/checkout@v4
- uses: actions/checkout@v5
- 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
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
# - 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') }}
@@ -48,88 +48,62 @@ jobs:
fail-fast: false
matrix:
include:
- postgres: 17
os: macos-15
- postgres: 18
os: macos-26
- postgres: 14
os: macos-13
os: macos-15-intel
steps:
- uses: actions/checkout@v4
- 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
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 == 17 && 'REL_17_2' || 'REL_14_15' }}
- 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: ${{ matrix.os == 'macos-15' && 18 || 15 }}
PG_CFLAGS: -DUSE_ASSERT_CHECKING
# - run: |
# brew install cpanm
# cpanm --notest IPC::Run
# wget -q https://github.com/postgres/postgres/archive/refs/tags/$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: windows-latest
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: actions/checkout@v5
- uses: ankane/setup-postgres@v1
with:
postgres-version: 14
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 && ^
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: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
runs-on: ubuntu-latest
container:
image: debian:12
options: --platform linux/386
steps:
- run: apt-get update && apt-get install -y build-essential git libipc-run-perl postgresql-15 postgresql-server-dev-15 sudo
- run: service postgresql start
- run: |
git clone https://github.com/${{ github.repository }}.git pgvector
cd pgvector
git fetch origin ${{ github.ref }}
git reset --hard FETCH_HEAD
make
make install
chown -R postgres .
sudo -u postgres make installcheck
sudo -u postgres make prove_installcheck
env:
PG_CFLAGS: -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@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

View File

@@ -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)
- Added support for Postgres 18 rc1

View File

@@ -31,6 +31,9 @@ LIBDIR = $(PGROOT)\lib
PKGLIBDIR = $(PGROOT)\lib
SHAREDIR = $(PGROOT)\share
# Use $(PGROOT)\bin\pg_regress for Postgres < 17
PG_REGRESS = $(LIBDIR)\pgxs\src\test\regress\pg_regress
CFLAGS = /nologo /I"$(INCLUDEDIR_SERVER)\port\win32_msvc" /I"$(INCLUDEDIR_SERVER)\port\win32" /I"$(INCLUDEDIR_SERVER)" /I"$(INCLUDEDIR)"
CFLAGS = $(CFLAGS) $(PG_CFLAGS)
@@ -54,11 +57,11 @@ install: all
copy $(SHLIB) "$(PKGLIBDIR)"
copy $(EXTENSION).control "$(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)"
installcheck:
"$(BINDIR)\pg_regress" --bindir="$(BINDIR)" $(REGRESS_OPTS) $(REGRESS)
"$(PG_REGRESS)" --bindir="$(BINDIR)" $(REGRESS_OPTS) $(REGRESS)
uninstall:
del /f "$(PKGLIBDIR)\$(SHLIB)"

View File

@@ -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), 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
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\17"
set "PGROOT=C:\Program Files\PostgreSQL\18"
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` to debug performance.
Use `EXPLAIN (ANALYZE, BUFFERS)` to debug performance.
```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
@@ -771,6 +771,8 @@ 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)
@@ -790,14 +792,18 @@ 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
@@ -812,7 +818,7 @@ 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 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?
@@ -1064,7 +1070,7 @@ 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/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:
@@ -1075,11 +1081,11 @@ sudo --preserve-env=PG_CONFIG make install
A few common paths on Mac are:
- 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`
- 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`
Note: Replace `17` with your Postgres server version
Note: Replace `18` with your Postgres server version
### Missing Header
@@ -1088,10 +1094,10 @@ 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-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
@@ -1138,13 +1144,15 @@ If installation fails with `Access is denied`, re-run the installation instructi
Get the [Docker image](https://hub.docker.com/r/pgvector/pgvector) with:
```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:
- `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`
@@ -1161,7 +1169,7 @@ 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=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.
@@ -1178,7 +1186,7 @@ With Homebrew Postgres, you can use:
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
@@ -1193,22 +1201,22 @@ 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-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
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_17
sudo yum install pgvector_18
# 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
@@ -1225,6 +1233,14 @@ 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:

View File

@@ -2,6 +2,7 @@
#include "bitutils.h"
#include "bitvec.h"
#include "fmgr.h"
#include "utils/varbit.h"
#include "vector.h"

View File

@@ -1,5 +1,7 @@
#include "postgres.h"
#include <math.h>
#include "halfutils.h"
#include "halfvec.h"

