Compare commits

..

23 Commits

Author SHA1 Message Date
Andrew Kane
3f75886bb2 Switched to runtime check [skip ci] 2024-01-28 22:08:53 -08:00
Andrew Kane
ebf7de0c16 Added comment [skip ci] 2024-01-28 21:35:53 -08:00
Andrew Kane
142a32af96 Update neighbor tuples in-place for HNSW vacuum 2024-01-28 21:27:17 -08:00
Andrew Kane
642ee1f423 Improved function names 2024-01-28 21:00:12 -08:00
Andrew Kane
270dd8189a Update neighbor tuples in-place for HNSW index build 2024-01-28 20:53:56 -08:00
Andrew Kane
ba2776850b Fixed Valgrind check for HNSW in-memory, parallel index builds - closes #441 2024-01-28 16:19:32 -08:00
Andrew Kane
0cc883b944 Removed checking neighbors for cached distance 2024-01-28 02:11:24 -08:00
Andrew Kane
e8e69278eb Fixed test name [skip ci] 2024-01-27 21:18:49 -08:00
Andrew Kane
a4893d9d5f Updated Postgres for Mac TAP tests on CI [skip ci] 2024-01-27 20:28:03 -08:00
Andrew Kane
10416b841f Fixed uninitialized memory in scan->xs_recheck [skip ci] 2024-01-27 19:54:16 -08:00
Andrew Kane
d0f13d4e7f Fixed possibly uninitialized memory in lists [skip ci] 2024-01-27 19:51:34 -08:00
Andrew Kane
705e71015f Fixed uninitialized entry point level - fixes #439 and closes #440 2024-01-27 19:49:40 -08:00
Heikki Linnakangas
121f53b8ff Remove unused heapRel arguments (#443) 2024-01-26 12:45:10 -08:00
Andrew Kane
8765e79ec2 Added note about max_parallel_maintenance_workers for HNSW [skip ci] 2024-01-25 13:32:35 -08:00
Andrew Kane
7fff6cd138 Moved Windows instructions [skip ci] 2024-01-25 02:11:21 -08:00
Heikki Linnakangas
571697fee7 Remove unnecessary UpdateProgress() wrappers (#433)
Now that we require PostgreSQL v12, we can use
pgstat_progress_update_param directly.
2024-01-25 00:07:57 -08:00
Andrew Kane
f7eda7bd20 DRY neighbor array size [skip ci] 2024-01-24 17:54:33 -08:00
Andrew Kane
2260e13315 Condensed code [skip ci] 2024-01-24 16:20:00 -08:00
Andrew Kane
90e0a14bda Moved allocating neighbor array to separate function [skip ci] 2024-01-24 16:17:34 -08:00
Andrew Kane
c816b5d0d1 Added note about --shm-size [skip ci] 2024-01-24 14:49:38 -08:00
Andrew Kane
b7b11cd8d5 Updated changelog [skip ci] 2024-01-24 14:31:41 -08:00
Andrew Kane
54c560c5cb Removed previous upgrade notes [skip ci] 2024-01-24 14:25:58 -08:00
Andrew Kane
8a6c52f649 Moved Docker image to pgvector org and added tags for each supported version of Postgres [skip ci] 2024-01-24 14:18:44 -08:00
17 changed files with 142 additions and 229 deletions

View File

@@ -57,9 +57,9 @@ jobs:
- run: |
brew install cpanm
cpanm --notest IPC::Run
wget -q https://github.com/postgres/postgres/archive/refs/tags/REL_14_5.tar.gz
tar xf REL_14_5.tar.gz
- run: make prove_installcheck PROVE_FLAGS="-I ./postgres-REL_14_5/src/test/perl" PERL5LIB="/Users/runner/perl5/lib/perl5"
wget -q https://github.com/postgres/postgres/archive/refs/tags/REL_14_10.tar.gz
tar xf REL_14_10.tar.gz
- run: make prove_installcheck PROVE_FLAGS="-I ./postgres-REL_14_10/src/test/perl" PERL5LIB="/Users/runner/perl5/lib/perl5"
- run: make clean && /usr/local/opt/llvm@15/bin/scan-build --status-bugs make PG_CFLAGS="-DUSE_ASSERT_CHECKING"
windows:
runs-on: windows-latest

View File

@@ -1,16 +1,17 @@
## 0.6.0 (unreleased)
If upgrading with Postgres < 13, see [this note](https://github.com/pgvector/pgvector#060).
