mirror of
https://github.com/pgvector/pgvector.git
synced 2026-07-22 20:15:46 +08:00
Compare commits
3 Commits
ivfflat-pa
...
hnsw-vacuu
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
3f75886bb2 | ||
|
|
ebf7de0c16 | ||
|
|
142a32af96 |
17
.github/workflows/build.yml
vendored
17
.github/workflows/build.yml
vendored
@@ -28,7 +28,7 @@ jobs:
|
||||
dev-files: true
|
||||
- run: make
|
||||
env:
|
||||
PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
|
||||
PG_CFLAGS: -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
|
||||
- run: |
|
||||
export PG_CONFIG=`which pg_config`
|
||||
sudo --preserve-env=PG_CONFIG make install
|
||||
@@ -49,7 +49,7 @@ jobs:
|
||||
postgres-version: 14
|
||||
- run: make
|
||||
env:
|
||||
PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter
|
||||
PG_CFLAGS: -Wall -Wextra -Werror -Wno-unused-parameter
|
||||
- run: make install
|
||||
- run: make installcheck
|
||||
- if: ${{ failure() }}
|
||||
@@ -97,15 +97,4 @@ jobs:
|
||||
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
|
||||
valgrind:
|
||||
if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- uses: actions/checkout@v4
|
||||
- uses: ankane/setup-postgres-valgrind@v1
|
||||
with:
|
||||
postgres-version: 16
|
||||
- run: make
|
||||
- run: sudo --preserve-env=PG_CONFIG make install
|
||||
- run: make installcheck
|
||||
PG_CFLAGS: -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
|
||||
|
||||
13
CHANGELOG.md
13
CHANGELOG.md
@@ -1,16 +1,11 @@
|
||||
## 0.6.1 (unreleased)
|
||||
## 0.6.0 (unreleased)
|
||||
|
||||
- Fixed error with `ANALYZE` and vectors with different dimensions
|
||||
- Fixed error with `shared_preload_libraries`
|
||||
If upgrading with Postgres 12 or Docker, see [these notes](https://github.com/pgvector/pgvector#060-unreleased).
|
||||
|
||||
## 0.6.0 (2024-01-29)
|
||||
|
||||
If upgrading with Postgres 12 or Docker, see [these notes](https://github.com/pgvector/pgvector#060).
|
||||
|
||||
- Added support for parallel index builds for HNSW
|
||||
- Added validation for GUC parameters
|
||||
- Changed storage for vector from `extended` to `external`
|
||||
- Improved performance of HNSW
|
||||
- Added support for parallel index builds for HNSW
|
||||
- Added validation for GUC parameters
|
||||
- Reduced memory usage for HNSW index builds
|
||||
- Reduced WAL generation for HNSW index builds
|
||||
- Fixed error with logical replication
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
ARG PG_MAJOR=16
|
||||
ARG PG_MAJOR=15
|
||||
FROM postgres:$PG_MAJOR
|
||||
ARG PG_MAJOR
|
||||
|
||||
|
||||
@@ -2,7 +2,7 @@
|
||||
"name": "vector",
|
||||
"abstract": "Open-source vector similarity search for Postgres",
|
||||
"description": "Supports L2 distance, inner product, and cosine distance",
|
||||
"version": "0.6.0",
|
||||
"version": "0.5.1",
|
||||
"maintainer": [
|
||||
"Andrew Kane <andrew@ankane.org>"
|
||||
],
|
||||
@@ -20,7 +20,7 @@
|
||||
"vector": {
|
||||
"file": "sql/vector.sql",
|
||||
"docfile": "README.md",
|
||||
"version": "0.6.0",
|
||||
"version": "0.5.1",
|
||||
"abstract": "Open-source vector similarity search for Postgres"
|
||||
}
|
||||
},
|
||||
|
||||
8
Makefile
8
Makefile
@@ -1,5 +1,5 @@
|
||||
EXTENSION = vector
|
||||
EXTVERSION = 0.6.0
|
||||
EXTVERSION = 0.5.1
|
||||
|
||||
MODULE_big = vector
|
||||
DATA = $(wildcard sql/*--*.sql)
|
||||
@@ -71,9 +71,11 @@ PG_MAJOR ?= 16
|
||||
.PHONY: docker
|
||||
|
||||
docker:
|
||||
docker build --pull --no-cache --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .
|
||||
docker build --pull --no-cache --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) .
|
||||
docker build --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .
|
||||
|
||||
.PHONY: docker-release
|
||||
|
||||
docker-release:
|
||||
docker buildx build --push --pull --no-cache --platform linux/amd64,linux/arm64 --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .
|
||||
docker buildx build --push --pull --no-cache --platform linux/amd64,linux/arm64 --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) .