View File

@@ -13,12 +13,16 @@
#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/numeric.h"
#include "utils/varbit.h"
#include "vector.h"
#if PG_VERSION_NUM >= 160000
#include "varatt.h"
#endif
#define STATE_DIMS(x) (ARR_DIMS(x)[0] - 1)
#define CreateStateDatums(dim) palloc(sizeof(Datum) * (dim + 1))

View File

@@ -1,18 +1,28 @@
#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/nodes.h"
#include "nodes/pathnodes.h"
#include "nodes/pg_list.h"
#include "storage/lockdefs.h"
#include "storage/shmem.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)

View File

@@ -3,6 +3,8 @@
#include "postgres.h"
#include <math.h>
#include "access/genam.h"
#include "access/parallel.h"
#include "lib/pairingheap.h"
@@ -12,6 +14,10 @@
#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
@@ -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 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 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 HnswSetNeighborTuple(char *base, HnswNeighborTuple ntup, HnswElement e, int m);
void HnswAddHeapTid(HnswElement element, ItemPointer heaptid);

View File

@@ -36,23 +36,51 @@
*/
#include "postgres.h"
#include <math.h>
#include "access/genam.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"
#include "catalog/pg_type_d.h"
#include "commands/progress.h"
#include "common/relpath.h"
#include "hnsw.h"
#include "miscadmin.h"
#include "nodes/execnodes.h"
#include "optimizer/optimizer.h"
#include "storage/block.h"
#include "storage/buf.h"
#include "storage/bufmgr.h"
#include "storage/bufpage.h"
#include "storage/condition_variable.h"
#include "storage/dsm.h"
#include "storage/itemid.h"
#include "storage/itemptr.h"
#include "storage/lockdefs.h"
#include "storage/lwlock.h"
#include "storage/off.h"
#include "storage/shm_toc.h"
#include "storage/shmem.h"
#include "storage/spin.h"
#include "tcop/tcopprot.h"
#include "utils/datum.h"
#include "utils/memutils.h"
#include "utils/rel.h"
#include "utils/relcache.h"
#include "utils/snapmgr.h"
#include "utils/snapshot.h"
#if PG_VERSION_NUM < 190000
#include "storage/item.h"
#endif
#if PG_VERSION_NUM >= 160000
#include "varatt.h"
#endif
#if PG_VERSION_NUM >= 140000
#include "utils/backend_progress.h"
@@ -62,6 +90,11 @@
#if PG_VERSION_NUM >= 140000
#include "utils/backend_status.h"
#endif
#if PG_VERSION_NUM >= 170000
#include "utils/wait_event_types.h"
#elif PG_VERSION_NUM >= 140000
#include "utils/wait_event.h"
#endif
@@ -398,7 +431,7 @@ UpdateNeighborsInMemory(char *base, HnswSupport * support, HnswElement e, int m)
* Update graph in memory
*/
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;
char *base = buildstate->hnswarea;
@@ -460,7 +493,7 @@ InsertTupleInMemory(HnswBuildState * buildstate, HnswElement element)
HnswFindElementNeighbors(base, element, entryPoint, NULL, support, m, efConstruction, false);
/* Update graph in memory */
UpdateGraphInMemory(support, element, m, efConstruction, entryPoint, buildstate);
UpdateGraphInMemory(support, element, m, entryPoint, buildstate);
/* Release entry lock */
LWLockRelease(entryLock);
@@ -545,7 +578,7 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heaptid, Hn
/* Copy the datum */
memcpy(valuePtr, DatumGetPointer(value), valueSize);
HnswPtrStore(base, element->value, valuePtr);
HnswPtrStore(base, element->value, (char *) valuePtr);
/* Create a lock for the element */
LWLockInitialize(&element->lock, hnsw_lock_tranche_id);
@@ -1054,7 +1087,7 @@ ComputeParallelWorkers(Relation heap, Relation index)
* Build graph
*/
static void
BuildGraph(HnswBuildState * buildstate, ForkNumber forkNum)
BuildGraph(HnswBuildState * buildstate)
{
int parallel_workers = 0;
@@ -1102,7 +1135,7 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
InitBuildState(buildstate, heap, index, indexInfo, forkNum);
BuildGraph(buildstate, forkNum);
BuildGraph(buildstate);
if (RelationNeedsWAL(index) || forkNum == INIT_FORKNUM)
log_newpage_range(index, forkNum, 0, RelationGetNumberOfBlocksInFork(index, forkNum), true);