If upgrading with Postgres 12 or Docker, see [these notes](https://github.com/pgvector/pgvector#060-unreleased).
- Changed storage for vector from `extended` to `external`
- Improved performance of HNSW
- Added support for parallel index builds for HNSW
- Added `hamming_distance` function
- Added validation for GUC parameters
- Reduced memory usage for HNSW index builds
- Reduced WAL generation for HNSW index builds
- Fixed error with logical replication
- Fixed `invalid memory alloc request size` error with HNSW index build
- Fixed `invalid memory alloc request size` error with HNSW index builds
- Moved Docker image to `pgvector` org
- Added Docker tags for each supported version of Postgres
- Dropped support for Postgres 11
## 0.5.1 (2023-10-10)
@@ -59,7 +60,7 @@ If upgrading with Postgres < 13, see [this note](https://github.com/pgvector/pgv
## 0.4.0 (2023-01-11)
If upgrading with Postgres < 13, see [this note](https://github.com/pgvector/pgvector#040).
If upgrading with Postgres < 13, see [this note](https://github.com/pgvector/pgvector/blob/v0.4.0/README.md#040).
- Changed text representation for vector elements to match `real`
- Changed storage for vector from `plain` to `extended`
@@ -76,7 +77,7 @@ If upgrading with Postgres < 13, see [this note](https://github.com/pgvector/pgv
## 0.3.1 (2022-11-02)
If upgrading from 0.2.7 or 0.3.0, [recreate](https://github.com/pgvector/pgvector#031) all `ivfflat` indexes after upgrading to ensure all data is indexed.
If upgrading from 0.2.7 or 0.3.0, [recreate](https://github.com/pgvector/pgvector/blob/v0.3.1/README.md#031) all `ivfflat` indexes after upgrading to ensure all data is indexed.
- Fixed issue with inserts silently corrupting `ivfflat` indexes (introduced in 0.2.7)
- Fixed segmentation fault with index creation when lists > 6500

View File

@@ -65,13 +65,17 @@ dist:
mkdir -p dist
git archive --format zip --prefix=$(EXTENSION)-$(EXTVERSION)/ --output dist/$(EXTENSION)-$(EXTVERSION).zip master
# for Docker
PG_MAJOR ?= 16
.PHONY: docker
docker:
docker build --pull --no-cache --platform linux/amd64 -t ankane/pgvector:latest .
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$(PG_MAJOR) .
.PHONY: docker-release
docker-release:
docker buildx build --push --pull --no-cache --platform linux/amd64,linux/arm64 -t ankane/pgvector:latest .
docker buildx build --push --platform linux/amd64,linux/arm64 -t ankane/pgvector:v$(EXTVERSION) .
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) .

View File

@@ -14,6 +14,8 @@ Plus [ACID](https://en.wikipedia.org/wiki/ACID) compliance, point-in-time recove
## Installation
### Linux and Mac
Compile and install the extension (supports Postgres 11+)
```sh
@@ -28,6 +30,28 @@ 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), 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:
```cmd
call "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvars64.bat"
```
Note: The exact path will vary depending on your Visual Studio version and edition
Then use `nmake` to build:
```cmd
set "PGROOT=C:\Program Files\PostgreSQL\16"
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
cd pgvector
nmake /F Makefile.win
nmake /F Makefile.win install
```
You can also install it with [Docker](#docker) or [conda-forge](#conda-forge).