|
||||
docker buildx build --push --platform linux/amd64,linux/arm64 --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
EXTENSION = vector
|
||||
EXTVERSION = 0.6.0
|
||||
EXTVERSION = 0.5.1
|
||||
|
||||
OBJS = src\hnsw.obj src\hnswbuild.obj src\hnswinsert.obj src\hnswscan.obj src\hnswutils.obj src\hnswvacuum.obj src\ivfbuild.obj src\ivfflat.obj src\ivfinsert.obj src\ivfkmeans.obj src\ivfscan.obj src\ivfutils.obj src\ivfvacuum.obj src\vector.obj
|
||||
HEADERS = src\vector.h
|
||||
|
||||
46
README.md
46
README.md
@@ -16,11 +16,11 @@ Plus [ACID](https://en.wikipedia.org/wiki/ACID) compliance, point-in-time recove
|
||||
|
||||
### Linux and Mac
|
||||
|
||||
Compile and install the extension (supports Postgres 12+)
|
||||
Compile and install the extension (supports Postgres 11+)
|
||||
|
||||
```sh
|
||||
cd /tmp
|
||||
git clone --branch v0.6.0 https://github.com/pgvector/pgvector.git
|
||||
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
|
||||
cd pgvector
|
||||
make
|
||||
make install # may need sudo
|
||||
@@ -28,7 +28,7 @@ make install # may need sudo
|
||||
|
||||
See the [installation notes](#installation-notes) 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), 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
|
||||
|
||||
@@ -44,7 +44,7 @@ Then use `nmake` to build:
|
||||
|
||||
```cmd
|
||||
set "PGROOT=C:\Program Files\PostgreSQL\16"
|
||||
git clone --branch v0.6.0 https://github.com/pgvector/pgvector.git
|
||||
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
|
||||
cd pgvector
|
||||
nmake /F Makefile.win
|
||||
nmake /F Makefile.win install
|
||||
@@ -264,7 +264,7 @@ HINT: Increase maintenance_work_mem to speed up builds.
|
||||
|
||||
Note: Do not set `maintenance_work_mem` so high that it exhausts the memory on the server
|
||||
|
||||
Starting with 0.6.0, you can also speed up index creation by increasing the number of parallel workers (2 by default)
|
||||
Starting with 0.6.0 [unreleased], you can also speed up index creation by increasing the number of parallel workers (2 by default)
|
||||
|
||||
```sql
|
||||
SET max_parallel_maintenance_workers = 7; -- plus leader
|
||||
@@ -430,15 +430,6 @@ To speed up queries with an IVFFlat index, increase the number of inverted lists
|
||||
CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 1000);
|
||||
```
|
||||
|
||||
## Vacuuming
|
||||
|
||||
Vacuuming can take a while for HNSW indexes. Speed it up by reindexing first.
|
||||
|
||||
```sql
|
||||
REINDEX INDEX CONCURRENTLY index_name;
|
||||
VACUUM table_name;
|
||||
```
|
||||
|
||||
## Languages
|
||||
|
||||
Use pgvector from any language with a Postgres client. You can even generate and store vectors in one language and query them in another.
|
||||
@@ -660,20 +651,20 @@ If compilation fails and the output includes `warning: no such sysroot directory
|
||||
|
||||
### Docker
|
||||
|
||||
Get the [Docker image](https://hub.docker.com/r/pgvector/pgvector) with:
|
||||
Get the [Docker image](https://hub.docker.com/r/ankane/pgvector) with:
|
||||
|
||||
```sh
|
||||
docker pull pgvector/pgvector:pg16
|
||||
docker pull ankane/pgvector
|
||||
```
|
||||
|
||||
This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (replace `16` with your Postgres server version, and run it the same way).
|
||||
This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (run it the same way).
|
||||
|
||||
You can also build the image manually:
|
||||
|
||||
```sh
|
||||
git clone --branch v0.6.0 https://github.com/pgvector/pgvector.git
|
||||
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
|
||||
cd pgvector
|
||||
docker build --build-arg PG_MAJOR=16 -t myuser/pgvector .
|
||||
docker build --build-arg PG_MAJOR=15 -t myuser/pgvector .