View File

@@ -1,13 +1,31 @@
#include "postgres.h"
#include <math.h>
#include "access/genam.h"
#include "access/generic_xlog.h"
#include "common/relpath.h"
#include "hnsw.h"
#include "nodes/execnodes.h"
#include "storage/block.h"
#include "storage/buf.h"
#include "storage/bufmgr.h"
#include "storage/bufpage.h"
#include "storage/itemid.h"
#include "storage/itemptr.h"
#include "storage/lmgr.h"
#include "storage/lockdefs.h"
#include "storage/off.h"
#include "utils/datum.h"
#include "utils/memutils.h"
#include "utils/rel.h"
#include "utils/relcache.h"
#if PG_VERSION_NUM < 190000
#include "storage/item.h"
#endif
#if PG_VERSION_NUM >= 160000
#include "varatt.h"
#endif
/*
* Get the insert page
@@ -660,7 +678,7 @@ FindDuplicateOnDisk(Relation index, HnswElement element, bool building)
* Update graph on disk
*/
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;
@@ -708,7 +726,7 @@ 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, DatumGetPointer(value));
HnswPtrStore(base, element->value, (char *) DatumGetPointer(value));
/* Prevent concurrent inserts when likely updating entry point */
if (entryPoint == NULL || element->level > entryPoint->level)
@@ -728,7 +746,7 @@ 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, efConstruction, entryPoint, building);
UpdateGraphOnDisk(index, support, element, m, entryPoint, building);
/* Release lock */
UnlockPage(index, HNSW_UPDATE_LOCK, lockmode);

View File

@@ -1,12 +1,25 @@
#include "postgres.h"
#include "access/genam.h"
#include "access/relscan.h"
#include "access/sdir.h"
#include "access/skey.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/itemptr.h"
#include "storage/lmgr.h"
#include "storage/lockdefs.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
@@ -193,6 +206,10 @@ 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)

View File

@@ -2,18 +2,32 @@
#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 "common/relpath.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/block.h"
#include "storage/buf.h"
#include "storage/bufmgr.h"
#include "storage/bufpage.h"
#include "storage/itemptr.h"
#include "storage/lwlock.h"
#include "storage/off.h"
#include "utils/datum.h"
#include "utils/memdebug.h"
#include "utils/rel.h"
#include "utils/relcache.h"
#include "vector.h"
#if PG_VERSION_NUM >= 160000
#include "varatt.h"
#endif
#if PG_VERSION_NUM < 170000
static inline uint64
@@ -256,7 +270,7 @@ HnswInitElement(char *base, ItemPointer heaptid, int m, double ml, int maxLevel,
HnswInitNeighbors(base, element, m, allocator);
HnswPtrStore(base, element->value, (Pointer) NULL);
HnswPtrStore(base, element->value, (char *) NULL);
return element;
}
@@ -282,7 +296,7 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
element->blkno = blkno;
element->offno = offno;
HnswPtrStore(base, element->neighbors, (HnswNeighborArrayPtr *) NULL);
HnswPtrStore(base, element->value, (Pointer) NULL);
HnswPtrStore(base, element->value, (char *) NULL);
return element;
}
@@ -508,7 +522,7 @@ HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHe
char *base = NULL;
Datum value = datumCopy(PointerGetDatum(&etup->data), false, -1);
HnswPtrStore(base, element->value, DatumGetPointer(value));
HnswPtrStore(base, element->value, (char *) DatumGetPointer(value));
}
}
@@ -922,7 +936,7 @@ HnswSearchLayer(char *base, HnswQuery * q, List *ep, int ef, int lc, Relation in
continue;
}
if (eElement == NULL || !(eDistance < f->distance || alwaysAdd))
if (!(eDistance < f->distance || alwaysAdd))
{
if (discarded != NULL)
{

View File

@@ -1,13 +1,30 @@
#include "postgres.h"
#include <math.h>
#include "access/genam.h"
#include "access/generic_xlog.h"
#include "commands/vacuum.h"
#include "common/relpath.h"
#include "hnsw.h"
#include "nodes/pg_list.h"
#include "storage/block.h"
#include "storage/buf.h"
#include "storage/bufmgr.h"
#include "storage/bufpage.h"
#include "storage/itemptr.h"
#include "storage/lmgr.h"
#include "storage/lockdefs.h"
#include "storage/off.h"
#include "utils/memutils.h"
#include "utils/rel.h"
#include "utils/relcache.h"
#if PG_VERSION_NUM < 190000
#include "storage/item.h"
#endif
#if PG_VERSION_NUM >= 160000
#include "varatt.h"
#endif
#if PG_VERSION_NUM >= 180000
#define vacuum_delay_point() vacuum_delay_point(false)
@@ -174,7 +191,12 @@ 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);
@@ -524,8 +546,9 @@ 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++)