## Getting Started
Enable the extension (do this once in each database where you want to use it)
@@ -240,6 +264,14 @@ 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 [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
```
For a large number of workers, you may also need to increase `max_parallel_workers` (8 by default)
### Indexing Progress
Check [indexing progress](https://www.postgresql.org/docs/current/progress-reporting.html#CREATE-INDEX-PROGRESS-REPORTING) with Postgres 12+
@@ -615,26 +647,6 @@ Note: Replace `16` with your Postgres server version
If compilation fails and the output includes `warning: no such sysroot directory` on Mac, reinstall Xcode Command Line Tools.
### Windows
Support for Windows is currently experimental. Ensure [C++ support in Visual Studio](https://learn.microsoft.com/en-us/cpp/build/building-on-the-command-line?view=msvc-170#download-and-install-the-tools) is installed, and run:
```cmd
call "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvars64.bat"
```
Note: The exact path will vary depending on your Visual Studio version and edition
Then use `nmake` to build:
```cmd
set "PGROOT=C:\Program Files\PostgreSQL\16"
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
cd pgvector
nmake /F Makefile.win
nmake /F Makefile.win install
```
## Additional Installation Methods
### Docker
@@ -731,7 +743,9 @@ SELECT extversion FROM pg_extension WHERE extname = 'vector';
### 0.6.0 [unreleased]
If upgrading with Postgres < 13, remove this line from `sql/vector--0.5.1--0.6.0.sql`:
#### Postgres 12
If upgrading with Postgres 12, remove this line from `sql/vector--0.5.1--0.6.0.sql`:
```sql
ALTER TYPE vector SET (STORAGE = external);
@@ -739,28 +753,20 @@ ALTER TYPE vector SET (STORAGE = external);
Then run `make install` and `ALTER EXTENSION vector UPDATE;`.
### 0.4.0
#### Docker
If upgrading with Postgres < 13, remove this line from `sql/vector--0.3.2--0.4.0.sql`:
The Docker image is now published in the `pgvector` org, and there are tags for each supported version of Postgres (rather than a `latest` tag).
```sql
ALTER TYPE vector SET (STORAGE = extended);
```sh
docker pull pgvector/pgvector:pg16
# or
docker pull pgvector/pgvector:0.6.0-pg16
```
Then run `make install` and `ALTER EXTENSION vector UPDATE;`.
Also, if youve increased `maintenance_work_mem`, make sure `--shm-size` is at least that size to avoid an error with parallel HNSW index builds.
### 0.3.1
If upgrading from 0.2.7 or 0.3.0, recreate all `ivfflat` indexes after upgrading to ensure all data is indexed.
```sql
-- Postgres 12+
REINDEX INDEX CONCURRENTLY index_name;
-- Postgres < 12
CREATE INDEX CONCURRENTLY temp_name ON table USING ivfflat (column opclass);
DROP INDEX CONCURRENTLY index_name;
ALTER INDEX temp_name RENAME TO index_name;
```sh
docker run --shm-size=1g ...