|
||||
```
|
||||
|
||||
### Homebrew
|
||||
@@ -716,21 +707,6 @@ sudo dnf install pgvector_16
|
||||
|
||||
Note: Replace `16` with your Postgres server version
|
||||
|
||||
### pkg
|
||||
|
||||
Install the FreeBSD package with:
|
||||
|
||||
```sh
|
||||
pkg install postgresql15-pg_vector
|
||||
```
|
||||
|
||||
or the port with:
|
||||
|
||||
```sh
|
||||
cd /usr/ports/databases/pgvector
|
||||
make install
|
||||
```
|
||||
|
||||
### conda-forge
|
||||
|
||||
With Conda Postgres, install from [conda-forge](https://anaconda.org/conda-forge/pgvector) with:
|
||||
@@ -765,7 +741,7 @@ SELECT extversion FROM pg_extension WHERE extname = 'vector';
|
||||
|
||||
## Upgrade Notes
|
||||
|
||||
### 0.6.0
|
||||
### 0.6.0 [unreleased]
|
||||
|
||||
#### Postgres 12
|
||||
|
||||
|
||||
@@ -8,7 +8,6 @@
|
||||
#include "commands/progress.h"
|
||||
#include "commands/vacuum.h"
|
||||
#include "hnsw.h"
|
||||
#include "miscadmin.h"
|
||||
#include "utils/guc.h"
|
||||
#include "utils/selfuncs.h"
|
||||
|
||||
@@ -29,7 +28,7 @@ static relopt_kind hnsw_relopt_kind;
|
||||
* this grows bigger, we should use a shmem_request_hook and
|
||||
* RequestAddinShmemSpace() to pre-reserve space for this.
|
||||
*/
|
||||
void
|
||||
static void
|
||||
HnswInitLockTranche(void)
|
||||
{
|
||||
int *tranche_ids;
|
||||
@@ -54,8 +53,7 @@ HnswInitLockTranche(void)
|
||||
void
|
||||
HnswInit(void)
|
||||
{
|
||||
if (!process_shared_preload_libraries_in_progress)
|
||||
HnswInitLockTranche();
|
||||
HnswInitLockTranche();
|
||||
|
||||
hnsw_relopt_kind = add_reloption_kind();
|
||||
add_int_reloption(hnsw_relopt_kind, "m", "Max number of connections",
|
||||
|
||||
@@ -60,7 +60,6 @@
|
||||
#define PROGRESS_HNSW_PHASE_LOAD 2
|
||||
|
||||
#define HNSW_MAX_SIZE (BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData)) - sizeof(ItemIdData))
|
||||
#define HNSW_TUPLE_ALLOC_SIZE BLCKSZ
|
||||
|
||||
#define HNSW_ELEMENT_TUPLE_SIZE(size) MAXALIGN(offsetof(HnswElementTupleData, data) + (size))
|
||||
#define HNSW_NEIGHBOR_TUPLE_SIZE(level, m) MAXALIGN(offsetof(HnswNeighborTupleData, indextids) + ((level) + 2) * (m) * sizeof(ItemPointerData))
|
||||
@@ -355,7 +354,6 @@ typedef struct HnswVacuumState
|
||||
/* Variables */
|
||||
struct tidhash_hash *deleted;
|
||||
BufferAccessStrategy bas;
|
||||
HnswNeighborTuple ntup;
|
||||
HnswElementData highestPoint;
|
||||
|
||||
/* Memory */
|
||||
@@ -389,7 +387,6 @@ void HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation i
|
||||
void HnswSetElementTuple(char *base, HnswElementTuple etup, HnswElement element);
|
||||
void HnswUpdateConnection(char *base, HnswElement element, HnswCandidate * hc, int lm, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation);
|
||||
void HnswLoadNeighbors(HnswElement element, Relation index, int m);
|
||||
void HnswInitLockTranche(void);
|
||||
PGDLLEXPORT void HnswParallelBuildMain(dsm_segment *seg, shm_toc *toc);
|
||||
|
||||
/* Index access methods */
|
||||
|
||||
@@ -148,6 +148,7 @@ CreateGraphPages(HnswBuildState * buildstate)
|
||||
{
|
||||
Relation index = buildstate->index;
|
||||
ForkNumber forkNum = buildstate->forkNum;
|
||||
Size etupAllocSize;
|
||||
Size maxSize;
|
||||
HnswElementTuple etup;
|
||||
HnswNeighborTuple ntup;
|
||||
@@ -159,11 +160,12 @@ CreateGraphPages(HnswBuildState * buildstate)
|
||||
char *base = buildstate->hnswarea;
|
||||
|
||||
/* Calculate sizes */
|
||||
etupAllocSize = BLCKSZ;
|
||||
maxSize = HNSW_MAX_SIZE;
|
||||
|
||||
/* Allocate once */
|
||||
etup = palloc0(HNSW_TUPLE_ALLOC_SIZE);
|
||||
ntup = palloc0(HNSW_TUPLE_ALLOC_SIZE);
|
||||
etup = palloc0(etupAllocSize);