View File

@@ -2,23 +2,56 @@
#include <float.h>
#include "access/attnum.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 "bitvec.h"
#include "access/xloginsert.h"
#include "catalog/index.h"
#include "catalog/pg_operator_d.h"
#include "catalog/pg_type_d.h"
#include "commands/progress.h"
#include "halfvec.h"
#include "common/relpath.h"
#include "executor/tuptable.h"
#include "fmgr.h"
#include "ivfflat.h"
#include "miscadmin.h"
#include "nodes/execnodes.h"
#include "optimizer/optimizer.h"
#include "storage/block.h"
#include "storage/buf.h"
#include "storage/bufmgr.h"
#include "storage/bufpage.h"
#include "storage/condition_variable.h"
#include "storage/dsm.h"
#include "storage/itemptr.h"
#include "storage/lockdefs.h"
#include "storage/off.h"
#include "storage/shm_toc.h"
#include "storage/shmem.h"
#include "storage/spin.h"
#include "tcop/tcopprot.h"
#include "utils/memutils.h"
#include "vector.h"
#include "utils/rel.h"
#include "utils/relcache.h"
#include "utils/sampling.h"
#include "utils/snapmgr.h"
#include "utils/snapshot.h"
#include "utils/tuplesort.h"
#if PG_VERSION_NUM < 190000
#include "storage/item.h"
#endif
#if PG_VERSION_NUM >= 160000
#include "varatt.h"
#endif
#if PG_VERSION_NUM >= 140000
#include "utils/backend_progress.h"
@@ -28,6 +61,11 @@
#if PG_VERSION_NUM >= 140000
#include "utils/backend_status.h"
#endif
#if PG_VERSION_NUM >= 170000
#include "utils/wait_event_types.h"
#elif PG_VERSION_NUM >= 140000
#include "utils/wait_event.h"
#endif
@@ -138,7 +176,7 @@ SampleRows(IvfflatBuildState * buildstate)
* Add tuple to sort
*/
static void
AddTupleToSort(Relation index, ItemPointer tid, Datum *values, IvfflatBuildState * buildstate)
AddTupleToSort(ItemPointer tid, Datum *values, IvfflatBuildState * buildstate)
{
double distance;
double minDistance = DBL_MAX;
@@ -215,7 +253,7 @@ BuildCallback(Relation index, ItemPointer tid, Datum *values,
oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
/* Add tuple to sort */
AddTupleToSort(index, tid, values, buildstate);
AddTupleToSort(tid, values, buildstate);
/* Reset memory context */
MemoryContextSwitchTo(oldCtx);
@@ -470,8 +508,8 @@ CreateMetaPage(Relation index, int dimensions, int lists, ForkNumber forkNum)
* Create list pages
*/
static void
CreateListPages(Relation index, VectorArray centers, int dimensions,
int lists, ForkNumber forkNum, ListInfo * *listInfo)
CreateListPages(Relation index, VectorArray centers, int lists,
ForkNumber forkNum, ListInfo * *listInfo)
{
Buffer buf;
Page page;
@@ -1004,7 +1042,7 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
/* Create pages */
CreateMetaPage(index, buildstate->dimensions, buildstate->lists, forkNum);
CreateListPages(index, buildstate->centers, buildstate->dimensions, buildstate->lists, forkNum, &buildstate->listInfo);
CreateListPages(index, buildstate->centers, buildstate->lists, forkNum, &buildstate->listInfo);
CreateEntryPages(buildstate, forkNum);
/* Write WAL for initialization fork since GenericXLog functions do not */

View File

@@ -3,14 +3,22 @@
#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/nodes.h"
#include "nodes/pathnodes.h"
#include "nodes/pg_list.h"
#include "storage/lockdefs.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)

View File

@@ -13,6 +13,10 @@
#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
@@ -21,6 +25,10 @@
#include "portability/instr_time.h"
#endif
#if PG_VERSION_NUM >= 190000
typedef Pointer Item;
#endif
#define IVFFLAT_MAX_DIM 2000
/* Support functions */