```
## Thanks
@@ -806,8 +812,8 @@ make prove_installcheck # TAP tests
To run single tests:
```sh
make installcheck REGRESS=functions # regression test
make prove_installcheck PROVE_TESTS=test/t/001_wal.pl # TAP test
make installcheck REGRESS=functions # regression test
make prove_installcheck PROVE_TESTS=test/t/001_ivfflat_wal.pl # TAP test
```
To enable benchmarking:

View File

@@ -3,6 +3,3 @@
-- remove this single line for Postgres < 13
ALTER TYPE vector SET (STORAGE = external);
CREATE FUNCTION hamming_distance(bytea, bytea) RETURNS float8
AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;

View File

@@ -290,8 +290,3 @@ CREATE OPERATOR CLASS vector_cosine_ops
OPERATOR 1 <=> (vector, vector) FOR ORDER BY float_ops,
FUNCTION 1 vector_negative_inner_product(vector, vector),
FUNCTION 2 vector_norm(vector);
-- bytea functions
CREATE FUNCTION hamming_distance(bytea, bytea) RETURNS float8
AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;

View File

@@ -64,6 +64,8 @@
#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))
#define HNSW_NEIGHBOR_ARRAY_SIZE(lm) (offsetof(HnswNeighborArray, items) + sizeof(HnswCandidate) * (lm))
#define HnswPageGetOpaque(page) ((HnswPageOpaque) PageGetSpecialPointer(page))
#define HnswPageGetMeta(page) ((HnswMetaPageData *) PageGetContents(page))
@@ -352,7 +354,6 @@ typedef struct HnswVacuumState
/* Variables */
struct tidhash_hash *deleted;
BufferAccessStrategy bas;
HnswNeighborTuple ntup;
HnswElementData highestPoint;
/* Memory */
@@ -379,7 +380,7 @@ void HnswUpdateMetaPage(Relation index, int updateEntry, HnswElement entryPoint
void HnswSetNeighborTuple(char *base, HnswNeighborTuple ntup, HnswElement e, int m);
void HnswAddHeapTid(HnswElement element, ItemPointer heaptid);
void HnswInitNeighbors(char *base, HnswElement element, int m, HnswAllocator * alloc);
bool HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel, bool building);
bool HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull, ItemPointer heap_tid, bool building);
void HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement e, int m, bool checkExisting, bool building);
void HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHeaptids, bool loadVec);
void HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec);

View File

@@ -65,8 +65,6 @@
#define CALLBACK_ITEM_POINTER HeapTuple hup
#endif
#define UpdateProgress(index, val) pgstat_progress_update_param(index, val)
#if PG_VERSION_NUM >= 140000
#include "utils/backend_status.h"
#include "utils/wait_event.h"
@@ -143,10 +141,10 @@ HnswBuildAppendPage(Relation index, Buffer *buf, Page *page, ForkNumber forkNum)
}
/*
* Create element pages
* Create graph pages
*/
static void
CreateElementPages(HnswBuildState * buildstate)
CreateGraphPages(HnswBuildState * buildstate)
{
Relation index = buildstate->index;
ForkNumber forkNum = buildstate->forkNum;
@@ -246,27 +244,24 @@ CreateElementPages(HnswBuildState * buildstate)
}
/*
* Create neighbor pages
* Write neighbor tuples
*/
static void
CreateNeighborPages(HnswBuildState * buildstate)
WriteNeighborTuples(HnswBuildState * buildstate)
{
Relation index = buildstate->index;
ForkNumber forkNum = buildstate->forkNum;
int m = buildstate->m;
HnswElementPtr iter = buildstate->graph->head;
char *base = buildstate->hnswarea;
HnswNeighborTuple ntup;
/* Allocate once */
ntup = palloc0(BLCKSZ);
while (!HnswPtrIsNull(base, iter))
{
HnswElement e = HnswPtrAccess(base, iter);
Buffer buf;
Page page;
Size ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m);
ItemId itemid;
HnswNeighborTuple ntup;
/* Update iterator */
iter = e->next;
@@ -278,18 +273,19 @@ CreateNeighborPages(HnswBuildState * buildstate)
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);
/* Check expected size */
Assert(ItemIdGetLength(itemid) == HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m));
/* Update page in-place */