|
||||
ntup = palloc0(BLCKSZ);
|
||||
|
||||
/* Prepare first page */
|
||||
buf = HnswNewBuffer(index, forkNum);
|
||||
@@ -176,13 +178,13 @@ CreateGraphPages(HnswBuildState * buildstate)
|
||||
Size etupSize;
|
||||
Size ntupSize;
|
||||
Size combinedSize;
|
||||
Pointer valuePtr = HnswPtrAccess(base, element->value);
|
||||
void *valuePtr = HnswPtrAccess(base, element->value);
|
||||
|
||||
/* Update iterator */
|
||||
iter = element->next;
|
||||
|
||||
/* Zero memory for each element */
|
||||
MemSet(etup, 0, HNSW_TUPLE_ALLOC_SIZE);
|
||||
MemSet(etup, 0, etupAllocSize);
|
||||
|
||||
/* Calculate sizes */
|
||||
etupSize = HNSW_ELEMENT_TUPLE_SIZE(VARSIZE_ANY(valuePtr));
|
||||
@@ -190,7 +192,7 @@ CreateGraphPages(HnswBuildState * buildstate)
|
||||
combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
|
||||
|
||||
/* Initial size check */
|
||||
if (etupSize > HNSW_TUPLE_ALLOC_SIZE)
|
||||
if (etupSize > etupAllocSize)
|
||||
elog(ERROR, "index tuple too large");
|
||||
|
||||
HnswSetElementTuple(base, etup, element);
|
||||
@@ -252,43 +254,38 @@ WriteNeighborTuples(HnswBuildState * buildstate)
|
||||
int m = buildstate->m;
|
||||
HnswElementPtr iter = buildstate->graph->head;
|
||||
char *base = buildstate->hnswarea;
|
||||
HnswNeighborTuple ntup;
|
||||
|
||||
/* Allocate once */
|
||||
ntup = palloc0(HNSW_TUPLE_ALLOC_SIZE);
|
||||
|
||||
while (!HnswPtrIsNull(base, iter))
|
||||
{
|
||||
HnswElement element = HnswPtrAccess(base, iter);
|
||||
HnswElement e = HnswPtrAccess(base, iter);
|
||||
Buffer buf;
|
||||
Page page;
|
||||
Size ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(element->level, m);
|
||||
ItemId itemid;
|
||||
HnswNeighborTuple ntup;
|
||||
|
||||
/* Update iterator */
|
||||
iter = element->next;
|
||||
|
||||
/* Zero memory for each element */
|
||||
MemSet(ntup, 0, HNSW_TUPLE_ALLOC_SIZE);
|
||||
iter = e->next;
|
||||
|
||||
/* Can take a while, so ensure we can interrupt */
|
||||
/* Needs to be called when no buffer locks are held */
|
||||
CHECK_FOR_INTERRUPTS();
|
||||
|
||||
buf = ReadBufferExtended(index, forkNum, element->neighborPage, RBM_NORMAL, NULL);
|
||||
buf = ReadBufferExtended(index, forkNum, e->neighborPage, RBM_NORMAL, NULL);
|
||||
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
|
||||
page = BufferGetPage(buf);
|
||||
itemid = PageGetItemId(page, e->neighborOffno);
|
||||
ntup = (HnswNeighborTuple) PageGetItem(page, itemid);
|
||||
|
||||
HnswSetNeighborTuple(base, ntup, element, m);
|
||||
/* Check expected size */
|
||||
Assert(ItemIdGetLength(itemid) == HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m));
|
||||
|
||||
if (!PageIndexTupleOverwrite(page, element->neighborOffno, (Item) ntup, ntupSize))
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
/* Update page in-place */
|
||||
HnswSetNeighborTuple(base, ntup, e, m);
|
||||
|
||||
/* Commit */
|
||||
MarkBufferDirty(buf);
|
||||
UnlockReleaseBuffer(buf);
|
||||
}
|
||||
|
||||
pfree(ntup);
|
||||
}
|
||||
|
||||
/*
|
||||
@@ -601,9 +598,6 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
|
||||
static void
|
||||
InitGraph(HnswGraph * graph, char *base, long memoryTotal)
|
||||
{
|
||||
/* Initialize the lock tranche if needed */
|
||||
HnswInitLockTranche();
|
||||
|
||||
HnswPtrStore(base, graph->head, (HnswElement) NULL);
|
||||
HnswPtrStore(base, graph->entryPoint, (HnswElement) NULL);
|
||||
graph->memoryUsed = 0;
|
||||
@@ -977,14 +971,6 @@ HnswBeginParallel(HnswBuildState * buildstate, bool isconcurrent, int request)
|
||||
/* Report less than allocated so never fails */
|
||||
InitGraph(&hnswshared->graphData, hnswarea, esthnswarea - 1024 * 1024);
