liuyu/pgvector
1
0
Fork 0
mirror of https://github.com/pgvector/pgvector.git synced 2026-07-11 23:26:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

Use datum for HNSW

Browse Source
This commit is contained in:
Andrew Kane
2023-09-12 22:18:19 -07:00
parent 9ac825d14e
commit d57a34b25c
5 changed files with 41 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/hnsw.h
View File

@@ -57,7 +57,7 @@
/* PROGRESS_CREATEIDX_SUBPHASE_INITIALIZE is 1 */ /* PROGRESS_CREATEIDX_SUBPHASE_INITIALIZE is 1 */
#define PROGRESS_HNSW_PHASE_LOAD 2 #define PROGRESS_HNSW_PHASE_LOAD 2
#define HNSW_ELEMENT_TUPLE_SIZE(_dim) MAXALIGN(offsetof(HnswElementTupleData, vec) + VECTOR_SIZE(_dim)) #define HNSW_ELEMENT_TUPLE_SIZE(_datum) MAXALIGN(offsetof(HnswElementTupleData, value) + VARSIZE_ANY(_datum))
#define HNSW_NEIGHBOR_TUPLE_SIZE(level, m) MAXALIGN(offsetof(HnswNeighborTupleData, indextids) + ((level) + 2) * (m) * sizeof(ItemPointerData)) #define HNSW_NEIGHBOR_TUPLE_SIZE(level, m) MAXALIGN(offsetof(HnswNeighborTupleData, indextids) + ((level) + 2) * (m) * sizeof(ItemPointerData))
#define HnswPageGetOpaque(page) ((HnswPageOpaque) PageGetSpecialPointer(page)) #define HnswPageGetOpaque(page) ((HnswPageOpaque) PageGetSpecialPointer(page))
@@ -96,12 +96,13 @@ typedef struct HnswElementData
List *heaptids; List *heaptids;
uint8 level; uint8 level;
uint8 deleted; uint8 deleted;
bool loaded;
HnswNeighborArray *neighbors; HnswNeighborArray *neighbors;
BlockNumber blkno; BlockNumber blkno;
OffsetNumber offno; OffsetNumber offno;
OffsetNumber neighborOffno; OffsetNumber neighborOffno;
BlockNumber neighborPage; BlockNumber neighborPage;
Vector *vec; Datum value;
} HnswElementData; } HnswElementData;
typedef HnswElementData * HnswElement; typedef HnswElementData * HnswElement;
@@ -200,7 +201,7 @@ typedef struct HnswElementTupleData
ItemPointerData heaptids[HNSW_HEAPTIDS]; ItemPointerData heaptids[HNSW_HEAPTIDS];
ItemPointerData neighbortid; ItemPointerData neighbortid;
uint16 unused2; uint16 unused2;
Vector vec; char value[FLEXIBLE_ARRAY_MEMBER];
} HnswElementTupleData; } HnswElementTupleData;
typedef HnswElementTupleData * HnswElementTuple; typedef HnswElementTupleData * HnswElementTuple;