HnswSetNeighborTuple(base, ntup, e, m);
if (!PageIndexTupleOverwrite(page, e->neighborOffno, (Item) ntup, ntupSize))
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Commit */
MarkBufferDirty(buf);
UnlockReleaseBuffer(buf);
}
pfree(ntup);
}
/*
@@ -303,8 +299,8 @@ FlushPages(HnswBuildState * buildstate)
#endif
CreateMetaPage(buildstate);
CreateElementPages(buildstate);
CreateNeighborPages(buildstate);
CreateGraphPages(buildstate);
WriteNeighborTuples(buildstate);
buildstate->graph->flushed = true;
MemoryContextReset(buildstate->graphCtx);
@@ -498,7 +494,7 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heaptid, Hn
{
LWLockRelease(flushLock);
return HnswInsertTupleOnDisk(index, value, values, isnull, heaptid, buildstate->heap, true);
return HnswInsertTupleOnDisk(index, value, values, isnull, heaptid, true);
}
/*
@@ -530,7 +526,7 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heaptid, Hn
LWLockRelease(flushLock);
return HnswInsertTupleOnDisk(index, value, values, isnull, heaptid, buildstate->heap, true);
return HnswInsertTupleOnDisk(index, value, values, isnull, heaptid, true);
}
/* Ok, we can proceed to allocate the element */
@@ -587,7 +583,7 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
{
/* Update progress */
SpinLockAcquire(&graph->lock);
UpdateProgress(PROGRESS_CREATEIDX_TUPLES_DONE, ++graph->indtuples);
pgstat_progress_update_param(PROGRESS_CREATEIDX_TUPLES_DONE, ++graph->indtuples);
SpinLockRelease(&graph->lock);
}
@@ -1048,7 +1044,7 @@ BuildGraph(HnswBuildState * buildstate, ForkNumber forkNum)
{
int parallel_workers = 0;
UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_HNSW_PHASE_LOAD);
pgstat_progress_update_param(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_HNSW_PHASE_LOAD);
/* Calculate parallel workers */
if (buildstate->heap != NULL)

View File

@@ -570,7 +570,7 @@ UpdateGraphOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement
* Insert a tuple into the index
*/
bool
HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel, bool building)
HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull, ItemPointer heap_tid, bool building)
{
HnswElement entryPoint;
HnswElement element;
@@ -625,7 +625,7 @@ HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull,
* Insert a tuple into the index
*/
static void
HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid)
{
Datum value;
FmgrInfo *normprocinfo;
@@ -642,7 +642,7 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
return;
}
HnswInsertTupleOnDisk(index, value, values, isnull, heap_tid, heapRel, false);
HnswInsertTupleOnDisk(index, value, values, isnull, heap_tid, false);
}
/*
@@ -671,7 +671,7 @@ hnswinsert(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid,
oldCtx = MemoryContextSwitchTo(insertCtx);
/* Insert tuple */
HnswInsertTuple(index, values, isnull, heap_tid, heap);
HnswInsertTuple(index, values, isnull, heap_tid);
/* Delete memory context */
MemoryContextSwitchTo(oldCtx);

View File

@@ -202,6 +202,7 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
MemoryContextSwitchTo(oldCtx);
scan->xs_heaptid = *heaptid;
scan->xs_recheck = false;
scan->xs_recheckorderby = false;
return true;
}

View File

@@ -7,6 +7,7 @@
#include "lib/pairingheap.h"
#include "storage/bufmgr.h"
#include "utils/datum.h"
#include "utils/memdebug.h"
#include "utils/rel.h"
#include "vector.h"
@@ -202,6 +203,19 @@ HnswInitPage(Buffer buf, Page page)
HnswPageGetOpaque(page)->page_id = HNSW_PAGE_ID;
}
/*
* Allocate a neighbor array
*/
static HnswNeighborArray *
HnswInitNeighborArray(int lm, HnswAllocator * allocator)
{
HnswNeighborArray *a = HnswAlloc(allocator, HNSW_NEIGHBOR_ARRAY_SIZE(lm));
a->length = 0;
a->closerSet = false;
return a;
}
/*
* Allocate neighbors
*/
@@ -209,22 +223,12 @@ void
HnswInitNeighbors(char *base, HnswElement element, int m, HnswAllocator * allocator)
{
int level = element->level;