|
||||
|
||||
/*
|
||||
* Avoid base address for relptr for Postgres < 14.5
|
||||
* https://github.com/postgres/postgres/commit/7201cd18627afc64850537806da7f22150d1a83b
|
||||
*/
|
||||
#if PG_VERSION_NUM < 140005
|
||||
hnswshared->graphData.memoryUsed += MAXALIGN(1);
|
||||
#endif
|
||||
|
||||
shm_toc_insert(pcxt->toc, PARALLEL_KEY_HNSW_SHARED, hnswshared);
|
||||
shm_toc_insert(pcxt->toc, PARALLEL_KEY_HNSW_AREA, hnswarea);
|
||||
|
||||
|
||||
@@ -355,7 +355,9 @@ HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, Hns
|
||||
Buffer buf;
|
||||
Page page;
|
||||
GenericXLogState *state;
|
||||
ItemId itemid;
|
||||
HnswNeighborTuple ntup;
|
||||
Size ntupSize;
|
||||
int idx = -1;
|
||||
int startIdx;
|
||||
HnswElement neighborElement = HnswPtrAccess(base, hc->element);
|
||||
@@ -394,7 +396,9 @@ HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, Hns
|
||||
}
|
||||
|
||||
/* Get tuple */
|
||||
ntup = (HnswNeighborTuple) PageGetItem(page, PageGetItemId(page, offno));
|
||||
itemid = PageGetItemId(page, offno);
|
||||
ntup = (HnswNeighborTuple) PageGetItem(page, itemid);
|
||||
ntupSize = ItemIdGetLength(itemid);
|
||||
|
||||
/* Calculate index for update */
|
||||
startIdx = (neighborElement->level - lc) * m;
|
||||
@@ -423,9 +427,13 @@ HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, Hns
|
||||
{
|
||||
ItemPointer indextid = &ntup->indextids[idx];
|
||||
|
||||
/* Update neighbor on the buffer */
|
||||
/* Update neighbor */
|
||||
ItemPointerSet(indextid, e->blkno, e->offno);
|
||||
|
||||
/* Overwrite tuple */
|
||||
if (!PageIndexTupleOverwrite(page, offno, (Item) ntup, ntupSize))
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Commit */
|
||||
if (building)
|
||||
MarkBufferDirty(buf);
|
||||
@@ -449,7 +457,9 @@ AddDuplicateOnDisk(Relation index, HnswElement element, HnswElement dup, bool bu
|
||||
Buffer buf;
|
||||
Page page;
|
||||
GenericXLogState *state;
|
||||
ItemId itemid;
|
||||
HnswElementTuple etup;
|
||||
Size etupSize;
|
||||
int i;
|
||||
|
||||
/* Read page */
|
||||
@@ -467,7 +477,9 @@ AddDuplicateOnDisk(Relation index, HnswElement element, HnswElement dup, bool bu
|
||||
}
|
||||
|
||||
/* Find space */
|
||||
etup = (HnswElementTuple) PageGetItem(page, PageGetItemId(page, dup->offno));
|
||||
itemid = PageGetItemId(page, dup->offno);
|
||||
etup = (HnswElementTuple) PageGetItem(page, itemid);
|
||||
etupSize = ItemIdGetLength(itemid);
|
||||
for (i = 0; i < HNSW_HEAPTIDS; i++)
|
||||
{
|
||||
if (!ItemPointerIsValid(&etup->heaptids[i]))
|
||||
@@ -483,9 +495,13 @@ AddDuplicateOnDisk(Relation index, HnswElement element, HnswElement dup, bool bu
|
||||
return false;
|
||||
}
|
||||
|
||||
/* Add heap TID, modifying the tuple on the page directly */
|
||||
/* Add heap TID */
|
||||
etup->heaptids[i] = element->heaptids[0];
|
||||
|
||||
/* Overwrite tuple */
|
||||
if (!PageIndexTupleOverwrite(page, dup->offno, (Item) etup, etupSize))
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Commit */
|
||||
if (building)
|
||||
MarkBufferDirty(buf);
|
||||
|
||||
@@ -60,7 +60,8 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
|
||||
/* Iterate over nodes */
|
||||
for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
|
||||
{
|
||||
HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, PageGetItemId(page, offno));
|
||||
ItemId itemid = PageGetItemId(page, offno);
|
||||
HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, itemid);
|
||||
int idx = 0;
|
||||
bool itemUpdated = false;
|
||||
|
||||
@@ -91,10 +92,15 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
|
||||
|
||||