25
src/hnswbuild.c
View File

@@ -8,6 +8,7 @@
#include "lib/pairingheap.h" #include "lib/pairingheap.h"
#include "nodes/pg_list.h" #include "nodes/pg_list.h"
#include "storage/bufmgr.h" #include "storage/bufmgr.h"
#include "utils/datum.h"
#include "utils/memutils.h" #include "utils/memutils.h"
#if PG_VERSION_NUM >= 140000 #if PG_VERSION_NUM >= 140000
@@ -106,8 +107,6 @@ CreateElementPages(HnswBuildState * buildstate)
{ {
Relation index = buildstate->index; Relation index = buildstate->index;
ForkNumber forkNum = buildstate->forkNum; ForkNumber forkNum = buildstate->forkNum;
int dimensions = buildstate->dimensions;
Size etupSize;
Size maxSize; Size maxSize;
HnswElementTuple etup; HnswElementTuple etup;
HnswNeighborTuple ntup; HnswNeighborTuple ntup;
@@ -119,10 +118,9 @@ CreateElementPages(HnswBuildState * buildstate)
/* Calculate sizes */ /* Calculate sizes */
maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData)); maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData));
etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions);
/* Allocate once */ /* Allocate once */
etup = palloc0(etupSize); etup = palloc0(maxSize);
ntup = palloc0(maxSize); ntup = palloc0(maxSize);
/* Prepare first page */ /* Prepare first page */
@@ -134,12 +132,14 @@ CreateElementPages(HnswBuildState * buildstate)
foreach(lc, buildstate->elements) foreach(lc, buildstate->elements)
{ {
HnswElement element = lfirst(lc); HnswElement element = lfirst(lc);
Size etupSize;
Size ntupSize; Size ntupSize;
Size combinedSize; Size combinedSize;
HnswSetElementTuple(etup, element); HnswSetElementTuple(etup, element);
/* Calculate sizes */ /* Calculate sizes */
etupSize = HNSW_ELEMENT_TUPLE_SIZE(element->value);
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(element->level, buildstate->m); ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(element->level, buildstate->m);
combinedSize = etupSize + ntupSize + sizeof(ItemIdData); combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
@@ -276,18 +276,15 @@ InsertTuple(Relation index, Datum *values, HnswElement element, HnswBuildState *
int m = buildstate->m; int m = buildstate->m;
/* Detoast once for all calls */ /* Detoast once for all calls */
Datum value = PointerGetDatum(PG_DETOAST_DATUM(values[0])); element->value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
/* Normalize if needed */ /* Normalize if needed */
if (buildstate->normprocinfo != NULL) if (buildstate->normprocinfo != NULL)
{ {
if (!HnswNormValue(buildstate->normprocinfo, collation, &value, buildstate->normvec)) if (!HnswNormValue(buildstate->normprocinfo, collation, &element->value, buildstate->normvec))
return false; return false;
} }
/* Copy value to element so accessible outside of memory context */
memcpy(element->vec, DatumGetVector(value), VECTOR_SIZE(buildstate->dimensions));
/* Insert element in graph */ /* Insert element in graph */
HnswInsertElement(element, entryPoint, NULL, procinfo, collation, m, efConstruction, false); HnswInsertElement(element, entryPoint, NULL, procinfo, collation, m, efConstruction, false);
@@ -363,7 +360,6 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
/* Allocate necessary memory outside of memory context */ /* Allocate necessary memory outside of memory context */
element = HnswInitElement(tid, buildstate->m, buildstate->ml, buildstate->maxLevel); element = HnswInitElement(tid, buildstate->m, buildstate->ml, buildstate->maxLevel);
element->vec = palloc(VECTOR_SIZE(buildstate->dimensions));
/* Use memory context since detoast can allocate */ /* Use memory context since detoast can allocate */
oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx); oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
@@ -371,9 +367,8 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
/* Insert tuple */ /* Insert tuple */
inserted = InsertTuple(index, values, element, buildstate, &dup); inserted = InsertTuple(index, values, element, buildstate, &dup);
/* Reset memory context */ /* Switch memory context */
MemoryContextSwitchTo(oldCtx); MemoryContextSwitchTo(oldCtx);
MemoryContextReset(buildstate->tmpCtx);
/* Add outside memory context */ /* Add outside memory context */
if (dup != NULL) if (dup != NULL)
@@ -381,9 +376,15 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
/* Add to buildstate or free */ /* Add to buildstate or free */
if (inserted) if (inserted)
{
element->value = datumTransfer(element->value, false, -1);
buildstate->elements = lappend(buildstate->elements, element); buildstate->elements = lappend(buildstate->elements, element);
}
else else
HnswFreeElement(element); HnswFreeElement(element);
/* Reset memory context */
MemoryContextReset(buildstate->tmpCtx);
} }
/* /*