HnswNeighborArrayPtr *neighborList = (HnswNeighborArrayPtr *) HnswAlloc(allocator, sizeof(HnswNeighborArrayPtr) * (level + 1));
HnswPtrStore(base, element->neighbors, neighborList);
for (int lc = 0; lc <= level; lc++)
{
HnswNeighborArray *a;
int lm = HnswGetLayerM(m, lc);
HnswPtrStore(base, neighborList[lc], (HnswNeighborArray *) HnswAlloc(allocator, offsetof(HnswNeighborArray, items) + sizeof(HnswCandidate) * lm));
a = HnswGetNeighbors(base, element, lc);
a->length = 0;
a->closerSet = false;
}
HnswPtrStore(base, neighborList[lc], HnswInitNeighborArray(HnswGetLayerM(m, lc), allocator));
}
/*
@@ -312,7 +316,10 @@ HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint)
if (entryPoint != NULL)
{
if (BlockNumberIsValid(metap->entryBlkno))
{
*entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
(*entryPoint)->level = metap->entryLevel;
}
else
*entryPoint = NULL;
}
@@ -731,7 +738,7 @@ HnswSearchLayer(char *base, Datum q, List *ep, int ef, int lc, Relation index, F
/* Create local memory for neighborhood if needed */
if (index == NULL)
{
neighborhoodSize = offsetof(HnswNeighborArray, items) + sizeof(HnswCandidate) * HnswGetLayerM(m, lc);
neighborhoodSize = HNSW_NEIGHBOR_ARRAY_SIZE(HnswGetLayerM(m, lc));
neighborhoodData = palloc(neighborhoodSize);
}
@@ -907,40 +914,10 @@ CompareCandidateDistancesOffset(const void *a, const void *b)
* Calculate the distance between elements
*/
static float
HnswGetDistance(char *base, HnswElement a, HnswElement b, int lc, FmgrInfo *procinfo, Oid collation)
HnswGetDistance(char *base, HnswElement a, HnswElement b, FmgrInfo *procinfo, Oid collation)
{
Datum aValue;
Datum bValue;
/* Look for cached distance */
if (!HnswPtrIsNull(base, a->neighbors))
{
HnswNeighborArray *neighbors = HnswGetNeighbors(base, a, lc);
for (int i = 0; i < neighbors->length; i++)
{
HnswElement element = HnswPtrAccess(base, neighbors->items[i].element);
if (element == b)
return neighbors->items[i].distance;
}
}
if (!HnswPtrIsNull(base, b->neighbors))
{
HnswNeighborArray *neighbors = HnswGetNeighbors(base, b, lc);
for (int i = 0; i < neighbors->length; i++)
{
HnswElement element = HnswPtrAccess(base, neighbors->items[i].element);
if (element == a)
return neighbors->items[i].distance;
}
}
aValue = HnswGetValue(base, a);
bValue = HnswGetValue(base, b);
Datum aValue = HnswGetValue(base, a);
Datum bValue = HnswGetValue(base, b);
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, aValue, bValue));
}
@@ -949,7 +926,7 @@ HnswGetDistance(char *base, HnswElement a, HnswElement b, int lc, FmgrInfo *proc
* Check if an element is closer to q than any element from R
*/
static bool
CheckElementCloser(char *base, HnswCandidate * e, List *r, int lc, FmgrInfo *procinfo, Oid collation)
CheckElementCloser(char *base, HnswCandidate * e, List *r, FmgrInfo *procinfo, Oid collation)
{
HnswElement eElement = HnswPtrAccess(base, e->element);
ListCell *lc2;
@@ -958,7 +935,7 @@ CheckElementCloser(char *base, HnswCandidate * e, List *r, int lc, FmgrInfo *pro
{
HnswCandidate *ri = lfirst(lc2);
HnswElement riElement = HnswPtrAccess(base, ri->element);
float distance = HnswGetDistance(base, eElement, riElement, lc, procinfo, collation);
float distance = HnswGetDistance(base, eElement, riElement, procinfo, collation);
if (distance <= e->distance)
return false;
@@ -1004,16 +981,20 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
/* Use previous state of r and wd to skip work when possible */
if (mustCalculate)
e->closer = CheckElementCloser(base, e, r, lc, procinfo, collation);
e->closer = CheckElementCloser(base, e, r, procinfo, collation);
else if (list_length(added) > 0)
{
/* Keep Valgrind happy for in-memory, parallel builds */
if (base != NULL)
VALGRIND_MAKE_MEM_DEFINED(&e->closer, 1);
/*
* If the current candidate was closer, we only need to compare it
* with the other candidates that we have added.