if (itemUpdated)
|
||||
{
|
||||
Size etupSize = ItemIdGetLength(itemid);
|
||||
|
||||
/* Mark rest as invalid */
|
||||
for (int i = idx; i < HNSW_HEAPTIDS; i++)
|
||||
ItemPointerSetInvalid(&etup->heaptids[i]);
|
||||
|
||||
if (!PageIndexTupleOverwrite(page, offno, (Item) etup, etupSize))
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
updated = true;
|
||||
}
|
||||
}
|
||||
@@ -192,8 +198,8 @@ RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswEleme
|
||||
FmgrInfo *procinfo = vacuumstate->procinfo;
|
||||
Oid collation = vacuumstate->collation;
|
||||
BufferAccessStrategy bas = vacuumstate->bas;
|
||||
HnswNeighborTuple ntup = vacuumstate->ntup;
|
||||
Size ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(element->level, m);
|
||||
ItemId itemid;
|
||||
HnswNeighborTuple ntup;
|
||||
char *base = NULL;
|
||||
|
||||
/* Skip if element is entry point */
|
||||
@@ -207,23 +213,23 @@ RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswEleme
|
||||
/* Find neighbors for element, skipping itself */
|
||||
HnswFindElementNeighbors(base, element, entryPoint, index, procinfo, collation, m, efConstruction, true);
|
||||
|
||||
/* Zero memory for each element */
|
||||
MemSet(ntup, 0, HNSW_TUPLE_ALLOC_SIZE);
|
||||
|
||||
/* Update neighbor tuple */
|
||||
/* Do this before getting page to minimize locking */
|
||||
HnswSetNeighborTuple(base, ntup, element, m);
|
||||
|
||||
/* Get neighbor page */
|
||||
buf = ReadBufferExtended(index, MAIN_FORKNUM, element->neighborPage, RBM_NORMAL, bas);
|
||||
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
|
||||
state = GenericXLogStart(index);
|
||||
page = GenericXLogRegisterBuffer(state, buf, 0);
|
||||
|
||||
/* Overwrite tuple */
|
||||
if (!PageIndexTupleOverwrite(page, element->neighborOffno, (Item) ntup, ntupSize))
|
||||
/* Get neighbor tuple */
|
||||
itemid = PageGetItemId(page, element->neighborOffno);
|
||||
ntup = (HnswNeighborTuple) PageGetItem(page, itemid);
|
||||
|
||||
/* Check expected size */
|
||||
if (ItemIdGetLength(itemid) != HNSW_NEIGHBOR_TUPLE_SIZE(element->level, m))
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Update page in-place */
|
||||
HnswSetNeighborTuple(base, ntup, element, m);
|
||||
|
||||
/* Commit */
|
||||
GenericXLogFinish(state);
|
||||
UnlockReleaseBuffer(buf);
|
||||
@@ -476,8 +482,11 @@ MarkDeleted(HnswVacuumState * vacuumstate)
|
||||
/* Update element and neighbors together */
|
||||
for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
|
||||
{
|
||||
HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, PageGetItemId(page, offno));
|
||||
ItemId itemid = PageGetItemId(page, offno);
|
||||
HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, itemid);
|
||||
HnswNeighborTuple ntup;
|
||||
Size etupSize;
|
||||
Size ntupSize;
|
||||
Buffer nbuf;
|
||||
Page npage;
|
||||
BlockNumber neighborPage;
|
||||
@@ -501,6 +510,10 @@ MarkDeleted(HnswVacuumState * vacuumstate)
|
||||
if (ItemPointerIsValid(&etup->heaptids[0]))
|
||||
continue;
|
||||
|
||||
/* Calculate sizes */
|
||||
etupSize = ItemIdGetLength(itemid);
|
||||
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(etup->level, vacuumstate->m);
|
||||
|
||||
/* Get neighbor page */
|
||||
neighborPage = ItemPointerGetBlockNumber(&etup->neighbortid);
|
||||
neighborOffno = ItemPointerGetOffsetNumber(&etup->neighbortid);
|
||||
@@ -527,10 +540,13 @@ MarkDeleted(HnswVacuumState * vacuumstate)
|
||||
for (int i = 0; i < ntup->count; i++)
|
||||
ItemPointerSetInvalid(&ntup->indextids[i]);
|
||||
|
||||
/*
|
||||
* We modified the tuples in place, no need to call
|
||||
* PageIndexTupleOverwrite
|