7
src/hnswinsert.c
View File

@@ -123,7 +123,6 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
Size minCombinedSize; Size minCombinedSize;
HnswElementTuple etup; HnswElementTuple etup;
BlockNumber currentPage = insertPage; BlockNumber currentPage = insertPage;
int dimensions = e->vec->dim;
HnswNeighborTuple ntup; HnswNeighborTuple ntup;
Buffer nbuf; Buffer nbuf;
Page npage; Page npage;
@@ -132,7 +131,7 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
BlockNumber newInsertPage = InvalidBlockNumber; BlockNumber newInsertPage = InvalidBlockNumber;
/* Calculate sizes */ /* Calculate sizes */
etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions); etupSize = HNSW_ELEMENT_TUPLE_SIZE(e->value);
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m); ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m);
combinedSize = etupSize + ntupSize + sizeof(ItemIdData); combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData)); maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData));
@@ -411,7 +410,7 @@ HnswAddDuplicate(Relation index, HnswElement element, HnswElement dup)
Buffer buf; Buffer buf;
Page page; Page page;
GenericXLogState *state; GenericXLogState *state;
Size etupSize = HNSW_ELEMENT_TUPLE_SIZE(dup->vec->dim); Size etupSize = HNSW_ELEMENT_TUPLE_SIZE(dup->value);
HnswElementTuple etup; HnswElementTuple etup;
int i; int i;
@@ -522,7 +521,7 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
/* Create an element */ /* Create an element */
element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m)); element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m));
element->vec = DatumGetVector(value); element->value = value;
/* Prevent concurrent inserts when likely updating entry point */ /* Prevent concurrent inserts when likely updating entry point */
if (entryPoint == NULL || element->level > entryPoint->level) if (entryPoint == NULL || element->level > entryPoint->level)

26
src/hnswutils.c
View File

@@ -4,6 +4,7 @@
#include "hnsw.h" #include "hnsw.h"
#include "storage/bufmgr.h" #include "storage/bufmgr.h"
#include "utils/datum.h"
#include "vector.h" #include "vector.h"
/* /*
@@ -187,7 +188,6 @@ HnswFreeElement(HnswElement element)
{ {
HnswFreeNeighbors(element); HnswFreeNeighbors(element);
list_free_deep(element->heaptids); list_free_deep(element->heaptids);
pfree(element->vec);
pfree(element); pfree(element);
} }
@@ -214,7 +214,7 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
element->blkno = blkno; element->blkno = blkno;
element->offno = offno; element->offno = offno;
element->neighbors = NULL; element->neighbors = NULL;
element->vec = NULL; element->loaded = false;
return element; return element;
} }
@@ -324,7 +324,7 @@ HnswSetElementTuple(HnswElementTuple etup, HnswElement element)
else else
ItemPointerSetInvalid(&etup->heaptids[i]); ItemPointerSetInvalid(&etup->heaptids[i]);
} }
memcpy(&etup->vec, element->vec, VECTOR_SIZE(element->vec->dim)); memcpy(&etup->value, DatumGetPointer(element->value), VARSIZE_ANY(element->value));
} }
/* /*
@@ -447,8 +447,10 @@ HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHe
if (loadVec) if (loadVec)
{ {
element->vec = palloc(VECTOR_SIZE(etup->vec.dim)); Datum value = PointerGetDatum(&etup->value);
memcpy(element->vec, &etup->vec, VECTOR_SIZE(etup->vec.dim));
element->value = datumCopy(value, false, -1);
element->loaded = true;
} }
} }
@@ -476,7 +478,7 @@ HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index,
/* Calculate distance */ /* Calculate distance */
if (distance != NULL) if (distance != NULL)
*distance = (float) DatumGetFloat8(FunctionCall2Coll(procinfo, collation, *q, PointerGetDatum(&etup->vec))); *distance = (float) DatumGetFloat8(FunctionCall2Coll(procinfo, collation, *q, PointerGetDatum(&etup->value)));
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
} }
@@ -487,7 +489,7 @@ HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index,
static float static float
GetCandidateDistance(HnswCandidate * hc, Datum q, FmgrInfo *procinfo, Oid collation) GetCandidateDistance(HnswCandidate * hc, Datum q, FmgrInfo *procinfo, Oid collation)
{ {
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, q, PointerGetDatum(hc->element->vec))); return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, q, hc->element->value));
} }
/* /*
@@ -722,7 +724,7 @@ HnswGetDistance(HnswElement a, HnswElement b, int lc, FmgrInfo *procinfo, Oid co
} }
} }
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(a->vec), PointerGetDatum(b->vec))); return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, a->value, b->value));
} }
/* /*
@@ -805,7 +807,7 @@ HnswFindDuplicate(HnswElement e)
HnswCandidate *neighbor = &neighbors->items[i]; HnswCandidate *neighbor = &neighbors->items[i];
/* Exit early since ordered by distance */ /* Exit early since ordered by distance */
if (vector_cmp_internal(e->vec, neighbor->element->vec) != 0) if (!datumIsEqual(e->value, neighbor->element->value, false, -1))
break; break;
/* Check for space */ /* Check for space */
@@ -880,13 +882,13 @@ HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int
/* Load elements on insert */ /* Load elements on insert */
if (index != NULL) if (index != NULL)
{ {
Datum q = PointerGetDatum(hc->element->vec); Datum q = hc->element->value;
for (int i = 0; i < currentNeighbors->length; i++) for (int i = 0; i < currentNeighbors->length; i++)
{ {
HnswCandidate *hc3 = &currentNeighbors->items[i]; HnswCandidate *hc3 = &currentNeighbors->items[i];
if (hc3->element->vec == NULL) if (!hc3->element->loaded)
HnswLoadElement(hc3->element, &hc3->distance, &q, index, procinfo, collation, true); HnswLoadElement(hc3->element, &hc3->distance, &q, index, procinfo, collation, true);
else else
hc3->distance = GetCandidateDistance(hc3, q, procinfo, collation); hc3->distance = GetCandidateDistance(hc3, q, procinfo, collation);
@@ -968,7 +970,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
List *w; List *w;
int level = element->level; int level = element->level;
int entryLevel; int entryLevel;
Datum q = PointerGetDatum(element->vec); Datum q = element->value;
HnswElement skipElement = existing ? element : NULL; HnswElement skipElement = existing ? element : NULL;
/* No neighbors if no entry point */ /* No neighbors if no entry point */