View File

@@ -2,11 +2,29 @@
#include <float.h>
#include "access/genam.h"
#include "access/generic_xlog.h"
#include "access/itup.h"
#include "common/relpath.h"
#include "fmgr.h"
#include "ivfflat.h"
#include "nodes/execnodes.h"
#include "storage/block.h"
#include "storage/buf.h"
#include "storage/bufmgr.h"
#include "storage/bufpage.h"
#include "storage/itemid.h"
#include "storage/itemptr.h"
#include "storage/lmgr.h"
#include "storage/lockdefs.h"
#include "storage/off.h"
#include "utils/memutils.h"
#include "utils/rel.h"
#include "utils/relcache.h"
#if PG_VERSION_NUM < 190000
#include "storage/item.h"
#endif
/*
* Find the list that minimizes the distance function
@@ -65,7 +83,7 @@ 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, Relation heapRel)
InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid)
{
const IvfflatTypeInfo *typeInfo = IvfflatGetTypeInfo(index);
IndexTuple itup;
@@ -204,7 +222,7 @@ ivfflatinsert(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid,
oldCtx = MemoryContextSwitchTo(insertCtx);
/* Insert tuple */
InsertTuple(index, values, isnull, heap_tid, heap);
InsertTuple(index, values, isnull, heap_tid);
/* Delete memory context */
MemoryContextSwitchTo(oldCtx);

View File

@@ -1,17 +1,19 @@
#include "postgres.h"
#include <float.h>
#include <limits.h>
#include <math.h>
#include "bitvec.h"
#include "halfutils.h"
#include "halfvec.h"
#include "access/genam.h"
#include "fmgr.h"
#include "ivfflat.h"
#include "miscadmin.h"
#include "utils/builtins.h"
#include "utils/datum.h"
#include "utils/memutils.h"
#include "vector.h"
#include "utils/relcache.h"
#if PG_VERSION_NUM >= 160000
#include "varatt.h"
#endif
/*
* Initialize with kmeans++

View File

@@ -2,15 +2,38 @@
#include <float.h>
#include "access/attnum.h"
#include "access/genam.h"
#include "access/itup.h"
#include "access/relscan.h"
#include "access/sdir.h"
#include "access/skey.h"
#include "access/tupdesc.h"
#include "catalog/pg_operator_d.h"
#include "catalog/pg_type_d.h"
#include "common/relpath.h"
#include "executor/tuptable.h"
#include "fmgr.h"
#include "lib/pairingheap.h"
#include "ivfflat.h"
#include "miscadmin.h"
#include "pgstat.h"
#include "storage/block.h"
#include "storage/buf.h"
#include "storage/bufmgr.h"
#include "storage/bufpage.h"
#include "storage/itemid.h"
#include "storage/itemptr.h"
#include "storage/off.h"
#include "utils/memutils.h"
#include "utils/rel.h"
#include "utils/relcache.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)
@@ -355,6 +378,10 @@ 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)

View File

@@ -1,13 +1,23 @@
#include "postgres.h"
#include "access/genam.h"
#include "access/generic_xlog.h"
#include "bitvec.h"
#include "catalog/pg_type.h"
#include "common/relpath.h"
#include "fmgr.h"
#include "halfutils.h"
#include "halfvec.h"
#include "ivfflat.h"
#include "storage/block.h"
#include "storage/buf.h"
#include "storage/bufmgr.h"
#include "storage/bufpage.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

View File

@@ -1,9 +1,18 @@
#include "postgres.h"
#include "access/genam.h"
#include "access/generic_xlog.h"
#include "access/itup.h"
#include "commands/vacuum.h"
#include "common/relpath.h"
#include "ivfflat.h"
#include "storage/block.h"
#include "storage/buf.h"
#include "storage/bufmgr.h"
#include "storage/bufpage.h"
#include "storage/itemptr.h"
#include "storage/off.h"
#include "utils/relcache.h"
#if PG_VERSION_NUM >= 180000
#define vacuum_delay_point() vacuum_delay_point(false)

View File

@@ -5,18 +5,23 @@
#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
typedef struct SparseInputElement
{
int32 index;

View File

@@ -16,10 +16,10 @@
#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/numeric.h"
#include "utils/varbit.h"
#include "vector.h"
#if PG_VERSION_NUM >= 160000