||||
*/
|
||||
/* Overwrite element tuple */
|
||||
if (!PageIndexTupleOverwrite(page, offno, (Item) etup, etupSize))
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Overwrite neighbor tuple */
|
||||
if (!PageIndexTupleOverwrite(npage, neighborOffno, (Item) ntup, ntupSize))
|
||||
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
|
||||
|
||||
/* Commit */
|
||||
GenericXLogFinish(state);
|
||||
@@ -575,7 +591,6 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
|
||||
vacuumstate->bas = GetAccessStrategy(BAS_BULKREAD);
|
||||
vacuumstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
|
||||
vacuumstate->collation = index->rd_indcollation[0];
|
||||
vacuumstate->ntup = palloc0(HNSW_TUPLE_ALLOC_SIZE);
|
||||
vacuumstate->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
|
||||
"Hnsw vacuum temporary context",
|
||||
ALLOCSET_DEFAULT_SIZES);
|
||||
@@ -595,7 +610,6 @@ FreeVacuumState(HnswVacuumState * vacuumstate)
|
||||
{
|
||||
tidhash_destroy(vacuumstate->deleted);
|
||||
FreeAccessStrategy(vacuumstate->bas);
|
||||
pfree(vacuumstate->ntup);
|
||||
MemoryContextDelete(vacuumstate->tmpCtx);
|
||||
}
|
||||
|
||||
|
||||
@@ -904,27 +904,6 @@ IvfflatBeginParallel(IvfflatBuildState * buildstate, bool isconcurrent, int requ
|
||||
WaitForParallelWorkersToAttach(pcxt);
|
||||
}
|
||||
|
||||
/*
|
||||
* Compute parallel workers
|
||||
*/
|
||||
static int
|
||||
ComputeParallelWorkers(Relation heap, Relation index)
|
||||
{
|
||||
int parallel_workers;
|
||||
|
||||
/* Make sure it's safe to use parallel workers */
|
||||
parallel_workers = plan_create_index_workers(RelationGetRelid(heap), RelationGetRelid(index));
|
||||
if (parallel_workers == 0)
|
||||
return 0;
|
||||
|
||||
/* Use parallel_workers storage parameter on table if set */
|
||||
parallel_workers = RelationGetParallelWorkers(heap, -1);
|
||||
if (parallel_workers != -1)
|
||||
return Min(parallel_workers, max_parallel_maintenance_workers);
|
||||
|
||||
return max_parallel_maintenance_workers;
|
||||
}
|
||||
|
||||
/*
|
||||
* Scan table for tuples to index
|
||||
*/
|
||||
@@ -944,7 +923,7 @@ AssignTuples(IvfflatBuildState * buildstate)
|
||||
|
||||
/* Calculate parallel workers */
|
||||
if (buildstate->heap != NULL)
|
||||
parallel_workers = ComputeParallelWorkers(buildstate->heap, buildstate->index);
|
||||
parallel_workers = plan_create_index_workers(RelationGetRelid(buildstate->heap), RelationGetRelid(buildstate->index));
|
||||
|
||||
/* Attempt to launch parallel worker scan when required */
|
||||
if (parallel_workers > 0)
|
||||
|
||||
30
src/vector.c
30
src/vector.c
@@ -866,10 +866,9 @@ vector_mul(PG_FUNCTION_ARGS)
|
||||
int
|
||||
vector_cmp_internal(Vector * a, Vector * b)
|
||||
{
|
||||
int dim = Min(a->dim, b->dim);
|
||||
CheckDims(a, b);
|
||||
|
||||
/* Check values before dimensions to be consistent with Postgres arrays */
|
||||
for (int i = 0; i < dim; i++)
|
||||
for (int i = 0; i < a->dim; i++)
|
||||
{
|
||||
if (a->x[i] < b->x[i])
|
||||
return -1;
|
||||
@@ -877,13 +876,6 @@ vector_cmp_internal(Vector * a, Vector * b)
|
||||
if (a->x[i] > b->x[i])
|
||||
return 1;
|
||||
}
|
||||
|
||||
if (a->dim < b->dim)
|
||||
return -1;
|
||||
|
||||
if (a->dim > b->dim)
|
||||
return 1;
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
@@ -897,9 +889,6 @@ vector_lt(PG_FUNCTION_ARGS)
|
||||
Vector *a = PG_GETARG_VECTOR_P(0);
|
||||
Vector *b = PG_GETARG_VECTOR_P(1);
|
||||
|
||||
/* TODO Remove in 0.7.0 */
|
||||
CheckDims(a, b);
|
||||
|
||||
PG_RETURN_BOOL(vector_cmp_internal(a, b) < 0);
|
||||
}
|
||||
|
||||
@@ -913,9 +902,6 @@ vector_le(PG_FUNCTION_ARGS)
|
||||