*/
if (e->closer)
{
e->closer = CheckElementCloser(base, e, added, lc, procinfo, collation);
e->closer = CheckElementCloser(base, e, added, procinfo, collation);
if (!e->closer)
removedAny = true;
@@ -1026,7 +1007,7 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
*/
if (removedAny)
{
e->closer = CheckElementCloser(base, e, r, lc, procinfo, collation);
e->closer = CheckElementCloser(base, e, r, procinfo, collation);
if (e->closer)
added = lappend(added, e);
}
@@ -1034,11 +1015,15 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
}
else if (e == newCandidate)
{
e->closer = CheckElementCloser(base, e, r, lc, procinfo, collation);
e->closer = CheckElementCloser(base, e, r, procinfo, collation);
if (e->closer)
added = lappend(added, e);
}
/* Keep Valgrind happy for in-memory, parallel builds */
if (base != NULL)
VALGRIND_MAKE_MEM_DEFINED(&e->closer, 1);
if (e->closer)
r = lappend(r, e);
else

View File

@@ -198,8 +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 */
@@ -213,20 +213,23 @@ RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswEleme
/* Find neighbors for element, skipping itself */
HnswFindElementNeighbors(base, element, entryPoint, index, procinfo, collation, m, efConstruction, true);
/* 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);
@@ -588,7 +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(BLCKSZ);
vacuumstate->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
"Hnsw vacuum temporary context",
ALLOCSET_DEFAULT_SIZES);
@@ -608,7 +610,6 @@ FreeVacuumState(HnswVacuumState * vacuumstate)
{
tidhash_destroy(vacuumstate->deleted);
FreeAccessStrategy(vacuumstate->bas);
pfree(vacuumstate->ntup);
MemoryContextDelete(vacuumstate->tmpCtx);
}

View File

@@ -29,8 +29,6 @@
#define CALLBACK_ITEM_POINTER HeapTuple hup
#endif
#define UpdateProgress(index, val) pgstat_progress_update_param(index, val)
#if PG_VERSION_NUM >= 140000
#include "utils/backend_status.h"
#include "utils/wait_event.h"
@@ -262,9 +260,9 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
TupleTableSlot *slot = MakeSingleTupleTableSlot(buildstate->tupdesc, &TTSOpsMinimalTuple);
TupleDesc tupdesc = RelationGetDescr(index);
UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_LOAD);
pgstat_progress_update_param(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_LOAD);
UpdateProgress(PROGRESS_CREATEIDX_TUPLES_TOTAL, buildstate->indtuples);
pgstat_progress_update_param(PROGRESS_CREATEIDX_TUPLES_TOTAL, buildstate->indtuples);
GetNextTuple(buildstate->sortstate, tupdesc, slot, &itup, &list);
@@ -300,7 +298,7 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
pfree(itup);
UpdateProgress(PROGRESS_CREATEIDX_TUPLES_DONE, ++inserted);
pgstat_progress_update_param(PROGRESS_CREATEIDX_TUPLES_DONE, ++inserted);
GetNextTuple(buildstate->sortstate, tupdesc, slot, &itup, &list);
}
@@ -400,7 +398,7 @@ ComputeCenters(IvfflatBuildState * buildstate)
{
int numSamples;
UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_KMEANS);
pgstat_progress_update_param(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_KMEANS);
/* Target 50 samples per list, with at least 10000 samples */
/* The number of samples has a large effect on index build time */