10
src/hnswvacuum.c
View File

@@ -93,7 +93,7 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
if (itemUpdated) if (itemUpdated)
{ {
Size etupSize = HNSW_ELEMENT_TUPLE_SIZE(etup->vec.dim); Size etupSize = HNSW_ELEMENT_TUPLE_SIZE(PointerGetDatum(&etup->value));
/* Mark rest as invalid */ /* Mark rest as invalid */
for (int i = idx; i < HNSW_HEAPTIDS; i++) for (int i = idx; i < HNSW_HEAPTIDS; i++)
@@ -485,6 +485,7 @@ MarkDeleted(HnswVacuumState * vacuumstate)
HnswNeighborTuple ntup; HnswNeighborTuple ntup;
Size etupSize; Size etupSize;
Size ntupSize; Size ntupSize;
Datum value;
Buffer nbuf; Buffer nbuf;
Page npage; Page npage;
BlockNumber neighborPage; BlockNumber neighborPage;
@@ -508,8 +509,11 @@ MarkDeleted(HnswVacuumState * vacuumstate)
if (ItemPointerIsValid(&etup->heaptids[0])) if (ItemPointerIsValid(&etup->heaptids[0]))
continue; continue;
/* Get datum */
value = PointerGetDatum(&etup->value);
/* Calculate sizes */ /* Calculate sizes */
etupSize = HNSW_ELEMENT_TUPLE_SIZE(etup->vec.dim); etupSize = HNSW_ELEMENT_TUPLE_SIZE(value);
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(etup->level, vacuumstate->m); ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(etup->level, vacuumstate->m);
/* Get neighbor page */ /* Get neighbor page */
@@ -532,7 +536,7 @@ MarkDeleted(HnswVacuumState * vacuumstate)
/* Overwrite element */ /* Overwrite element */
etup->deleted = 1; etup->deleted = 1;
MemSet(&etup->vec.x, 0, etup->vec.dim * sizeof(float)); MemSet(&etup->value, 0, VARSIZE_ANY(value));
/* Overwrite neighbors */ /* Overwrite neighbors */
for (int i = 0; i < ntup->count; i++) for (int i = 0; i < ntup->count; i++)