Vector *a = PG_GETARG_VECTOR_P(0);
|
||||
Vector *b = PG_GETARG_VECTOR_P(1);
|
||||
|
||||
/* TODO Remove in 0.7.0 */
|
||||
CheckDims(a, b);
|
||||
|
||||
PG_RETURN_BOOL(vector_cmp_internal(a, b) <= 0);
|
||||
}
|
||||
|
||||
@@ -929,9 +915,6 @@ vector_eq(PG_FUNCTION_ARGS)
|
||||
Vector *a = PG_GETARG_VECTOR_P(0);
|
||||
Vector *b = PG_GETARG_VECTOR_P(1);
|
||||
|
||||
/* TODO Remove in 0.7.0 */
|
||||
CheckDims(a, b);
|
||||
|
||||
PG_RETURN_BOOL(vector_cmp_internal(a, b) == 0);
|
||||
}
|
||||
|
||||
@@ -945,9 +928,6 @@ vector_ne(PG_FUNCTION_ARGS)
|
||||
Vector *a = PG_GETARG_VECTOR_P(0);
|
||||
Vector *b = PG_GETARG_VECTOR_P(1);
|
||||
|
||||
/* TODO Remove in 0.7.0 */
|
||||
CheckDims(a, b);
|
||||
|
||||
PG_RETURN_BOOL(vector_cmp_internal(a, b) != 0);
|
||||
}
|
||||
|
||||
@@ -961,9 +941,6 @@ vector_ge(PG_FUNCTION_ARGS)
|
||||
Vector *a = PG_GETARG_VECTOR_P(0);
|
||||
Vector *b = PG_GETARG_VECTOR_P(1);
|
||||
|
||||
/* TODO Remove in 0.7.0 */
|
||||
CheckDims(a, b);
|
||||
|
||||
PG_RETURN_BOOL(vector_cmp_internal(a, b) >= 0);
|
||||
}
|
||||
|
||||
@@ -977,9 +954,6 @@ vector_gt(PG_FUNCTION_ARGS)
|
||||
Vector *a = PG_GETARG_VECTOR_P(0);
|
||||
Vector *b = PG_GETARG_VECTOR_P(1);
|
||||
|
||||
/* TODO Remove in 0.7.0 */
|
||||
CheckDims(a, b);
|
||||
|
||||
PG_RETURN_BOOL(vector_cmp_internal(a, b) > 0);
|
||||
}
|
||||
|
||||
|
||||
@@ -24,56 +24,6 @@ SELECT '[1e37]'::vector * '[1e37]';
|
||||
ERROR: value out of range: overflow
|
||||
SELECT '[1e-37]'::vector * '[1e-37]';
|
||||
ERROR: value out of range: underflow
|
||||
SELECT '[1,2,3]'::vector = '[1,2,3]';
|
||||
?column?
|
||||
----------
|
||||
t
|
||||
(1 row)
|
||||
|
||||
SELECT '[1,2,3]'::vector = '[1,2]';
|
||||
ERROR: different vector dimensions 3 and 2
|
||||
SELECT vector_cmp('[1,2,3]', '[1,2,3]');
|
||||
vector_cmp
|
||||
------------
|
||||
0
|
||||
(1 row)
|
||||
|
||||
SELECT vector_cmp('[1,2,3]', '[0,0,0]');
|
||||
vector_cmp
|
||||
------------
|
||||
1
|
||||
(1 row)
|
||||
|
||||
SELECT vector_cmp('[0,0,0]', '[1,2,3]');
|
||||
vector_cmp
|
||||
------------
|
||||
-1
|
||||
(1 row)
|
||||
|
||||
SELECT vector_cmp('[1,2]', '[1,2,3]');
|
||||
vector_cmp
|
||||
------------
|
||||
-1
|
||||
(1 row)
|
||||
|
||||
SELECT vector_cmp('[1,2,3]', '[1,2]');
|
||||
vector_cmp
|
||||
------------
|
||||
1
|
||||
(1 row)
|
||||
|
||||
SELECT vector_cmp('[1,2]', '[2,3,4]');
|
||||
vector_cmp
|
||||
------------
|
||||
-1
|
||||
(1 row)
|
||||
|
||||
SELECT vector_cmp('[2,3]', '[1,2,3]');
|
||||
vector_cmp
|
||||
------------
|
||||
1
|
||||
(1 row)
|
||||
|
||||
SELECT vector_dims('[1,2,3]');
|
||||
vector_dims
|
||||
-------------
|
||||
|
||||
@@ -6,17 +6,6 @@ SELECT '[1,2,3]'::vector * '[4,5,6]';
|
||||
SELECT '[1e37]'::vector * '[1e37]';
|
||||
SELECT '[1e-37]'::vector * '[1e-37]';
|
||||
|
||||
SELECT '[1,2,3]'::vector = '[1,2,3]';
|
||||
SELECT '[1,2,3]'::vector = '[1,2]';
|
||||
|
||||
SELECT vector_cmp('[1,2,3]', '[1,2,3]');
|
||||
SELECT vector_cmp('[1,2,3]', '[0,0,0]');
|
||||
SELECT vector_cmp('[0,0,0]', '[1,2,3]');
|
||||
SELECT vector_cmp('[1,2]', '[1,2,3]');
|
||||
SELECT vector_cmp('[1,2,3]', '[1,2]');
|
||||
SELECT vector_cmp('[1,2]', '[2,3,4]');
|
||||
SELECT vector_cmp('[2,3]', '[1,2,3]');
|
||||
|
||||
SELECT vector_dims('[1,2,3]');
|
||||
|
||||
SELECT round(vector_norm('[1,1]')::numeric, 5);
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
comment = 'vector data type and ivfflat and hnsw access methods'
|
||||
default_version = '0.6.0'
|
||||
default_version = '0.5.1'
|
||||
module_pathname = '$libdir/vector'
|
||||
relocatable = true
|
||||
|
||||
Reference in New Issue
Block a user