@@ -475,7 +473,7 @@ CreateListPages(Relation index, VectorArray centers, int dimensions,
IvfflatList list;
listSize = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
list = palloc(listSize);
list = palloc0(listSize);
buf = IvfflatNewBuffer(index, forkNum);
IvfflatInitRegisterPage(index, &buf, &page, &state);
@@ -921,7 +919,7 @@ AssignTuples(IvfflatBuildState * buildstate)
Oid sortCollations[] = {InvalidOid};
bool nullsFirstFlags[] = {false};
UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_ASSIGN);
pgstat_progress_update_param(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_ASSIGN);
/* Calculate parallel workers */
if (buildstate->heap != NULL)

View File

@@ -310,6 +310,7 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
ItemPointer heaptid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
scan->xs_heaptid = *heaptid;
scan->xs_recheck = false;
scan->xs_recheckorderby = false;
return true;
}

View File

@@ -10,7 +10,6 @@
#include "lib/stringinfo.h"
#include "libpq/pqformat.h"
#include "port.h" /* for strtof() */
#include "port/pg_bitutils.h"
#include "utils/array.h"
#include "utils/builtins.h"
#include "utils/float.h"
@@ -1135,43 +1134,3 @@ vector_avg(PG_FUNCTION_ARGS)
PG_RETURN_POINTER(result);
}
/*
* Ensure same number of bytes
*/
static inline void
CheckByteLengths(uint32 aLen, uint32 bLen)
{
if (aLen != bLen)
ereport(ERROR,
(errcode(ERRCODE_DATA_EXCEPTION),
errmsg("different byte lengths %u and %u", aLen, bLen)));
}
/*
* Get the Hamming distance between two binary strings
*/
PGDLLEXPORT PG_FUNCTION_INFO_V1(hamming_distance);
Datum
hamming_distance(PG_FUNCTION_ARGS)
{
bytea *a = PG_GETARG_BYTEA_PP(0);
bytea *b = PG_GETARG_BYTEA_PP(1);
char *ax = VARDATA_ANY(a);
char *bx = VARDATA_ANY(b);
uint32 aLen = VARSIZE_ANY_EXHDR(a);
uint32 bLen = VARSIZE_ANY_EXHDR(b);
uint64 distance = 0;
CheckByteLengths(aLen, bLen);
for (uint32 i = 0; i < aLen; i++)
{
unsigned char diff = (unsigned char) (ax[i] ^ bx[i]);
distance += pg_number_of_ones[diff];
}
/* TODO Decide on return type */
PG_RETURN_FLOAT8((double) distance);
}

View File

@@ -158,32 +158,6 @@ SELECT l1_distance('[3e38]', '[-3e38]');
Infinity
(1 row)
SELECT hamming_distance('\xFFFF', '\xFFFF');
hamming_distance
------------------
0
(1 row)
SELECT hamming_distance('\xFFFF', '\xFFFE');
hamming_distance
------------------
1
(1 row)
SELECT hamming_distance('\xFFFF', '\xFFFC');
hamming_distance
------------------
2
(1 row)
SELECT hamming_distance('\xFFFF', '\x0000');
hamming_distance
------------------
16
(1 row)
SELECT hamming_distance('\xFFFF', '\x00');
ERROR: different byte lengths 2 and 1
SELECT avg(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]']) v;
avg
-----------

View File

@@ -37,12 +37,6 @@ SELECT l1_distance('[0,0]', '[0,1]');
SELECT l1_distance('[1,2]', '[3]');
SELECT l1_distance('[3e38]', '[-3e38]');
SELECT hamming_distance('\xFFFF', '\xFFFF');
SELECT hamming_distance('\xFFFF', '\xFFFE');
SELECT hamming_distance('\xFFFF', '\xFFFC');
SELECT hamming_distance('\xFFFF', '\x0000');
SELECT hamming_distance('\xFFFF', '\x00');
SELECT avg(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]']) v;
SELECT avg(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]', NULL]) v;
SELECT avg(v) FROM unnest(ARRAY[]::vector[]) v;